Advanced Techniques in Diagnostic Microbiology

Advanced Techniques in Diagnostic Microbiology

Yi-Wei Tang • Charles W. Stratton Editors

Advanced Techniques in Diagnostic Microbiology Volume 1: Techniques Third Edition

Editors Yi-Wei Tang Departments of Laboratory Medicine and Internal Medicine Memorial Sloan Kettering Cancer Center New York, NY, USA

Charles W. Stratton Department of Pathology, Microbiology and Immunology and Medicine Vanderbilt University Medical Center Nashville, TN, USA

Department of Pathology and Laboratory Medicine Weill Medical College of Cornell University

New York, NY, USA

ISBN 978-3-319-33899-6 ISBN 978-3-319-33900-9 (eBook) https://doi.org/10.1007/978-3-319-33900-9 Library of Congress Control Number: 2018952488 © Springer International Publishing AG, part of Springer Nature 2006, 2013, 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

Medical microbiology is a branch of medical science that deals with the prevention, diagnosis, and therapy of infectious diseases. A clinical microbiologist is a professional within the field of medical microbiology who is knowledgeable about the characteristics of microbial pathogens, including their modes of transmission as well as their mechanisms of infection and growth. Clinical microbiologists often practice in a clinical microbiology laboratory or a public health laboratory where they may direct these laboratories. Clinical microbiology laboratories are concerned mainly with the diagnostic aspects of the practice of medical microbiology, whereas public health laboratories are more concerned with the epidemiology of infectious diseases within given populations. There is, and must be, a strong link between clinical microbiology laboratories and public health laboratories. Clinical microbiology laboratories primarily determine whether pathogenic microorganisms are present in clinical specimens collected from individuals with suspected infections; if such microbial pathogens are found, these microorganisms are identified and susceptibility profiles, when indicated, are determined. Clinical microbiologists work closely with and serve as consultants with physicians who are caring for infected individuals. The importance of the field of medical microbiology can be appreciated by noting that hospitals in the United States annually report over five million cases of infectious disease-related illnesses. Diagnostic microbiology within the clinical microbiology laboratory continues to undergo a quiet revolution that already has resulted in many benefits for microbiologists, clinicians, and most importantly patients. This revolution was initially made possible by the elucidation of the structure of DNA and the genetic code, which allowed scientific advances centered around hybridization probes, the polymerase chain reaction, genomics, transcriptomics, proteomics, and metabolomics. These technical advances in molecular microbiology over the first decade of the twenty-first century have profoundly altered every aspect of the clinical microbiology laboratory, including their staffing patterns, work flow, and turnaround time. More recently, fully automated sample processing systems with digital plate reading technology have emerged as a second wave of technical advances, and have enabled clinical microbiology laboratories to cope with large numbers of samples v

vi

Preface

with improved quality despite limited personnel and financial resources. Moreover, total laboratory automation in the clinical microbiology laboratory also provides superior and more rapid detection of microbial growth. The total laboratory automation system combined with molecular microbiology technical advances such as MALDI-TOF MS and rapid phenotypic susceptibility methods promises to markedly reduce the turnaround time and ultimately reduce the length of stay for hospitalized patients with infections. The knowledge base required to stay current in the rapidly changing and advancing technology involved in molecular microbiology, as well as similar advances in total laboratory automation in the clinical microbiology laboratory, is enormous. In 2006 and 2013, the first and second editions of Advanced Techniques in Diagnostic Microbiology were published and were well received. According to its “Book Performance Report 2017,” since its online publication on August 06, 2012, there has been a total of 145,240 chapter downloads for the second edition eBook by the end of 2017 on SpringerLink. This means the second edition has been one of the top 25% most downloaded eBooks in the relevant SpringerLink eBook Collection for 5 consecutive years. In order to continue to provide this kind of relevant and current information for microbiologists, the third edition of Advanced Techniques in Diagnostic Microbiology has been extensively revised and extended with a total of 55 chapters covering all current stateof-the art techniques and applications in the field of diagnostic microbiology. Advanced Techniques in Diagnostic Microbiology thus provides a comprehensive, well-referenced, and up-to-date description of these rapidly evolving advanced methods for the diagnosis of infectious diseases in the routine clinical microbiology laboratory. The third edition is extended to two volumes. The first volume covers the principles and characteristics of important molecular techniques; these techniques include rapid antigen testing, advanced antibody detection, real-time/ digital nucleic acid amplification techniques, state-of-the-art molecular typing techniques, and MALDI-TOF MS. New chapters on advanced techniques have been added; these include chapters on total laboratory automation systems, rapid phenotypic antimicrobial susceptibility methods, metabolic techniques, and transcriptomic methods. The second volume describes the application of these advanced techniques; new and evolving molecular applications such as molecular detection and characterization of carbapenem-resistant Enterobacteriaceae, advanced typing techniques applied to molecular epidemiology investigations, and multiplex approaches for syndromic testing are covered. Like the first two editions, a diverse team of authors provides authoritative, comprehensive, and well-referenced information on clinically relevant topics; these include sequencebased bacterial identification, blood and blood product screening, automated blood culture systems, molecular diagnosis of sexually transmitted diseases, advances in the molecular diagnosis of fungal/mycobacterial infections, metagenomic sequencing of microbiomes, and application of advanced techniques for antimicrobial stewardship.

Preface

vii

We hope our readers like this technique- and application-based approach and their feedback is greatly appreciated. We want to again thank the authors who devoted their time and effort to produce their chapters. We also thank the staff at Springer. Finally, we continue to appreciate the constant encouragement of our wives and family members throughout this long effort. They are, indeed, the “wind in our sails.” New York, NY, USA Nashville, TN, USA

Yi-Wei Tang Charles W. Stratton

Contents

utomated Blood Cultures�� 1 A Xiang Y. Han Laboratory Automation in Clinical Bacteriology�� 15 Antony Croxatto iochemical Profile-Based Microbial Identification Systems�� 33 B Safina Hafeez and Jaber Aslanzadeh dvanced Phenotypic Antimicrobial Susceptibility Testing Methods�� 69 A Charles W. Stratton Rapid Microbial Antigen Tests�� 99 Sheldon Campbell and Marie L. Landry Antibody Detection: Principles and Applications �� 127 Yun F. (Wayne) Wang rocalcitonin and Other Host-Response-Based Biomarkers P for Evaluation of Infection and Guidance of Antimicrobial Treatment�� 149 Philipp Schuetz, Ramon Sager, Yannick Wirz, and Beat Mueller unctional Assessment of Microbial, Viral, and Parasitic Infections F Using Real-Time Cellular Analysis�� 161 Dazhi Jin, Xiao Xu, Min Zheng, Alex Mira, Brandon J. Lamarche, and Alex B. Ryder ellular Fatty Acid-Based Microbial Identification and Antimicrobial C Susceptibility Testing �� 199 Nicole Parrish and Stefan Riedel ALDI-TOF Mass Spectrometry-Based Microbial Identification M and Beyond �� 211 Alexander Mellmann and Johannes Müthing ix

x

Contents

ranscriptomic Techniques in Diagnostic Microbiology�� 235 T Zachary E. Holcomb and Ephraim L. Tsalik he Use of Microbial Metabolites for the Diagnosis T of Infectious Diseases �� 261 Mahesh J. Thalavitiya Acharige, Seena S. Koshy, and Sophia Koo ucleic Acid Extraction and Enrichment�� 273 N Jeong Hwan Shin Nonamplified Probe-Based Microbial Detection and Identification �� 293 Fann Wu, Tao Hong, and Phyllis Della-Latta olecular Typing Techniques: State of the Art�� 305 M Richard V. Goering PCR and Its Variations�� 327 Eleanor A. Powell and Michael Loeffelholz Non-PCR Amplification Techniques�� 347 Rosemary C. She, Ted E. Schutzbank, and Elizabeth M. Marlowe eal-Time and Digital PCR for Nucleic Acid Quantification�� 377 R Alexander J. McAdam irect Nucleotide Sequencing for Amplification Product D Identification�� 389 Tao Hong olid and Suspension Microarrays for Detection and Identification S of Infectious Diseases �� 403 Sherry Dunbar, Janet Farhang, Shubhagata Das, Sabrina Ali, and Heng Qian eal-Time Detection of Amplification Products Through R Fluorescence Quenching or Energy Transfer�� 451 Caitlin Otto and Shihai Huang CR/Electrospray Ionization-Mass Spectrometry as an Infectious P Disease Diagnostic Tool�� 481 Volkan Özenci and Kristoffer Strålin ucleic Acid Amplicons Detected and Identified N by T2 Magnetic Resonance�� 491 Jessica L. Snyder, Heather S. Lapp, Zhi-Xiang Luo, Brendan Manning, and Thomas J. Lowery Molecular Contamination and Amplification Product Inactivation �� 505 Susan Sefers and Jonathan E. Schmitz Index�� 527

Contributors

Sabrina Ali Luminex Corporation, Toronto, ON, Canada Jaber Aslanzadeh Division of Clinical Microbiology, Hartford Hospital, Hartford, CT, USA Sheldon Campbell Pathology and Laboratory Medicine Service, VA Connecticut, West Haven, CT, USA Department of Laboratory Medicine, Yale University School of Medicine, New Haven, CT, USA Antony Croxatto Institute of Microbiology, Lausanne University Hospital and Lausanne University, Lausanne, Switzerland Shubhagata Das Luminex Corporation, Austin, TX, USA Phyllis Della-Latta Department of Pathology & Cell Biology, Columbia University Medical Center, New York-Presbyterian Hospital, New York, NY, USA Sherry Dunbar Luminex Corporation, Austin, TX, USA Janet Farhang Luminex Corporation, Austin, TX, USA Richard V. Goering Department of Medical Microbiology and Immunology, Creighton University School of Medicine, Omaha, NE, USA Safina Hafeez Department of Pathology and laboratory Medicine, Divsion of Clinical Microbiology, Hartford Hospital, Hartford, CT, USA Xiang Y. Han Department of Laboratory Medicine, Unit 84, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Zachary E. Holcomb Duke University School of Medicine, Durham, NC, USA Tao Hong Department of Pathology, Hackensack University Medical Center, Hackensack, NJ, USA

xi

xii

Contributors

Tao Hong Department of Pathology, Hackensack University Medical Center, Hackensack, NJ, USA Shihai Huang Abbott Molecular Inc., Des Plaines, IL, USA Dazhi Jin Zhejiang Provincial Center for Disease Control and Prevention, Hangzhou, Zhejiang, China Sophia Koo Division of Infectious Diseases, Brigham and Women’s Hospital, Boston, MA, USA Harvard Medical School, Boston, MA, USA Dana-Farber Cancer Institute, Boston, MA, USA Seena S. Koshy Division of Infectious Diseases, Brigham and Women’s Hospital, Boston, MA, USA Harvard Medical School, Boston, MA, USA Brandon J. Lamarche ACEA Biosciences, Inc, San Diego, CA, USA Marie L. Landry Department of Laboratory Medicine, Yale University School of Medicine, New Haven, CT, USA Heather S. Lapp T2 Biosystems, Lexington, MA, USA Michael Loeffelholz Department of Pathology, University of Texas Medical Branch, Galveston, TX, USA Thomas J. Lowery T2 Biosystems, Lexington, MA, USA Zhi-Xiang Luo T2 Biosystems, Lexington, MA, USA Brendan Manning T2 Biosystems, Lexington, MA, USA Elizabeth M. Marlowe Roche Molecular Systems, Inc., Pleasanton, CA, USA Alexander J. McAdam Infectious Disease Diagnostics Laboratory, Department of Laboratory Medicine, Boston Children’s Hospital and Harvard Medical School, Children’s Hospital Boston, Boston, MA, USA Alexander Mellmann Institute of Hygiene, University Hospital Münster, Münster, Germany Alex Mira FISABIO Foundation; Center for Advanced Research in Public Health, Valencia, Spain Beat Mueller Medical University Department, Division of General Internal and Emergency Medicine, Kantonsspital Aarau, Aarau, Switzerland University of Basel, Basel, Switzerland Johannes Müthing Institute of Hygiene, University Hospital Münster, Münster, Germany Caitlin Otto SUNY Downstate Medical Center, Brooklyn, NY, USA

Contributors

xiii

Volkan Özenci Department of Clinical Microbiology, Karolinska University Hospital, Stockholm, Sweden Division of Clinical Microbiology, Department of Laboratory Medicine, Karolinska Institute, Stockholm, Sweden Nicole Parrish Department of Pathology, Johns Hopkins University, Baltimore, MD, USA Eleanor A. Powell Department of Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA Heng Qian Luminex Corporation, Toronto, ON, Canada Stefan Riedel Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, MA, USA Alex B. Ryder University of Tennessee Health Science Center, Memphis, TN, USA Ramon Sager Medical University Department, Division of General Internal and Emergency Medicine, Kantonsspital Aarau, Aarau, Switzerland University of Basel, Basel, Switzerland Jonathan E. Schmitz Department of Pathology, Microbiology, and Immunology, Vanderbilt University Medical Center, Nashville, TN, USA Philipp Schuetz Medical University Department, Division of General Internal and Emergency Medicine, Kantonsspital Aarau, Aarau, Switzerland University of Basel, Basel, Switzerland Ted E. Schutzbank Ascension – St. John Providence, Grosse Pointe, MI, USA Susan Sefers Department of Pathology, Microbiology, and Immunology, Vanderbilt University Medical Center, Nashville, TN, USA Rosemary C. She Keck School of Medicine of the University of Southern California, Los Angeles, CA, USA Jeong Hwan Shin Department of Laboratory Medicine, Busan Paik Hospital, Inje University College of Medicine, Busan, South Korea Jessica L. Snyder T2 Biosystems, Lexington, MA, USA Kristoffer Strålin Department of Infectious Diseases, Karolinska University Hospital, Stockholm, Sweden Unit of Infectious Diseases, Department of Medicine Huddinge, Karolinska Institutet, Stockholm, Sweden Charles W. Stratton Department of Pathology, Microbiology and Immunology, Vanderbilt University School of Medicine, Nashville, TN, USA

xiv

Contributors

Mahesh J. Thalavitiya Acharige Division of Infectious Diseases, Brigham and Women’s Hospital, Boston, MA, USA Harvard Medical School, Boston, MA, USA Ephraim L. Tsalik Emergency Medicine Service, Durham VAMC, Durham, NC, USA Center for Applied Genomics & Precision Medicine, Duke University Medical Center, Durham, NC, USA Division of Infectious Diseases, Department of Medicine, Duke University Medical Center, Durham, NC, USA Yun F. (Wayne) Wang Emory University and Grady Memorial Hospital, Atlanta, GA, USA Yannick Wirz Medical University Department, Division of General Internal and Emergency Medicine, Kantonsspital Aarau, Aarau, Switzerland University of Basel, Basel, Switzerland Fann Wu Department of Pathology & Cell Biology, Columbia University Medical Center, New York-Presbyterian Hospital, New York, NY, USA Xiao Xu ACEA Biosciences, Inc, San Diego, CA, USA Min Zheng State Key Laboratory of Diagnostic and Treatment for Infectious Diseases, Hangzhou, Zhejiang, China

Automated Blood Cultures Xiang Y. Han

Introduction Cultivation or otherwise detecting an infectious agent typically confirms a clinically suspected infection. Timely identification of a cultured microorganism along with antimicrobial susceptibility testing is used to ensure effective antimicrobial therapy. Bloodstream infections, being systemic, are the most severe forms of infection. They are frequently life-threatening, and blood cultures to detect circulating microorganisms have been the diagnostic standards. Much of the scientific and technologic advances for blood culture methods have been made through the 1970s–1990s, with further refinements and improvements accomplished in the past two decades. This chapter briefly reviews important aspects of blood culture methodology with emphasis on automated culturing systems.

Principles The principles and scientific basis for optimizing the diagnostic yield of blood cultures have been reviewed and summarized (for adult patients) [1, 2]. Most parameters were initially established for manual blood culture systems that used basal culture media. A recent study addressed some of these parameters for newer automated culturing systems and media and found them to be still valid for the most part [3]. Major characteristics of blood cultures are summarized as follows.

X. Y. Han (*) Department of Laboratory Medicine, Unit 84, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_1

1

2

X. Y. Han

Host and Microbial Factors Invasion of the bloodstream by microorganisms reflects the failure of initial host defense, such as the loss of integrity of skin and mucosa and weakening of the innate and acquired immunity, to prevent such invasion or spread from a localized infection site. Among certain patients having an intravascular device or using recreational drugs intravenously, direct seeding of the bloodstream with microorganism is also possible. Once in the bloodstream, microbes are constantly attacked by host defenses, such as complements, phagocytic leukocytes, antibodies, and other factors, and are filtered by the liver and the spleen. The ability of invading microorganisms to evade host defenses (or antimicrobial agents) favors their survival and dissemination in the bloodstream. On the other hand, if the host defenses are paralyzed, such as seen with leucopenia or immune suppression by medications or other means, even the least pathogenic organisms are able to cause fatal infections. Therefore, both the host and microbial factors determine the occurrence, severity, and duration of septic episodes; these factors also influence the yield of blood cultures. Moreover, the presence of antimicrobial agents in the circulation may reduce the yield of blood cultures. Timing, Blood Volume, and Frequency of Cultures Timing of the blood draw may influence the yield of blood cultures. Most of the time, bacteria or fungi are not constantly distributed or evenly circulated in the bloodstream except in the case of endocarditis; thus, the host responses, such as rising fever, are likely to herald the best time to draw a blood culture. Blood culture should also be obtained, if at all possible, before the initiation of empiric antimicrobial therapy. For each septic episode, two or three sets of cultures over a 24-h period provide the maximum recovery for the offending microorganism(s). A set of blood cultures usually means one aerobic broth bottle and one anaerobic broth bottle with each inoculated with 10 ml blood for an adult patient and incubated under aerobic and anaerobic conditions, respectively. This practice requires draws of a total of 40–60 ml blood from two to three venipunctures from different arms. In children and infants, the volume will be less and should be based upon the weight of the child/infant (see below). In a culture bottle, the blood sample is diluted by the culture broth to reach a blood-broth ratio of 1:2.5–1:10, which may dilute inhibitory blood components to favor microbial growth. It is generally accepted and hence practiced that 40 ml for two culture sets in adults offers good culture recovery while maintaining cost-effective microbiology. A few laboratories have recently demonstrated that a 60 ml blood draw yielded consistently higher culture recoveries than did lower volume draws [4, 5]. However, drawing higher blood volumes may add to the cost and increase the likelihood of iatrogenic anemia for the patient, particularly one who was already anemic. The need to draw repeat blood cultures hinges on the patient’s response to initial treatment, the identity and hence expected behavior of the cultured microbe, and antimicrobial susceptibility test results. It may take a few days for a patient receiving adequate therapy to show obvious clinical improvement. The patient thus may still spike a fever for 2–3 days while clearing the killed and dying microorganisms

Automated Blood Cultures

3

from the circulation. The underlying condition of the patient, such as profound neutropenia, may also affect the response time to therapy. Persistence of fever during therapy is a common reason for repeating blood cultures. Atmosphere and Length of Incubation Traditionally, both aerobic and anaerobic blood cultures have been obtained and are thus recommended. However, the declining proportion of bacteremias due to obligate anaerobes has led to suggestion that routine anaerobic cultures may not be needed and should be tailored to the needs of individual institution, patient population, and even the individual patient. Anaerobic cultures are valuable for patients who have had recent surgery or gynecologic/ obstetric procedures because of the high number of anaerobes in the lower gastrointestinal and urogenital tracts. How long should blood cultures be incubated? Several studies on different culturing systems have shown that an incubation period up to 5 days is sufficient to recover nearly all significant microorganisms (~99%) [3, 6–9]. The vast majority of organisms become culture positive in the first 3 days, and most fastidious bacteria can be recovered during the extra 2 days, including the HACEK organisms (Aggregatibacter (Haemophilus) aphrophilus, Aggregatibacter (Actinobacillus) actinomycetemcomitans, Cardiobacterium species, Eikenella corrodens, and Kingella kingae), Brucella spp., and nutritionally variant streptococci (currently known as Abiotrophia species and Granulicatella species) [10]. A new species, Cardiobacterium valvarum, proposed by us, is a cause of endocarditis and can be cultured within 3 days [11]. The length of culture for Brucella spp. had been controversial until studies done in the past two decades by Bannatyne et al. [12] in which 90 of 97 such bacteremic patients became culture positive within 5 days and by Baysallar et al. [13] who noted that 30 of 30 were positive within 4 days. Bloodstream infections due to Francisella tularensis are rare today, with fewer than a dozen such cases per year in the United States, yet blood cultures usually become positive upon 3–8 days of incubation [10, 14]. Yeasts, such as Candida species that are among the most common ten microorganisms isolated from blood cultures, can also be isolated within 5 days of incubation [3, 6, 8]. Considerations of Blood Cultures in Pediatric Patients The blood culture methodology has been developed and refined in adult patients who have adequate circulating blood volumes for evaluation. In pediatric patients, however, the smaller body blood volume has contributed to far fewer studies. In a systematic review, Bard and TeKippe summarized current consensus for pediatric blood cultures [15]. Compared to the blood cultures in adult population, blood cultures in pediatric patients recover far more contaminants, accounting for 25–69% of all positive blood cultures. In order to reduce contaminants and to recover true pathogens, adequate blood culture volumes are important. Yet, for safety consideration, 90%. In the past several years, matrix-assisted laser desorption ionization time-of- flight (MALDI-TOF) mass spectrometry has been used to identify microorganisms

6

X. Y. Han

directly in positive blood culture bottles. Briefly, a microorganism in a positive broth is pelleted by centrifugation, washed with a saline solution, lysed with an organic solvent, and applied to MALDI-TOF. Identity of the organism can be achieved within 30 min. In several clinical evaluations [25–29], this methodology works well for most bacteria and fungi in the setting of monomicrobic positive blood cultures. As such, it is gaining popularity for its speed, simplicity, and accuracy. Other Trends Some noticeable trends in the past decades are increasing numbers of as well as increasing life span for immune defective or suppressed patients and, thus, emergence of more opportunistic or rare pathogens; more frequent use of antibiotics and associated resistance, in fact, up to 29% of blood cultures come from patients with active antimicrobial therapy; increasing use of indwelling devices, such as intravascular catheters and others; and emergence of more Candida and other fungal infections in the bloodstream [3, 30].

Automated Culturing Systems Blood culture methodology has evolved over the past four decades from manual methods (currently rarely being used) to automated blood culturing systems. The major advantage of an automated system, such as the earlier BACTEC NR660, is to automatically detect the growth of microorganisms through the presence and accumulation of CO2 produced by the metabolism of the organisms. The automation obviates manual inspection or examination that is required periodically to detect microbial growth. Automated agitation of culture bottles also improves mixing and aeration to promote the growth of aerobes and facultative anaerobes. These features make blind subcultures of negative bottles unnecessary, as shown in a few studies reviewed by Reimer et al. [2] Automation has improved the practice of blood culture enormously in terms of timely report of positive culture and more laboratory efficiency and consequently better patient care. Continuously monitoring blood culturing systems (CMBCS) currently are the most commonly used blood culture methods. Introduced in the early 1990s, CMBCS have added nearly continuous (every 10–12 min) monitoring of microbial growth to the automated systems. Currently, three CMBCS are available in the United States, and they are briefly shown in Table 1. More detailed description can be found elsewhere [31]. New versions of these CMBCS, available for the past decade or so, have kept the key elements from earlier versions while refined the hardware, computer system, data processing, and report. The trend is to increase user-friendly features for space, operation, and flexibility. Figure 1 illustrates a culture and detection curve using the BACTEC FX system. Numerous studies have been published to evaluate the performance of the CMBCS and associated media with or without various lytic agents or additives to remove blood antimicrobics; a number of representative ones are summarized as


7

Table 1 Commercial continuously monitoring blood culturing systems (CMBCS) in the United States Test interval Current (min) system, year BacT/Alert 3D, 10 2001

Microbial detection mechanism Colorimetric for CO2 production

Becton- Dickinson

BACTEC FX, 2009

10

Fluorescent for CO2 production

Trek diagnostic systems

VersaTREK, 2004

12

Manometric for gas (CO2 and other gases)

Manufacturer BioMerieuxa

Initial system since early 1990s BacT/Alert series for varying holding capacity BACTEC series for varying holding capacity ESP series for varying holding capacity

A new system from this manufacturer, called Bact/Alert Virtuo, was introduced in Europe in 2014 and is being evaluated in the United States and Canada

a

Fig. 1 Culture and detection of Klebsiella pneumoniae. The blood culture was set up in BACTEC FX system with Aerobic Plus broth bottle and the release of CO2 from bacterial growth monitored every 10 min by fluorescence. The culture turned positive with an incubation of 0 day 12 h and 40 min. A typical sigmoid curve was seen

follows (Table 2). McDonald et al. [32] compared the BacT/Alert standard bottle with BacT/Alert FAN bottle that contains Ecosorb™, an antimicrobic-absorbing substance, and they found that FAN bottle recovered significantly more microbes from all septic episodes, especially S. aureus, CoNS, and members of Enterobacteriaceae. Along with this, however, recovery of all contaminants, including CoNS, was also higher. The performance of the BacT/Alert FAN bottle and BACTEC Plus aerobic/F bottle (with resins to absorb antimicrobics) was also compared, and the two systems were found to be comparable [33]. In a study comparing

8

X. Y. Han

Table 2 Performance evaluations of automated culturing systems and media with or without lytic agents or additives Compared systems and/or media (broth bottle) BacT/Alert FAN vs. BacT/ Alert standard BacT/Alert FAN vs. BACTEC Plus/F BacT/Alert FAN vs. TREK ESP 80A

Main findings BacT/Alert FAN improved recovery of S. aureus, CoNS, and enterics Comparable

BacT/Alert FAN improved recovery of S. aureus, enterics, and non-Pseudomonas aeruginosa gram-negative rods BacT/Alert FAN vs. TREK Overall comparable. BacT/Alert FAN better for S. aureus and antibiotic-treated samples; ESP ESP 80A, in pediatric 80A better for streptococci and enterococci. patients BacT/Alert FAN vs. Comparable to detect fungemia. BACTEC fungal medium BACTEC Plus Anaerobic/F bottles detected BACTEC Plus more microorganisms Anaerobic/F bottles vs. Standard Anaerobic/F bottles BacT/Alert standard vs. BacT/Alert standard improved recovery of S. BACTEC 9240 standard aureus, CoNS, and yeasts BacT/Alert FA Medium in Comparable plastic vs glass bottles BacT/Alert 3D SA and SN vs. VersaTREK Redox media BacT/Alert FA plus and FN plus vs. BacT/Alert FA and FN media Bact/Alert Virtuo vs Bact/ Alert 3D

Overall comparable for bacteria and fungi. VersaTrek better in detecting streptococci and enterococci. Overall better culture detection and Gram stain interpretation with BacT/Alert FA plus and FN plus media. Similar recovery rates of but significantly shorter time to detection with Virtuo

Reference no. (author, year) [32] (McDonald, 1996) [33] (Jorgensen, 1997) [34] (Doern, 1998) [35] (Welby- Sellenriek, 1997) [36] (McDonald, 2001) [37] (Wilson, 2001)

[38] (Mirrett, 2003) [39, 40] (Petti 2005; Riley 2005) [41] (Mirrett 2007) [42] (Kirn 2014)

[43] (Jacobs 2017)

BacT/Alert FAN versus Trek ESP 80A, Doern et al. [34] found that BacT/Alert FAN recovered more S. aureus, enteric bacilli, and non-Pseudomonas aeruginosa Gram-negative rods, along with more contaminants too. In a similar study in pediatric patients [35], the two systems were found to be overall comparable with BacT/ Alert FAN recovering more S. aureus and better for antibiotic-containing samples and ESP 80A recovering more streptococci and enterococci. Since yeasts are an increasing cause of nosocomial bloodstream infections, McDonald et al. [36] compared BacT/Alert FAN with BACTEC fungal medium for their recovery, and the two systems were found comparable. The anaerobic culture media have also been compared; Wilson et al. [37] found that the BACTEC Plus Anaerobic/F bottles detected more microorganisms and episodes of bacteremia and fungemia than the BACTEC Standard Anaerobic/F bottles. Mirrett et al. [38] compared BacT/Alert standard bottle and BACTEC standard bottle and found the former significantly


9

improved the recovery of S. aureus, CoNS, and yeasts. Two recent studies found that, in the BacT/Alert system, the plastic culture bottles were comparable to the glass bottles [39, 40]. Recent studies have also evaluated the newer versions of CMBCS and their media. Mirrett et al. [41] compared the BacT/Alert 3D standard media and VersaTREK Redox media and found that they were overall comparable. The same team [42] recently compared the BacT/Alert FA plus broth and the FN plus broth with the BacT/Alert FA and FN media; they found slightly better performance of the newer FA plus broth and FN plus broth that contain adsorbent polymeric beads. A recent BacT/Alert system, BacT/Alert Virtuo, that was introduced in 2014 in Europe, was evaluated in the United States and Canada in 2017 [43]. This clinical trial compared BacT/Alert Virtuo with BacT/Alert 3D and found nearly identical recovery rates for the systems yet significantly shorter time to detection, by a mean of 1.8 h, with Virtuo. Together, these data show that CMBCS, each with similar cost, strengths, and weaknesses, perform well overall in delivering timely and accurate diagnosis of bloodstream infections. Incorporation of lytic or antimicrobic-absorbing substances in these systems has consistently improved the recovery of S. aureus and members of Enterobacteriaceae, particularly from treated patients.

Blood Culture and CMBCS for Mycobacteria Bacteremia due to rapidly growing mycobacteria (RGM) can be detected by blood cultures, similar to other fastidious organisms [44, 45]. In our experience with 115 consecutive clinical RGM strains [44], 46 (40%) were isolated from blood cultures using the BACTEC 9240 and/or the Isolator 10 system (Wampole Laboratories, Princeton, NJ). These RGM typically grow in 3–5 days; these bacteremias are usually associated with use of intravascular catheter. Among the 46 blood RGM isolates, Mycobacterium mucogenicum was the most common (24 of 46 or 52%), followed by M. abscessus complex and M. fortuitum complex. RGM can be recognized as beaded Gram-positive rods and Kinyoun stain-positive acid-fast bacilli. In contrast to RGM, Mycobacterium avium and M. tuberculosis are two species of many slowly growing mycobacteria. Blood culture has been useful to detect and monitor M. avium bacteremia in patients with AIDS. M. avium bacteremia usually occurs when the CD4+ cell count falls below 50/mm3 [46]. Circulating M. avium, exclusively within monocytes, usually range in 10–103 colony-forming units (CFU) per ml blood but can be as high as 106 CFU/ml [46]. A number of blood culture systems have been used: the earlier BACTEC 13A radiometric system and Isolator 10 system and the more recent CMBCS, such as BACTEC 9240 with MYCO/F LYTIC medium and BacT/Alert MB. Several studies have evaluated these systems, for example, in a controlled comparison of the performance of these systems. Crump et al. [47] found that these systems perform comparably well in detecting

10

X. Y. Han

M. avium bacteremia and other mycobacterial and fungal sepsis. In addition, the two CMBCS detect M. avium bacteremia 2–3 days sooner than the earlier systems. On average, an incubation of 14 days is required. Blood cultures also are able to detect M. tuberculosis bacteremia [47]. M. tuberculosis bacteremia appears to be particularly common in AIDS patients in developing countries. For examples, in Tanzania, it is the most common organism of all sepsis-causing microorganisms, accounting for 48% (57 of 118 patients) [48]. In Thailand, it ranks the second (27 of 114 patients), following Cryptococcus neoformans (30 of 114) and surpassing M. avium (24 of 114) [49]. In Brazil, it is also the most common cause of mycobacterial sepsis [50]. Blood cultures are as sensitive as bone marrow cultures for the detection of M. tuberculosis, and the role of this method appears to be expanding [51]. M. tuberculosis bacteremia in patients with AIDS is associated with a rapid and high mortality [52]. Developing countries recently have evaluated the performance of several blood culture systems for detecting M. tuberculosis bacteremia; these include the Isolator 10 system and the BACTEC using MYCO/F LYTIC medium, the conventional BacT/Alert FA, and the BacT/Alert MB [52–55]. Crump et al. [55] found that BacT/Alert MB detected M. tuberculosis bacteremia more rapidly than did manual methods, the BACTEC with MYCO/F LYTIC medium, and the Isolator 10 system. However, the mean time to positivity exceeded 3 weeks. The mean colony-forming units (CFU) per milliliter blood were found to be 30 CFU/ml in one study [55] and 8 CFU/ml in another study [52]. Together, these studies may provide some assistance with the initiation of timely empiric antibiotic coverage against tuberculosis in patients with AIDS in these countries in order to reduce the immediate mortality once the patient is seen at the hospital. These data may also impact public health policies and healthcare priorities in their respective countries.

Concluding Remarks In conclusion, automated blood cultures have become the diagnostic mainstay for bloodstream infections. The systems are refined and capable to cultivate various bacteria, fungi, and mycobacteria. The laboratories have seen improved efficiency through automation and a 5-day culturing cycle (except those for mycobacteria). With the vast majority of significant isolates being detected within the first 72 h of culture, the timely care of patients is facilitated. Recent integration of MALDI-TOF in the rapid identification of bacteria from positive culture broth has further shortened the turnaround time in the culture diagnosis of bloodstream infections.


11

References 1. Weinstein MP. Current blood culture methods and systems: clinical concepts, technology, and interpretation of results. Clin Infect Dis. 1996;23:40–6. 2. Reimer LG, Wilson ML, Weinstein MP. Update on detection of bacteremia and fungemia. Clin Microbiol Rev. 1997;10:444–65. 3. Cockerill FR 3rd, Wilson JW, Vetter EA, Goodman KM, Torgerson CA, Harmsen WS, Schleck CD, Ilstrup DM, Washington JA 2nd, Wilson WR. Optimal testing parameters for blood cultures. Clin Infect Dis. 2004;38:1724–30. 4. Lee A, Mirrett S, Reller LB, Weinstein MP. Detection of bloodstream infections in adults: how many blood cultures are needed? J Clin Microbiol. 2007;45:3546–8. 5. Patel R, Vetter EA, Harmsen WS, Schleck CD, Fadel HJ, Cockerill FR 3rd. Optimized pathogen detection with 30- compared to 20-milliliter blood culture draws. J Clin Microbiol. 2011;49:4047–51. https://doi.org/10.1128/JCM.01314-11. 6. Wilson ML. Clinically relevant, cost-effective clinical microbiology. Strategies to decrease unnecessary testing. Am J Clin Pathol. 1997;107:154–67. 7. Doern GV, Brueggemann AB, Dunne WM, Jenkins SG, Halstead DC, McLaughlin JC. Four- day incubation period for blood culture bottles processed with the Difco ESP blood culture system. J Clin Microbiol. 1997;35:1290–2. 8. Han XY, Truant AL. The detection of positive blood cultures by the Accumed ESP-384 system: the clinical significance of three day testing. Diagn Microbiol Infect Dis. 1999;33:1–6. 9. Bourbeau PP, Foltzer M. Routine incubation of BacT/ALERT FA and FN blood culture bottles for more than 3 days may not be necessary. J Clin Microbiol. 2005;43:2506–9. 10. Doern GV, Davaro R, George M, Campognone P. Lack of requirement for prolonged incubation of Septi-Chek blood culture bottles in patients with bacteremia due to fastidious bacteria. Diagn Microbiol Infect Dis. 1996;24:141–3. 11. Han XY, Meltzer MC, Woods JT, Fainstein V. Endocarditis with ruptured cerebral aneurysm caused by Cardiobacterium valvarum sp. nov. J Clin Microbiol. 2004;42:1590–5. 12. Bannatyne RM, Jackson MC, Memish Z. Rapid diagnosis of Brucella bacteremia by using the BACTEC 9240 system. J Clin Microbiol. 1997;35:2673–4. 13. Baysallar M, Aydogan H, Kilic A, Kucukkaraaslan A, Senses Z, Doganci L. Evaluation of the BacT/ALERT and BACTEC 9240 automated blood culture systems for growth time of Brucella species in a Turkish tertiary hospital. Med Sci Monit. 2006;12:BR235–8. 14. Han XY, Ho LX, Safdar A. Francisella tularensis peritonitis in stomach cancer patient. Emerg Infect Dis. 2004;10:2238–40. 15. Bard JD, Tekippe EM. Diagnosis of bloodstream infections in children. J Clin Microbiol. 2016;54:1418–24. 16. Woods-Hill CZ, Fackler J, Nelson McMillan K, Ascenzi J, Martinez DA, Toerper MF, Voskertchian A, Colantuoni E, Klaus SA, Levin S, Milstone AM. Association of a clinical practice guideline with blood culture use in critically ill children. JAMA Pediatr. 2017;171:157– 64. https://doi.org/10.1001/jamapediatrics.2016.3153. 17. Mirrett S, Weinstein MP, Reimer LG, Wilson ML, Reller LB. Relevance of the number of positive bottles in determining clinical significance of coagulase-negative staphylococci in blood cultures. J Clin Microbiol. 2001;39:3279–81. 18. Weinstein MP. Blood culture contamination: persisting problems and partial progress. J Clin Microbiol. 2003;41:2275–8. 19. Han XY, Kamana M, Rolston KV. Viridans streptococci isolated by cultures from blood of cancer patients: clinical and microbiologic analysis of 50 cases. J Clin Microbiol. 2006;44:160–5. 20. Li L, Tarrand JJ, Han XY. Microbiologic and clinical features of four cases of catheter-related infection by Methylobacterium radiotolerans. J Clin Microbiol. 2015;53:1375–9. https://doi. org/10.1128/JCM.00380-15. 21. Chapin K, Musgnug M. Evaluation of three rapid methods for the direct identification of Staphylococcus aureus from positive blood cultures. J Clin Microbiol. 2003;41:4324–7.

12

X. Y. Han

22. Oliveira K, Brecher SM, Durbin A, et al. Direct identification of Staphylococcus aureus from positive blood culture bottles. J Clin Microbiol. 2003;41:889–91. 23. Qian Q, Eichelberger K, Kirby JE. Rapid identification of Staphylococcus aureus in blood cultures by use of the direct tube coagulase test. J Clin Microbiol. 2007;45:2267–9. 24. Qian Q, Venkataraman L, Kirby JE, Gold HS, Yamazumi T. Direct detection of methicillin resistance in Staphylococcus aureus in blood culture broth by use of a penicillin binding protein 2a latex agglutination test. J Clin Microbiol. 2010;48:1420–1. 25. La Scola B, Raoult D. Direct identification of bacteria in positive blood culture bottles by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. PLoS One. 2009;4:e8041. https://doi.org/10.1371/journal.pone.0008041. 26. Stevenson LG, Drake SK, Murray PR. Rapid identification of bacteria in positive blood culture broths by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2010;48:444–7. https://doi.org/10.1128/JCM.01541-09. 27. Prod'hom G, Bizzini A, Durussel C, Bille J, Greub G. Matrix-assisted laser desorption ionization- time of flight mass spectrometry for direct bacterial identification from positive blood culture pellets. J Clin Microbiol. 2010;48:1481–3. https://doi.org/10.1128/ JCM.01780-09. 28. Christner M, Rohde H, Wolters M, Sobottka I, Wegscheider K, Aepfelbacher M. Rapid identification of bacteria from positive blood culture bottles by use of matrix-assisted laser desorption- ionization time of flight mass spectrometry fingerprinting. J Clin Microbiol. 2010;48:1584–91. https://doi.org/10.1128/JCM.01831-09. 29. Patel R. MALDI-TOF MS for the diagnosis of infectious diseases. Clin Chem. 2015;61:100– 11. https://doi.org/10.1373/clinchem.2014.221770. 30. Weinstein MP, Towns ML, Quartey SM, Mirrett S, Reimer LG, Parmigiani G, Reller LB. The clinical significance of positive blood cultures in the 1990s: a prospective comprehensive evaluation of the microbiology, epidemiology, and outcome of bacteremia and fungemia in adults. Clin Infect Dis. 1997;24:584–602. 31. Wilson ML, Weinstein MP, Reller LB. Commercial blood culture systems and methods. In: Truant AL, editor. Manual of commercial methods in clinical microbiology. 2nd ed. Hoboken: Wiley-Blackwell; 2016. p. 63–79. 32. McDonald LC, Fune J, Gaido LB, Weinstein MP, Reimer LG, Flynn TM, Wilson ML, Mirrett S, Reller LB. Clinical importance of increased sensitivity of BacT/Alert FAN aerobic and anaerobic blood culture bottles. J Clin Microbiol. 1996;34:2180–4. 33. Jorgensen JH, Mirrett S, McDonald LC, Murray PR, Weinstein MP, Fune J, Trippy CW, Masterson M, Reller LB. Controlled clinical laboratory comparison of BACTEC plus aerobic/F resin medium with BacT/Alert aerobic FAN medium for detection of bacteremia and fungemia. J Clin Microbiol. 1997;35:53–8. 34. Doern GV, Barton A, Rao S. Controlled comparative evaluation of BacT/Alert FAN and ESP 80A aerobic media as means for detecting bacteremia and fungemia. J Clin Microbiol. 1998;36:2686–9. 35. Welby-Sellenriek PL, Keller DS, Ferrett RJ, Storch GA. Comparison of the BacT/Alert FAN aerobic and the Difco ESP 80A aerobic bottles for pediatric blood cultures. J Clin Microbiol. 1997;35:1166–71. 36. McDonald LC, Weinstein MP, Fune J, Mirrett S, Reimer LG, Reller LB. Controlled comparison of BacT/Alert FAN aerobic medium and BACTEC fungal blood culture medium for detection of fungemia. J Clin Microbiol. 2001;39:622–4. 37. Wilson ML, Mirrett S, Meredith FT, Weinstein MP, Scotto V, Reller LB. Controlled clinical comparison of BACTEC plus anaerobic/F to standard anaerobic/F as the anaerobic companion bottle to plus aerobic/F medium for culturing blood from adults. J Clin Microbiol. 2001;39:983–9. 38. Mirrett S, Reller LB, Petti CA, Woods CW, Vazirani B, Sivadas R, Weinstein MP. Controlled clinical comparison of BacT/Alert standard aerobic medium with BACTEC standard aerobic medium for culturing blood. J Clin Microbiol. 2003;41:2391–4.


13

39. Petti CA, Mirrett S, Woods CW, Reller LB. Controlled clinical comparison of plastic and glass bottles of BacT/ALERT FA medium for culturing organisms from blood of adult patients. J Clin Microbiol. 2005;43:1960–2. 40. Riley JA, Heiter BJ, Bourbeau PP. Comparative recovery of microorganisms from BacT/ ALERT plastic and glass FA and FN blood culture bottles. J Clin Microbiol. 2005;43:3244–6. 41. Mirrett S, Hanson KE, Reller LB. Controlled clinical comparison of VersaTREK and BacT/ ALERT blood culture systems. J Clin Microbiol. 2007;45:299–302. 42. Kirn TJ, Mirrett S, Reller LB, Weinstein MP. Controlled clinical comparison of BacT/alert FA plus and FN plus blood culture media with BacT/alert FA and FN blood culture media. J Clin Microbiol. 2014;52:839–943. https://doi.org/10.1128/JCM.03063-13. 43. Jacobs MR, Mazzulli T, Hazen KC, Good CE, Abdelhamed AM, Lo P, Sum B, Roman KP, Robinson DC. Multicenter clinical evaluation of Bact/Alert Virtuo blood culture system. J Clin Microbiol. 2017;55:2413–21. 44. Han XY, De I, Jacobson KL. Rapidly growing mycobacteria: clinical and microbiologic analyses of 115 cases. Am J Clin Pathol. 2007;128:612–21. 45. El Helou GE, Viola GM, Hachem R, Han XY, Raad II. Rapidly growing mycobacterial bloodstream infections. Lancet Infect Dis. 2013;13:166–74. 46. Inderlied CB, Kemper CA, Bermudez LM. The Mycobacterium avium complex. Clin Microbiol Rev. 1993;6:266–310. 47. Crump JA, Reller LB. Two decades of disseminated tuberculosis at a university medical center: the expanding role of mycobacterial blood culture. Clin Infect Dis. 2003;37:1037–43. 48. Archibald LK, den Dulk MO, Pallangyo KJ, Reller LB. Fatal Mycobacterium tuberculosis bloodstream infections in febrile hospitalized adults in Dar es Salaam, Tanzania. Clin Infect Dis. 1998;26:290–6. 49. Archibald LK, McDonald LC, Rheanpumikankit S, Tansuphaswadikul S, Chaovanich A, Eampokalap B, Banerjee SN, Reller LB, Jarvis WR. Fever and human immunodeficiency virus infection as sentinels for emerging mycobacterial and fungal bloodstream infections in hospitalized patients >/=15 years old, Bangkok. J Infect Dis. 1999;180:87–92. 50. Grinsztejn B, Fandinho FC, Veloso VG, Joao EC, Lourenco MC, Nogueira SA, Fonseca LS, Werneck-Barroso E. Mycobacteremia in patients with the acquired immunodeficiency syndrome. Arch Intern Med. 1997;157:2359–63. 51. Crump JA, Tanner DC, Mirrett S, McKnight CM, Reller LB. Controlled comparison of BACTEC 13A, MYCO/F LYTIC, BacT/Alert MB, and Isolator 10 systems for detection of mycobacteremia. J Clin Microbiol. 2003;41:1987–90. 52. Munseri PJ, Talbot EA, Bakari M, Matee M, Teixeira JP, von Reyn CF. The bacteraemia of disseminated tuberculosis among HIV-infected patients with prolonged fever in Tanzania. Scand J Infect Dis. 2011;43:696–701. 53. Hanscheid T, Monteiro C, Cristino JM, Lito LM, Salgado MJ. Growth of Mycobacterium tuberculosis in conventional BacT/ALERT FA blood culture bottles allows reliable diagnosis of Mycobacteremia. J Clin Microbiol. 2005;43:890–1. 54. Gopinath K, Kumar S, Singh S. Prevalence of mycobacteremia in Indian HIV-infected patients detected by the MB/BacT automated culture system. Eur J Clin Microbiol Infect Dis. 2008;27:423–31. 55. Crump JA, Morrissey AB, Ramadhani HO, Njau BN, Maro VP, Reller LB. Controlled comparison of BacT/Alert MB system, manual Myco/F lytic procedure, and Isolator 10 system for diagnosis of Mycobacterium tuberculosis bacteremia. J Clin Microbiol. 2011;49:3054–7.

Laboratory Automation in Clinical Bacteriology Antony Croxatto

Introduction About 70% of medical decisions depend on laboratory results, which indicates that diagnostic tests have a great impact on health care [1, 2]. However, most routine diagnostic laboratories currently face a steady increase in sample numbers despite a limited budget as well as personnel shortages; this dichotomy significantly impacts quality. Thus, a number of diagnostic disciplines, including chemistry, hematology, and molecular diagnostics, have implemented automated methods over the past several decades; in contrast, diagnostic microbiology has been considered as too complex and having an insufficient number of samples to justify the development of automated solutions. Thus, conventional methods in bacteriology did not change much over these several decades until the relatively recent introduction of advanced approaches such as automated antibiotic susceptibility testing (AST), molecular diagnostic microbiology, and matrix-assisted laser desorption/ionization time-of- flight mass spectrometry (MALDI-TOF). The introduction of these advanced technologies has greatly improved the productivity of the clinical microbiology laboratory, but is not sufficient to deal with the continued increase in the volume of samples. Fortunately, the development of advanced technologies such as MALDI- TOF together with the introduction of liquid-based transport devices as well as the consolidation of clinical microbiology laboratories have finally triggered the development of “total automated systems” for diagnostic bacteriology. Even though the first automated modules for sample inoculation such as the Autostreaker were introduced in the 1970s [3, 4], the demand for such automated solutions has only emerged at the beginning of the twenty-first century. One particular company, Kiestra Lab

A. Croxatto (*) Institute of Microbiology, Lausanne University Hospital and Lausanne University, Lausanne, Switzerland e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_2

15

16

A. Croxatto

Automation (the Netherlands), has developed and proposed automated incubators since 2006. In 2008, Kiestra introduced the first automated line in bacteriology called “total laboratory automation” (TLA); this TLA connected an agar plate sorting and barcoding instrument with automated incubators and workbenches through a bidirectional conveyor system. The same year, three other companies, Becton- Dickinson (Baltimore, USA), bioMérieux (Marcy l’Etoile, France), and Copan (Brescia, Italy), commercialized their automated inoculation systems Innova, Previ- Isola, and WASP (Walk-Away Specimen Processer), respectively. Kiestra Lab Automation launched a semiautomated sample processor (InoqulA) in 2009 and a second generation that was fully automated in 2011. Eventually, Becton-Dickinson (BD) acquired Kiestra and formed BD Kiestra in 2012, which allowed the improvement of the TLA system by introducing updated inoculation and incubator systems, while Copan commercialized the first WASPLab in 2012. Thus, today there are several different automated systems for bacteriology, each with different technological architectures and workflows and each with their own advantages and disadvantages. Importantly, automated solutions represent a new revolution for diagnostic bacteriology with the promise of significant impact for productivity and quality in the clinical microbiology laboratory.

The Different Automated Systems Currently, four major diagnostic bacteriology activities can be automated: (1) inoculation, (2) plate management with conveyor systems and robotic arms, (3) incubation, and (4) digital imaging which allows plate reading by telebacteriology (Fig. 1). There are a number of commercially available different systems that address partially or totally these four processes (Fig. 1). Thus, the automated systems for bacteriology are commonly classified in three levels of automation: Level 1, inoculation only; Level 2, partial automation; and Level 3, complete automation (Fig. 1). The first level only includes specimen processors (Fig. 2). Currently, there are four automated inoculation systems commercially available: (1) the Autoplak (NTE-SENER), (2) the InoqulA (BD Kiestra), (3) the PreLUD (I2A diagnostics), and (4) the WASP (Copan). The Previ-Isola (bioMérieux) is no longer commercially available but has been installed in many laboratories during the past decade. The second level, partial automation, includes the Work Cell Automation (WCA; BD) and WASPLab (Copan) and consists of an inoculation module connected to incubators by a unidirectional conveyor system (Fig. 3). The third level, complete automation, only includes the TLA from BD and consists of an inoculation module, incubators, and workbenches that are connected by a bidirectional conveyor system (Fig. 3). The MAESTRO system from I2A, which belongs to the second level of automation, is still in a research and development phase and will not be discussed. This chapter will focus only on the current commercially available automated systems by BD and Copan.

Laboratory Automation in Clinical Bacteriology

17

Automated Inoculation

Incubation

Manual Imaging

Reading

Follow up work

AUTOPLAK Previ-Isola WASP

WASPLab / WCA

PRELUD

MAESTRO

InoqulA Specimen processors

TLA Incubators Digital imaging

TelePlates bacteriology delivery Level of automation

Inoculation

Partial lab automation

Complete lab automation

Fig. 1 Automated and manual laboratory activities. Different levels of automation are commercialized: Inoculation only (AUTOPLAK, Previ-Isola, WASP, PreLUD, InoqulA), partial automation (WCA, WASPLab, MAESTRO), and total automation (TLA). (Adapted with permission from Croxatto et al. [6])

Specimen Processors The InoqulA and WASP specimen processors (also called inoculation systems) utilize different technological approaches with different features and different workflows (Fig. 2 and Table 1). The InoqulA is composed of a fully automated (FA) and semiautomated (SA) module which are linked to the SorterA (media storage) and the BarcodA (barcode labeling) modules (Fig. 2). The FA module can only process liquid-based samples and uses a pipetting system that has the ability to inoculate 10–800 ul on agar plates, on slides, and/or in enrichment broth. The FA element is equipped with two robotic arms to manage sample containers and transportation, including decapping and recapping during the sample inoculation process. The system is also equipped with a vortex for sample agitation prior to pipetting. The SA module can be equipped with a safety hood and is used to inoculate nonliquid samples or liquid samples with insufficient homogenization or containing aggregates that can clog the pipetting system. In the SA mode, samples are deposited on the agar, but the streaking is automatically processed. The InoqulA is characterized by an innovative streaking approach consisting of a rolling magnetic bead. The magnetic bead covered with sample material streaks the microbes using different rolling movements such as

18

A. Croxatto

Fig. 2 Example of inoculation systems. Left: BD Kiestra inoculation system composed of the SorterA (plate storage and delivery), BarcodA (barcoding), InoqulA FA (full automated), and InoqulA SA (semiautomated). The FA module is equipped with two robotic arms and a pipetting system for sample container transportation, decapping, recapping, and pipetting during the inoculation process. Following sample material deposition on the agar by the pipetting system in the FA module or by a technician in the SA module, the streaking is processed by a rolling magnetic bead on the agar surface. Different streaking patterns can be used, including zigzag patterns as shown on the plate image. Both pre-barcoded slides and tubes can be also inoculated by the pipetting system. Right: Copan WASP (Walk Away Specimen Processor) inoculation system. The system is composed of three robotic arms to manage plates, slide, and broth inoculation and labeling. The system is using reusable loops to perform all its activities. Different patterns can be used for plate streaking, including semiquantitative and four quadrants (Adapted with permission from Croxatto et al. [6])

zigzag or four-quadrant patterns. This system is capable of generating a higher number of isolated colonies compared to manual streaking and compared to many other automated inoculation systems, even with high microbial loads [5]. Because of this technology, the InoqulA can streak up to five plates at once, allowing thus a maximum throughput up to 235 inoculations/hr. Plates are automatically labeled with barcode (BarcodA module), patient, and sample information, whereas slides and broths need to be pre-barcoded (manually or with dedicated external devices). Samples inoculated by the FA module are loaded on racks with dedicated sizes and having different capacities. Depending on the size of the containers, a maximum


19

WCA (BD) ReadA incubators

TLA (BD)

External workstation

ProceedA conveyor ErgonomicA workstation

InoqulA SA InoqulA FA BarcodA SorterA

BarcodA SorterA

WASPLab (Copan)

ReadA

incubators

ErgonomicA workstation InoqulA SA InoqulA FA

WASPLa bincubators WASP

Output stacks

Conveyor

External workstation

Fig. 3 Description of the different modules composing partial (WCA and WASPLab) and total lab automation (TLA) systems. FA, full automation; SA, semiautomation. (Images courtesy of BD Kiestra and Copan) (Adapted with permission from Croxatto et al. [6])

number of 270 samples at once can be loaded on the system. The racks carrying the samples are introduced in the FA module by pausing the system, indicating that the BD Kiestra inoculation system is mainly designed to process samples in batches. Finally, 12–48 different media types can be loaded in the storage module (SorterA module), depending on the system (InoqulA Stand-Alone Solution, WCA, TLA) and the number of inoculation devices. The WASP can only process liquid-based transport devices but includes a very flexible specimen processor that can vortex, centrifuge, decap, and recap the containers before inoculation. The WASP is composed of two robotic arms to manage both sample containers and plates and an additional robotic arm to inoculate agar plates, enrichment broth, and slides. The method for streaking is a reusable loop that mimics the human streaking movements throughout multiple streaking patterns including both semiquantitative and four-quadrant patterns (Fig. 2). The inoculation volumes are between 1 and 30 ul depending on the size of the loop used. An optional pipetting module can be integrated to inoculate larger volumes in enrichment broths or other growth media. The system can only streak one plate at a time with a maximum throughput of about 130–180 inoculations/hr. The storage carrousel of the WASP can store and use nine different media types. The WASP is designed to accept

20

A. Croxatto

Table 1 Technical features of automated inoculation systems manufactured by BD Kiestra and Copan Features Method of inoculation Streaking method Streaking patterns Consumable/waste Automatic decapping/recapping Number of different media at oncea Number of samples at once Continuous loading/unloading of the system Number of plates streaked at once Non liquid samples Automatic gram slide processing Automatic broth inoculation Throughputb Inoculation volume Image inoculum on platec Sample vortex Sample centrifugation Possibility to change loop/bead between quadrants Manual interaction Possibility of using biplates

InoqulA (BD Kiestra) Pipette Rolling bead Multiple Bead, pipette tip Yes 12–48 Up to 270 No for FA Yes for SA 1–5 Yes (SA mode) Yes (optional module) Yes (open platform) Up to 235 inoculations/hr 10–250 ul Yes Yes No No Yes (SA mode) Yes

WASP (Copan) Loop Loop Multiple Reusable loops (30,000 inoculations) Yes 9 72 Yes 1 No Yes (optional module) Yes (optional module. Copan tubes 5 ml and 10 ml) ~130 inoculations/hr (180) 1, 10 and 30 ul Yes (loop) Yes Yes Yes NO Yes

The WCA has a capacity of 12 different media types in the SorterA, while the TLA has a capacity up to 48 different media types at once depending on the number of installed InoqulA (1 or 2) b The throughput is dependent on multiple factors including the number of plates streaked per sample, the streaking pattern, and the inoculation of additional enrichment broth and slides c A sensor is detecting when the sample material is dropped by the pipette on the media, WASP: A sensor is detecting that the loop contains the sample material but not that the loop is touching the agar during the streaking process. FA, full automated; SA, semiautomated. (Adapted from Croxatto et al. [6]) a

a continuous sample loading of the system without pausing or interruption of the automated processes, allowing a continuous and flexible inoculation workflow. Moreover, the system is able to automatically label plates, slides, and broth with barcodes as well as patient and sample relevant information. It is important to appreciate that the sample throughput of the two systems largely depends on several factors including the number of media plates inoculated per sample, the streaking patterns, and the additional inoculation of slides and/or enrichment broths (Table 1). An insufficient throughput can result in significant delays in


21

the sample inoculation processing step and is an important factor that has to be thoroughly understood during a workflow analysis of a specific laboratory’s activities and needs.

Partial Automation: WCA and WASPLab The WCA and WASPLab both utilize a similar combination of automated modules with similar workflows (Figs. 3 and 4). However, the number and the types of inoculation systems and incubators differ between the WCA and WASPLab. The WCA contains one inoculation system (InoqulA, SorterA, and BarcodA) connected to 1–3 incubators, whereas the WASPLab can integrate 2 WASP inoculation systems connected to 1–3 single or double capacity incubators (Table 2). The plates, broth, and tubes are inoculated by the inoculation modules (Fig. 2). The plates are then automatically transported to incubators for incubation and imaging, whereas broth and slides are processed externally. The reading of digitalized plate images and followup work such as identification by MALDI-TOF and AST are done on external workstations. When needed, the plates are called and delivered to dedicated output stacks or carrousels that can be defined according (1) to the laboratory tasks that have to be performed such as microbial ID, AST, and small tests or (2) according to the sample types such as urines, biopsies, and respiratory samples. Plates that need additional incubation or plates that have been inoculated manually can be inserted into the system using input stacks or carrousels for incubation or can be inserted just for imaging in the automated incubators.

WASP Media storage

Samples

Inoculation

CO2

Normal atmosphere

Image acquisition

Image acquisition

Identification (bar code) Manuel Input

Workbench

Output Stacks or carrousel

Fig. 4 Partial laboratory automation (WCA and WASPLab). Example of the Copan WASPLab workflow. Plates, enrichment broth, and slides are inoculated and labeled with barcodes in the WASP. Plates are transported to the incubators by a unidirectional conveyor system. Upon request for follow-up work, plates are delivered to dedicated output stacks or carrousel. The technicians collect the plates and take them for follow-up work to external workbenches. (Adapted with permission from Croxatto et al. [6])

22

A. Croxatto

Table 2 Technical features of incubators manufactured by BD Kiestra and Copan Feature Capacity per incubator Plate loading Plate unloading Plate loading + picture Plate unloading + picture Plate unloading + picture + plate loading (Plates incubated in the lab automation incubators) Plate loading + picture + plate unloading (Plates incubated in external incubators) Maximal days of plates incubation Imaging time 0 Definition of the camera Size of the image files Light sources/background

ReadA compact (BD Kiestra) 1152 plates 600 plates/h 600 plates/h 300 plates/h 300 plates/h 150 plates/h

WASPLab incubator (Copan) 882/1764 platesa 360 plates/h 250 plates/h 250 plates/h ND 120 plates/h

163 plates/h

100 plates/h

ND Yes 5 Mp 3 Mb • Front, back, side lights • No or black background

6 Yes 48 Mp 20–25 Mb • Front, back lights • No or black background

The different indicated throughputs represent the maximal performance of stand-alone incubators. These values can greatly vary when the incubators are connected to the automated systems (inoculation modules and conveyors) a Single and double capacity WASPLab incubators. Mp, megapixel; Mb, megabytes; ND, not determined. (Adapted from Croxatto et al. [6])

Full Automation: TLA The TLA is composed of several modules, the SorterA (media storage and distribution), the BarcodA (barcodes labeling), the InoqulA (Inoculation FA/SA), the ReadA compact (incubation and imaging), and the ErgonomicA (workbenches) that are all connected with the ProceedA, a bidirectional conveyor system (Figs. 3 and 5). Thus, all the components are connected allowing plate distribution between all modules. Inoculated plates can be delivered from the InoqulA to incubators or to output stacks for external incubation. Automatically barcoded plates can also be delivered from the BarcodA to ErgonomicA workbenches for manual streaking (nonliquid samples, purity plates for AST, subcultures). When needed, plates are transported from ReadA compact incubators to ErgonomicA workbenches for follow-up work such as identification and AST. Plates can also be sent back to incubators or output stacks directly from workbenches. Manually inoculated plates and external plates can be inserted in the system by placing them on the conveyor for incubation or can be inserted only for imaging (e.g., anaerobic or microaerophilic cultures). The management of plates on the bidirectional conveyor systems is regulated by stacker/destacker regulatory nods that modulate plate workflow and distribution.

Laboratory Automation in Clinical Bacteriology SorterA

ErgonomicA

Media storage and distribution BarcodA

Identification (barcode)

Workbench

ProceedA

ProceedA ProceedA ErgonomicA Stacker Destacker

Workbench

InoqulA Samples

23

ReadA

Inoculation FA/SA Manuel input

ProceedA

ReadA ReadA

Normal atmosphere

CO2

Image acquisition

Image acquisition

Stacker Destacker

Fig. 5 Total laboratory automation. Example of the BD Kiestra TLA workflow. Plates, enrichment broth, and slides are inoculated by the InoqulA. Barcode labeling of plate is automatically performed in the BarcodA module, whereas enrichment broth tubes and slides have to be pre-barcoded (manually). The system is equipped with a bidirectional conveyor system that can transport inoculated plates from the InoqulA (automated inoculation) or workbenches (manual inoculation) to the incubators and from the incubators to workbenches for follow-up work. Plates can be sent back to the incubators when an additional incubation is required following reading or follow-up work. The ProceedA conveyor system is regulated by stacker/destacker regulatory nods for barcode reading and plate workflow and distribution. (Adapted with permission from Croxatto et al. [6])

Incubators As described previously, media plate management and automated digital imaging are performed in automated incubators that utilize both normal and CO2 atmospheres. The incubators of the BD Kiestra and WASPLab systems have a different architecture and throughput but share the same functionalities (Table 2) [6]. The incubators can operate automated storage and incubation of plates that are stored in unique cells in a carrousel with improved traceability and security [6]. They also provide a constant and uniform temperature allowing increased growth efficiency and thus decreased turnaround time (TAT). Automated incubators allow rapid plate delivery upon request for follow-up works, avoiding thus any major delay in the laboratory workflow. These incubators integrate an automated digital imaging system consisting of industrial camera that can take high-quality images with a resolution of 5 Mp (ReadA compact) to 48 Mp (WASPLab incubator). The plate images can be taken with various illuminating conditions thanks to several light sources (above, below, side) and different backgrounds (black and white), allowing an enhancement of colonies physiological and morphological traits such as opacity, blood hemolysis, and colors [6].

24

A. Croxatto

Recent and Future Hardware Developments Both BD Kiestra and Copan are working on the development of new automated modules with a potential to further improve laboratory productivity and quality (Table 3). The two companies are developing automated colony picking modules with several functionalities such as preparation of bacterial suspension for AST and deposition of sample material on MALDI plate targets. However, each company is also developing unique functionalities. The BD Kiestra system will include an automated AST panel preparation and loading of the BD Phoenix M50 systems, offering a full automation from colony picking to AST results. BD Kiestra will also release a disk dispensing module (DiskA) that can be connected to the BD Kiestra conveyor system. The WASPLab module, called Colibri, is integrating several functionalities, such as subculture, AST plates streaking with disk dispensing, as well as preparation of bacterial suspension for external automated AST such as Vitek (bioMérieux) and Phoenix (BD). The Colibri from Copan is already functional and has been recently introduced in a few laboratories for preliminary testing, whereas most of the BD Kiestra solutions are under development with a release planned in the next several years (Table 3). Finally, Copan is already proposing a sample input track dedicated to bacteriology (SIR; sample input rail) for dispatching different sample types to laboratories equipped with multiple WASPLab systems.

Imaging and Telebacteriology Digital imaging and telebacteriology are a major advance for routine diagnostic bacteriology [6]. The manual reading of plate is replaced by a reading of plate images on screens (telebacteriology). Laboratory technicians can quickly screen plate images for reading and results validation, with the ability to rapidly report negative results to the laboratory LIS. When microbial growth is observed, relevant microbial colonies are chosen by touch screen or mouse selection, which results in downstream actions such as MALDI-TOF ID, AST, and rapid biochemical tests (oxidase, spot indole, coagulase, etc.). Presumptive identification and/or normal flora results can also be directly sent to the LIS for rapid reporting of laboratory results in the electronic medical record, which adds considerable clinical value to these automated systems. The reading can be performed on a reading station near the automated system or on delocalized stations installed in “reading rooms” that offer a quiet background for both reading of routine sample and training of medical technologists. When follow-up testing has been ordered, technicians can call the plates directly to the workbenches with a BD Kiestra TLA or on dedicated output stacks or carrousels with the WCA or WASPLab systems. Upon scanning of the plate barcode, the technician can visualize on the digital image, the selected colonies, and the follow-up testing that have been assigned to these colonies by the reader. Telebacteriology offers thus an increased reading and follow-up testing


25

Table 3 Newly released automated modules or systems in research and development Feature Automated colony picking Bacterial suspension for AST (all platforms), subculture, purity plates MALDI plates Plates streaking + disk dispensing Automated AST panel preparation and loading Release Sample input track

BD Kiestra Yes Yes

WASPLab Yes Yes

Yes (Bruker) NO (separated modules) Yes (BD Phoenix M50™) 2018/2019 No

Yes (Bruker+Vitek MS) Yes

Broth incubator Release Disk dispenser

No NA Yes Streaking with InoqulA

Release

Unknown

Specimen processor pipetting system Release Next-generation middleware solution

Yes Available Yes (cloud based, for all BD solutions) 2018/2019 Next-generation InoqulA Unknown

Release Next-generation Inoculation module Release

No April 2017 Yes Inpeco SIR (sample input rail) Yes Unknown Yes WASP integrated Dedicated module (Automated colony picking) April 2017 (dedicated module) Yes April 2017 Yes Available

Release: Released date as communicated by the manufacturers in 2017

productivity but represents a significant change compared with conventional m anual plate reading, which combines visual information with the ability to smell and touch the colonies in order to gather multiple pieces of information for colonies identification. Telebacteriology represents a major advance in clinical bacteriology, but the reading process still relies on human-based decision and thus on technician’s experience and skills. Thus, further improvement of laboratory quality and productivity can only be achieved with the development of intelligent algorithms and expert systems that can replace human-based decision or offer a support to help laboratory technicians for both reading and results interpretation. Thus, new applications or algorithms should be developed for each analytical process, from digital imaging to detection, quantification, and identification of microbial growth (Table 4).

26

A. Croxatto

Table 4 Newly released applications or applications in research and development Applications Growth/no growtha Batch release Auto release (Semi) quantificationa Identification Sortinga Detection of sister colonies Quantification per isolate Color recognition Morphology recognition

BD Kiestra

WASPLab

Yes (any kind of plates) Yes Yes

Yes (any kind of plates) NO

3D imaginga

Yes Chromagar (MRSA, VRE) Yes Yes Yes Yes Yes • MH • MHF NO

Expert systems

Yes

Zone reading with expert system for ASTa

Yes Yes Chromagar (MRSA, VRE) Yes Yes Yes Yes Yes • MH Yes For reading but mainly for automated colony picking Yes

Newly released applications that can be introduced in routine diagnostic. The other applications are in research and development (R&D). Yes, newly released applications of applications in R&D. NO, application not in the R&D pipeline

a

Manufacturers have developed improved image acquisition algorithms to capture most valuable optical information across each red, green, and blue channels for accurate colony information extraction. Thus, industrial camera combined to different light sources and intensities as well as contrasting background can capture images for the optimization of both signal-to-noise ratio (SNR) and contrast. Copan uses separate algorithms for each plate type to optimize the quality and the information of the acquired image, which requires a calibration and optimization for each incubators’camera and each media type. BD has developed an algorithm (OPTIS™) allowing a standardization of acquired data using multi-sources image information. The system is performing a real-time image acquisition analysis to optimize both SNR and contrast for each pixel of the image by capturing 22 images for each media plate acquisition [7]. Thus, no specific calibration is required with this approach, which facilitates the installation of the BD Kiestra ReadA compact in routine d iagnostic laboratories. Copan has also developed the capacity to acquire three-dimensional (3D) images which represents a benefit for image acquisition and morphology recognition. However, 3D imaging has been developed primarily for automated colony picking, allowing a very fine regulation on the z axis for the picking system to collect a maximum of microbial objects without agar material.


27

Because of this high-quality image acquisition, there are new applications that allow rapid detection and quantification of microbial growth. These algorithms offer the capability to automatically sort plates by colony count (no growth, 102–103, 103–104, 104–105, >105) and also can be programmed to either auto-release negative plates as well as provide automated quantification or can be programmed to require human validation before release of these results. These growth/no growth applications clearly offer a significant increase in laboratory productivity by allowing batch release with rapid validation or auto-release of negative plates but also provide improved quality by exhibiting a much higher accuracy in detection and quantification of microbial growth compared to human visual inspection of media plates [7–9]. Moreover, these imaging applications also can be dedicated for automated disk diffusion inhibition zone measurement for AST, with or without human validation. The most recent imaging algorithmic developments allow identification of microbial colonies on chromogenic plates (color and morphological recognition) but also on conventional media plates such as blood, chocolate, and MacConkey agar (morphological recognition). These algorithms will trigger the development of several applications such as colony sorting, detection of sister colonies, colony identification, and quantification per isolate. In a recent study, the developed algorithms for the identification of colonies on chromogenic agar exhibited very high performances with accuracies ranging from 94% to 99% for the identification of bacterial species such as Escherichia coli and Enterococcus spp. or of bacterial groups (Klebsiella spp., Enterobacter spp., Serratia spp., Citrobacter spp. KESC group) [7]. The performance is even higher when these algorithms are trained for specific and defined detection and recognition. Thus, algorithms developed to automatically detect and segregate positive methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococci (VRE) exhibited a 100% sensitivity and 90–96% specificity [8, 9]. Moreover, these algorithms detected multiple plates that were missed following human visual inspection, allowing significant improvement of screening sensitivity compared to manual procedures. The data of image acquisition including detection, quantification, and identification provide valuable information for the development of expert systems. Expert systems are developed to integrate several critical information for result inter pretation including image information, patient demographic, clinical information (age, sex, hospitalization unit, etc.), and sample information (type of urine, sputum, biopsies, etc.). Based on expert laboratory rules that can be user-defined, these expert systems will be able to classify culture results for human manual review by providing an interpretation support or to release automatically the results to the laboratory LIS (auto-release). Thus, the integrations of intelligent algorithms for image acquisition and reading linked to expert systems will provide the capability in the near future to fully automatize routine diagnostic bacteriology, from sample reception to results report (Table 4).

28

A. Croxatto

Performance, Impact, and Gain For many years, the performance, impact, and improvements from laboratory automation were only seen in marketing operations with hoped-for performance and inferred expectations. Thus, there was a need for a thorough investigation of the true performance of total lab automation using objective, comparative, and prospective studies performed by independent laboratories. Because automated inoculation systems were commercialized before partial and total lab automation systems, these inoculation systems have already been tested in multiple studies. These studies focused on a comparison of automated inoculation system with conventional manual streaking or between different automated inoculation systems. These studies demonstrated that automated inoculation systems provided improved performance in a number of parameters, which included higher yield of discrete colonies, higher number of detected morphologies in polymicrobial samples, increased reproducibility, decreased analytical variation (sample processing, disk diffusion), increased accuracy, decreased hands-on time (≥50%), shortening of time to results (ID/AST), reduction of laboratory cost, and better laboratory workflows [5, 6, 10–16]. Because partial and totally automated systems have only been recently introduced in routine diagnostic laboratories, there are only a few studies that have demonstrated the impact of these laboratory automation systems. The efficiency of smart incubators has allowed a significant reduction of incubation time (from 24 h to 16 h), thus allowing a significant decrease of the TAT [17]. Compared to conventional methods, Mutter et al. demonstrated that optimized incubation conditions and rapid identification provided by the combination of total lab automation and MALDI-TOF resulted in a significant shortening of time to results of up to 30 h for microbial identification from positive blood cultures [18]. Moreover, this significant reduction of TAT resulted in a rapid adjustment of the antibiotic regimen in 12% of patients. Two studies presented at the European Congress of Clinical Microbiology and Infectious Diseases (ECCMID) in 2011 demonstrated that the introduction of BD Kiestra total lab automation and MALDI-TOF identification increased the laboratory productivity index (LPI, samples number/staff member/day) from 2- to 2.6-fold by allowing a reduction of required full-time equivalents (FTEs) for several laboratory activities [19, 20]. Being already equipped with a MALDI-TOF and a WASP, the diagnostic laboratory of the Lausanne University Hospital (Switzerland) conducted a study to estimate the gain of FTEs that could be expected following the sole introduction of partial or full automation [6]. Five main tasks including inoculation, plate reading, identification, AST, and plate management, representing 36% of the laboratory activities, could be impacted by total lab automation, representing a gain of 16.6% (2.4/14.5) FTEs. The reduction of FTEs conferred by laboratory automation (inoculation, plate management and reading) should allow a relocation of these FTEs to other essential laboratory activities including sample reception and preparation, microscopy, and follow-up works with improved quality and productivity.


29

The introduction of automation in clinical microbiology has thus far had a very positive impact on laboratory activities and TAT, but the clinical impact resulting from this increased efficiency remains to be determined by studies that prospectively evaluate the management and treatment of infected patients before and after the implementation of laboratory automation in clinical microbiology.

utomation Installation and Implementation: Difficulties A and Threats As demonstrated, the introduction of automation in diagnostic bacteriology should significantly improve laboratory productivity, quality, and time to report results. However, the installation of automation in routine diagnostic laboratories is complex, involves multiple stakeholders, and leads to many changes in both laboratory workflow and laboratory organization. Thus, the automation of a diagnostic laboratory requires a detailed project plan as well as change management in order to guarantee a successful achievement of the laboratory objectives [21].

Laboratory Reorganization The automation of bacteriology requires a profound reorganization of the laboratory utilizing a lean management approach in order to fully optimize the use of these automated systems [21]. Moreover, the conventional workflow and processes are in most clinical microbiology laboratories poorly adapted to automation and will need to be redesigned to improve laboratory productivity while maintaining a high- quality level of laboratory results. Thus, a failure in project planning and/or in change management will jeopardize the implementation of automation; thus the laboratory’s expectations such as increased productivity and quality as well as reduction of time to results will not be met. Ideally, laboratory reorganization should be conducted at every possible level of the laboratory activities from analytical workflows and task modifications to increased opening hours (24/7) with the introduction of staff shifts. In addition, the reorganization of the clinical microbiology laboratory is a complex process that can last several years and is subjected to both institutional regulatory rules and personal adhesion. The user needs time to fully understand the automated system and to define what can be or not be done with their chosen automated system. In addition, the adaptation of the laboratory staff to the automated system will take time, and misuse of this automated system during the first months of usage can lead to multiple system failures, which usually leads to a decreased confidence on the system by the laboratory technologists. It is thus essential to plan an efficient training program and change management before and during the first months of utilization of these complex systems which are revolutionizing

30

A. Croxatto

conventional bacteriology procedures. Moreover, laboratory automation needs to be a laboratory project that includes everybody, from technologists to managers and medical directors, in order to guarantee continued bacteriological interest as well as to avoid staff dismissals and/or resignations.

System Failures The automated microbiology systems are managing key laboratory tasks, which means that a dramatic impact on laboratory activities can be expected following a system failure. It is thus essential to ensure an effective support and maintenance (preventive and restorative) contract upon acquisition of an automation system. The laboratory technologists can be trained to prevent and fix some easy and frequently occurring small failures such as plate stocking and barcode reading, but more important failures require a manufacturer’s technician or engineer that ideally should be on site as quickly as possible (in a time frame usually defined in the maintenance contract). However, since sample processes cannot be totally interrupted, the laboratory together with the manufacturer must plan backup procedures to deal with both hardware and software failures. For instance, these procedures should include the need to keep external incubators for plate incubation in both normal and CO2 atmosphere as well as the possibility of performing manual inoculation and/or other laboratory activities. Depending on the type of system failure (central with an impact on the entire automated system or local with only isolated modules), the automated system can usually be run in a degraded mode offering the possibility to keep a low level of activity. It is thus important to clearly describe the procedures to avoid stress and time delay by defining a person in charge, whom to contact, and what to do for each type of failure that could be encountered with the automated system.

Connectivity The bidirectional interface between the automated systems and the laboratory information system (LIS) is crucial and complex. This interface always represents a high risk in a laboratory automation project since standard procedure is lacking. The interface between the automated system and the LIS is usually unique or at least exhibits differences between laboratories with similar LIS but different organization. Often, the automated system management software package offers functionalities that are not compatible with all laboratories LIS. Moreover, only a few technologists (if any) have a good working knowledge of both automated systems and LIS information technology (IT) specifications. In addition, each laboratory has strict IT and network security issues that need to be integrated with the interface of the automated system to the LIS. Thus, the interface has to be identified as a high


31

risk for project management planning and requires a thorough identification of the most appropriate stakeholders, from both the manufacturer and the laboratory, that need to work together to solve any and all of the complex IT and LIS issues that will be encountered during the installation and implementation of the automated system in the diagnostic laboratory. A failure to establish an efficient bidirectional interface will dramatically and negatively impact the success of the automation project and will make it impossible to achieve the laboratory expectations in a timely manner.

Conclusion Laboratory automation is a major paradigm shift in the field of diagnostic bacteriology. The commercialized systems and their future developments promise to provide a significant improvement of laboratory productivity, quality, and time to report results while decreasing errors, analytical variations, and laboratory costs. However, the introduction and implementation of automation requires optimal project planning and change management in order to avoid project failures and misuse of this automated tool. The lack of reorganization of the laboratory workflow and staff resignations can significantly impair all of the potential advantages and progress associated with laboratory automation. Moreover, the decision to automatize the laboratory should always be based on the thorough analysis of the laboratory activities and needs supported by a well-defined business plan to avoid the acquisition of any system not compatible with the laboratory requirements in terms of sample volumes, sample diversity, laboratory staff, and laboratory organization as well as meeting the expectations of laboratory management.

References 1. Forsman RW. Why is the laboratory an afterthought for managed care organizations? Clin Chem. 1996;42(5):813–6. 2. Wians FH. Clinical laboratory tests: which, why, and what do the results mean? Labmedicine. 2009;40(2):105. 3. Williams RE, Trotman RE. Automation in diagnostic bacteriology. J Clin Pathol Suppl. 1969;3:8–13. 4. Tilton RC, Ryan RW. Evaluation of an automated agar plate streaker. J Clin Microbiol. 1978;7(3):298–304. 5. Croxatto A, Dijkstra K, Prod'hom G, Greub G. Comparison of inoculation with the InoqulA and WASP automated systems with manual inoculation. J Clin Microbiol. 2015;53(7):2298–307. 6. Croxatto A, Prod'hom G, Faverjon F, Rochais Y, Greub G. Laboratory automation in clinical bacteriology: what system to choose? Clin Microbiol Infect. 2016;22(3):217–35. 7. Croxatto A, Marcelpoil R, Orny C, Morel D, Prod’hom G, Greub G. Towards automated detection, quantification and identification of microbial growth in clinical bacteriology: a proof of concept. Biomed J. 2017;40(6):317–28.

32

A. Croxatto

8. Faron ML, Buchan BW, Coon C, Liebregts T, van Bree A, Jansz AR, et al. Automatic digital analysis of chromogenic media for vancomycin-resistant-enterococcus screens using Copan WASPLab. J Clin Microbiol. 2016;54(10):2464–9. 9. Faron ML, Buchan BW, Vismara C, Lacchini C, Bielli A, Gesu G, et al. Automated scoring of chromogenic media for detection of methicillin-resistant Staphylococcus aureus by use of WASPLab image analysis software. J Clin Microbiol. 2016;54(3):620–4. 10. Bourbeau PP, Swartz BL. First evaluation of the WASP, a new automated microbiology plating instrument. J Clin Microbiol. 2009;47(4):1101–6. 11. Froment P, Marchandin H, Vande Perre P, Lamy B. Automated versus manual sample inoculations in routine clinical microbiology: a performance evaluation of the fully automated InoqulA instrument. J Clin Microbiol. 2014;52(3):796–802. 12. Iversen J, Stendal G, Gerdes CM, Meyer CH, Andersen CO, Frimodt-Moller N. Comparative evaluation of inoculation of urine samples with the Copan WASP and BD Kiestra InoqulA instruments. J Clin Microbiol. 2016;54(2):328–32. 13. Jones G, Matthews R, Cunningham R, Jenks P. Comparison of automated processing of flocked swabs with manual processing of fiber swabs for detection of nasal carriage of Staphylococcus aureus. J Clin Microbiol. 2011;49(7):2717–8. 14. Mischnik A, Mieth M, Busch CJ, Hofer S, Zimmermann S. First evaluation of automated specimen inoculation for wound swab samples by use of the Previ Isola system compared to manual inoculation in a routine laboratory: finding a cost-effective and accurate approach. J Clin Microbiol. 2012;50(8):2732–6. 15. Quiblier C, Jetter M, Rominski M, Mouttet F, Bottger EC, Keller PM, et al. Performance of Copan WASP for routine urine microbiology. J Clin Microbiol. 2016;54(3):585–92. 16. Hombach M, Jetter M, Blochliger N, Kolesnik-Goldmann N, Bottger EC. Fully automated disc diffusion for rapid antibiotic susceptibility test results: a proof-of-principle study. J Antimicrob Chemother. 2017;72(6):1659–68. 17. Bielli A, Lacchini C, Vismara C, Lombardi G, Sironi MC, Gesu G. WASPLab urine validation study: comparison between 16 and 24 hours of incubation. 25th European Congress of Clinical Microbiology and Infectious Disease 2015: Abstract EVO535. 18. Mutters NT, Hodiamont CJ, de Jong MD, Overmeijer HP, van den Boogaard M, Visser CE. Performance of Kiestra total laboratory automation combined with MS in clinical microbiology practice. Ann Lab Med. 2014;34(2):111–7. 19. Bentley N, Farrington M, Doughton R, Pearce D. Automating the bacteriology laboratory. 21st European Congress of Clinical Microbiology and Infectious Disease 2011: Abstract P-1792. 20. Humphrey G, Malone C, Gough H, Awadel-Kariem FM. Experience with KIESTRA’s total lab automation solution to meet the challenge of universal MRSA screening for Lister Hospital, a large UK district general hospital. 21st European Congress of Clinical Microbiology and Infectious Disease 2011: Abstract P-1793. 21. Croxatto A, Greub G. Project management: importance for diagnostic laboratories. Clin Microbiol Infect. 2017;23(7):434–40.

Biochemical Profile-Based Microbial Identification Systems Safina Hafeez and Jaber Aslanzadeh

Introduction The introduction of multiplex amplification-based syndromic panels, matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS), and next-generation sequencing (NGS) have decreased the utility of biochemical- based microbial identification test systems and are expected to eventually replace biochemical-based microbial identification methods in most clinical microbiology laboratories in the USA [1]. Overall, biochemical identification tests may be classified into two major groups: (1) the conventional bench microbial biochemical-based identification tests systems and (2) commercial microbial biochemical-based identification systems. The bench identification schemes used by various laboratories are not uniform in part due to the availability of numerous choices, varied complexity of the testing laboratories, volume, experience of technical staff, and cost. In general, most laboratories rely on a combination of both conventional bench testing biochemical-based test systems and commercial biochemical-based identification systems.

S. Hafeez Department of Pathology and Laboratory Medicine, Divsion of Clinical Microbiology, Hartford Hospital, Hartford, CT, USA e-mail: [email protected] J. Aslanzadeh (*) Division of Clinical Microbiology, Hartford Hospital, Hartford, CT, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_3

33

34

S. Hafeez and J. Aslanzadeh

Single Enzyme Rapid Tests The single enzyme rapid tests are a group of tests that detect the presence or absence of a single enzyme or a biochemical reaction within seconds to minutes. These tests are fairly inexpensive and easy to perform at the bench and often provide important initial information that can be used to determine the subsequent steps in the microbial identification scheme. In addition, these tests may be used for presumptive identification of certain organisms to the genus or even species level. For example, a positive catalase test can establish that a Gram-positive cocci is a Staphylococcus spp.; a subsequent positive coagulase test can then establish that the catalase- positive cocci is S. aureus.

Catalase Test Catalase, an enzyme within the cytochrome enzyme system, is responsible for the decomposition of hydrogen peroxide (H2O2) formed during aerobic respiration. All organisms using the cytochrome system of respiration will give a positive catalase reaction when tested. Those organisms using a different system will not produce catalase and will yield a negative reaction. The mechanism of action is as follows:

H 2 O2 + catalase = H 2 O + 1 / 2O2

The possession of the catalase enzyme helps to distinguish staphylococci from streptococci and is useful in the identification of many other bacteria. A positive test is a rapid bubbling reaction caused by the release of O2 from the H2O2 in the presence of catalase. A negative test is the absence of bubbling. Despite the simplicity of the test, a false-positive reaction may be seen if the test is performed on colonies selected from blood agar plate (BAP), colonies that are selected from the first quadrant of a culture plate used to subculture blood cultures or use of nickel loops to select colonies [2].

Oxidase Test The oxidase test is based on the production of the enzyme indophenol oxidase by microorganisms that utilize cytochrome C. Indophenol oxidase, in the presence of atmospheric oxygen, oxidizes a redox dye (N,N,N′,N′-tetramethyl-p- phenylenediamine dihydrochloride) to form a dark-purple compound, indophenol. Filter paper impregnated with the reagent is allowed to dry completely; the test is done by placing a loopful of bacteria from a nonselective plate onto the paper using

Biochemical Profile-Based Microbial Identification Systems

35

an inoculating loop or wooden applicator stick and then examining the paper for development of a violet or purple color within 30 sec (positive reaction). No color change or a slight color change of a light pink/light purple after 30 sec indicates a negative result. Steel loops, nichrome loops, or wire loops containing iron may give a false- positive reaction, and reactions from weak oxidase-positive organisms, e.g., Pasteurella may be inaccurate. Colonies growing on selective media or differential media can carry over the indicator and thus cause inaccurate results. Colonies grown on media containing high glucose concentration cannot be used for oxidase determination, since fermentation inhibits indophenol oxidase activity resulting in false- negative results [2, 3].

Spot Indole Test The indole test is based on the ability of an organism to hydrolyze tryptophan to glycine and indole. Certain organisms are able to remove the glycine radical from tryptophan resulting in the production of indole. This test can be performed on organisms grown on a BAP after 24 h of incubation. Filter paper is placed in a Petri plate and saturated with 3–4 drops of 1% solution of p-dimethylaminocinnamaldehyde. An isolated colony from a 24-h-old culture grown on a BAP is rubbed into the filter paper using a wooden applicator stick or inoculating loop. Appearance of a blue color immediately or within 30 sec of inoculation indicates a positive reaction; no blue color seen within 30 sec indicates negative reaction. The test must be performed from BAP. False-negative results will occur from MacConkey agar and Triple Sugar Iron (TSI) slants since there is no sufficient source of tryptophan in these media. False positives will occur if indole positive organisms are present in mixed cultures [4].

Slide Coagulase Test Coagulase is a thermostable enzyme found primarily in Staphylococcus aureus and is used to differentiate S. aureus from other commonly isolated staphylococci. Two forms of coagulase exist: one is bound to the bacterial cell wall, and one is liberated by the cell and is known as “free coagulase.” The slide coagulase test detects the bound coagulase (clumping factor), which acts directly on the fibrinogen in plasma and causes clumping of bacteria. The results of the slide coagulase test agree approximately 96% with the results of the tube coagulase test. Coagulase-positive organisms form clumps within 10 sec, but coagulase-negative organisms remain uniformly suspended.

36


The test is done as follows: using a sterile pipette, a drop of sterile saline is placed on a glass slide. One to two colonies of the microorganism is emulsified in the saline and tested for autoagglutination. A drop of rabbit plasma is placed on the slide and mixed for few seconds and observe for clumping within 10 sec. A positive slide coagulase test result is valid only for strains of Staphylococcus spp. that are negative for autoagglutination or stickiness. Coagulase is also present in S. intermedius and S. hyicus, but these species are infrequent clinical isolates. Similarly, clumping factor is produced by S. schleiferi and S. lugdunensis and may result in false-positive reactions [2, 5]. All negative slide coagulase tests must be confirmed using the tube coagulase test.

Microdase The Microdase disk is a reagent-impregnated disk used in the differentiation of Staphylococcus from Micrococcus by the detection of the oxidase enzyme. In the presence of atmospheric oxygen, the oxidase enzyme reacts with tetramethyl-p- phenylenediamine (TMPD) in the disk and cytochrome C in the organism to form a colored compound. All micrococci contain cytochrome C, whereas most staphylococci lack cytochrome C. The oxidase reagent substantiates the presence of type C cytochrome. The Microdase disk is placed on a glass slide and inoculated with several isolated colonies of an 18–24-h pure culture grown on BAP. The disk is examined for up to 2 min for development of a blue color to purple blue (positive reaction). No color change or white to gray color after 2 min is considered a negative reaction. Microdase is not designed for routine testing for oxidase activity in organisms other than Staphylococcus and Micrococcus. Staphylococcus sciuri is the only Staphylococcus species recognized to give a positive Microdase reaction [1, 4].

Bile Solubility Test Gross morphology alone is insufficient to differentiate between Streptococcus pneumoniae and alpha-hemolytic streptococci spp. S. pneumoniae lyse when treated with a 10% solution of sodium desoxycholate, while other streptococci and Gram- positive cocci are not bile soluble. Lysis occurs because bile soluble organisms contain an autolytic enzyme, an amidase that when activated by bile salts cleaves the bond between alanine and muramic acid in the cell wall. One drop of desoxycholate is placed on a well-isolated 18–24 h of incubation culture of an alpha-hemolytic colony on a BAP and incubated at room temperature, agar side down for 15–30 min. The area where the reagent was applied is examined


37

for evidence of colony disintegration or lysis. Dissolving of the colony is a positive test for S. pneumoniae. Colonies remaining intact indicates a negative test for S. pneumoniae. False negative may occur when testing isolated colonies older than 18–24 h. Occasionally, alpha-hemolytic colonies do not dissolve but merely lift off the surface of the agar, float away, and settle elsewhere on the plate. The plate should be carefully examined for evidence of this [6].

PYR PYR is a chromogenic substrate (l-pyrrolidonyl-β-naphthylamide or PYR) which when hydrolyzed by PYRase (l-pyroglutamyl aminopeptidase) produces a red color upon the addition of p-dimethylaminocinnamaldehyde. PYR is a substrate that is hydrolyzed by 100% of the enterococci and group A streptococci but not by any other streptococcal species. Two to four drops of a buffer reagent are applied to the PYR test strip circle. (Do not flood the disk.) The strip is then inoculated with 3–5 colonies of the organism (culture grown on BAP; 18–24 h old) and incubated at room temperature for 2 min. Two drops of p-dimethylaminocinnamaldehyde are applied to the test strip circle. An intense red color develops immediately around the colonies in the presence of hydrolyzed PYR. The PYR test is negative if no color, an orange color, or a weak pink color develops. Some Staphylococcus species may cause a positive PYR reaction [1, 4, 5].

Leucine Aminopeptidase (LAP) Test The LAP Test is a rapid assay for the detection of the enzyme leucine aminopeptidase in bacteria cultured on laboratory media. It is used as one of the tests for the presumptive identification of catalase-negative Gram-positive cocci. Leucine-β- naphthylamide impregnated disks serve as a substrate for the detection of leucine aminopeptidase. Following hydrolysis of the substrate by the enzyme, the resulting β-naphthylamine produces a red color upon the addition of cinnamaldehyde reagent. A moistened LAP disk is placed on a glass slide or in a Petri dish and inoculated with isolated colonies of catalase-negative, Gram-positive cocci. The disk is incubated at room temperature for 5 min before a drop of the color developer is added and examined for up to 1 min for pink to red color development. Pink/red color indicates a positive reaction. No color change/slight yellow indicates a negative reaction. The LAP test is only part of the overall scheme for identifying catalase-negative, Gram-positive cocci. Further biochemical characterization and serological grouping may be necessary for specific identification. False negatives may result from using an inoculum that is too small [7].

38


MUG Test The MUG test is a rapid test used for the presumptive identification of Escherichia coli. E. coli produces the enzyme beta-d-glucuronidase which hydrolyzes beta-d- glucopyranoside-uronic derivatives to aglycons and d-glucuronic acid. The substrate 4-methylumbelliferyl-beta-d-glucoronide is impregnated into a disk which, when hydrolyzed by the enzyme yields the 4-methylumbelliferyl moiety which fluoresces blue under UV light. Moisten a disk with water and with a wooden applicator stick, rub a portion of a colony from a pure culture onto the disk. Incubate in a close container at 35 °C for 2 h. Observe the disk under long wavelength (366 nm) UV light. An electric blue fluorescence is a positive reaction [2].

Indoxyl Butyrate Disk Moraxella catarrhalis produces the enzyme butyrate esterase. This property can be used as a rapid test in the identification of M. catarrhalis. Indoxyl is liberated from indoxyl butyrate by the enzyme butyrate esterase, forming an indigo color in the presence of oxygen. Smear several colonies of oxidase-positive, Gram-negative diplococci across the disk surface using a loop or wooden applicator. Incubate at room temperature for 5 min and observe for a blue-green color development where the colonies were applied indicating a positive test for butyrate esterase production. A negative reaction is indicated by no color change. Interpretation of results is based on testing only oxidase-positive, Gram-negative diplococci. Some strains of Moraxella spp. other than M. catarrhalis may produce a positive or weak positive reaction. Acinetobacter, Staphylococcus, and Pseudomonas may also yield a positive reaction [1, 5].

Chromogenic Enzyme Substrate Test The chromogenic enzyme substrate test is used for rapid identification of different Neisseria species, which are detected by a colored end product after synthetic chromogenic substrate hydrolysis by bacterial enzymes. The two commercially available systems that use this approach include the Gonochek II and the BactiCard Neisseria.

Gonochek II The Gonochek II consists of three synthetic chromogenic substrates contained in a single tube to detect preformed enzymes associated with different Neisseria species.


39

Oxidase-positive, Gram-negative diplococci from a pure culture growing on Martin-Lewis agar are emulsified in the tube with a wooden applicator. The tube is capped with a stopper and incubated for 30 min at 35 °C. Specific color reactions confirm the identity of N. lactamica (blue) and N. meningitidis (yellow). If neither color develops, the stopper is split apart, and the top part is inserted into the tube. The tube is inverted so that the suspension comes in contact with the diazo dye coupler (o-aminoazotoluene diazonium) on the stopper. Development of a pink-red color indicates the isolate is N. gonorrhoeae; absence of a colored product, or a pale-yellow color, is presumptive for Moraxella catarrhalis. The identification of M. catarrhalis can be confirmed by a positive M. catarrhalis butyrate test. The active chemical ingredients used in the tube and the enzymatic reactions detected are: (a) 5-Bromo-4-chloro-indolyl-β-d-galactopyranoside. Hydrolysis of the β-d- galactoside bond by β-galactosidase yields a blue color from the colorless substrate. (b) Gamma-glutamyl-para-nitroanilide. Hydrolysis of this substrate by gamma- glutamyl aminopeptidase releases yellow p-nitroaniline from the colorless substrate. (c) l-Proline-beta-naphthylamide. Hydrolysis of this substrate by hydroxy prolyl aminopeptidase releases colorless free beta-naphthylamine derivative. Coupling of the beta-naphthylamine derivative with a diazo dye coupler (o- aminoazotoluene diazonium salt – Fast Garnet, GBC Salt) produces a pink to red color. The Gonochek II should be used on Gram-negative diplococci isolated from media such as Martin-Lewis agar. Do not use on isolates only grown on nonselective media such as chocolate agar since other Neisseria species (N. sicca, N. mucosa) may grow and lead to incorrect results. Similarly, Kingella species may be found on Martin-Lewis medium. It is essential to perform a Gram stain prior to selecting organisms for identification and if the morphology of the organism selected is questionable, it is suggested that a catalase test be performed. Kingella species are catalase negative, and Neisseria and Moraxella species are catalase positive. N. cinerea will be pink after the addition of PRO (prolyl aminopeptidase) reagent [8].

BactiCard Neisseria This test uses an identification strip that contains the four chromogenic enzyme substrate tests for the identification of the different Neisseria species. The four test circles are rehydrated with buffer solution; growth from a selective media is applied to each of the four test circles area. If a blue-green color develops in the IB (butyrate esterase) within 2 min, the organism is identified as M. catarrhalis. If no color develops, the strip is incubated for another 13 min. If a blue-green color develops in the BGAL (β-galactosidase), the organism is N. lactamica. If the strip still remains

40


colorless in that time, a single drop of color-developing reagent is added to the PRO (prolyl aminopeptidase) and GLUT (gamma glutamyl aminopeptidase) test area. The development of a red color in the PRO test area identifies the isolate as N. gonorrhea while similar color change in the GLUT test area identifies the organism as N. meningitidis [9].

Hippurate The hippurate hydrolysis test may be used to identify Campylobacter jejuni, Gardnerella vaginalis, and Listeria monocytogenes or differentiate Streptococcus agalactiae from other beta-hemolytic streptococci. The assay is based on hydrolysis of the sodium hippurate by the enzyme hippuricase to sodium benzoate and glycine. Glycine is detected by oxidation with ninhydrin reagent that results in production of a deep purple color. Hippurate tubes are inoculated with a heavy suspension of the organism, or a hippurate disk could be added to the suspension and incubated at 35 °C for 2 h. The tube is then inoculated with 0.2 mL of ninhydrin and re-incubated for additional 15–30 min. The presence of deep purple color indicates a positive hippurate, and no color change indicates negative hippurate. A light inoculum or the use of old culture may give false-negative results [1, 4].

Lysostaphin The lysostaphin test is used to differentiate members of Staphylococcus spp. from Micrococcus spp. and is based on the activity of lysostaphin, which is an endopeptidase that cleaves the glycine-rich pentapeptide bridges of peptidoglycan. These crossbridges are found in all Staphylococcus spp. but not in Micrococcus spp. or Stomatococcus spp. A suspension of the organism equivalent to a 3.0 McFarland is prepared, and 0.2 mL of the working lysostaphin solution is added to the tube and mixed. The tube is allowed to stand undisturbed for 2 h at 35 °C. Clearing of the solution indicates susceptibility to lysostaphin. A turbid solution indicates resistance to lysostaphin. Micrococcus, Stomatococcus, and Streptococcus spp. are resistant to lysostaphin. Reading the test beyond 2 h of incubation may result in false-positive tests. Lysostaphin susceptibility also can be determined using the disk diffusion method. A plate of Mueller Hilton agar is inoculated, and a lysostaphin disk (10 μg) is placed on the plate. The plate is incubated for 24 h at 35 °C. A zone of inhibition of 10–16 mm indicates a Staphylococcus species. Micrococcus and related species will show no zones. To obtain optimal result, the organism must be grown in media containing beef peptone rather than casein peptone as the glycine content of the media is crucial [2].


41

CLO Test The CLO test is a rapid test for identification of Helicobacter pylori. The test uses a sealed plastic slide holding an agar gel that contains urea, phenol red, buffers, and bacteriostatic agents. If the urease enzyme of H. pylori is present in the inserted gastric tissue biopsy, the urea in the gel is degraded resulting in an increased pH; the color of the gel changes from yellow to bright magenta. Inoculate the CLO test slide with the specimen and incubate at 37 °C in the nonCO2 incubator for 3 h. The slide is examined for color change from yellow to magenta pink after 1 h of incubation and again at 2 and 3 h. A magenta pink color indicates a positive reaction. If the biopsy contains urease, the change first appears around the sample and eventually colors all of the gel. The pH change in a positive test is first seen at the interface of the gel and the biopsy. If a significant amount of urease is present, the visible change is rapid. Any color change of the whole gel to a shade other than yellow (i.e., red, magenta, pink, deep orange) indicates the presence of H. pylori. The test is considered negative if the medium remains yellow 24 h after insertion of the biopsy. False-negative CLO tests may occur when very low numbers of H. pylori are present or if the bacteria are focally distributed. False-positive CLO tests can occur in patients with achlorhydria. This is because commensal organisms such as Proteus spp. that also produce urease will grow in the absence of acid [22]. (Figs. 1, 2, 3, 4, 5, and 6).

Overnight Biochemical-Based Tests The overnight biochemical-based tests are a group of tests that require inoculating one or more culture media containing specific substrates and chemical indicators that detect pH change or detect specific microbial by-product(s). Similar to rapid tests, the choice of which overnight biochemical-based test is selected is based on Gram stain morphology and the results of preliminary testing with rapid enzyme tests. These tests are also inexpensive and easy to perform and may be used in three different ways. They may be used to obtain important initial information with respect to the identity of an unknown organism, such as the MILS test, which is used to screen for the presence of enteric pathogens. They may be used to verify the result of a preliminary positive/negative test result, or they may be used to assess an indeterminate finding. For example, Taxo P is an overnight test that will demonstrate if an isolate with an equivocal bile solubility result is S. pneumoniae. Similarly, a tube coagulase test will substantiate if a suspicious isolate, that is, slide coagulase negative, is truly a coagulase-negative Staphylococcus species. Finally, these tests may be used as the sole identification system (classical biochemical identification) to identify an unknown organism. This is generally labor intensive and requires the technologist to inoculate, incubate, read, interpret, and chart a number of

42

S. Hafeez and J. Aslanzadeh Gram positive Cocci

aerobe/facultative anaerobe

catalase

anaerobe

Peptostreptococci

+

yellowish to white colcony

-

bright yellow pigment colony

Coaghulase

Streptococci

Microdase

+

-

+

-

Staphylococcus aureus

novobiocin resistance

Micrococcus

Staphylococcus sp.

+

S. saprophyticus

-

alpha hemolytic

Beta hemolytic

bile soluble or optochin sensitive

bacitracin sensitive

+

S.pneumoniae

Staphylococcus sp.

-

+

-

PYR

Group A Strep

hippurate or CAMP

+

-

LAP, BE, 6.5% salt vancomycin

+, +, +,v

Enterococci

varible

Growth at 10oC

Vancomycin

-, v, +,s

R

Aerococcus

S

Streptococcus sp.

LAP

+

Pediococcus

+

-

Group B Strep

Streptococcus sp.

+

-

Lactococcus

Gamella

-

Leuconostoc

Fig. 1 Flow chart for presumptive identification of Gram-positive cocci

biochemical reactions over several days. This is then followed by using various identification schemes or flow charts to generate final identification. As a rule, the classical biochemical identification system is used to identify fastidious or slowgrowing organisms in the reference laboratories. These isolates are by and large rare biotypes that are not part of the commercial identification system’s database. Commonly used biochemical tests for identification of a Gram-negative organism include motility, OF glucose (oxid), OF glucose (Ferm), xylose, mannitol, lactose, sucrose, maltose, catalase, oxidase, macConkey, citrate, sodium acetate, urea, nitrate, nitrate to gas, indole, TSI slant, TSI butt, H2S (TSI butt), H2S (Pb ac paper), gelatin, pigment, arginine, lysine, and growth at 42 °C [10].


43 Gram Positive Bacilli

growth in air

+

-

Branch

spore

-

+

+

-

spore

Modified AFB

Lecithinase

Actinomyces Proppionibacterium Eubacterium Lactobacillus

+

-

+

-

+

-

beta hemolytic

catalase

Nocardia Rhodococcus Gordona Tsukamurella

Streptomyces Actinomadura Oreskovia Rothia

Reverse CAMP Dounble-zone beta hemolysis

C. septicum C. difficile

+

-

+

Bacillus sp.

motile

motile

-

beta hemolytic

H2S + in TSI

Beta hemolytic on HBT

Erysiplothrix rhousiopathiae

Gardnerella vaginalis

+

-

+

-

reverse CAMP positive

Bacillus sp.

? B. anthracis

Esculin

Corynebacterium sp. Brevibacterium Turicella Dermabacter

Archanobacterium hemolyticum

+

-

Clostridium perfringens

Urea

+

-

C. sordellii

Spot indole

+

-

+

-

Listeria monocytogenes

Kurthia

C. bifermentans

C. barati

Fig. 2 Flow chart for presumptive identification of Gram-positive bacilli

Tube Coagulase Test The tube coagulase test detects free coagulase (liberated by the cell) that forms a complex with coagulase-reacting factor found in plasma. The complex reacts with fibrinogen to form a fibrin clot. Several colonies of Staphylococcus are emulsified in 0.5 mL of rabbit plasma (with EDTA) to give a milky suspension, and then incubate at 35 °C for 4 h. Examine for the presence of a clot. If negative for clot, re-incubate the tube and reexamine at 24 h. Any degree of clot formation at 4 or 24 h is considered a positive reaction. No clot formation at 24 h is considered negative coagulase reaction [1, 2, 4].

DNA Hydrolysis The DNA hydrolysis test detects the presence of enzyme, deoxyribonuclease (DNase) in an organism. Using this media, DNase-positive coagulase-positive staphylococci are differentiated from other Staphylococcus spp. The media contains either toluidine blue or methyl green, which upon hydrolysis of the incorporated DNA turns colorless if methyl green is used in the media or pink if toluidine blue is used instead.

+

Proteus

+

-

-

Edwardsiella

Salmonella

Citrate

+

Urea

+

Hafnia

-

-

Shigella

+

Yersinia

Motlie at 22oC

Morganella

Urea

-

-

PAD

+

Citrobacter

H2S

+

Providancia

+

VP

-

delayed

-

+

Serratia

? A. actinomycet emcomitans

Esculin

+

P. oryzih abitsans ( -)

Insoluble yellow pigment

P. luteola (+)

? H. aphrophilus Capnocytophaga Dysgomonas

-

Soluble tan to brown pigment

S. matophilia

Maltose +

Motile

+

Bordetella parapertussis

Fig. 3 Flow chart for presumptive identification of aerobic and facultative anaerobic oxidase negative Gram-negative rod

-

Klebsiella

+

Enterobacter

-

Motility

+

H2S

Indole

E. coli

-

Catalase

ferement Lactose

+

+

Urea

-

+

-

Requires cycteine to grow

-

? F. tularensis

Rapid Urea

? Brucella sp.

Acinetobcater sp. Bordetella sp.

-

MacConkey

-

MacConkey

+

Oxidized/Inactive

feremented

Glucose

Aerobic and Faculatative Anaerobic Oxiadase Negative Gram Negative Rod

44 S. Hafeez and J. Aslanzadeh

N. meningitidis

N. gonorrhoeae

+

Moraxella sp Oligella sp.

-,-,-

Suttonella indologenes

?Kingella sp.

+

Bordetella parapertussis

Urea

non motile

+

-

Acinetobacter Bordetella sp.

Fig. 4 Flow chart for presumptive identification of Gram-negative cocci and coccobacilli

+,+, -

+,-,-

-

Kingella denitrificans

+

Indole

Catalase

Acid from Glu, Malt, Lac

-

?D. gladei

+

MacConkey

Growth on Martin Lewis

+

-

oxidase

+

Veillonella Bacteroides Prevotella Prophyromonas

+

growth in air

Gram Negative Cocci and Coccobacilli

+

-

?Dysgonomonas sp.

Catalase

+

-

Requires cycteine to grow

-

? F. tularensis

Rapid Urea

? Brucella sp.

Glucose fermenter

-

Biochemical Profile-Based Microbial Identification Systems 45

46

S. Hafeez and J. Aslanzadeh Aerobic and Facultative Anaerobic Oxidase Positive Gram Negative Rod

Glucose

feremented

inactive

MacConkey

+oxidized

MacConkey

MacConkey

+

-

+

-

+

-

Growth on TCBS

Curved rod

Alcaligenes Comamonas Bordetella Brucella

Afipia Brucella Bordettella Eikenella

Pseudomonas Burkholderia Ralstonia Pandoraea

Afipia Brucella Sphingomonas Roseomonas

Achromobacter Chryseobacterium Flavobacterium Shewanella

Chryseobacterium Flavobacterium

+

-

+

-

colony color

Lysine, Argenine, Ornithine

Campylobacter Helicobacter

Pasturella Kingella Capnocytophaga Cardiobacterium

Yellow

Green

+, +, +

v, v, v

Growth on 0% NaCl

Growth on 6 % NaCl

Plesiomonas shigelloides

violet pigment

+

-

+

-

+

-

?Vibrio cholera V. mimicus

Vibrio sp.

Vibrio sp.

Vibrio sp.

Chromobacterium violaceum

Aeromonas sp.

Fig. 5 Flow chart for presumptive identification of aerobic and facultative anaerobic oxidase- positive Gram-negative rods Preliminary Identification of Anaerobic Gram Negative Rod growth on KV, BBE agar plates -, -

+, +

Kan,Van, Col

Kan, Van, Col

Kan, Van, Col

Kan, Van, Col

R,S, R

R, R, R

S, R, S

R, R, V

brick red/orange/pink fluorescence

Bacteroides fragilis group

Catalase

brick red fluorescence black pigment

Prophyromonas sp.

+, -

+

-

+

-

Biolophila

requires formate fumarate

Pimented Prevotella Prevotella melininogenica

Prevotella sp.

+

-

B. ureolyticus

Fusobacterium sp.

Fig. 6 Flow chart for presumptive identification of anaerobic Gram-negative rods


47

The media is inoculated with the organism (generally, the organisms are boiled as some S. epidermis have a DNase, but the DNase is not heat stable) and incubated overnight at 35 °C. The plate is examined for evidence of growth and loss of color or a pink color around the inoculum (positive reaction). No color change indicates negative reaction [1].

Vancomycin Disk Test The vancomycin disk test is performed as a susceptibility procedure to help differentiate the Gram-positive, catalase-negative cocci. Aerococcus, Gemella, Lactococcus, Streptococcus, and some enterococci are susceptible to vancomycin. Leuconostoc, Pediococcus, Lactobacillus, and some enterococci are resistant to vancomycin. A 0.5 McFarland suspension of the organism is prepared in sterile saline. Using a sterile swab, the bacterial suspension is inoculated onto a BAP. A vancomycin disk is placed in the center of the inoculated plate and incubated at 35 °C in a CO2 incubator for 18–24 h. The plate is observed for the presence of a zone of inhibition around the vancomycin disk. Leuconostoc spp., Pediococcus spp., Lactobacillus spp., and some Enterococcus spp. are resistant to vancomycin with growth to the edge of the disk ≤9 mm. Aerococcus spp., Gemella spp., Lactococcus spp., Streptococcus spp., and some Enterococcus spp. are susceptible to vancomycin and produce a zone of inhibition ≥12 mm [1, 2].

Bacitracin Inhibition Test (Taxo A Disk) The bacitracin inhibition test presumptively differentiates Streptococcus pyogenes, group A streptococci (GAS) from other beta-hemolytic streptococci. The bacitracin at concentration of 0.04 units will selectively inhibit growth of GAS. While there are rare strains of GAS that are bacitracin resistant, approximately 5–10% of strains of non-group A beta-hemolytic streptococci (C, F, and G) are bacitracin susceptible. Using a pure culture of the test organism, inoculate a BAP with the bacterial suspension. Using a sterile forceps, place a bacitracin disk in the first quadrant (the area of heaviest growth of the inoculated BAP) and incubate at 35°C for 18–24 h. Any zone of inhibition around the bacitracin disk is considered a positive test. Uniform lawn of growth right up to the rim of the disk indicates negative bacitracin inhibition test [1, 2, 5].

48


Bacitracin and STX Susceptibility Test Susceptibility to low concentrations of bacitracin and sulfonamide trimethoprim- sulfamethoxazole (SXT) is a relatively inexpensive method for the presumptive identification of both group A and group B β-hemolytic streptococci. Group A streptococci are susceptible to bacitracin but resistant to SXT, and Group B streptococci are resistant to both antibiotics. Other β-hemolytic streptococci may show varying susceptibility to bacitracin but are also susceptible to STX so the combination increases the sensitivity and predictive value of the bacitracin test. Streak three or four isolated colonies of β-hemolytic streptococci down the center of a BAP. Using a loop, spread the inoculum across the plate to create a confluent lawn. Place a Taxo A bacitracin disk and a SXT disk on the inoculated area. Incubate the plate in ambient air at 35 °C for 18–24 h. Susceptibility is defined as any zone around either disk while resistant is growth up to the edge of the disk [2].

Taxo P Disks (Optochin) Ethylhydrocupreine hydrochloride (optochin) at the concentration 5.0 μg or less selectively inhibits the growth of S. pneumoniae, but not of other streptococci. S. pneumoniae may, therefore, be differentiated from other alpha-hemolytic streptococci by the formation of a zone of inhibition around a disk impregnated with this compound. Three or four well-isolated colonies of alpha-hemolytic Streptococcus isolate are streaked onto one half of a BAP plate. Using a flamed forceps, place a Taxo P disk (optochin) firmly in the upper one-third of the streaked areas and incubate the plate aerobically at 35°C for 24 h in 5–7% CO2. If using a 6 mm disk, a zone of inhibition of 14 mm or greater is considered sensitive. With a 10 mm disk, a zone of 16 mm or greater is sensitive. Other organisms may show zone sizes less than 14 mm in diameter. A diameter between 6 and 14 mm is questionable for S. pneumoniae, and the strain should be tested for bile solubility [1].

CAMP Test The CAMP test is based on the fact that group B streptococci produce an extracellular protein known as the CAMP factor that acts synergistically with a staphylococcal beta-hemolysin (β-lysin) on sheep erythrocytes to produce an enhanced zone of hemolysis [11]. Streak a loopful of β-toxin-producing S. aureus in a straight line across the center of a BAP. Streak a loopful of group B streptococci perpendicular to and nearly touching the streak line of the staphylococci (positive control). Streak a loopful of group A streptococci perpendicular to and nearly touching the streak line of the


49

staphylococci (negative control). Streak a loopful of unknown isolate perpendicular to and nearly touching the streak line of the staphylococci, and incubate the plate at 35 °C for 18–24 h in the aerobic non-CO2 incubator. Following the incubation, if the unknown isolate demonstrates an arrowhead zone of enhanced hemolysis, the isolate is identified as group B streptococci. If the unknown isolate does not demonstrate an arrowhead of enhanced hemolysis, the isolate is not a group B streptococcus. Do not incubate the CAMP test plate in the presence of 5–10% CO2 incubator as this can cause false-positive results with non-group B Streptococcus spp.

Reverse CAMP Test The reverse CAMP test is based on the fact that some organisms such as Arcanobacterium haemolyticum completely inhibit the effect of S. aureus β-hemolysin on sheep erythrocytes. The β-hemolysin inhibition zone in the form of a triangle is formed. A loopful of β-toxin-producing S. aureus is streaked in a straight line across the center of a BAP. Group B streptococci and group A streptococci are streaked perpendicular to and nearly touching the streak line of the staphylococci. Similarly, A. haemolyticum and the test isolate are streaked perpendicular to and nearly touching the line of the staphylococci. The plate is incubated at 35 °C for 24 h in the aerobic non-CO2 incubator. Following the incubation, if the test isolate demonstrates a triangular-shaped inhibition of β-hemolysis, it is reverse camp test positive. If the test isolate does not demonstrate a triangular-shaped inhibition of β-hemolysis, it is reverse camp test negative. Do not incubate the reverse CAMP test plate in the 5–10% CO2 incubator. This may result in an incorrect interpretation [5, 10].

Bile Esculin Agar Slant Group D streptococci (including Enterococcus spp.) and a few other bacteria, such as Listeria spp., can grow in the presence of 40% bile and produce esculinase, which hydrolyzes esculin to esculetin. Esculetin reacts with ferric ions, supplied by ferric citrate in the agar medium, to form a diffusible black complex. Most strains of viridans streptococci that are capable of hydrolyzing esculin will not grow in the presence of 40% bile. Streak the surface of the bile esculin agar slant with several colonies of the organism to be tested. Incubate at 35°C in non-CO2 for 24–48 h. A diffuse blackening of more than half of the slant within 24–48 h is considered positive. No growth or growth without blackening of the medium after 48 h is considered negative test.

50


If the inoculum is too heavy, viridans streptococci may give a false-positive test result. Approximately 3% of viridans streptococci are able to hydrolyze esculin in the presence of bile. Growth in the presence of 6.5% salt is used to differentiate enterococci from non-enterococcal group D streptococci [2, 5].

6.5% Salt Broth Heart infusion broth is a general-purpose medium for the cultivation of both fastidious and non-fastidious organisms. With the addition of 6.5% sodium chloride, the medium can be used to differentiate between salt-tolerant and salt-intolerant organisms. It is especially useful for distinguishing Enterococcus spp., which are salt- tolerant, from non-enterococcal group D streptococci, such as S. gallolyticus (previously known as S. bovis) and S. equinus. This broth contains small amount of glucose and a bromocresol purple as an indicator of acid production. Inoculate the tube containing 6.5% sodium chloride with the organism and incubate at 35 °C in non-CO2 for 24–48 h. A visible turbidity with or without color change is considered positive, and no growth or color change is considered negative. If the medium is inoculated too heavily, the inoculum may be interpreted as growth, resulting in a false-positive reaction. Aerococci, Pediococci, Staphylococci, and up to 80% of group B Streptococci can grow in 6.5% salt broth. In addition, Aerococci may also be bile-esculin positive [2, 5].

Indole Test Indole, a benzyl pyrrole, is one of the metabolic degradation products of the amino acid tryptophan. Bacteria that possess the enzyme tryptophanase are capable of hydrolyzing and deaminating tryptophan with the production of indole, pyruvic acid, and ammonia. The indole test is based on the formation of a red color complex when indole reacts with the aldehyde group of p-dimethylaminobenzaldehyde, the active chemical in Kovac’s or Ehrlich’s reagent. In order to perform this test, the organism must be grown on a medium rich in tryptophan such as indole nitrate broth. Inoculate the indole nitrate broth medium with 2–3 colonies of the organism to be tested. Incubate the tubes at 35°C in a non-CO2 incubator for 18–24 h. Examine the tubes for growth. When the broth is visibly turbid, using a sterile pipette, transfer 3 mL into a sterile tube. Add 0.5 mL of Kovac’s reagent to tube and observe for the development of a bright fuchsia red color at the interface between the reagent and the broth. If Ehrlich’s reagent is used instead, add 1 mL of xylene to the contents of the tube, which extracts the indole, if present, from the broth into the xylene. Wait 1–2 min, and add 0.5 mL Ehrlich’s reagent, and observe for the production of a pink to red color in the xylene layer. A pink to red color change after addition of either reagent indicates positive reaction. No color change indicates negative reaction [2].


51

Nitrite Test Organisms that reduce nitrate have the ability to extract oxygen from nitrates to form nitrites and other reduction products. The presence of nitrites in the medium is detected by the formation of a red diazonium dye, p-sulfobenzeneazo-α- naphthylamine, following the addition of α-naphthylamine and sulfanilic acid. If no color develops after adding the reagents, this indicates that nitrates have not been reduced (a true negative reaction) or that they have been reduced beyond the oxidation level of nitrite to products such as ammonia, nitrogen gas (denitrification), nitric oxide (NO), or nitrous oxide (N2O) and hydroxylamine. Since the test reagents detect only nitrites, the latter process would lead to a false-negative result. Therefore, it is necessary to add a small amount of zinc dust to all negative reactions. Because zinc ions reduce nitrates to nitrites, the development of a red color after adding zinc dust indicates the presence of nitrates and confirms a true negative reaction. Using a sterile inoculating loop, an indole nitrate broth medium is inoculated with 2–3 colonies of the organism to be tested and incubated at 35°C in a non-CO2 incubator for 18–24 h. When the broth is visibly turbid, 3 mL of the broth culture is transferred into a sterile tube and 1 mL of α-naphthylamine (Nitrate Reagent A) is added to the broth. One mL of sulfanilic acid (Nitrate Reagent B) is then added to the broth and observed for the production of a pink to red color within 30 sec. If no color change occurs within 30 sec, a small amount of zinc dust is added and looked for the production of a pink to red color within 10 min [2].

ALA (Haemophilus Influenzae Porphyrin Test) The porphyrin test is used in direct assessment of the ability of Haemophilus to synthesize protoporphyrins intermediates in the production of hemin (Factor X) from substrate, δ-aminolevulinic acid. Haemophilus species (H. parainfluenzae and H. parahaemolyticus) that produce the enzyme porphobilinogen synthase have the ability to synthesize heme (factor X) and therefore do not require an exogenous source of factor X for growth. Porphobilinogen and porphyrin, precursors in heme synthesis, can be detected by inoculating the Haemophilus strain in δ-aminolevulinic acid (which can be incorporated in a disk, agar, or liquid) and by the addition of Kovac’s reagent or by examination with a Wood’s lamp. In the tube method, a loopful of organisms is suspended in 0.5 mL of δ-aminolevulinic acid. Incubate at 35°C for 4 h if the suspension is heavy or 18–24 h if the suspension is light. After incubation, add an equal volume of Kovac’s reagent and vortex the mixture. Allow substrate and reagent to separate. After the addition of Kovac’s reagent, a red (pink) color will form in the aqueous phase, indicating the presence of porphobilinogen, and therefore a positive test for Haemophilus spp. not requiring factor X. Alternatively, a Wood’s lamp can be used to detect fluorescence in the reagent phase, indicating the presence of porphyrins, also a positive test. No

52


coloration or fluorescence indicates a factor X-dependent Haemophilus spp. and a negative test. In the disk method, the ALA impregnated disk is moistened with water and a loopful of organism is gently rubbed onto the disk. The disk is incubated for 4 h and observed under ultraviolet light for pink fluorescence [2, 12].

Motility Indole Lysine (MILS) MILS medium is a semisolid medium useful in the identification of members of the Enterobacteriaceae, specifically for screening suspicious colonies from stool cultures for potential pathogens. This test is used to demonstrate motility, indole production, lysine decarboxylase and deaminase activity, as well as hydrogen sulfide production. A small amount of agar is added to the media for demonstration of motility along a stab line of inoculation. Growth of motile organisms extends out from the line of inoculation, while non-motile organisms grow along the stab line. The pH indicator bromocresol purple is used to facilitate detection of decarboxylase activity. When inoculated with an organism that ferments dextrose, acids are produced that lower the pH, causing the indicator in the medium to change from purple to yellow. The acidic pH also stimulates enzyme activity. Organisms that possess a specific decarboxylase degrade the amino acid provided in the medium, yielding a corresponding amine. Lysine decarboxylation yields cadaverine. The production of these amines elevates the pH and causes the medium in the bottom portion of the tube to return to a purple color. The medium in the upper portion of the tube remains acidic because of the higher oxygen tension. Lysine deamination produces a color change in the upper portion of MILS Medium. Oxidative deamination of lysine yields a compound that reacts with ferric ammonium citrate, producing a burgundy red color on the top of the medium, and the bottom portion of the medium remains acidic. This reaction can only be detected if lysine decarboxylation is not produced, which is the case with Proteus, Morganella, and Providencia species. Indole is produced in MILS Medium by organisms that possess the enzyme tryptophanase. Tryptophanase degrades the tryptophan present in the casein peptone, yielding indole. Indole can be detected in the medium by adding Kovac’s reagent to the agar surface. MILS Medium is also used in the demonstration of hydrogen sulfide production. Hydrogen sulfide, which is produced by some enteric organisms from sulfur compounds contained in the medium, reacts with ferric ion, producing a characteristic black precipitate. (Note: Kovac’s reagent is not added until the final lysine carboxylation, lysine deamination, and motility results are interpreted.) Occasionally, the indole test produces a false-negative or false-weak reaction [13].


53

O-Nitrophenyl-β-d-Galactopyranoside Test In order for an organism to ferment lactose, it must have enzymes permease to transport the lactose inside the cell and β-galactosidase to cleave the transported sugar. Some organisms (delayed lactose fermenters), though possess β-galactosidase, do not have the enzyme permease. These organisms can utilize the enzyme β-galactosidase to hydrolyze O-nitrophenyl-β-d-galactopyranoside (ONPG). ONPG is a colorless compound similar to lactose. In the presence of β-galactosidase, ONPG is hydrolyzed to galactose and a yellow compound o-nitrophenyl. Bacteria grown on lactose-rich medium such as KIA and TIA gives optimal results with the ONPG test. A loopful of bacteria is suspended in 0.5 mL of saline to produce a heavy suspension. Add an equal amount of buffered ONPG solution to the suspension and incubate at 37 °C. Periodically, examine the color change for up to 24 h. Yellow color indicates positive reaction, and no color change (colorless) indicates negative reaction. Alternatively, an ONPG disk can be added to suspension and observe for color change [1].

Methyl Red (MR) Test This assay determines if an organism metabolizing glucose utilizes mixed acid fermentation pathway and produces strong acid end products (lactic, acetic, formic) that are detected by the indicator methyl red. A 5 mL MR-VP broth tube is inoculated with the organism and incubated at 35 °C for 48–72 h. 2.5 mL of the broth culture is transferred to a fresh tube and inoculated with five drops of methyl red indicator. Positive MR is indicated if the methyl red reagent remains red. Negative result is indicated if the reagent turns yellow orange [1, 2].

Voges-Proskauer (VP) Test Organisms such as Klebsiella, Enterobacter, Hafnia, and Serratia spp. that utilize the butylene glycol fermentation pathway produce acetoin, an intermediate in the fermentation of butylene glycol. The VP test detects the production of acetoin by these organisms. In the presence of air and potassium hydroxide, acetoin is oxidized to diacetyl, which produces a red-colored complex. The addition of α- naphthol increases the sensitivity of the test. A 5 mL MR-VP broth tube is inoculated with pure culture of organism and incubated at 35 °C for 18–24 h. 2.5 mL of the broth culture is transferred to a fresh tube and inoculated with six drops of α-naphthol followed by three drops of KOH. Shake the tube gently to expose the suspension to oxygen and leave undisturbed for 10–15 min. Positive result is indicated by the presence of red color that develops within 15 min. No color change indicates negative result [1, 2].

54


Pseudosel Agar Slant Pseudosel agar is a medium used for the identification of Pseudomonas aeruginosa. Magnesium chloride and potassium sulfate in the medium enhance the production of pyocyanin, a blue-green, water-soluble, nonfluorescent phenazine pigment. P. aeruginosa is the only Gram-negative rod known to excrete pyocyanin. In addition to the promotion of pyocyanin production, Pseudosel agar also enables the detection of fluorescent products by some Pseudomonas species other than P. aeruginosa. Streak the surface of the Pseudosel agar slant, and incubate at 35 °C in non-CO2 for 18–24 h. A blue-green pigmentation surrounding the growth on the agar slant indicates positive reaction. No pigmentation indicates negative reaction. Negative Pseudosel slants should be examined under short-wavelength (254 nm) ultraviolet light to check for fluorescent products produced by some Pseudomonas species. P. aeruginosa typically produces fluorescein as well as pyocyanin [14].

Urea Agar Slant Microorganisms that possess the enzyme urease are capable of hydrolyzing urea, which releases ammonia. This reaction raises the pH of the medium and is detected by phenol red, which turns pink red when the pH is above 8.0. The color change first appears in the slant since the oxidative decarboxylation of amino acids in the air- exposed portion of the medium enhances the alkaline reaction. The color change eventually spreads deeper into the medium. Stuart’s urea broth and Christensen’s urea agar are the two most common media used in the detection of urease activity. Streak the surface of the urea agar slant with a heavy inoculum of a pure culture. Incubate at 35 °C in non-CO2 for 18–24 h. Production of intense pink-red color on the slant, which may penetrate into the butt, is considered positive reaction. No color change (yellow) indicates negative reaction. The medium is not specific for urease. The utilization of peptones or other proteins in the medium by some urease-negative organisms may raise the pH due to protein hydrolysis and release of amino acid residues, resulting in false-positive reactions [2, 14].

Citrate Agar Slant Some organisms have the ability to utilize citrate, an intermediate metabolite in the Krebs cycle, as the sole external source of carbon. These organisms also utilize inorganic ammonium salts in the medium as the sole source of nitrogen. The resulting production of ammonia creates an alkaline environment that turns the bromothymol blue indicator to an intense blue.


55

Using an inoculating loop, select a well-isolated colony and streak the surface of the Simmons’s citrate slant (do not stab the agar). Incubate at 35–37 °C in non-CO2 incubator and examine daily for up to 7 days. Growth with an intense blue color on the agar slant indicates positive reaction, and no growth and no color change (green) indicate negative reaction. Luxuriant growth on the slant without an accompanying color change may indicate a positive test. This should be confirmed by incubating the tube for an additional 24 h. The biochemical reaction requires oxygen. Therefore, the medium should not be stabbed, and the cap must be kept loose during incubation. Carryover of protein and carbohydrate substrates from previous media may provide additional sources of carbon and therefore cause false-positive reactions [14].

Cetrimide Agar Cetrimide agar is a selective differential medium used for the identification of P. aeruginosa. The principle of the test is to determine the ability of an organism to grow in the presence of cetrimide. Cetrimide acts as a detergent and inhibits the growth of most other organisms. The magnesium chloride and potassium sulfate of the medium stimulate the production of pyocyanin and pyoverdin (fluorescein). Using an inoculating loop, select a well-isolated colony and streak the surface of the cetrimide slant (do not stab the agar). Incubate at 35 °C in non-CO2 incubator and examine daily for up to 7 days. Growth on the agar slant indicates positive reaction, and no growth indicates negative reaction [4, 14].

Gelatin The gelatin test is used to identify bacteria that produce the proteolytic enzyme, gelatinase. Organisms that produce gelatinase are capable of hydrolyzing gelatin and cause it to lose its gelling characteristics. Inoculate several well-isolated colonies deep into the gelatin and repeat to inoculate heavily. Incubate the inoculated tube and an uninoculated control at 35 °C in ambient air. The tubes are then removed daily and incubated at 4 °C to check for liquefaction. Liquefaction is determined only after the control has hardened. Alternatively, strips of exposed but undeveloped X-ray film are placed in the bacterial suspension of equivalent to at least 2.0 McFarland standard, and incubate at 35 °C in a non-CO2 incubator for 48 h. Prepare an uninoculated tube as control. The strip is examined after 24 and 48 h for loss of gelatin coating that leave the X-ray clear [1].

56


Acetate Utilization Some organisms have the ability to utilize acetate as a sole external source of carbon. Acetate slants contain a mixture of salts and sodium acetate in a medium without organic nitrogen. Organisms that can utilize acetate as a sole carbon source break down sodium acetate causing the pH of the medium to shift toward the alkaline range, turning the bromothymol blue indicator blue. Organisms that cannot utilize acetate as a sole carbon source do not grow on the medium. Acetate differential agar is useful in the differentiation of Shigella spp. and Escherichia coli. Streak the surface of the acetate differential agar slant (do not stab the agar), with a colony and cap the tube loosely. Incubate at 35 °C in non-CO2 and examine daily for up to 7 days. Growth with an intense blue color on the agar slant indicates positive test, and no growth or no color change (green) indicates negative test. Luxuriant growth on the slant without an accompanying color change may indicate a positive test. This should be confirmed by incubating the tube for an additional 24 h. The biochemical reaction requires oxygen. Therefore, the medium should not be stabbed, and the cap must be kept loose during incubation. Carryover of protein and carbohydrate substrates from previous media may provide additional sources of carbon and therefore cause false-positive reactions [14].

Lead Acetate for Hydrogen Sulfide Detection Some organisms are capable of enzymatically liberating sulfur from sulfur- containing amino acids or inorganic sulfur compounds. The released hydrogen sulfide reacts with lead acetate to yield lead sulfide, an insoluble black precipitate. Lead acetate is the most sensitive H2S indicator reagent and is useful with organisms that produce trace amounts of H2S, especially organisms that are not in the family Enterobacteriaceae. Inoculate a TSI medium with the isolate (stab once through the center, into the butt of the tube to within 3–5 mm of the bottom, withdraw the inoculating needle, and streak the surface of the TSI agar slant.). Place the lead acetate strip so that it hangs down approximately 1 in. inside the TSI tube. Incubate at 35°C in non-CO2 for 18–24 h. A brownish-black coloration of the paper strip indicates positive reaction. No coloration of the strip indicates negative reaction. Lead acetate is toxic to bacterial growth. Do not allow the strip to touch the medium. The TSI medium must support the growth of the test organism for H2S production to occur [1, 2].


57

Lysine Iron Agar (LIA) Lysine iron agar is a differential medium used for the identification of enteric bacilli based on their ability to decarboxylate or deaminate lysine and produce hydrogen sulfide. Dextrose serves as a source of fermentable carbohydrate. The pH indicator, bromocresol purple, is changed to a yellow color at or below pH 5.2 and is purple at or above pH 6.8. Ferric ammonium citrate and sodium thiosulfate are indicators of hydrogen sulfide formation. Lysine serves as the substrate for detecting the enzymes lysine decarboxylase and lysine deaminase. LIA is designed for use with TSI (triple sugar iron agar) for the identification of enteric pathogens. Using a sterile inoculating needle, stab the butt of the LIA slant twice then streak back and forth along the surface of the agar with the organism. Incubate at 35 °C ± 2 °C in non-CO2 for 18–24 h. Alkaline (purple) reaction in butt indicates lysine decarboxylation; red slant indicates lysine deamination; and black precipitate indicates H2S production. H2S may not be detected in this medium by organisms, which are negative for lysine decarboxylase activity since acid production in the butt may suppress H2S formation. For this reason, H2S-producing Proteus species do not blacken this medium [14].

Triple Sugar Iron Agar Slant (TSI) TSI agar is a medium that differentiates Gram-negative bacilli on the basis of the ability to ferment carbohydrates and liberate H2S. The medium contains one part of glucose to ten parts each of lactose and sucrose. Phenol red serves as an indicator to detect pH change, and ferrous sulfate detects the formation of H2S. If the organism ferments glucose, the butt and slant of the agar will become acidic and turn yellow. If the organism ferments lactose and/or sucrose, the slant will remain acidic (yellow). If the organism is unable to ferment lactose or sucrose, the slant will revert to alkaline (red) when the glucose is used up, and alkaline amines are produced in the oxidative decarboxylation of peptides (derived from protein in the medium) near the surface of the agar. Organisms unable to ferment glucose will not change the pH of the medium or will produce alkaline products, and the TSI tube will remain red. Blackening of the medium indicates H2S production. Gas production is indicated by splits or cracks in the butt of the agar. Gas may also push the agar up the tube. Using a sterile inoculating needle, stab the butt of the TSI slant twice then streak back and forth along the surface of the agar with the organism. Incubate at 35 °C ± 2 °C in non-CO2 for 18–24 h. If acid slant-acid butt (yellow-yellow): glucose and sucrose and/or lactose fermented. If alkaline slant-acid butt (redyellow): glucose fermented only. If alkaline slant-alkaline butt (red-red): glucose not fermented. The presence of black precipitate (butt) indicates hydrogen sulfide production, and the presence of splits or cracks or air bubbles indicates gas production.

58


Early readings may result in false acid-acid results, while delayed readings may result in false alkaline-alkaline results. Copious amounts of H2S may mask the glucose reaction. If this occurs, glucose has been fermented even if it is not observable. The utilization of sucrose may suppress the enzyme mechanism that results in the production of H2S. Trace amounts of H2S may not be detectable with the ferrous sulfate indicator in the agar [2, 14].

Phenylalanine Deaminase This assay is used to detect the ability of an organism to oxidatively deaminate phenylalanine to phenylpyruvic acid. The phenylpyruvic acid is detected by adding a few drops of 10% ferric chloride. Inoculate a phenylalanine agar slant with the organism, and incubate at 35 °C in non-CO2 incubator for 18–24 h with the cap loose. Following the incubation, add 4–5 drops of 10% ferric chloride solution to the slant. The development of green color on the surface of the slant indicates positive reaction. No color change indicates negative reaction [1, 4].

Decarboxylase (Moeller’s Method) Decarboxylases are a group of substrate-specific enzymes that are capable of decarboxylate (or hydrolyze) amino acids to form amines produces alkaline pH. Each decarboxylase enzyme is specific for an amino acid. Lysine, ornithine, and arginine are the three amino acids used routinely in the identification of Enterobacteriaceae, Aeromonas, Plesiomanas, and Vibrio species. The decarboxylation of lysine and ornithine yield cadaverine and putrescine, respectively. Arginine is converted to citrulline by a dihydrolase reaction. A control tube containing the base without an added amino acid to verify that the organism utilizes glucose must accompany all decarboxylase tests. Since decarboxylation is an anaerobic reaction, the tubes must be overlaid with mineral oil prior to incubation. If the organism is viable, both the control and the test tube with amino acid should turn yellow because of fermentation of the small amount of glucose in the medium. If the amino acid is decarboxylated, the alkaline amines cause the indicator (bromocresol purple) in the acid medium to revert back to its original purple color. Inoculate a Moeller decarboxylase broth containing ornithine, lysine, and/or arginine. Overlay the contents of all tubes with 1 mL of sterile mineral oil, and incubate in a non-CO2 incubator at 35°C for 18–24 h. Examine for a color change. Negative reactions are examined daily for no more than 4 days [14].


59

OF Glucose Medium Bacteria can utilize glucose and other carbohydrates by using various metabolic cascades. Some are fermentative routes; others are oxidative. Oxidation-fermentation (OF) medium permits classification of Gram-negative bacilli by a simple method that differentiates aerobic and anaerobic degradation of carbohydrates. The low protein to carbohydrate (P/C) ratio in the medium prevents the neutralization of weak acids by the alkaline products if the protein is utilized, thus allowing small quantities of acid to be detected. Acid production results in a pH shift that changes the color of the bromothymol blue indicator from green to yellow. Using inoculating needle, two tubes of OF glucose medium are stabbed, inoculating halfway to the bottom of the tubes. The content of one tube is overlaid with 1 mL of sterile mineral oil. Both tubes are incubated at 35 °C in non-CO2, and examine daily for 72 h or longer for slow-growing organisms. Yellow color indicates the production of acid. Acid production in tube without oil overlay is considered oxidative reaction. Acid production in both tubes is considered fermentative. No acid production in either tube is considered non-saccharolytic. Nonsaccharolytic organisms produce slight alkalinity (blue-green color) in the tube without oil overlay, but the tube with oil will not exhibit a color change and will remain green [14].

OF Sugars OF basal medium, when supplemented with an appropriate carbohydrate, is used to determine an organism’s ability to utilize sugars such as lactose, xylose, sucrose, maltose, or mannitol. The low protein to carbohydrate (P/C) ratio in OF basal medium prevents the neutralization of small quantities of weak acids by the alkaline products of protein metabolism, which makes this medium ideal for determining carbohydrate utilization. Acid production from carbohydrate metabolism results in a pH shift that changes the color of the bromothymol blue indicator from green to yellow. Yellow color indicates carbohydrate metabolism. Using an inoculating needle, touch the center of one colony and stab the OF medium with the appropriate carbohydrate once halfway to the bottom of the tube. Cap the tubes loosely and incubate at 35°C in non-CO2, examining the tube daily for 72 h or longer for slow-growing organisms. A yellow color indicates carbohydrate utilization, and no color change (green) or blue color indicates no carbohydrate utilization. The acid reaction produced by oxidative organisms is detected first at the surface and gradually extends throughout the medium. When oxidation is weak or slow, it is common to observe an initial alkaline reaction at the surface of the tube that may persist for several days. This must not be mistaken for a negative test. If the organism is unable to grow in the OF medium, add either 2% serum or 0.1% yeast extract prior to inoculation [14].

60 Table 1 Commercial systems commonly used in clinical laboratories

S. Hafeez and J. Aslanzadeh Product API systems BBL Crystal systems BBL Phoenix systems Vitek MicroScan MIdI Sherlock Sensititre AP80 Biolog micro plate

Manufacturer bioMerieux Inc. Becton Dickinson Becton Dickinson bioMerieux Inc. Dade international MIDI Trek Biolog

Turnaround time 2 h to over night 4 h to over night 2 h to over night 2 h to over night 2 h to over night Over night 5 h to over night 2 h to over night

Commercial Microbial Identification Systems The commercial microbial identification systems are the backbone of microbial identification in most clinical microbiology laboratories. These identification systems provide an advantage over conventional identification systems by requiring little storage space, having an extended shelf life, rapid turnaround, low cost, standardized quality control, and ease of use. They range from manual to semiautomated to fully automated systems. These systems require simultaneous inoculation and incubation of a series of miniaturized biochemical reactions which are either based on detecting bacterial enzymes or cellular products that do not require microbial growth and have fairly rapid turnaround time (2–4 h) or are based on metabolic activity that requires microbial growth and require several hours to overnight incubation. In either case, the enzymatic or biochemical end results are combined, and using the Bayer’s theorem with the aid of a computer program, the identity of the test organism is determined. The majority of metabolic based automated commercial identification systems also incorporate antimicrobial susceptibilities testing. In fact, over the years, the growing numbers of clinically significant pathogens and their rapidly emerging resistance to various antimicrobial agents have led to innovation of several commercial identification (ID) and antimicrobial susceptibility testing (AST) systems. For the most part, these systems have a fairly extensive database. They are fast accurate and have significantly improved turnaround time for ID and AST of the common organisms. Despite their extensive database, they remain less than optimal in identifying fastidious slow-growing esoteric organisms. Table 1 presents the list of most commonly used commercial identification systems.

API Identification System The API identification systems (bioMerieux Inc. Hazelwood MO) consist of series of microcapsules on a plastic strip that contains dehydrated substrates for the demonstration of enzymatic activity or the fermentation of carbohydrates. Depending on type of the organism and the API strip utilized, it may or may not require microbial growth. API systems are manual and do not incorporate AST [15, 16].


61

API Gram-Negative Identification 1. API 20E is an 18 to 24-h identification test for identification of Enterobacteriaceae and group/species of non-fermenting Gram-negative rods. 2. API Rapid 20E is a 4-h identification test for identification of Enterobacteriaceae. 3. API 20NE is a 24–48 h identification test for identification of Gram-negative non-Enterobacteriaceae. 4. API NH is a 2-h test for identification of Neisseria, Haemophilus, and Moraxella. API Gram-Positive Identification 5. API Staph is an overnight test for identification of clinical staphylococci and micrococci. 6. RAPIDEC Staph is a 2-h identification of the commonly occurring staphylococci. 7. API 20 Strep is a 4- or 24-h test for identification of streptococci and enterococci. 8. API Coryne is a 24-h test for identification of corynebacteria and coryne-like organisms. API Anaerobe Identification 9. API 20A is a 24-h test for identification of anaerobic organisms. 1 0. Rapid ID 32 is a 4-h test for identification of anaerobes. API Yeast Identification 1. API 20C AUX is a 48 to 72-h test for identification of yeasts.

BBL™ Crystal™ Identification System The BBL™ Crystal™ System (Becton Dickinson, Cockeysville, MD) is a manual method that utilizes miniaturized fluorogenic and or chromogenic linked substrates to detect enzymes that microbes use to metabolize a variety of substrates. These kits consist of BBL Crystal panel lids, bases, and inoculum fluid tubes. A suspension of the test organism is prepared in the inoculum fluid and then used to fill the reaction wells in the base. The substrates are rehydrated when the base and lid are aligned and snapped into place. Following the recommended incubation time, the wells are manually examined for color changes or the presence of fluorescence. The resulting pattern of positive and negative test scores is the basis for identification [16, 17]. 1. BBL™ Crystal™ Enteric/Nonfermenter (E/NF) Identification System is an overnight identification method utilizing modified conventional and chromogenic substrates. The E/NF identifies clinically significant aerobic Gram-negative Enterobacteriaceae isolates and non-fermenting Gram-negative rod.

62


2. The BBL™ Crystal™ Rapid Stool/Enteric (RS/E) Identification System is a miniaturized 3-h identification method employing modified conventional and chromogenic substrates. It is intended for the identification of clinically significant aerobic Gram-negative bacteria that belong to the family Enterobacteriaceae as well as most pathogens isolated from stool specimens. 3. The BBL™ Crystal™ Neisseria/Haemophilus (N/H) Identification System is a miniaturized 4-h identification method employing modified conventional, fluorogenic, and chromogenic substrates. It is intended for the identification of Neisseria, Haemophilus, Moraxella, Gardnerella vaginalis, as well as other fastidious bacteria. 4. The BBL™ Crystal™ Gram-Positive ID System is a miniaturized 18-h identification method employing modified conventional, fluorogenic, and chromogenic substrates. It is intended for the identification of both Gram-positive cocci and bacilli. 5. The BBL™ Crystal™ Rapid Gram-Positive ID System is a miniaturized 4-h identification method employing modified conventional, fluorogenic, and chromogenic substrates. It is intended for the identification of Gram-positive bacteria isolated from clinical specimens. 6. The BBL™ Crystal™ Anaerobe ID kit is a miniaturized 4-h identification method employing modified conventional, fluorogenic, and chromogenic substrates to identify clinically significant anaerobic organisms.

BBL Phoenix Identification and Susceptibility System The BBL Phoenix™ (Becton Dickinson, Cockeysville, MD) is an automated identification and susceptibility system that can identify clinically significant Gram- negative or Gram-positive microorganisms. The Phoenix ID panel utilizes a series of conventional, chromogenic, and fluorogenic biochemical tests to determine the identification of the organism. Both growth-based and enzymatic substrates are employed to cover the different types of reactivity. The tests are based on microbial utilization and degradation of specific substrates detected by various indicator systems. Acid production is indicated by a change in phenol red indicator when an isolate is able to utilize a carbohydrate substrate. Chromogenic substrates produce a yellow color upon enzymatic hydrolysis of either p-nitrophenyl or p-nitroanilide compounds. Enzymatic hydrolysis of fluorogenic substrates results in the release of a fluorescent coumarin derivative. Organisms that utilize a specific carbon source reduce the resazurin-based indicator. Currently, there are three types of BD Phoenix™ Identification panels: BD Phoenix™ NID panel, BD Phoenix PID panel, and BD Phoenix yeast ID panel. The AST method is a broth-based microdilution test. The system utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. Continuous measurements of changes to the indicator as well as bacterial turbidity are used in the determination of bacterial growth. Each AST panel configu-


63

ration contains several antimicrobial agents with a wide range of twofold doubling dilution concentrations. Organism identification is used in the interpretation of the MIC values of each antimicrobial agent resulting in susceptible, intermediate, or resistant (S, I, R) result classification. The system includes an inoculation station for panel setup and an incubator/ reader carousel module. The carousel houses 4 horizontal tiers of 26 panel carriers to accommodate a tier-specific Normalizer and 25 Phoenix Panels. Phoenix Panel utilizes up to 51 micro-wells for ID and up to 85 micro-wells for AST. A bacterial inoculum concentration approximately equivalent to a 0.5 McFarland Standard is required for the identification of either Gram-negative or Gram-positive bacteria. Susceptibility testing is performed with an inoculum concentration of 3–7 × 105 cfu/ mL Kinetic measurements of bio-reactivity within individual micro-wells via red, green, blue, and fluorescence readings are collected and comparatively analyzed with the Phoenix database [17]. Currently, seven BD Phoenix™ Susceptibility panels are known: BD Phoenix Emerge™ NMIC-300 panel, BD Phoenix™ NMIC-303 panel, BD Phoenix NMIC- 304 panel, BD Phoenix PMIC-106 panel, BD Phoenix PMIC-107 panel, BD Phoenix PMIC-108 panel, BD Phoenix PMIC-109 panel, and BD Phoenix SMIC- 101 panel.

VITEK and VITEK 2 Identification System The Vitek (bioMerieux Inc. Hazelwood MO) is an automated ID and AST system that utilizes identification cards with miniaturized wells. The system is fairly automated. It requires the user to prepare a suspension of the isolate in saline and verify the organism concentration with a densitometer. The inoculum tube is then placed into a rack called the cassette. The sample identification number is entered into the carrier via barcode or keypad and electronically linked to the supplied barcode on each test card. ID and AST test cards can be mixed and matched in the cassette. All information entered at the bench is then transported to the instrument in a memory chip attached to the cassette. VITEK 2 is the fully automated version that all processing steps are completely autonomous including test setup verification, AST inoculum dilution test inoculation, card sealing, incubator loading, optical reading and data transmission, and card disposal. The VITEK 2 optical system reads all the wells every 15 min. There are several card that are designed for ID and susceptibility testing with these systems including Vitek GPI (Vitek 1), Vitek GPC (Vitek 2), Vitek EPS, GNI Plus, UID and UID, Vitek 2 ID-GNB, and Vitek NHI and AST panels for Gram-positive and Gram- negative organism [15].

64


MicroScan WalkAway The MicroScan WalkAway (Beckman Coulter Inc. Brea CA) is an automated ID and AST system that requires the ID and/or AST panels (96 well plates) be manually inoculated with bacteria isolated from clinical specimens and inserted into the WalkAway System. The panels are then incubated at 35 °C for 16–42 h, depending on panel, organism type, and results of readings. At the appropriate time, the WalkAway System automatically dispenses reagents into the appropriate biochemical wells and incubates the panels for an additional period of time (approximately 2–20 min, depending on the panel type). The WalkAway System then reads the panels. The identification of bacteria is based on measuring a series of biochemical test contained in panels designed for the speciation of most medically significant bacteria. The panels contain identification media consisting of substrates and/or growth inhibitors, which, depending on the species of the bacteria present, will exhibit color changes or increases in turbidity after incubation. The panel may also contain series of antibiotic that are present in specified concentrations in the wells of applicable MicroScan panels. The WalkAway System reads the minimum inhibitory concentrations (MIC) and certain biochemicals, and if the criteria are met for adding reagents, reagents are added. The panel is then incubated for an additional period of time (approximately 5–30 min) depending on the panel type. The readings for the biochemicals needing no reagents and MIC wells (for Combo panels) are stored prior to reagent addition. If additional incubation is necessary for the biochemicals, the susceptibilities and certain biochemical tests will be read first and stored. The reagents will not be added until after additional incubation, at which time biochemical tests not previously read will be determined. The following is the list of commonly used MicroScan panels: MicroScan Gram Pos ID panel, MicroScan Rapid Gram Pos ID panel, MicroScan Neg Type 2, MicroScan Rapid Neg ID Types 2 and 3, and MicroScan NHID [18].

Sensititre® Microbiology Systems The Sensititre ARIS 2X (Thermo Scientific Inc. TREK Diagnostic Systems, Oakwood Village, OH) is an automated ID and AST system. The Sensititre ID and AST panels (96- well plates) may be inoculated manually or by autoinoculation, which is designed to automatically deliver inoculum in multiples of 50 μl to the 96-well Sensititre plate. The Sensititre ID system is based on 32 biochemical tests pre-dosed and dried in the Sensititre plate that are formulated to allow fluorometric reading along with unique fluorescent tests. The AST plate may be read manually or using the automated system. The automated system is fluorescent based and detect bacterial growth by monitoring the activity of specific surface enzyme produced by the test organism. Growth is determined by generating a fluorescent product from a nonfluorescent substrate. Presumptive ID of Gram-negative organisms can be


65

obtained in 5 h; identification to species level for both Gram-negatives and Gram- positives can be obtained after overnight incubation. The Sensititre ARIS 2X is a combined incubation and reading system that fits onto an autoreader. Sensititre uses an internal barcode scanner to identify each plate type, assign the appropriate incubation time, and when this assigned time has elapsed, transport the plate to the autoreader for fluorescence measurement. The system has capacity to accommodate up to 64 ID or AST plates. The following is the list of ID and AST plates with this system: GNID (AP80) for Gram negative, GPID for Gram-positive Identification, Gram-positive and Gram-negative MIC plates, Sensititre Haemophilus influenzae or Streptococcus pneumoniae susceptibility Plates, Anaerobe MIC Plate, EBSL Confirmatory MIC plate, and S. pneumoniae MIC plate [16, 19].

MIDI Sherlock The MIDI Sherlock ID system (MIDI, Inc. Newark, DE) is based on Gas Chromatographic (GC) analysis of the bacterial fatty acids. Branched-chain acids are known to predominate in most Gram-positive bacteria, while short-chain hydroxy acids often characterize the lipopolysaccharides of the Gram-negative organisms. The system is fairly labor intensive and is designed for use in reference laboratories to identify isolates that are not easily identified by the routine identification systems. The Sherlock system detects the presence or the absence of more than 300 fatty acids and related compounds (9–20 carbons in length) as well as quantity of these compounds. The peaks are automatically named and quantified by the system. Initially, the organism undergoes saponification, methylation, extraction, and base wash before GC analysis. A GC with phenyl methyl silicone fused silica capillary column is injected with the final prep, and the temperature program in GC ramps the temperature from 170 °C to 270 °C at 5 °C per minute. Following the analysis, a ballistic increase to 300 °C allows cleaning of the column. The electronic signal from the GC detector is then passed to the computer where the integration of peaks is performed. The electronic data is stored on the hard disk, and the fatty acid methyl ester composition of the sample is compared to a stored database using the Sherlock pattern recognition software [16, 20].

BioLog ID System The Biolog Micro Plate ID Systems (Biolog, Inc. Hayward CA) relies on carbon source utilization test methodology in a 96-well format. The system is based on 95 reactions from 6 to 8 different classes of carbon sources with redox indicator (tetrazolium dye) and 1 negative control well with no carbon source. The isolate is inoculated to the micro-well plate and incubated. If the isolate oxidizes any of

66


the carbon sources, the net electron will reduce the tetrazolium to highly colored formazin (purple color). The carbon source utilization produces a characteristic pattern or “fingerprint” that are then compared to the Biolog database for identification. The system can identify environmental as well as fastidious organisms. In addition to the original Microlog manual and a semiautomated version, the manufacturer has recently introduced a fully automated version (OmniLog) with a database to identify over 700 species of Gram-positive and Gram-negative organisms [16, 21]. The new OmniLog Plus ID System with a database of over 2650 species adds testing capabilities for anaerobic bacteria, yeasts, and filamentous fungi.

References 1. Jorgensen JH, Pfaller MA, et al., editors. Manual of clinical microbiology. 11th ed. Washington, DC: American Society for Microbiology; 2015. 2. Procop GW, Church DL, Hall GS, et al., editors. Koneman’s color atlas and textbook of diagnostic microbiology. 6th ed. Philadelphia: J.B. Lippincott Co; 2017. 3. Oxidase Reagent Package Insert, Becton Dickinson Microbiology Systems, Cockeysville. Date of issue June 2010. 4. Forbes BA, Sahm DF, Weissfeld AS, editors. Bailey and Scott’s diagnostic microbiology. 14th ed. St. Louis: Mosby; 2016. 5. Isenberg HD. Clinical microbiology procedures handbook. 3rd ed. Washington, DC: American Society for Microbiology; 2007. 6. York M, Traylor MM, Hardy J, Henry M. Biochemical tests for the identification of aerobic bacteria: bile solubility tests. In: Isenberg HD, editor. Clinical microbiology procedures handbook. 2nd ed. Washington, DC: American Society for Microbiology; 2004. p. 3.17.6. 7. Coleman G, Ball LC. Identification of streptococci in the medical laboratory. J Appl Microbiol. 1984;57:1–14. 8. D’Amato RF, et al. Rapid identification of Neisseria gonorrhoeae and Neisseria meningitidis by using enzymatic profiles. J Clin Microbiol. 1978;7:77–81. 9. Thermo Scieentific, Remel™ BactiCard™ Neisseria. www.thermofisher.com. 2017. 10. Weyant RS, Moss CW, Weaver RE, et al., editors. Identification of unusual pathogenic gram- negative aerobic and facultatively anaerobic bacteria. 2nd ed. Baltimore: Williams & Wilkins Co; 1996. 11. Wilkinson HW. Camp-disk test for presumptive identification of group B streptococci. J Clin Microbiol. 1977;6:42–5. 12. Killian M. A rapid method of differentiation of Haemophilus strains-the porphyrin test. Acta Pathol Microbiol Scand B. 1974;82:835–42. 13. Motility Indole Lysine (MILS) Package Insert 88-0655-1. BD Microbiology Systems, Cockeysville, January 1999. 14. Quality Control and Product Information Manual for Tubed Media, BD Microbiology Systems, Cockeysville. 2009. 15. bioMerieux Inc. http://www.biomerieux-usa.com. 2017. 16. Traunt AL, editor. Manual of commercial methods in clinical microbiology. Washington, DC: American Society for Microbiology; 2002.


67

1 7. Becton Dickinson Inc. http://www.bd.com. 2017. 18. Beckman Coulter MicroScan Dried Gram Negative and Dried Gram Positive Procedure Manual package insert. April 2016. 19. Thermo Scientific TREK Diagnostic Systems. http://www.trekds.com. 2017. 20. MIDI Inc. http://www.midi-inc.com. 2017. 21. Biolog, Inc. http://www.biolog.com. 2017. 22. CLO Test Package Insert. Delta West Ply Ltd., Bentley, Western Australia, 2001.

Advanced Phenotypic Antimicrobial Susceptibility Testing Methods Charles W. Stratton

Introduction The goal of antimicrobial therapy, as first appreciated and described by Joseph Lister and Paul Ehrlich, is to destroy the microorganism without harming the host. Lister in the 1860s was the first scientist to investigate the inhibitory effect of chemicals on bacteria and to directly apply this knowledge to the practice of medicine by using phenol to sterilize surgical instruments [1]. Ehrlich also recognized the interaction of chemical agents with microorganisms and in 1910 reported the 606th arsenic compound tested to be active against the treponemal cause of syphilis [2]. The discovery of penicillin by Fleming in 1929 [3] and of the azo dye prontosil rubrum by Domagk in 1932 [4] led to the introduction of the penicillin and sulfonamide classes of antibiotics as well as the dawn of the antibiotic era in 1935 [5].

istory and Evolution of Phenotypic Antimicrobial H Susceptibility Testing Methods The effectiveness of an antimicrobial agent has historically been measured by its ability to inhibit and/or kill microorganisms; this ability is usually assessed using phenotypic antimicrobial susceptibility testing (AST) methods [6–9]. Phenotypic AST remains the gold standard for clinical microbiology laboratories [8] although supplemental genotypic AST methods are increasingly being used for syndromic testing [10]. Theoretically, there are three basic ways to kill a bacterial cell; these are causing irreversible damage to the genome, to the envelope, and/or to critical C. W. Stratton (*) Department of Pathology, Microbiology and Immunology, Vanderbilt University School of Medicine, Nashville, TN, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_4

69

70 Table 1 Key factors that influence the outcome of antimicrobial therapy

C. W. Stratton Host immune system Site of infection Pharmacokinetics Microbial pathogen Pharmacodynamics Timing for initiation of appropriate antimicrobial therapy

proteins [9, 11]. Each of these bacterial killing mechanisms has been found to work best on bacteria during their growth phase [9, 12–14]. Phenotypic AST methods therefore involve actively replicating microorganisms, even though this may not be the usual state of microorganisms in infected tissues [15]. Table 1 briefly lists the key factors that influence the outcome of antimicrobial therapy; these are reviewed in detail in the cited reference. Of particular importance are the host immune system and the timing for the initiation of appropriate antimicrobial therapy. Clearly the drug must reach the infected tissues and must be active against the infecting microorganism. Suffice it to say that the results of susceptibility testing are an in vitro approximation of the likely outcome of antimicrobial therapy using a specific antimicrobial agent against a specific microbial pathogen [16]. This approximation has greatly improved due to the following improvements in phenotypic antimicrobial susceptibility testing methods that have occurred over the past 50 years as these AST methods have evolved. The first of these improvements involves the standardization of phenotypic AST methods in both the Unities States [17–23] and Europe [24–33]. Of particular importance in this standardization process has been the standardization and control of the inoculum. This standardization process is one of the major advantages of current phenotypic AST methods. The second and equally important improvement is the integration of pharmacokinetics and pharmacodynamics with antimicrobial susceptibility testing [34–49]. Pharmacokinetics describes the basic processes of absorption, distribution, metabolism, and elimination of the antimicrobial agent administered to a patient; these factors determine the ability of the agent to reach the infected tissue site [15, 36, 43]. Pharmacokinetics, in other words, is the effect of the drug on the patient. Pharmacodynamics is the effect of the antimicrobial agent on the microorganism: in other words, the effect of the drug on the bug [15, 41, 43]. An important aspect of pharmacodynamics that has been recognized as critical for reducing the emergence of resistance is bactericidal activity [50–54]. Bactericidal activity can be further characterized as time-dependent, concentration-dependent, or some combination of these two types; bactericidal indices can be evaluated in vivo, in vitro, and using population modeling that employs Monte Carlo simulations, which are computer algorithms in which repeated random sampling is done to obtain a probability distribution in order to predict the probability of target attainment [49, 55]. The integration of PK-PD with antimicrobial susceptibility testing allows rational and optimal antimicrobial therapy while minimizing the emergence of resistance [56–60]. Along with standardization, this integration is among the key advantages of phenotypic antimicrobial susceptibility testing. Another advantage of

Advanced Phenotypic Antimicrobial Susceptibility Testing Methods

71

phenotypic AST is that such testing allows quantification of the level of susceptibility of a bacterial isolate to an individual antimicrobial agent. Low cost and being easy to perform are two additional advantages of traditional phenotypic antimicrobial susceptibility testing methods. The major disadvantage of traditional phenotypic antimicrobial susceptibility testing methods is the requirement for the bacterial isolate to be in pure culture, which may take several days to achieve in the clinical microbiology laboratory and then 3–4 h for the inoculum to achieve logarithmic growth before being exposed to the antimicrobial agents being tested. Once the inoculum is ready, there is an additional 18–24 h of exposure before the results are available. The total time necessary for a susceptibility testing result thus ranges from as little as 2 or 3 days to as many as 4 or 5 days. This delay clearly has consequences in terms of patient management and requires broad-spectrum empiric antimicrobial therapy followed by de- escalation of antimicrobial therapy when susceptibility results are available. Antimicrobial stewardship programs are confronted with this limitation of traditional phenotypic AST methods. Not surprisingly, rapid phenotypic and genotypic antimicrobial susceptibility test methods are being developed in order to provide more timely patient management as well as to prevent overuse of potent broad- spectrum antimicrobial agents. Direct detection of the pathogen in the clinical sample along with direct detection of resistance mechanisms using genomic molecular methods is the most rapid approach as it bypasses culture methods entirely. It is important at this point to understand the difference between rapid phenotypic AST methods and direct AST methods. Direct AST methods mean that the clinical specimens serve as the “inoculum”; this can be problematic if the specimens involve multiple microbial pathogens. In contrast, rapid phenotypic AST methods mean rapid methods for the detection of growth from pure cultures with defined inoculums. Direct AST methods usually involve genomic molecular methods although urine cultures have been used with direct AST methods with the assumption that infected urine specimens have a large inoculum with one bacterial pathogen such as Escherichia coli. Genomic molecular methods are reviewed in other chapters. Instead, this chapter will focus on rapid phenotypic methods for antimicrobial susceptibility testing and will include innovative/alternate rapid phenotypic antimicrobial susceptibility testing methods.

cceleration of Traditional Phenotypic Antimicrobial A Susceptibility Testing Methods Traditional phenotypic AST methods that include disk diffusion, antibiotic gradient diffusion, agar dilution, broth dilution, and broth microdilution involve continuous exposure of a bacterial isolate to a set of antimicrobial agents followed by visual determination of growth after 18–24 h of incubation [7, 8]. Acceleration of these traditional AST methods generally has focused on increasing the sensitivity of growth detection [61–66]; these methods often involve semiautomatic instruments that use

72 Table 2 Traditional phenotypic antimicrobial susceptibility testing methods

C. W. Stratton Broth dilution Broth microdilution Agar dilution Disk diffusion Gradient diffusion Semiautomatic instruments with optical detection systems Breakpoint test methods (agar or semiautomatic instruments testing two concentrations)

optical systems for growth detection. Some of these semiautomatic instruments do “breakpoint testing,” in which there are two concentrations tested as opposed to serial dilutions involving eight to ten concentrations tested. Table 2 lists these traditional phenotypic AST methods. Semiautomated instruments such as the Thermo Fisher Sensititre System, Beckman Coulter MicroScan, Biomerieux Vitek, and BD Phoenix use optical systems to measure subtle changes in bacterial growth in order to produce susceptibility test results in 6–12 h rather than 18–24 h [61, 62, 67–75]. The basis for these growth-dependent automated instruments is the broth microdilution antimicrobial susceptibility method [61, 62, 68], which has replaced the agar dilution method as the reference standard for which all AST methods are currently compared during development, verification, validation, and clinical trials [76]. These automated AST instruments are precise, are reliable, and provide quantitative results; moreover, they have matured into state-of-the-art technologies and are the predominant methods used in clinical microbiology laboratories today [66]. They do require the use of isolated bacteria in pure cultures but can provide results within 6–12 h. Another approach in which a traditional AST method has been accelerated to provide more rapid results involves the disk diffusion AST method [77] coupled with an automated video image zone reader [78–81]. The disk diffusion method uses solid agar media that allows visualization of the bacterial growth on agar plates and represents an alternative to broth microdilution; this method usually requires 16–20 h before reading the zone diameters and has the advantage of being a widely used low-cost method [8, 77]. The use of instrumentation for reading zone sizes as well as for reading broth microdilution MIC endpoints has evolved in parallel for both AST methods [68, 70, 81, 82], although the ability to automate both identification and AST has led to the predominance of broth microdilution AST methods [61, 62]. High-resolution scanners now allow disk diffusion zone sized to be read at 8–10 h rather than 16–20 h [78, 79, 81]. The overall frequency of discrepant results in a recent study using the Scan 1200 [81] was 1.3%, which is within current US FDA guidelines [8]. The disk diffusion method can also be accelerated by using broth from positive blood cultures as the inoculum; this method has been termed the direct disk diffusion or direct from blood culture disk diffusion method [83–85]. Results can be read at 16–18 h or can be read earlier at 6 h; reading at 6 h further accelerates this direct disk diffusion method [86]. A recently reported study compared this method with


73

reference disk diffusion results for 20 challenge isolates of Enterobacteriaceae, Acinetobacter baumannii, and Pseudomonas aeruginosa isolated seeded into human blood [87]. Categorical agreement values for this direct disk diffusion method read at 16–18 h were 87.8%, 88.4%, and 92.2% for the BacT/Alert, Bactec, and VersaTREK systems, respectively. No very major errors were observed, and major error rates were 3.0%, 2.3%, and 1.7%, respectively. Reading at 6 hours resulted in 19.9% of tests having too light of growth to allow reading of zones of inhibition; for those tests that could be read, the categorical agreement was 58.9%, 76.6%, and 73.2%, respectively. For the 6-h reading, very major error rates were 2.2%, 1.8%, and 3.0%, respectively; major error rates were 25.4%, 6.1%, and 2.8%, respectively. The best performance was noted for blood cultures with bacterial concentrations in the range of 7.6 × 107 to 5.0 × 108 CFU/ml; categorical agreement values ranged from 94.7% to 96.2% for these concentrations read at 18 h of incubation and from 76.9% to 84.1% for these concentrations read at 6 h of incubation. These results demonstrate the potential accuracy of such direct disk diffusion methods but also confirm that a shorter 6-h reading time is more problematic than the 18-h reading time. Laboratory automation in the clinical microbiology laboratory [88–90] provides yet another option for acceleration of the traditional disk diffusion AST method. The two commercially available systems for automation in the clinical microbiology laboratory are BD Kiestra and Copan [89]; both systems incorporate digital plate reading [91] to facilitate the analysis of positive cultures and improve the quality and efficiency within the clinical microbiology laboratory. The use of these digital plate readers allows the traditional disk diffusion plates to be read after 6–12 h of incubation rather than at 18 h [92, 93]. This approach could possibly be used with the E-test [94] when MICs rather than zone sizes were desired. Another example of an accelerated approach to phenotypic susceptibility testing methods is the Accelerate PhenoTest™ for blood cultures; this system has been approved for use in the clinical microbiology laboratory by the FDA [95–97]. The Accelerate PhenoTest™ BC kit is used with the Accelerate Pheno™ system to provide fast microbial identification (ID) and antimicrobial susceptibility testing (AST) of appropriate detected bacterial targets directly from positive blood cultures using fully automated sample preparation, digital microscopy, and image analysis. Fluorescence in situ hybridization (FISH) provides ID results within approximately 1.5 h, while morphokinetic cellular analysis provides phenotypic AST results after 6.5 h. The Accelerate PhenoTest™ BC kit consists of a sample vial, cassette, and reagent cartridge. The cassette contains 48 test channels for performing ID and AST assays, while the reagent cartridge contains gel electrofiltration (GEF) wells, FISH probes, antibiotics, and other reagents (Romney Humphries, personal communication). After the operator loads the sample vial, cassette, and reagent cartridge and starts the run, all operations are automated within the system. Sample preparation by GEF uses electrochemistry and a proprietary gel formulation to separate bacterial and fungal cells from other sample debris (such as lysed blood cells). A fluorescence and dark-field microscope with camera captures emissions from the FISH ID probes in

74

C. W. Stratton

each test channel. Co-localized signals from the target probes (green fluorescently labeled DNA probes that bind to the ribosomal RNA of target organisms) and universal bacterial or eukaryotic probes (red fluorescently labeled probes that bind to the DNA in bacteria or yeast cells, respectively) confirm the presence and identity of the target organism and distinguish it from non-specific staining. The ID result determines the antibiotics that are used for subsequent AST. Prior to AST, a portion of the sample combined with growth media is incubated during the FISH step to allow for normalization of growth rates. Cells then undergo dilution to the appropriate concentration range for AST and are immobilized in test channels containing a single concentration of test antibiotics. Images of each test channel are captured every 10 min for up to 4.5 h to create a record of bacterial growth. The system uses morphokinetic cellular analysis to measure, track, and analyze changes to morphological and kinetic features of bacterial cells (e.g., cell morphology, mass, division rate, anomalous growth patterns, heterogeneity) as these microbial cells respond to the antibiotics over time. Software algorithms derive MIC values and report categorical interpretations (susceptible, intermediate, or resistant). A key factor for reducing the time required for the Accelerate PhenoTest™ method is that this system, like the direct disk diffusion method [83–87], uses the positive blood culture medium as the inoculum for a phenotypic broth microdilution susceptibility test; digital microscopy with morphokinetic cellular analysis to confirm that the growth does not involve more than one microorganism as well as to derive an MIC value. As pointed out by Dr. Christopher Doern in an insightful commentary [98]: “The prospect of shortening growth-based AST is, biologically, a difficult task. Most, if not all antibiotics exert their activity by targeting some aspect of organism viability; such as cell-wall assembly, protein synthesis, replication of nucleic acids, or metabolic pathways. Therefore, in order for AST to assess the activity of an antibiotic, the organism in question must be actively replicating. One of the most fundamental elements of bacteriology is the three phases of an organism growth curve; lag phase, logarithmic (or exponential) phase, and stationary phase. Unfortunately, traditional AST methods require that an organism transition from lag phase to an actively growing phase, or logarithmic growth, in order to fully assess antimicrobial susceptibility. Duration of lag phase growth varies, but can take several hours for many organisms. This reality poses a significant challenge to those growth-based technologies which are attempting to shorten the turnaround time for phenotypic AST results.”

ear-Future Novel Approaches to Rapid Phenotypic N Susceptibility Testing Methods Laboratory automation in the clinical microbiology laboratory [88–90] generally includes matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) as part of the technologically novel solutions used to cope


75

with increasing numbers of clinical specimens versus limited personnel and limited financial resources [99, 100]. MALDI-TOF MS, in particular, has been widely embraced by clinical microbiology laboratories due to its being an inexpensive, easy, rapid, and accurate method for identification of grown bacteria and fungi based on automated analysis of the mass distribution of bacterial proteins [101– 104]. Because of its widespread use for microbial identification in clinical microbiology laboratories, it is reasonable to ask what novel role MALDI-TOF MS might have in the near future for antimicrobial susceptibility testing. One obvious novel approach is to start with positive blood cultures, which allows the growth in the blood culture to provide the inoculum; this inoculum can be used for MALDI-TOF as well as for a separate accelerated AST method such as Vitek 2 [105]. This method does require either an in-house (inexpensive, but labor-intensive) or commercial (expensive, but labor-saving) extraction method [101] such as the Bruker Sepsityper kit [106, 107]. This method can also be combined with a genotypic AST probe method such as the Verigene assay [108]. Other novel approaches for using MALDI-TOF MS for detection of antimicrobial resistance mechanisms are being evaluated and will be briefly reviewed [104]. Detection of beta-lactamase activity has been evaluated using MALDI-TOF MS to detect degradation products of a beta-lactam agent when exposed to the microorganism for a brief period of time [110–114]. The detection of beta-lactamase is done by using a fresh overnight bacterial culture, which is suspended in a buffer and centrifuged [109]. The pellet is then resuspended in a buffer containing the beta- lactam agent; after incubation at 35°C for 1–3 h, the reaction mixture is centrifuged, and the supernatant is mixed with the proper matrix and then evaluated by MALDI- TOF MS; the spectra are then analyzed for the beta-lactam agent, its salts, and/or its degradation products [110–114]. Different beta-lactam agents have been investigated; it is important to note that some agents such as amoxicillin and piperacillin may spontaneously degenerate during the incubation time [113]. This methodological approach has considerable potential to become a routine method; special software for the interpretation of the raw spectra would be required as would a fresh culture [111]. The detection of carbapenemase directly from clinical specimens without enhancement in specific cultivation media is likely to be very difficult, if not impossible [109]. Methylation and dimethylation of ribosomal RNA are a recognized resistance mechanism for protein synthesis inhibitors [9]. Kirpekar et al. [115] in 2000 reported a method for using MALDI-TOF MS to detect modification of ribosomal RNA. This MALDI-TOF method subsequently has been used by Kehrenberg et al. [116] to characterize a new mechanism for resistance to chloramphenicol, florfenicol, and clindamycin; this mechanism be methylation of the 23S ribosomal RNA at position A2503. Savic et al. [117] also have used this MALDI-TOF method to determine the target nucleosides for 16S ribosomal RNA methyltransferases that confer self- resistance in aminoglycoside-producing microorganisms. Clearly, this MALDI- TOF MS method has the potential to become a routine diagnostic technique at reference centers provided the special software for interpretation of the raw spectra was developed and commercially available [109].

76

C. W. Stratton

Yet another approach is to use MALDI-TOF MS to analyze intact cell mass s pectrometry (ICMS) in order to identify the bacteria as well as to identify specific markers involved in resistance [109, 118–123]. The first study to use this method to differentiate MSSA from MRSA was reported by Edward-Jones and colleagues in 2000 [118]; these investigators used intact staphylococcal cells and 5-chloro- 2mercaptobenzothiazole as the matrix [118–120]. This study has been confirmed by Du et al. in 2002 [121] and Wolters et al. in 2011 [122]. One study using bacterial lysates for analysis was unable to differentiate MSSA from MRSA [124] although another study using bacterial lysates was able to do so [122]. The most recent study by Sogawa et al. published in 2017 [123] was the largest study to date; this study tested 160 clinical isolates of S. aureus and reported identification rates of 93.3% for MSSA and 86.7% for MRSA. This study used intact staphylococcal cells and alpha-cyano-4-hydroxycinnamic acid in 50% acetonitrile/2.5% trifluoroacetic acid as the matrix [123]. MALDI-TOF MS for differentiating MSSA from MRSA clearly has the potential to become a routine diagnostic technique at reference centers provided the special software for interpretation of the raw spectra was developed and commercially available [109]. MALDI-TOF MS also has been used to differentiate vanB-positive Enterococcus faecium from vanB-negative E. faecium [125] as well as vanA-positive E. faecium from vanA-negative E. faecium [126]. Resistance to vancomycin mediated by both vanA and vanB strains of E. faecium involves substitution of the high-affinity terminal d-Ala-d-Ala peptide on N-acetyl-muramic acid subunits with d-Ala-d-Lac peptide; this amino acid substitution results in a 1000-fold decrease in the affinity of the pentapeptide for vancomycin [127]. Both of the reported methods [125, 126] used a standard ethanol-formic acid extraction for the MALDI-TOF MS measurements; direct application of some Gram-positive microorganisms to the target without an extraction procedure may not provide proper results due to the thickness of the peptidoglycan in the cell wall [109, 125]. This MALDI-TOF MS method for determining vanA and vanB resistance in Enterococcus has the potential to become a routine diagnostic technique at reference centers provided the special software for interpretation of the raw spectra was developed and commercially available [109]. MALDI-TOF MS has proven useful for determining the effects of antifungal agents on the protein spectral profiles of susceptible fungal isolates. The spectral profiles of Candida albicans grown in the presence of increasing concentrations of fluconazole have been assessed by MALDI-TOF MS [128]. These profiles were noted to change at a concentration of fluconazole that was defined as the lowest concentration of fluconazole at which a change in the profile could be documented. The investigators noted a high degree of agreement between this concentration and the minimal inhibitory concentration (MIC) obtained by a reference broth dilution method. In a similar study, De Carolis et al. investigated the use of MALDI-TOF MS to evaluate caspofungin resistance due to fks mutations in Candida and Aspergillus species [129]. The spectral profiles of Candida and Aspergillus fungal strains grown in the presence of increasing concentrations of caspofungin were assessed by MALDI-TOF MS [130]. The lowest concentration of caspofungin at


77

which a change in the spectral profile could be documented was calculated for each fungal strain and compared with minimal inhibitory concentrations (MIC) and minimal effective concentrations (MEC) of caspofungin determined by the reference broth dilution method. There was a 100% agreement for all isolated between the minimal concentration for a spectral profile change and the minimal inhibitory or minimal effective concentration. Of note was that two Candida isolates were incorrectly interpreted as nonsusceptible based on MALDI-TOF MS spectral profiles and MIC/MEC results although these isolates did not have a fks mutation. This result suggests a different resistance mechanism for these two Candida isolates. Finally, a universal phenotypic method of rapid determination of antimicrobial susceptibility testing independently of underlying resistance mechanisms recently has been described for MALDI-TOF MS [130]. This MALDI-TOF MS method uses a novel direct-on-target microdroplet growth assay for detecting antimicrobial resistance within several hours by incubating the microorganism with and without an antimicrobial agent in nutrient broth as a microdroplet directly on the MALDI- TOF target plate. After several hours for growth to occur, the droplet broth is separated from the microorganisms by contracting these broth droplets with an absorptive material. If the microorganisms incubated in the presence of the antimicrobial agent could be detected and identified by the MALDI-TOF MS spectra, it meant that the microorganism was “nonsusceptible,” whereas if the microorganism incubated in the presence of the antimicrobial agent could not be detected and identified by the MALDI-TOF MS spectra, it meant that the microorganism was “susceptible”; the identification of “susceptible” microorganisms was accomplished by MALDI-TOF MS spectra from that microorganism grown without an antimicrobial agent. This method is not yet available commercially, but clearly this is a potentially useful approach for the use of MALDI-TOF MS for susceptibility testing [130].

uture Novel Approaches to Rapid Phenotypic Susceptibility F Testing Methods Future novel approaches to rapid phenotypic susceptibility testing methods will involve innovative/alternate methods that ultimately will need to be correlated with traditional phenotypic AST methods and/or with clinical outcomes [63–66, 76, 131]. In contrast to traditional phenotypic susceptibility testing methods that involve continuous exposure to antimicrobial agents as various concentrations and use macro-detection of growth, these future methods are likely to use a short exposure to antimicrobial agents and rely on innovative methods to detect either growth or lysis. Table 3 is a list of a number of innovative rapid phenotypic susceptibility methods that may prove useful in the future. Many of these rapid AST methods have published “proof of concept” articles, but few have large corroboration studies published in peer reviewed articles. Some of these methods will be briefly reviewed in this chapter.

78

C. W. Stratton Table 3 Innovative/alternate rapid phenotypic antimicrobial susceptibility testing methods Flow cytometry with bacterial fluorescence probes Dielectric impedance measurement Single-frequency dielectric impedance measurement Multi-frequency dielectric impedance measurement Microfluidics with growth detection Electrochemical reduction detection by redox-active resazurin assay Single-cell microscopy growth detection by imaging technologies Forward laser light-scattering technology Digital time-lapse microscopy pH sensor Microbial cell mass and density measurement Ultrahigh resolution mass measurement using microchannel cantilevers Isothermal microcalorimetry Determination by flow rate Determination by total production of heat over time Two-photon excitation fluorescence assays Two-photon assay for MRSA Direct susceptibility testing Direct antimicrobial susceptibility testing from urine and blood Cell-free immunoassay with field-effect enzymatic detection/quantification Synthetic biology approaches for antimicrobial susceptibility testing Bacteriophage-based MRSA detection assay Luciferase reporter phage excited system Toehold switches for RNA detection Padlock probes for DNA detection

Flow Cytometry with Bacterial Fluorescence Probes Flow cytometry is an analytical procedure in which rapid measurement of light-scattering and probe-mediated fluorescence emission produced by each cell within a large population of cells results in a multidimensional histogram that can be analyzed by quantitative statistical methods [132]. Flow cytometry has been used to evaluate the physiological effects of antimicrobial agents on bacterial cells due to their effect on various parameters such as membrane potential, cell size, and cell shape [133]. Therefore, a number of investigators have been investigating flow cytometry as a potential method for rapid antimicrobial susceptibility testing [134–140]. One advantage of flow cytometry is that both bacteriostatic and bactericidal effects can be determined. Bacteriostatic effects can be detected within 30 min, while bactericidal effects can be detected within 3 h. Another advantage of flow cytometry is that a large population of cells can be assessed, which allows visualization of the heterogeneity of the response of the cells by detecting subpopulations that are less susceptible to the


79

antimicrobial agent. Several companies have recognized the potential value of flow cytometry for AST and are developing kits for this method. One of these products, the FASTinov® kit, has been evaluated in the clinical setting [140]. The FASTinov® kit contains lyophilized antimicrobial agents, a specific fluorochrome, and dedicated software that allows the selection of both EUCAST and CLSI classification rules. Use of this FASTinov® kit provided AST results in 2 h, which was 24–48 h sooner than the results of Vitek 2 AST method; there was categorical agreement with a low number of errors [140]. Dielectric Impedance Measurement Dielectric impedance measurement for AST is based on the phenomenon that electrical changes (i.e., impedance, conductance, or capacitance) will occur in broth bacterial culture media provided the living test microorganism is able to grow to a population of approximately 106 to 107 CFU/ml [141, 142]. In early dielectric impedance studies, two electrical probes were used with the broth bacterial broth culture being situated between these probes in order to create an electrical circuit. When an electrical current was applied at a single frequency to this electrical circuit, the current initially was not be able to pass through the broth bacterial culture due to resistance. As the bacteria in the broth bacterial culture grow, the current will gradually be able to pass through the culture with less resistance. The dielectric permittivity is defined as a measure of the overall polarizability of both inorganic and organic molecules within the bacterial cell suspension. The dielectric impedance is defined as the total opposition faced by an electrical current as it attempts to pass through this bacterial broth culture circuit. Dielectric measurements were initially done using a single frequency and were able to predict biomass most accurately during the growth phase of the cultures. During the stationary and decline phases of the bacterial cultures, these dielectric measurements were decreased in accuracy. If an antimicrobial agent is added to the bacterial broth cultures, the dielectric permittivity will immediately decrease if the bacteria are susceptible to the antimicrobial agent. The change will be proportional to the concentration of antimicrobial agent used. If the bacteria are resistant to the antimicrobial agent used, the dielectric permittivity will continue to increase as the bacteria continues to grow. Early dielectric impedance methods were described as single-frequency impedance measurements as only a single frequency was used. Single-frequency dielectric impedance measurements have been applied to broth dilution AST [143] as well as to direct AST methods for positive blood cultures [144, 145]. Dielectric impedance is able to measure both bacteriostatic and bactericidal effects and can also measure microbial stress, which appears sooner that either bacteriostatic or bactericidal effects [146, 147]. The principle of dielectric impedance measurement of living microorganism was first described at the turn of the nineteenth century [141]; the first use of this technique for susceptibility testing was in 1977 [143]. There have been several major advances in dielectric impedance techniques in the past several decades that promise to make this AST methodology more adaptable for use in the clinical

80

C. W. Stratton

icrobiology laboratory. The first advance is the development of an electrochemim cal screen-printed biosensor array [148]. This development was an important first step for the miniaturization of the dielectric impedance technology. The next step in this miniaturization process was the development of a microscale multi-frequency reactance measurement technique that allowed the detection of bacterial growth at low bio-particle concentrations [149]. The principle of dielectric impedance technology relies on bacterial metabolism to produce a measurable change in electrical conductivity [141–143]. The “bulk” of bacteria becomes important when measuring the change in dielectric impedance over time; the measurement is generally done by applying a single-frequency current through a bacterial broth culture circuit. However, the use of a multi-frequency current measurement technique allows the use of frequencies lower than 1 MHz to measure dielectric impedance of bacterial growth at low bio-particle concentrations (i.e., “low bulk”) and in less time (i.e., minutes versus hours) [149]. When electrochemical screen-printed biosensor arrays (i.e., “printed electrodes”) [148] are combined with microscale multi-frequency reactance measurement techniques [149] along with microfluidic methods [150], the results are rapid real-time AST methods on plastic microchips with printed electrodes [151, 152]. This evolution of dielectric impedance technologies has the potential to provide point-of-care platform technologies with rapid real-time guidance for the selection of antimicrobial therapy [141–152]. Microfluidic Assays Microfluidic assays separate individual microorganisms using stochastic confinement into nanoliter volume droplet plugs where bacterial growth can be measured [150]. This method enables elimination of the preincubation step while allowing detection of growth in less time (i.e., 2 h). Bacteria can be detected at concentrations of 1 CFU/μL; MIC profiles have been shown to be comparable to those obtained with conventional culture-based methods [150, 153, 154]. These plug-based microfluidic agarose channels are able to track the growth of single cells in the presence of antimicrobial agents by measuring the electrochemical reduction of a redox-active molecule such as resazurin [150, 153, 154] as well as by using microscopy [155–157], forward laser light-scattering technology [158], digital time-lapse microscopy [159, 160], or pH sensing [161]. By confining single bacteria in nanoliter plugs, the time required to detect growth is reduced to several hours. Moreover, the use of antimicrobial gradients in microfluidic-based systems allows a quantitative investigation of antimicrobial effects. The results of this approach are similar to those produced by conventional culture-based susceptibility testing methods. In particular, microfluidic assays are particularly useful when combined with single-cell bacterial growth detection using advanced imaging technologies. Single-Cell Bacterial Growth Detection by Imaging Technologies Direct imaging of bacterial growth in the presence of antimicrobial agents at a single-cell level using imaging technologies is an emerging tool for rapid bacterial antimicrobial susceptibility testing [65, 155–160]. In particular, multiplexed automated digital microscopy technology and single-cell morphological analysis are two of the advance imaging technologies that are making rapid AST methods possible.


81

When these image-based technologies are combined with microfluidics, the detection times are greatly shortened (hours versus days) although these technologies still use replication-dependent methods that require a preliminary growth step from blood cultures or from primary culture plates. Microfluidic agarose chips containing antimicrobial agents and nutrients are used to immobilize microorganisms upon gel solidification such that bacteria can be easily imaged. Time-lapse bright-field imaging of single cells with automated image processing and data interpretation can be utilized to profile the response of these microorganisms to the antimicrobial agents [159, 160]; morphological changes such as dividing, filamentary formation, swelling, and lysis are used to assess such responses [155–157]. Results can be obtained from as little as 30 min up to 3–4 h; these results are comparable to those obtained with standard broth microdilution testing [157–160]. Microbial Cell Mass and Density Measurement Cantilevers with small canals that allow microbial passage can be vibrated continuously [162]. When bacteria pass through these canals, their weight will produce a change in the frequency of the cantilever movement [163, 164]. Less-dense bacterial cells will produce a different change than more-dense bacterial cells. Bacterial cells treated with antimicrobial agents will have a change in their buoyant mass density that can be measured by a vibrating cantilever. Cantilevers can be multiplexed using microfluidic nanotechnology to produce embedded nanochannels that can be vibrated using centrifugal force. In this manner, multiple antimicrobial agents in various concentrations can be tested simultaneously for a single growing culture. Bacteria usually require a pre- enrichment culture step as well as sample processing steps in order to provide single cells that can be used in cell mass and density measurements [65, 163, 164]. Isothermal Microcalorimetry Isothermal microcalorimetry measures heat production, either as a flow rate or as total accumulation over time; heat production stems from the metabolism of actively growing cells, with cumulative heat production usually paralleling conventional growth curves with a slope and shape that corresponds with classical lag, log, and stationary phases [165]. Maximum heat values correspond to the total number of cells grown over time. Two-Photon Excitation Fluorescence Assays Two-photon excitation fluorescence detection is a fluorescence microscopy technique that allows fluorescence imaging of living cells [166–168]. The two-photon excitation fluorescence assay has been used for screening of MRSA; different progressions of fluorescence signals are seen for MRSA versus MSSA when the growth of S. aureus is monitored in the presence of cefoxitin [169]. Fluorescence monitoring of the broth sample is accomplished by polystyrene microparticles with capture antibodies and a fluorescently labeled tracer. Tracer molecules are bound to the polystyrene microparticles only via S. aureus antigens; the number of fluorescent tracer molecules bound to the microparticles rapidly increases with time only if viable MRSA cells are present in the broth sample [169]. A rapid increase in the fluorescence signal over time means that the isolate is MRSA.

82

C. W. Stratton

Direct Susceptibility Testing Rapid and direct identification of microbial pathogens along with direct antimicrobial susceptibility testing of these microbes has long been a goal of clinical microbiologists. Ideally, this would direct inoculation of the clinical specimen into an instrument for both identification and AST determinations, with the direct susceptibility testing (DST) being considered more clinically relevant due to the impact on selection/modification of antimicrobial therapy. Urine cultures have been the obvious place to start as urinary tract infections often have a single microorganism (e.g., E. coli) at a suitable inoculum (i.e., 105 CFU/ml) [170– 172]. These early studies determined that direct susceptibility testing on urine was reliable for monobacterial Gram-negative urinary tract infections. More recently, direct susceptibility testing has been used on other clinical samples such as respiratory tract specimens, bile fluid specimens, and abdominal abscess discharge specimens [173]. Many of these clinical specimens would be expected to contain more than one microorganism and would have a less well-defined inoculum. Yet, the results in this study [173] for both AST and DST results were comparable for Gram- negative bacilli with 93.4% total agreement and 1.6% minor discordances, 4.6% major discordances, and 0.4% very major discordances. Another clinical scenario in which rapid DST results are desired is with septic patients and blood cultures [174]. Blood culture continues to be the optimal method for identifying microorganisms causing sepsis; accordingly, a number of investigators have studied direct antimicrobial susceptibility testing from blood culture broth in order to reduce the time to results [83–87, 175]. Such DST from blood culture broth is usually done when the blood cultures become positive. A more rapid approach would be to use the blood itself for the inoculum. This approach is being evaluated for molecular methods and has been evaluated for one culture-free detection, identification, and rapid phenotypic AST method [176]. In this report, bacterial identification was achieved directly from blood in 84 min (sample to result), and rapid phenotypic AST was achieved directly from blood in 204 minutes (sample to result) [176]. This method utilizes an immunoassay platform [177] for the direct detection of microorganisms in blood without the need for culture enrichment. The detection platform is based upon the technology of field-effect enzymatic detection (FEED) [178, 179] in which the target is immobilized on the detecting electrode and the detection signal is obtained by measuring the reduction peak current of an enzyme that is used to label an antibody. This method requires an antibody against a potential pathogen; however, the specificity of the antibody against the microbial pathogen reduces the concern for mixed microorganisms in the clinical sample. AST testing is accomplished by quantifying the changes in bacteria concentration (CFU/ml) after exposure to antimicrobial agents using FEED. The immune reaction between a microorganism and its antibody is highly specific and therefore provides selection of a specific microorganism in a population of mixed microorganisms. Synthetic Biology Approaches for Antimicrobial Susceptibility Testing Synthetic biology approaches using bacteriophages or engineered gene circuits are another emerging field in which a number of innovative tools for rapid antimicrobial susceptibility testing are being developed [180–183]. Among these synthetic biology


83

approaches are bacteriophage-based methods [184–186]. One of these is the MicroPhage KeyPath MRSA/MSSA blood culture test that identifies Staphylococcus aureus and can differentiate between MSSA and MRSA using samples taken directly from signal-positive blood culture bottles [186]. Differentiation is done by adding cefoxitin to the assay such that MRSA can grow, but MSSA cannot; the growth of MRSA thus amplifies phages and results in a positive readout [186]. Bacteriophages also can be combined with the firefly luciferase gene in order create a “luciferase reporter phage” [187]. This has been done using luciferase reporter mycobacteriophages [188] in order to perform antimicrobial susceptibility testing on Mycobacterium tuberculosis strains [188–190]. Production of light by the luciferase gene requires metabolically active M. tuberculosis in which reporter phages replicate allowing the luciferase gene to be expressed. When drug-susceptible strains of M. tuberculosis are incubated with an effective antituberculosis drug, these strains fail to produce light because they are not metabolically active. In contrast, drug-resistant strains are able to replicate in the presence of the antituberculosis drug and produce light at levels equal to the untreated controls. The luciferase reporter phage technology has been adapted by GeneWEAVE (“Smarticles”) and will be further adapted by Roche, who recently have acquired this technology (Michael Lewinski, personal communication). Rapid detection of bacterial RNA or DNA is another synthetic biology approach for rapid phenotypic antimicrobial susceptibility testing; the detection of RNA or DNA in the presence of a specific antimicrobial agent indicates growth and implies resistance. Toehold switches that respond to endogenous RNA [191] and padlock probes with rolling circle amplification for DNA detection [192] are two synthetic biology methods that are being evaluated for rapid phenotypic antimicrobial susceptibility testing [65, 183]. For example, padlock probes for the most common mutations associated with rifampin resistance in M. tuberculosis were combined with a probe for the identification of M. tuberculosis complex and used to determine rifampin resistance strains [193]. Another study employed an initial short incubation step in the absence and presence of different antimicrobial agents combined with a species-specific padlock probe for detection of the bacterial target DNA has been shown to be capable of determining the antimicrobial susceptibility profiles of E. coli with 100% accuracy in 3.5 h [194].

Summary and Concluding Remarks Antimicrobial susceptibility testing has always been an elusive goal. As Dr. David Greenwood pointedly queried in 1981 [195], In vitro veritas?, the issue of extrapolating laboratory results to the clinical scenario continues to be difficult, although significant improvements have been made since Dr. Greenwood made his query. In particular, standardization of traditional AST methods with integration of pharmacokinetics and pharmacodynamics with breakpoints along with clinical correlations

84

C. W. Stratton

studies has brought AST a long way since the 1980s. Rapid phenotypic antimicrobial susceptibility testing will undoubtedly progress through the same laborious steps as have traditional AST methods, but I have no doubts that in the end, clinical correlations will be demonstrated and strengthened as they have been for traditional methods. Some of the major issues with rapid phenotypic AST methods that will need to be addressed and will likely require resolution are as follows. Pure Culture Versus Mixed Culture Direct susceptibility testing using the clinical specimen as the inoculum has been proposed in order to save time by circumventing isolation and identification of microorganisms. However, good clinical practice has been to obtain the appropriate cultures and start empirical treatment of life- threatening infections without delay [8, 196]. One very important issue for direct antimicrobial susceptibility testing directly is the issue of pure culture versus mixed culture and the potential effect on AST results. Pure cultures have been the gold standard for susceptibility testing for almost 50 years [196, 197]. This recommendation is, in part, related to a careful study of the effect of mixed cultures on disk diffusion antimicrobial susceptibility testing that such results were completely unreliable [197]. In this study, a number of susceptible species gave results interpreted as resistant when tested in combination [197]. As stated by the authors of this study [197], “It is obvious from the results obtained in this study that the dictum so often stated (and so often ignored) that antimicrobial susceptibility tests should be performed only with pure cultures has a sound factual basis.” Although studies evaluating direct antimicrobial susceptibility from urine cultures have been reported as being in agreement with standardized AST methods [170–172], these studies addressed monomicrobial urinary tract infections. When a similar study [198] compared direct and standardized disk diffusion susceptibility testing of urine cultures and determined the number of major discrepancies for mixed cultures, it was 42.6%. Another clinical study [199] noted that antimicrobial testing of mixed cultures from infected prosthetic implant in bone resulted in 67–100% showing complete resistance. The use of antibodies as a detection method [176, 177] is one way to deal with this issue. Single-cell growth detection AST methods [65, 155–160] is another way to deal with this issue. A similar issue is control/standardization of the inoculum [196], which will be discussed next. Control/Standardization of Inoculum The control and standardization of the inoculum is another important issue that needs to be considered in the quest for novel approaches to rapid phenotypic AST methods. The inoculum has long been recognized as an important variable in phenotypic antimicrobial susceptibility testing methods [6, 196, 200, 201]; the inoculum effect is particularly important when determining bactericidal activity [202, 203]. The inoculum effect can be defined as an increase in the MIC that is seen with an increase in inoculum size [204]. A high inoculum size may be important clinically. For example, a high inoculum size at the site of an infection has been shown in an in vivo murine model to decrease the activity of the antimicrobial agent by selecting resistant subpopulations [205]. A high inoculum size also may result in decreased antimicrobial effectiveness due to a


85

reduced ratio of available drug molecules per target because of a relative reduction of the antimicrobial concentration [206]. Another issue with a high inoculum size may be tolerance [207], which is the self-limiting effect of bacterial growth in a high inoculum size. Reduced growth rate at a high inoculum size such as seen during the stationary phase of bacterial growth can result in decreased activity of an antimicrobial agent. This is the reason that the inoculum should be achieved during the logarithmic phase of growth [208, 209]. The rate of bacterial inhibition/killing is well recognized to be dependent on the rate of growth of the inoculum [12–14]. The use of stationary-phase microbial cells in the inoculum will include an increased number of dormant microbial cells that are not as susceptible to the agent and thus result in reduced efficacy of the antimicrobial agent. These factors related to the inoculum are the main reasons that the inoculum has been standardized and tightly controlled in susceptibility test methods [17–33]. Breakpoints Phenotypic antimicrobial susceptibility testing in an attempt to phenotypically recognize microbial isolates as susceptible or resistant according to an epidemiological cutoff value that is known as the breakpoint [7, 210–212]. Clinical microbial resistance can be defined as a condition in which antimicrobial therapy did not achieve a clinical cure in an infected patient. Ideally, clinical resistance would be determined by the clinical breakpoints; these breakpoints should separate clinically resistant microbes (those with a high likelihood of therapeutic failure) from clinically susceptible microbes (those with a high likelihood of therapeutic success) [44]. Breakpoints are influenced by pharmacokinetic and pharmacodynamic factors; determination of breakpoints thus requires integration of these pharmacokinetic/pharmacodynamic factors as well as knowledge of the wild-type distribution of MICs of the microbe being tested and the results of clinical studies in which the clinical outcomes of infections caused by the microbe being tested have been treated with the antimicrobial agent being tested [15, 34–49, 211]. If such clinical studies have not been done, then the results of in vivo animal model studies and/or Monte Carlo simulation can be used to determine breakpoints [41, 46, 213]. Finally, clinical breakpoints are usually determined by criteria established by the Clinical and Laboratory Standards Institute and the European Committee on Antimicrobial Susceptibility Testing [44]. Clearly, new approached to rapid phenotypic susceptibility testing will need similar breakpoints. Validation and Clinical Correlation The need for correlation of clinical results with the results of laboratory-based phenotypic antimicrobial susceptibility testing was recognized early in the evolution of antimicrobial therapy [6, 214]. Similarly, the need for validation of phenotypic susceptibility testing methods has driven much of the standardization of AST methods over the past 40 years [6, 17–33]. Although much progress has been made, there is still a disconnect between in vitro susceptibility testing results and in vivo clinical results of antimicrobial therapy [15, 16, 215]. Validation and clinical correlation of new approaches to rapid phenotypic susceptibility testing are equally important and will be required as these new methods continue to be developed. Validation must include a determination

86

C. W. Stratton

of the wild-type distribution of the results of these new approaches when applied to large populations of a single species of a pathogenic microbe known to cause infection. Clinical correlation must be made using the results of new approaches to rapid phenotypic susceptibility testing; both clinical studies and studies using in vivo animal models must be correlated. It is possible that comparing the results of new approaches with those already obtained using standardized AST methods will be sufficient, but his remains to be seen. Some would argue that these past results of standardized AST methods should not become the gold standard [215]. Single Cell/Low Inoculum Versus Antimicrobial Heteroresistance “Heteroresistanc e” is defined as the phenomenon where supposedly isogenic bacteria exhibit a range of susceptibilities to a particular antimicrobial agent [216]. Heteroresistance has been recognized since 1947 and occurs in both Gram-positive and Gram-negative microorganisms. The clinical relevance of heteroresistance is likely to be high due to the potential selection of resistance subpopulations during antimicrobial therapy [205, 207]. The use of single-cell microscopy or low inoculums in microfluidic AST methods may or may not identify such heteroresistance [216]. Heteroresistance is an important phenomenon and is related to phenotypic resistance [217]. Both must be appreciated and dealt with as new approaches to rapid phenotypic antimicrobial susceptibility testing methods are developed, validated, and clinically correlated. Persistence, Tolerance, and Phenotypic Resistance Persistence occurs in a microbial subpopulation (classically defined as 95%

Specificity

(continued)

Often performed at POC. Negatives must be evaluated by culture Provides adjunct, but not definitive, diagnostic information in patients at risk for S. pneumoniae disease Requires FA microscope. Cross-reactions with some other bacteria, especially with polyclonal reagents. No gold standard for comparison Test characteristics well-established only for L. pneumophilia group 1 Toxin tests are insensitive relative to more laborious and expensive methods such as PCR and cytotoxicity. Combination algorithms for GDH and toxin antigen testing are more sensitive, still may not approach PCR, but appear to be more specific Used as an alternative to serology and urea breath testing; usable as test of cure Being phased out, but POC versions might be valuable if sensitivity improves. No single-test format available. Not useful for screening low-prevalence populations due to poor specificity

Comments

108 S. Campbell and M. L. Landry

Cryptococcus

ELISA

Fungi Aspergillus

Agglutination, ELISA, CSF, serum LFIA

Serum

Stool

EIA; LFIC

Campylobacter antigen

CSF, urine

Stool

Agglutination

Shiga-toxin-producing E. EIA; LFIC coli

Meningitis panel (H. influenzae, N. meningitidis, S. pneumoniae, group B Streptococcus)

Table 2 (continued)

99%+;

Varies

Despite high specificity, positive predictive values are low

(continued)

Used for surveillance in neutropenic patients, allows for early initiation of antifungal therapy. Roughly 2/3 of patients have positive antigenemia before diagnosis Very high if heat or Sensitivity may exceed culture. Cross- pronase pretreatment reactivity with (very rare) systemic Trichosporon infections. Prozone is a used problem in high-level infections. New LFIA is highly sensitive and specific

Varies

High Higher than culture; no gold standard defined 79.6–87.6% 95.9–99.5%

Inadequate sensitivity/specificity for routine clinical use Empiric therapy given for CSF neutrophilia covers these pathogens, until culture results available. Positive predictive value of antigen tests is very low in patients without CSF leukocytosis Does not detect all strains, so culture still recommended

Rapid Microbial Antigen Tests 109

Blood

Genital

IF, LA, LFIC

LFIC and other rapid formats

Stool

ELISA

Entamoeba histolytica/ dispar group Trichomonas vaginalis

Plasmodium species

Stool

IF, LFIC, ELISA, rapid EIA

Cryptosporidium

Specimen Respiratory

Stool

Methods IF

IF, LFIC, ELISA, rapid EIA

Parasites Giardia

Pathogen Pneumocystis jirovecii (formerly P. carinii)

Table 2 (continued)

Similar to microscopy

Higher than microscopy 85%

Higher than microscopy

Higher than microscopy

Sensitivitya Variable

High

High

> 95%

93–100%

Near 100%

Specificity High

No “gold standard” available for comparison. Specimen treatment (e.g., fixed, unfixed, or frozen) varies with different tests No “gold standard” available for comparison. Some kits detect both Giardia and Cryptosporidium Reagents are available to distinguish between E. histolytica and E. dispar Alternatives include wet prep (60% sensitivity relative to culture), culture, molecular detection. Wet prep is limited by specimen stability One FDA-approved test available. Cost limits use in endemic areas, but development is a high priority

Comments Requires fluorescence microscope. Sensitivity is highest for antibodies that detect antigens present in trophozoites and cysts. No significant sensitivity or specificity advantages over conventional and calcofluor-white stains


NP swab or aspirate, BAL, sputum, throat swab

NP swab, NP aspirate, nasal wash, BAL

LFIC, EIA

IF

NP swab or aspirate, BAL, sputum

IF

Influenza A and B

NP swab or aspirate, BAL, sputum

LFIC, EIA

Viruses Respiratory syncytial

Blood

ELISA, LFIC

Lymphatic filariases

Table 2 (continued)

97–99%

> 99%

95–99% 48–98%b

95–99%

90–100%

50–90%a 10–79%b

85–98%a 46–93%b

> 95%

80–95%

Equivalent to microscopy; similar to or higher than concentration methods

(continued)

Very sensitive in young infants who shed high titers of virus. Mucoid samples may not disperse properly and may give rise to erroneous results More sensitive than culture or other antigen tests. Can be multiplexed with other antibodies. IF allows assessment of sample quality Sensitivity higher in children, and with NP aspirates and washes. Some kits require the use of special swab. LFIC suitable for POC. Mucoid samples may not disperse properly and may give rise to erroneous results Performance must be established in each laboratory. Can be more sensitive than other rapid tests. Cytospin preparation of slides improves results. Pooled antibodies can be used to screen a single-cell spot for multiple respiratory viruses. IF allows assessment of sample quality

No “gold standard.” cost limits use in endemic areas


Stool

EIA, agglutination, LFIC

EIA EIA

Rotavirus

Astrovirus Norovirus

Stool Stool

Stool

EIA

IF

NP swab or aspirate, BAL, sputum NP swab or aspirate NP swab or aspirate, BAL, sputum Stool

Specimen NP swab or aspirate, BAL, sputum

Adenovirus, enteric types EIA 40,41 Agglutination

Adenovirus

IF

Human metapneumovirus

LFIC

Methods IF

Pathogen Parainfluenza

Table 2 (continued)

97% 30–40%

90–98%

98%

99% 94–99%

90–98%

99%

99%

99%

50–70%

90%

97%

99%

85%b

63%b

Specificity 95–99%

Sensitivitya 80–95%

Comments Only rapid method available. Cytospin preparation of slides improves results. Antibodies to types 1, 2, and 3, but not type 4, are included in commercial antibody pool Available as a separate reagent or in a multiplex reagent with seven other respiratory viruses Since HMPV grows poorly in culture, RT-PCR is the reference method IF for adenovirus not as sensitive as for other respiratory viruses. Cytospin preparation of slides improves results Test available for detection of all adenovirus types in culture fluids or stools; does not differentiate among types Test available to detect only enteric types 40 and 41 which do not grow in routine cell cultures Rotavirus does not grow in routine cell cultures, so rapid tests historically compared to EM. Rotavirus shed in high titers in stools of infants and young children Rapid tests compared to EM Sensitivity compared to RT-PCR. Limited by antigenic variation and rapid onset and resolution of illness


80% of antibody negative, RNA-positive infections

95–99%

Blood

CLIA, EIA, LFIC

Human immunodeficiency virus fourth-generation antigen/antibody combo

> 99%

> 99%

Blood leukocytes 90–97%

Cytomegalovirus

Nuchal biopsy, brain tissue

> 99%

IF IP

IF

Rabies

Skin lesions, BAL

> 99%

> 99%

IF

Varicella zoster

Skin, genital, or 80–95% oral lesions, BAL, brain tissue

60–100%

IF

Herpes simplex

Table 2 (continued)

(continued)

Sensitivity enhanced by cytospin preparation of slides. Sensitivity is higher for skin lesions than for mucosal lesions. Pooled HSV and VZV antibodies labeled with different fluorochromes can be used to test for both viruses in a single-cell spot IF for VZV in skin lesions is more sensitive than culture. VZV and HSV antibodies can be pooled for dual detection Sensitivity of antemortem nuchal biopsy depends on stage of disease and abundance of large hair follicles with nerve fibers. Brain biopsy has a sensitivity of 95–100% but is most often obtained postmortem Quantitative detection of CMV pp65 antigenemia is useful for rapid diagnosis and monitoring therapy. More sensitive than culture. Not suitable for severely neutropenic patients Fourth-generation HIV screening tests combine p24 antigen with antibody detection to detect early HIV-1 infection before anti-HIV antibody appears. Antibody differentiation and molecular tests needed to confirm specificity. Rapid LFIC test available


CLIA, EIA

EIA

EIA, LFIC

LFIC

Hepatitis C

Hepatitis D

Dengue

Zika virus

Blood

Blood

Blood

Blood

Specimen Blood

Not available

> 80%

40–100%

80%

Sensitivitya 99%

Not available

> 95%

99%

Specificity > 99%

Comments Free HBsAg is produced in 100–1000-fold excess over complete virus particles. Thus HBsAg sensitivity can approach NAAT. HBeAg is a marker for high levels of viral replication. Viral mutations can lead to falsely negative IA results HCV antigen tests can be useful and cost-effective at POC and in low-resource settings In acute infection, detected early and transiently. In chronic infection present in high titers Useful at POC in low-resource areas. New standard for diagnosis. Detects NS1 antigen excreted by infected cells. Often combined with IgM rapid test. Approaches sensitivity of NAAT in some studies In development. Potentially useful at POC in low-resource areas. Detects NS1 antigen

b

a

Test performance compared to culture unless otherwise stated Test performance for 2009 pandemic influenza compared to RT-PCR Abbreviations POC point of care, ELISA enzyme linked immunosorbent assay, EIA enzyme immunoassay, IF immunofluorescence, IA immunoassay, CLIA chemiluminescence immunoassay, LFIC lateral flow immunochromatography, RT-PCR reverse transcription polymerase chain reaction

Methods CLIA, EIA

Pathogen Hepatitis B surface antigen (HBsAg) and e antigen (HBeAg)

Table 2 (continued)


Rapid Microbial Antigen Tests

115

time to result; tests which detect both toxin A and toxin B are more sensitive than methods which detect only toxin A. As molecular methods have become more widely adopted, the role of toxin detection versus toxin gene detection by PCR has become controversial [18, 19]. Toxin-only antigen immunoassays are insensitive relative to toxigenic culture or PCR; on the other hand, PCR testing detects patients with asymptomatic colonization with Clostridium difficile [20]. Optimal antigen- based algorithms involve a screening C. difficile glutamate dehydrogenase (GDH) antigen test, which is more sensitive than a toxin assay but detects non-toxigenic strains, followed by a toxin antigen test for confirmation. PCR or repeat testing may be used to address discrepant results [21, 22]. A cartridge-type rapid test incorporating both GDH and toxin antigen detection is available. In addition, more sensitive toxin antigen tests, which approach the sensitivity of cytotoxin neutralization, are in development [23]. Rapid tests for Helicobacter pylori include tests for antibody and antigen. H. pylori antibody testing is unspecific, as many seropositive patients do not harbor the bacterium, and is no longer recommended for routine use. Antigen testing of stool for Helicobacter pylori is available both in laboratory-based EIA and a rapid format. It serves as a diagnostic option to the urea breath test, serology, and endoscopy. It may be particularly useful in children, where the urea breath test may be difficult to perform. False-negative results are common in patients on proton-pump inhibitor therapy, bismuth, or antibiotics; it is recommended that patients be off therapy for 2 weeks prior to testing. The stool antigen test can also be used as a test of cure, though the time required after treatment is still unclear [24, 25]. Antigen testing for genital Chlamydia and gonococcal infections has been almost entirely replaced by nucleic acid testing, which is substantially more sensitive and specific. Rapid tests have the potential for POC use, but none is FDA-approved, and sensitivity remains an issue [26]. The emergence of rapid POC NAAT tests is likely to eliminate any meaningful role of antigen testing for chlamydia and gonorrhea, barring significant improvements in the antigen tests [27]. Bacterial antigen testing for meningitis is rapid but has fallen out of use in recent years due to inadequate sensitivity and specificity, and the use of empiric antibiotic therapy. The presence of neutrophils in CSF generally leads to therapy in patients with compatible syndromes, while the positive predictive value of antigen testing performed on patients with acellular CSF is dismal. Empiric antibiotic choices cover the organisms detected by the antigen tests [28, 29]. Current guidelines for detection of Shiga-toxin-producing E. coli (STEC) recommend the use of a Shiga-toxin antigen test (or molecular test for the gene) in addition to culture for the O:157 strain. Antigen testing may be performed directly on stool, but improved sensitivity is available if an overnight enrichment in selective broth is performed. The sensitivity of antigen tests, even after broth enrichment, falls short of 100%, so selective culture on sorbitol-MacConkey agar still recommended. Sorbitol-negative colonies of E. coli may be tested with a rapid antigen test for the O:157 antigen. For cases of STEC detected either by culture or antigen testing, the stool sample or the isolate should be forwarded to public health laboratory for analysis, confirmation, and strain typing for outbreak investigation [30, 31].

116

S. Campbell and M. L. Landry

Campylobacter antigen tests have also become widely used for detection of that difficult-to-culture organism from stool. The stool antigen tests ranged from 79.6% to 87.6% in sensitivity, 95.9 to 99.5% in specificity, and 41.3 to 84.3% in positive predictive value [32]. While the properties of individual tests vary, it has become evident that the low incidence of Campylobacter infections leads to poor positive predictive values even from tests with high specificity, and the value of Campylobacter antigen tests in the overall assessment of GI pathogens is still evolving.

Fungi Detection of invasive Aspergillus infections using the galactomannan ELISA has become an important aspect of management of bone marrow transplant and other profoundly neutropenic patient populations. Historically, the test was used both for periodic (biweekly) surveillance and diagnostically in patients who become symptomatic. More recently, the test is less used for surveillance due to the routine use of antifungal prophylaxis in neutropenic patients. Galactomannan is insensitive and not recommended for routine surveillance in patients on mold-active agents. The combination of radiologic and antigen testing allows early initiation of antifungal therapy and improves outcome in neutropenic patients who develop symptomatic disease. The test is less sensitive in non-neutropenic patients, including solid organ transplant recipients, due likely to lower organism loads, and is not recommended for surveillance in those populations, though it is still a useful test for diagnosis of symptomatic disease. False-positives occur, occurring frequently in patients who receive amoxicillin-clavulanate and other antibiotics, and in patients with other fungal infections, though the latter can be clinically useful. Galactomannan testing is also recommended on BAL specimens and improves sensitivity of diagnosis relative to culture. The use of galactomannan antigen for therapeutic monitoring is promising; increases in Aspergillus antigen levels over time signify a poor prognosis, but return to normal levels is not sufficient to stop therapy [33]. A similar marker, (1 → 3)-β-D-glucan, is also useful for diagnosing invasive mycoses; it is not specific for Aspergillus, but instead is positive with many mold species. For Cryptococcus, antigen testing is the mainstay of diagnosis. The sensitivity in cryptococcal meningitis approaches that of culture while providing more rapid diagnosis. Cryptococcal antigen testing of both serum and CSF is important in diagnosis, staging, and monitoring of cryptococcal disease in at-risk patients [34]. A rapid lateral flow cryptococcal antigen test (in place of the latex agglutination assays first developed) has sensitivities in the high 90% range in whole blood or serum and nearly 100% in CSF in cryptococcal meningitis. The lateral flow format is simpler for laboratory-based use and is highly deployable to low-resource settings for global health [35, 36]. By contrast, IF staining of respiratory specimens for Pneumocystis jirovecii is one of the several techniques of similar sensitivity for detection of Pneumocystis pneumonia. The choice of IF, conventional stains, or calcofluor-white depends on


117

the laboratory. IF and calcofluor require fluorescent microscopes, and IF reagents are expensive. Calcofluor and conventional staining methods require the reader to discriminate between Pneumocystis, yeasts, other pathogens, and cellular structures morphologically, which requires more interpretive skill than IF staining [37]. IF appears to be insensitive compared with PCR for Pneumocystis, especially in non- AIDS patients who frequently have lower organism burdens, but PCR is positive in persons who are most likely asymptomatically colonized [38].

Parasites For infections by Giardia and Cryptosporidium, antigen testing has become a method of choice, with sensitivities that exceed that of routine microscopic exam [39]. Many different formats are available, and laboratories select a method based on technical (e.g., availability of fluorescence microscope, test format) and operational (e.g., specimen requirements, test volume) differences [40]. Antigen detection is superior to most other non-molecular methods for detection of Entamoeba histolytica. E. histolytica is morphologically indistinguishable from nonpathogenic relatives including the relatively common E. dispar. ELISA format and rapid point-of-care LFAs are available, and some are comparatively specific for pathogenic E. histolytica and useful for distinguishing it from E. dispar. Because E. histolytica is comparatively rare in the USA, antigen tests are not as widely used as for Giardia and Cryptosporidium; additionally, the available tests require fresh, rather than fixed, stools, which may create workflow challenges in laboratories used to working primarily with fixed samples. Antigen tests appear to be less sensitive than PCR, but the relative performance in different patient populations in endemic and non-endemic regions varies widely, and it remains unclear whether the added sensitivity of PCR-based testing provides significant clinical yield, especially in resource-limited settings [41, 42]. Since Trichomonas rapidly loses motility below body temperature, the wet prep has always been an insensitive approach to diagnosis, particularly if specimens needed to be transported prior to viewing. Commercially available DFA and latex agglutination methods provide better sensitivity; nucleic acid amplification, direct nucleic acid hybridization, and culture are more sensitive but more complex. The OSOM TV Trichomonas Rapid Test is a CLIA-waived rapid lateral flow test that detects Trichomonas membrane proteins, with an additional internal control. It is clinical laboratory. Sensitivities compared with various gold standards range from 80% to 90%, with 95% + specificities, meaningfully better than wet mount microscopy, but short the performance of nucleic acid amplification [27]. Rapid diagnosis of malaria by antigen detection, primarily using lateral flow immunochromatography (LFIC), is a vital component of global malaria control. The proliferation of chloroquine-resistant strains and the expense of newer antimalarial drugs make these tests economical in endemic regions, and WHO recommends that all malaria treatment be based on parasite diagnosis. As of January

118


Table 3 WHO requirements for malaria rapid diagnostic tests Products should be selected in line with the following set of criteria, based on the results of the assessment of the WHO Malaria RDT Product Testing Program For the detection of Plasmodium falciparum (Pf) in all transmission settings, the panel detection score (PDS) against Pf samples should be at least 75% at 200 parasites/μl For the detection of Plasmodium vivax (Pv) in all transmission settings, the panel detection score (PDS) against Pv samples should be at least 75% at 200 parasites/μl The false-positive rate should be less than 10% The invalid rate should be less than 5%

1, 2018, WHO will require that malaria RDTs fulfill specific requirements (see Table 3) [43]. An FDA-approved test in the USA has separate channels for a specific P. falciparum antigen and for an interspecies common antigen. It does not replace thick and thin smears but provides a simple, rapid diagnostic test usable in virtually any laboratory, with a sensitivity of roughly 95% for P. falciparum and 69% for P. vivax. The higher analytical sensitivity of this test for P. falciparum than for other malaria species allows for separate reporting of P. falciparum or non-P. falciparum malaria. Worldwide, rapid malarial antigen tests vary significantly in sensitivity, ease of use, and quality control, and dissemination of high-quality tests is a priority for worldwide malaria control [44]. Rapid antigen tests have also been utilized for Wuchereria bancrofti infections. The Binax immunochromatography card (Binax, Inc.: http://www.binax.com) and the TropBio ELISA are more sensitive than blood smears for detection of infection and can use the blood collected during the day or night in contrast to microscopy, which requires nighttime blood collection. There appear to be some cross-reactions between Wuchereria and other filaria with antigen testing [45, 46].

Viruses Prior to the availability of nucleic acid amplification tests (NAAT), antigen tests were the diagnostic mainstay for viruses that do not grow in routine cell culture and for rapid virus detection. Since antigen tests do not amplify the target, they are inherently less sensitive than culture (biological amplification) and NAAT (biochemical amplification) and perform best in clinical scenarios in which the target antigen is present in samples in large amounts, such as hepatitis B surface antigen (HBsAg) in serum, rotavirus in stools in young children, and respiratory syncytial virus (RSV) in nasopharyngeal samples of infants [47, 48]. Despite availability of NAAT, HBsAg remains central to diagnosis: it is rapid, inexpensive, and is present in serum in 1000-fold excess over HBV DNA. In assessing rotavirus vaccine efficacy, positives detected by EIA have been reported to correlate better with clinical disease and vaccine failure, than do low positive samples detected only by NAAT [49].


119

For respiratory virus infections, RSV antigen tests approach or exceed the sensitivity of culture and have long been the standard of care in pediatric populations. When neuraminidase inhibitor therapies became available, rapid influenza antigen testing increased dramatically in clinics, emergency departments, and hospitals. Having an influenza test result within 10–30 min at POC for pediatric patients was shown to reduce the use of antibiotics and diagnostic tests, allow earlier discharge, and permit early administration of antiviral agents [50–52]. In adults, rapid influenza antigen tests have not performed as well due to lower viral titers. Indeed, the 2009 H1N1 influenza pandemic focused attention on the insensitivity of rapid flu antigen tests in adults and led to increased demand for NAAT [53–55]. Current antigen tests utilize predominantly LFIC and require approximately 100,000 viral copies for a positive result, a titer found in samples from very young children but not typically in adults. LFIC sensitivity can be increased by using fluorescent immunoassay technology and a small instrument to analyze results [56–58]. Effective January 2018, the US Food and Drug Administration (FDA) performance standards for rapid influenza tests now require 80% sensitivity compared to RT-PCR and annual reactivity testing using circulating influenza virus strains [59]. Despite much greater costs, 20–30-min POC NAAT tests for influenza can be expected to replace antigen detection methods in high-resource settings unless sensitivity of the antigen tests can be dramatically increased. However, both NAAT and antigen detection tests can fail to detect influenza as a consequence of the virus unique ability to continually change its genetic and antigenic composition [60, 61]. IF, another common rapid viral antigen method, requires microscopic examination and samples with abundant target cells. Advantags are that IF can detect a variety of viruses, employ multiplex reagents, and permit visual quantitation of infected cells. Cytocentrifugation to prepare slides further enhances test performance. For respiratory viruses, IF, using pooled monoclonal antibodies on nasopharyngeal and BAL samples, can detect and identify up to eight viruses in a single-cell spot with results in 1–2 h [62–64]. When done well, IF is significantly more sensitive than other rapid RSV and influenza virus antigen tests. For the 2009 influenza pandemic, IF sensitivities in adults of 82–93% compared to RT-PCR were achieved by some laboratories [62, 65]. Fixation and staining of cells in suspension rather than affixed to slides can shorten time to result. Antibodies labeled with different fluorochromes can also be pooled to screen skin lesion samples for HSV and VZV [66, 67]. When good quality lesion samples are tested, IF can give excellent results [67]. Of note, IF was the first method used to rapidly detect and quantitate CMV antigen in peripheral blood cells and it thus revolutionized the diagnosis and monitoring of CMV infections [68, 69]. Although IF has largely been replaced by real-time PCR in high- volume laboratories, CMV antigenemia quantification by IF retains advantages for on-site testing in smaller laboratories: IF is less expensive than PCR, takes 1–2 h to complete, and can be done “on demand”, including testing a single urgent sample when needed. IF can also be used to detect viral antigens in touch preps of tissue, such as HSV or rabies virus antigens in brain tissue. However, the interpretation

120


of IF performed directly on clinical samples is challenging, and both the sensitivity and specificity of IF for direct detection of viral antigens in clinical specimens are highly variable between laboratories due to differences in expertise and quality control. In particular, nonspecific staining can be misinterpreted as positive, and small numbers of true positive cells can be overlooked. A more recent advance for rapid viral antigen detection can be found in fourth- generation human immunodeficiency virus (HIV) combination tests in which HIV p24 antigen detection has been added to the detection of IgM and IgG antibodies to reduce the “seronegative window” in acute HIV infection [70]. Results are available within 45 min when CLIA is used or within 30 min when a fourth-generation LFIC test is used [71, 72]. Although HIV RNA amplification is more sensitive, it is more expensive and time-consuming and not suitable for routine large-scale screening. Thus, the p24 antigen/antibody combo assays have the clear advantage. New uses for rapid antigen testing continue to emerge in resource-limited settings and at POC. Recent examples include dengue virus (DENV), Zika virus (ZIKV), and hepatitis C virus (HCV). For DENV, rapid diagnostic tests based on viral nonstructural protein 1 (NS1) have revolutionized DENV diagnosis by providing a simple, low technology assay with high sensitivity and specificity, thus allowing early diagnosis and effective patient management [73, 74]. Indeed, NS1 antigen tests have become the new standard for dengue diagnosis. A limitation is the rapid rise in cross-reacting IgG antibodies to NS1 in secondary infections, which narrows the days that NS1 is detectable. Recently, a rapid ZIKV antigen test based on ZIKV NS1 has been developed, which does not cross-react with DENV NS1 [75]. Lastly, effective drugs for HCV are increasingly available, and an accurate viral load by NAAT may no longer needed to predict treatment duration or outcome. Consequently, simple low-cost HCV antigen and antigen/antibody combination tests are being considered as an alternative to NAAT, especially where the expense of NAAT can be a barrier to care [76–79].

Rapid Antigen Testing in Global Health Rapid antigen tests are well adapted to deployment in global health for low and moderate income settings. The WHO ASSURED list (Table 4) describes criteria a rapid test must meet to address disease control needs [80–82]. Rapid antigen tests, Table 4 WHO ASSURED criteria for rapid tests A = Affordable S = Sensitive S = Specific U = User-friendly (simple to perform in a few steps with minimal training) R = Robust and rapid (can be stored at room temperature; results available in 107 molecules per minute, represents a millionfold increase. Direct Chemiluminescence (CLIA) is a nonenzymatic system. Substrate linked to an antibody/antigen complex is the label. One oxidation event liberates one molecule of label with release of a set number of photons. A nonenzymatic system uses direct chemiluminescent labels, which have lower background signals than the enzyme systems, and will typically give rise to very fast reaction times for eliciting signals. The luminol reaction is widely used as a chemiluminescent fast or “flash” reaction, but unlike the peroxyoxalate systems does not require an organic/mixed solvent system. The chemiluminescent emitter is a “direct descendent” of the oxidation of luminol by an oxidant in basic aqueous solution. Probably, the most useful oxidant is hydrogen peroxide (H2O2). With the acridinium ester system, the signal takes only 2 s to develop after the immunological binding and subsequent wash step, compared with 30 min or longer for an enzyme-generated systems. The acridinium ester molecule produces chemiluminescence that will speed up most assays by an assay of tenfold or more sensitive and can be easily detected [10]. The major reason accounting for the increasing popularity of chemiluminescent assays is their exquisite detection sensitivity. Unlike absorbance (colorimetric) or fluorescent measurements, assay samples typically contribute little or no native background chemiluminescence. The lack of inherent background and the ability to easily measure very low and very high light intensities with simple instrumentation provide a large potential dynamic range of measurement. Linear measurements over a dynamic range of 106 or 107 using purified compounds and standards are routinely achieved. Enhanced Chemiluminescence Like indirect or enzymatic CLIA, the HRP enzyme is the label, and luminol is the substrate. In addition, this method employs a so-called enhancer, which acts as a catalyst. Enhancers speed the oxidation of the luminol by HRP by as much as 1,000 times. Thus, HRP oxidation of luminol as enhancement leads to eventual light by luminol.

134

Y. F. (Wayne) Wang

Electrochemiluminescent (ECL) Electrochemiluminescence (ECL) is another antibody detection method, which is a technology similar to ELISA except that the secondary antibody is labeled with a chemiluminescent label ruthenium (Ru) and magnetic beads provide greater surface for soluble and target capture and separation [2]. Electron transfer between the Ru atom and the substrate tripropylamine (TPA) results in photon production. Excitation resulting in light emission can be detected by photon detector that detects electrochemiluminescent signal in an electrochemical flow cell for the magnetic-bead-Ru-tagged immune complex. The magnetic beads are usually small and spherical, ranging from a few nanometers to micrometers in sizes. The advantage of magnetic beads that contain paramagnetic magnetite (FE3O4) is the capability of rapid separation of captured antigen-antibody complex when placed in a magnetic field. The ECL assay, for example, uses immunomagnetic separation (IMS by ORIGEN system, IGEN) and the magnetic ECL detection system [11–13]. Detection of ECL is accompanied by a heavy metal chelate ruthenium (Ru) conjugated to a detector antibody. Initially, Ru and tripropylamine (TPA) in the buffer are oxidized at the surface of an anode when an electric field is applied to the electrode. TPA loses a proton and becomes a reducer, which causes Ru to enter a high-energy state by a high-energy electron transfer from the electron carrier TPA. A rapid electron transfer reaction between the substrate TPA and the Ru atom occurs, resulting in the production of photons in light transmission, which in turn is sensed by the photon detector at 620 nm. A linear dynamic ranges spanning six orders of magnitude [14] and thus makes the ECL assay an excellent method for antibody detection.

Fluorescent Immunoassays Fluorescent immunoassays use fluorescent methods for detection and can be categorized into four groups: (1) direct fluorescent assay (DFA), (2) indirect immunofluorescence (IFA), (3) time-resolved fluorescence (TRF), and (4) flow cytometry (FC). DFA is commonly used for antigen testing and thus will not be discussed in this chapter. IFA such as the slide method for microscopic examination under the UV light is used mainly for antibody detection rather than antigen detection and will not be discussed in this chapter. However, IFA techniques such as those used in the TRF and FC (Table 1) will be discussed briefly. Time-Resolved Fluorescence (TRF) assays use an lanthanide chelate containing an europium (Eu3+) label having unique properties such as a long fluorescence decay time in order to lower background signals. TRF is similar to ELISA, except that the capture antigen is affixed to the solid phase and mixed with the specimen and the complex, if any, is further mixed to diluted detector antibody that has been labeled with a lanthanide chelate such as europium or samarium. Addition of a low-pH

Antibody Detection: Principles and Applications

135

enhancement solution results in dissociation of lanthanide from the labeled compound and becoming highly fluorescent [15, 16]. TRF exploits the differential fluorescence life span of lanthanide chelate labels compared to background fluorescence. Lanthanide chelate labels have an intense long-lived fluorescence signal and a large stokes shift that allows assays to have a very high signal-to-noise ratio and excellent sensitivity [17]. TRF produces its signal through the excitation of the lanthanide chelate by a specific wavelength of light. Fluorescence with a pulse of excitation energy, repeatedly and reproducibly, is initiated in TRF. Flow Cytometry (FC) is commonly used IFA method. The first use of flow cytometry for analysis of microsphere-based immunoassays was published in 1977 [18, 19]. Initially, different-sized microspheres were used for simultaneous analysis of different analytes [19]. A fluorescent probe is added to a liquid suspension with sample, which is then streamed past a laser beam where the probe is excited. A detector analyzes the fluorescent properties of the sample as it passes through the laser beam. Using the same laser excitation source, the fluorescence may be split into different color components so that several different fluorophores can be measured simultaneously and analyzed by specialized software. A flow cytometer has the ability to discriminate different particles on the basis of size or color, thus make the multiplexed analysis possible with different microsphere populations in a single tube and in the same sample at the same time. The FC technology is the basis of the multiplex flow immunoassay.

Multiplex Flow Technology Multiplex flow immunoassay/multiplex bead immunoassay has been called various names; these include multi-analyte profile (xMAP) technology, multiplexed particle- based flow cytometric assays, fluorescent microsphere immunoassay (MIA), fluorescence covalent microbead immunosorbent assay (FCMIA), multiplexed indirect immunofluorescence assay, and/or multiplex flow cytometry. This two-step suspension method is based on fluorescent detection technique known as the FlowMetrix analysis system [20]. Systems that use xMAP technology are able to perform assays on the surface of color-coded beads (microspheres) that are covered with capture antigens that react with the target antibodies. These microbeads with special surface binding characteristics are created using a dying process that allows up to 100 unique dye ratios to be used. These specific dyes permeate the polystyrene microspheres that are 5.5 μm (5.5 micron) in diameter and are composed of polystyrene and methacrylate and thus are able to provide surface carboxylate functional groups on the surface. Each antigen is covalently linked by a carbodiimide conjugation method [21] such as EDC coupling method to beads of uniform size, which were colored with different amounts of red and orange fluorescent dyes (in a unique ratio) to allow for discrimination based upon the relative emission intensities at the wavelengths of the two fluorescent dyes.

136

Y. F. (Wayne) Wang

There are 64 different ratios of red and orange fluorescence, which thus are able to identify 64 distinctly colored sets of microspheres. Differently colored microsphere sets can be individually coupled via the surface carboxylate moiety to a specific probe for a unique target. The flow cytometer analyzes individual microspheres by size and fluorescence, distinguishing three fluorescent colors: green (530 nm), orange (585 nm), and red (> 650 nm) simultaneously. Microsphere size, determined by 90-degree light scatter, is used to eliminate microsphere aggregates from the analysis. All fluorescent molecules are labeled with a green-emitting fluorophore. Each fluorochrome has a characteristic emission spectrum, requiring a unique compensation setting for spillover into the orange fluorescence channel. Green-emitting fluorochrome can be used as a reporter. Microspheres conjugated with antigen can be added to well, in the sample, as well as the fluorescein-conjugate [red-phycoerythrin (R-PE) through biotin and streptavidin] antispecies detector or secondary antibody [22]. The red laser excites specific dyes to identify the analyte [red and orange fluorescent dyes (detected by FL2/FL [3]; the green laser excites a different dye to quantify the result (a green fluorescent reporter dye FL1) [23, 24]. The fluorescence emission of each bead of the specific antigen was determined with a fluorescence-activated cell scanner (FACScan, Becton-Dickinson), a benchtop flow cytometer (multiparameter flow cytometer that is based on a single 488-nm excitation laser), with FlowMetrix hardware for data acquisition and analysis (Luminex Corp., Austin, Texas). The software allows rapid classification of microsphere sets on the basis of the simultaneous gating on orange and red fluorescence. The Luminex instrument is a dual-laser flow analyzer. The first laser excites the fluorochrome mixture intrinsic to the microspheres, enabling the bead identity to be determined as the beads pass single file through the laser path in the flow cell. The second laser excites the extrinsic fluorochrome (R-PE) that is covalently attached to the secondary antibodies. The dual lasers allow the operator to mix beads with different antigens together in a well of a filter plate, thus enabling multiplex analysis of different antibody specificities at one time. Orange and red fluorescence are used for microsphere classification, and green fluorescence is used for analyte measurement [20]. The multiplex flow immunoassay has been useful for detection of multiple antibodies in a single assay.

Comparison and Contrast of These Techniques A comparison and contrast of these immunoassay techniques is shown in Table 1. • EIA or ELISA are relatively inexpensive, can be adapted for high-throughput use, and used to be commonly performed in both research and clinical laboratories. The enzymes and substrates used for ELISA frequently are unstable and require specialized storage to maintain activity. Most commercial products have been validated and have successfully dealt with this issue. The manual format of ELISA has more hands-on time and can only measure one analyte at a time. Automated


•

•

•

•

137

EIA system using the batched samples can be useful for large screening purpose and can be used as open platform for different tests; however, automated EIA systems cannot meet the demand of random access and “stat test” requirements. Rapid test or handheld system: The signal-to-noise ratio and limit of detection of antibody has been improved for handheld systems to overcome the concern of lower sensitivity as compared to regular EIA. Many handheld systems have been approved and are in use for rapid diagnosis of HIV. The better the avidity and affinity of the antibody, the more sensitive and specific the assay. The handheld system makes point-of-care antibody detection available for certain clinical and in the resource-limited settings. In general, the limitation is that only one or two agents that can be detected per assay strip with certain sensitivity levels. Another limitation is that assessment of a result is qualitative and subject to the interpretation. The improved assay can be read by using the reader. Chemiluminescent (CLIA) detection: The use of a more sensitive detection method such as chemiluminescence allows for a faster assay system, as well as a lower detection limits. These assays are often more sensitive than enzyme-based immunoassays. CLIA techniques have been widely accepted and implemented for automation because assay samples typically contribute little or no native background chemiluminescence and because detection procedure is simple. It requires no excitation source (as does fluorescence and phosphorescence), only a single light (photon) detector such as a photomultiplier tube. Most samples have no “background” signal, i.e., they do not themselves emit light. No interfering signal limits sensitivity [25–27]. Most chemiluminescent reactions are labeling either with a chemiluminescent compound or with an enzyme and using a chemiluminescent substrate, so as most commercially developed immunoassays that are of CLIA type [4, 9] as shown in Table 2. CLIA-based method can be used for automation, random access, and stat detection of antibody in the clinical setting and linked to automated specimen track system. Electrochemiluminescent (ECL): ECL background signal is constant with time, and steady-state ECL signal is proportional to rate of substrate turnover, which is different with the colorimetric background signal that accumulates with time. Light is the consequence of chemical reaction, luminol undergoes oxidative bond cleavage to yield an excited state species that decays by a radioactive pathway, and HRP (in the conjugate reagent) catalyzes the one-electron oxidation of luminol and expends hydrogen peroxide. The reaction does not suffer from the same surface steric and diffusion limitations as compared to conventional EIA or ELISA as the magnetic beads provide a greater surface area. Fluorescence immunoassay and indirect fluorescence assay (IFA): Fluorescent detection will allow more sensitive or faster detection than colorimetric methods. However, it might suffer from possible high background contamination due to the intrinsic fluorescent of some proteins and light-scattering effects. While simple to perform and requiring minimal equipment and reagents, significant expertise is necessary to interpret the results of IFA by slide microscopic method [28]. IFA can be performed if batching is not required. Individual or semiautomated enzyme-linked fluorescent immunoassay has been used in clinical laboratories.

138

Y. F. (Wayne) Wang

Table 2 Types of commercially available automated antibody detection systems Method Colorimetric

Detection method Enzyme colorimetric

Chemiluminescence (CL) CLIA immunoassay

Enhanced CLIA Electro-CLIA (ECL) Fluorescence

Fluorescence

Flow cytometry (FC) Multiplex (bead) flow immunoassay

Automated system (company) Evolis (Bio-Rad) ETI-Max (DiaSorin) Triturus (Grifols) DSX (Alere) DS2 (INOVA) QUANTA-Lyser (INOVA) ACCESS (Beckman) UniCel DxI (Beckman) ADVIA Centaur (Siemens) Architect (Abbott) Immulite (Siemens) Liaison (DiaSorin) VITROS (Ortho) Cobas (Roche) ORIGEN (IGEN, Roche) AxSYM (Abbott) VIDAS (bioMerieux) Nexgen Four (Adaltis) FACScan (Becton- Dickinson) AtheNA Multi- Lyte (Zeus, Alere) Bio-Plex (Bio-Rad)

High throughput Yes

Full automation Yes

Yes

Yes

Yes Yes

Yes Yes

Yes

Yes

Note: Many can handle antibody detection assays such as anti-HIV, anti-HAV, anti-HCV, anti- HBs, anti-HBc, CMV, and rubella

• Time-resolved fluorescence (TRF) and flow cytometry (FC): The limitation for TRF is similar to that of ELISA. In addition, dedicated measuring instrument and rigorous washing techniques are important to avoid lanthanide contamination, since lanthanide label is highly fluorescent [16]. A major strength of FC technology is its ability to be multiplexed with little or no loss of sensitivity [23, 29]. The system itself is relatively complicated and requires training and expertise to operate. • Multiplex flow immunoassay: Unlike traditional ELISA and other immunoassays that allow one test for each specific antibody at one time, many antibodies can be measured at the same time, in a single well or tube by using multiplex flow


139

technology. Unlike general flow cytometry on different sizes of beads, this technology detects identically sized microspheres with 2 different dyes, emitting in 2 different wavelengths, allows aggregates to be distinguished, and permits discrimination of at least 64 different sets of microspheres. Due to multiplexing, this technology delivers more data with comparable results to ELISA within the same sample. Automation of multiplex flow immunoassay has been developed and has been used in the clinical setting.

pplication of Immunodiagnostic Techniques in Diagnostic A Microbiology Clinical Applications Clinical application of immunodiagnostic techniques can be best demonstrated by immunoassays for HIV [7, 30] and hepatitis. Immunoassays have been developed to detect anti-HIV antibodies or viral antigens present in serum, plasma, dried blood spots, urine, and saliva. Assay formats range from EIAs, ELISA-based Western blot assays, IFA assays, and even rapid handheld immunoassays. In general, EIA remains the most widely used serologic test for detecting antibodies to HIV. Thus, HIV-1, 2, and O or HIV antigen-antibody combo immunoassays represent the advances in antibody detection technologies to detect and identify infectious agents [31]. Immunoanalyzers for broad application range will help meet the challenges of diagnosis of infectious diseases through the use of automation, random access, multiplexing, and high throughput. The main focus of this section of clinical application will be general utilization of technologies and automation in terms of methods for antibody detection. EIA Detection EIA or ELISA are still the methods of choice for large-scale investigations during outbreaks or for epidemiological surveillance studies. Because of its relative simplicity and good sensitivity, ELISA has been used for screening large numbers of small-volume samples and has had great impact in epidemiology and in the diagnosis of infection, particularly in the diagnosis of the difficult bacteria and viruses such as West Nile (WN) virus [32, 33], not to mention that these assays have been used extensively in HIV and hepatitis testing [7]. There are continuous developments in EIAs for screening and diagnosis of HIV. Third-generation EIAs include recombinant or synthetic peptide antigens derived from both HIV-1 and HIV-2 as capture antigens and allow detection of both IgG and IgM, which enabled improved sensitivity to early detection of HIV antibody [34]. FDA approved the first fourth-generation assay for use in the United States in June 2010, which detects both HIV antibodies and HIV p24 antigen. The fourthgeneration HIV antigen-antibody combo assay is designed to detect acute HIV infection even in those who have not yet begun to produce HIV-specific antibodies [35].

140

Y. F. (Wayne) Wang

Other than HIV, IgM and IgG capture ELISAs have been developed and used as the most useful and widely used tests for diagnosis of arboviral encephalitis [36]. Serum or CSF immunoglobulin M (IgM) testing has been recommended in the diagnosis of West Nile virus (WNV) infection [36]; however, issue of cross-reactivity especially with the flaviviruses has been reported [37, 38]. The plaque reduction neutralization test is considered as the gold standard test and can be used to confirm a positive ELISA test [37, 39]. This combination of assays is highly sensitive and specific, but performing a complete panel of ELISAs requires 2–3 working days to complete, as overnight incubations are deemed necessary to enhance sensitivity. IFAs may also be used for diagnosis, but they are not suitable for a high throughput of specimens and they are less sensitive than ELISA. Alternative methods that allow a more rapid detection of anti-WNV virus antibodies now exist, which include an anti-WNV immunoglobulin G (IgG) optical fiber immunoassay that uses biosensors and chemiluminescence [37, 40]. Immunoblotting Method Cross-reactivity could result from an antibody that binds to structurally distinct but similar epitopes present on different antigens or result from an antibody that binds to structurally identical epitopes on different antigens. This is why confirmatory tests are needed for certain tests such as HIV and have used more specific assays such as the Western blot [41].The separated HIV-1 proteins are electrotransferred to a nitrocellulose membrane. If antibodies to any of the major HIV-1 antigens are present in the specimen, bands corresponding to the HIV-1 proteins (p) or glycoproteins (gp) such as gp24, gp41, or gp120 will be seen on the nitrocellulose strip. Antibodies can thus be detected by using enzyme- conjugated secondary antibody (to human IgG) and demonstrated by darkly colored lines on the membrane under the substrate. However, EIAs with integration of p24 antigen often detect HIV infection earlier than Western blotting, and with the improvement in the sensitivity and the specificity of EIAs and rapid tests, the role of Western blotting in HIV diagnosis has been diminished and replaced with rapid supplemental test [42]. The recombinant immunoblot assay (RIBA) or strip immunoblot assay (SIA) for detecting NS5 and c33c recombinant proteins and c100p, 5-1-1p, and c22p synthetic peptides of hepatitis C virus (HCV) is intended as a supplemental test that was originally developed to confirm the results of a positive EIA test. However, the specificity of the EIA results that exceed particular signal/cutoff ratios is extremely high, and given the availability of nucleic acid testing, the role for RIBA testing in HCV diagnosis has been limited or not recommended [43]. Rapid or Handheld Assay Rapid immunoassays/handheld immunoassays have evolved significantly in the past decade. Development of self-contained miniaturized devices allows an immunoassay to be performed in the field or at a point-of- care setting. Currently available handheld immunoassays include those used for detection of antibodies against Epstein-Barr virus, Helicobacter pylori, and HIV. Starting in 2002, the FDA has approved many rapid HIV tests for use in the United States; these rapid tests utilize either immunochromatography (lateral flow)


141

or immunoconcentration (flow-through) techniques [44]. A study evaluating six rapid HIV tests showed that all tests present high sensitivity and specificity compared to the third-generation EIAs [44]. In 2010, the OraQuick® HCV Rapid Antibody Test was approved by FDA for HCV diagnosis. This assay utilizes antigens from the core and NS3 and NS4 regions of the HCV genome and has shown to have sensitivity and specificity that are comparable to laboratory EIA [45]. Handheld rapid HIV testing can be improved by a combined antigen and antibody detection, which has been designed for detection of acute infection [46–48]. Chemiluminescence Chemiluminescence will be discussed in the automation section. Fluorescence Immunoassay Although TRF has been used to detect various viral antigens, recent studies have utilized this method to measure immunity or antibody response after vaccination against varicella zoster virus [49–51]. Another type of fluorescent technology in use is fluorescent polarization (FP), which is a phenomenon seen when polarized light excites a fluorescent dye causing photons to be emitted in the same plane as the exciting light. Because of the limited need for sample processing, FP antibody detection assays are particularly useful. Multiplex Flow Immunoassay Diagnosis of infection often requires testing for multiple antibodies or multiple markers. Multiplex flow or bead immunoassays have been developed and are designed for clinical use [52–55]. Bead-based immunoassays allow a quantitative and qualitative analysis of multiple targets rapidly with excellent sensitivity and specificity [56]. This technique includes detection of antibodies against infectious agents, disease surveillance, and screening of donated blood [57]. It has been used to test a panel of respiratory viruses, including influenza A and B viruses; adenovirus; parainfluenza viruses 1, 2, and 3; and respiratory syncytial virus [58]. When compared with the ELISA method [59, 60], this technology usually has good positive correlation, better sensitivity and speed, and enhanced dynamic range. It uses smaller sample volume and can be multiplexed, that is, measures more than one analyte simultaneously [61, 62]. The multiplex technology has been successful in detecting targets from dried blood spot specimens: antibodies to HIV-1 p24, gp160, and gp120 eluted from dried blood spot specimens from newborns were detected simultaneously [63, 64], as well as the HCV antibody and hepatitis B virus surface (HBs) antigen with HIV antibodies [65]. The multiplex technology is also applied to vaccine development by testing antibody response. Simultaneous measurement of antibodies to 23 pneumococcal capsular polysaccharides (PnPs) has been developed [61], which demonstrated results similar to those of another xMAP assay developed for antibodies to PnPs [66]. The assay simultaneously determines serum IgG concentrations to 14PnPs serotypes. A multiplexed bead-based immunoassay to quantify 17 pneumococcal proteins was developed in pursuit of a pneumococcal vaccine that would provide protection regardless of serotypes, as opposed to the currently available vaccines [67].

142

Y. F. (Wayne) Wang

An advantage of the 96-well plate Luminex assay format is that it lends itself to automation, such as the Tecan Genesis liquid handler; this allows the assay to be automated. This automation, such as seen with the Bio-Plex system (Bio-Rad Laboratories, Hercules, CA, USA), employs multiple flow technology and allows individual and multiplex analysis for up to 100 different analytes in a single microtiter well [23]. A multiplexed bead assay was evaluated for assessment of Epstein- Barr virus (EBV) immunologic status using FDA-cleared IgM and IgG EBV assays on a fully automated Bio-Plex 2200 system (Bio-Rad). Concordance between results generated by the Bio-Plex system and conventional assays showed 97.9%, 91.4%, and 96.9% agreement for viral capsid antigen (VCA) IgM, VCA IgG, and EBNA-1 IgG assays, respectively [68].

Automation Commercially available immunoanalyzers have been widely used to facilitate the analysis of large numbers of samples because of improvements in the throughput achieved by automation (Table 2). The first generation of immunoassay systems was developed more than a decade ago in order to automate what previously had been multiple labor-intensive manual laboratory tests. Advances in clinical immunology, as well as the demand for faster turnaround times and reduced costs, have prompted technology developments in immunoassays and in integrated immunochemistry analyzers. These high-volume immunoassays have made a significant impact on laboratory performance by reducing errors, reducing turnaround times, and reducing the labor requirements for those tests. Capability of interfacing with laboratory information system (LIS) further enables the utilization of these commercial systems. The ideal immunoassay system would have the following capabilities in order to provide optimal productivity and a comprehensive disease-focused menu: (1) no- pause loading of all reagents, samples, and supplies, (2) continuous sample loading for fast turnaround time, (3) high-throughput process efficiency, (4) random access and reduced operator intervention, (5) minimal hands-on time with large onboard capacity for reagents, and (6) ability to interface with the LIS for increased efficiency with easy-to-use software. The above features are particularly critical for HIV assays and comprehensive hepatitis antibody detection assays. Most chemiluminescent reactions can be adapted to this assay format by labeling either with a chemiluminescent compound or with an enzyme and using a chemiluminescent substrate. Most commercially developed immunoassays use a chemiluminescent format (Table 2). For example, Lumi-Phos 530 of Luminol CLIA is used as the detection reagent in the access immunoassay analyzer (Beckman Coulter Inc., Fullerton, CA, USA). Lumigen PPD and enhancer are incorporated in the chemiluminescent detection reagent used in the Immulite Immunoassay Analyzer from


143

Siemens. The AxSYM immunoassay system (Abbott) is based on the microparticle enzyme immunoassay technology [69–71]. Immulite (Siemens, Deerfield, IL, USA) is a benchtop immunoassay analyzer with continuous random-access capabilities that uses enzyme-amplified chemiluminescence chemistry for antibody or antigen detection [72]. Multiple high-throughput systems that can provide streamlined operations to reduce total processing time are available in the market, and some are capable in running different types of immunoassays. Many types of immunoassays can be developed on the automated systems for hepatitis virus A, B, and C, cytomegalovirus, and HIV assays. With the availability of EIA, CLIA, and multiplex flow immunoassay that allow automated high-throughput testing for syphilis, some laboratories have adopted reverse screening algorithm in which a treponemal EIA or CLIA is followed by a nontreponemal testing for positive specimens in order to reduce the time and labor required for syphilis screening [73]. This could create problems, especially when the treponemal-specific screening test is positive but the nontreponemal tests that follow are negative [59]. The recommendation of the CDC on such discordant results is to use the Treponema pallidum particle agglutination (TP-PA) assay, and if the TP-PA is nonreactive, syphilis is considered to be unlikely [73]. The fluorescent treponemal antibody (FTA) is no longer widely used now [73, 74].

Summary Antibody detection technologies have been developed to identify host response to the infectious agents or microorganisms and are widely used for the laboratory diagnosis of infectious diseases with improved sensitivity and specificity. Antibody detection methods have been utilized for detection and immune response of slow- growing, difficult-to-culture, uncultivatable, or emerging infectious agents, which include but not limit to (1) viruses including HIV, hepatitis A virus, hepatitis B virus, hepatitis C virus, cytomegalovirus, Epstein-Barr virus, herpes simplex virus, mumps, rubella, rubeola virus, and West Nile virus; (2) parasite including Toxoplasma gondii; and (3) bacteria including Helicobacter pylori, Mycoplasma pneumoniae, and syphilis. Antibody detection methods such as CLIA, ECL, and TRF detection formats have become the predominant technologies for the diagnosis of infectious diseases as opposed to conventional EIA or ELISA methods. Emerging antibody detection methods such as rapid or handheld assay and multiplexed flow cytometry have proven to be useful in the clinical setting.

144

Y. F. (Wayne) Wang

References 1. Yalow RS, Berson SA. Immunoassay of endogenous plasma insulin in man. J Clin Invest. 1960;39:1157–75. 2. Peruski AH, Peruski LF Jr. Immunological methods for detection and identification of infectious disease and biological warfare agents. Clin Diagn Lab Immunol. 2003;10:506–13. 3. Andreotti PE, Ludwig GV, Peruski AH, Tuite JJ, Morse SS, Peruski LF Jr. Immunoassay of infectious agents. BioTechniques. 2003;35:850–9. 4. Kricka LJ. Chemiluminescent and bioluminescent techniques. Clin Chem. 1991;37:1472–81. 5. Engvall E, Perlmann P. Enzyme-linked immunosorbent assay, ELISA: quantitation of specific antibodies by enzyme-labeled anti-immunoglobulin in antigen-coated tubes. J Immunol Methods. 1972;109:129–35. 6. Engvall E, Perlmann P. Enzyme-linked immunosorbent assay (ELISA). Quantitative assay of immunoglobulin G. Immunochemistry. 1971;8:871–4. 7. Wang YF, Eaton ME, Schuetz AN, Nesheim SR. Human immunodeficiency virus. Washington, DC: ASM Press; 2009. 8. Zuk RF, Ginsberg VK, Gouts T, et al. Enzyme immunochromatography: a quantitative immunoassay requiring no instrumentation. Clin Chem. 1985;31:1144–50. 9. Kricka LJ. The clinical and research potential of bioluminescence and chemiluminescence in medicine. Chichester: Wiley; 1996. 10. Weeks I, Wodhead JS. Chemiluminescent assays based on acridinium labels. Chichester: Wiley; 1991. 11. Yu H. Comparative studies of magnetic particle-based solid phase fluorogenic and electrochemiluminescent immunoassay. J Immunol Methods. 1998;218:1–8. 12. Haukanes BL, Kyam B. Application of magnetic beads in bioassays. Biotechnology (N Y). 1993;11:60–3. 13. Blackburn GF, Shah HP, Kenten JH, et al. Electrochemiluminescence detection for development of immunoassays and DNA probe assays for clinical diagnosis. Clin Chem. 1991;37:1534–9. 14. Yang H, Leland JK, Massey RJ. Electrochemiluminescence: a new diagnostic and research tool. ECL detection technology promises scientist new “yardsticks” for quantification. Biotechnology (N Y). 1994;12:193–4. 15. Peruski AH, Johnson LH, Peruski LF Jr. Rapid and sensitive detection of biological warfare agents using time-resolved fluorescence assays. J Immunol Methods. 2002;263:35–41. 16. Aggerbeck H, Norgaard-Pedersen B, Heron I. Simultaneous quantitation of diphtheria and tetanus antibodies by double antigen, time-resolved fluorescence immunoassay. J Immunol Methods. 1996;190:171–83. 17. Hemmila I, Dakubu S, Mukkala VM, Siitari H, Lovgren T. Europium as a label in time- resolved immuofluorometric assays. Anal Biochem. 1984;137:335–43. 18. McHugh TM. Flow microsphere immunoassay for the quantitative and simultaneous detection of multiple soluble analytes. Methods Cell Biol. 1994;42:575–95. 19. Horan PK, Wheeless LL. Quantitative single cell analysis and sorting. Science. 1977;198:149–57. 20. Fulton RJ, McDade RL, Smith PL, Kienker LJ, Kettman JR. Advanced multiplexed analysis with the FlowMetrix system. Clin Chem. 1997;43:1749–56. 21. Staros JV, Wright RW, Swingle DM. Enhancement by N-hydroxysulfosuccinimide of water- soluble carbodiimide-mediated coupling reactions. Anal Biochem. 1986;156:220–2. 22. Jones LP, Zheng HQ, Karron RA, Peret TCT, Tsou C, Anderson LA. Multiplex assay for detection of strain-specific antibodies against the two variable regions of the G protein of Respiratory Syncytial Virus. Clin Diagn Lab Immunol. 2002;9:633–8. 23. Vignali DA. Multiplexed particle-based flow cytometric assays. J Immunol Methods. 2000;243:243–55. 24. Mandy FF, Nakamura T, Bergeron M, Sekiguchi K. Overview and application suspension array technology. Clin Lab Med. 2001;21:713–29.


145

25. Nieman T. Chemiluminescence: theory and instrumentation, overview. In: Encyclopedia of analytical science. Orlando: Academic Press; 1995. p. 608–13. 26. Campbell AK. Detection and quantification of chemiluminescence. Chichester: Ellis Horwood; 1988. 27. Berthrold F. Instrumentation for chemiluminescence immunoassays. Boca Raton: CRC Press; 1990. 28. Nuwayhid NF. Laboratory tests for detection of human immunodeficiency virus type 1 infection. Clin Diag Lab Immunol. 1995;2:637–45. 29. Carson RT, Vignali DA. Simultaneous quantitation of 15 cytokines using a multiplexed flow cytometric assay. J Immunol Methods. 1999;227:41–52. 30. Nielsen K, Bryson YJ. Diagnosis of HIV infection in children. Pediatr Clin N Am. 2000;47:39–63. 31. Stafylis C, Klausner JD. Evaluation of two 4th generation point-of-care assays for the detection of Human Immunodeficiency Virus infection. PLoS One. 2017;12:e0183944. 32. MMWR. Guidelines for surveillance, prevention, and control of West Nile virus infection— United States. MMWR Morb Mortal Wkly Rep. 2000;49:25–8. 33. Nash D, Mostashari F, Fine A, et al. The outbreak of West Nile virus infection in the New York City area in 1999. N Engl J Med. 2001;344:1807–14. 34. Louie B, Pandori MW, Wong E, Klausner JD, Liska S. Use of an acute seroconversion panel to evaluate a third-generation enzyme-linked immunoassay for detection of human immunodeficiency virus-specific antibodies relative to multiple other assays. J Clin Microbiol. 2006;44:1856–8. 35. Pandori MW, Hackett J Jr, Louie B, et al. Assessment of the ability of a fourth-generation immunoassay for human immunodeficiency virus (HIV) antibody and p24 antigen to detect both acute and recent HIV infections in a high-risk setting. J Clin Microbiol. 2009;47:2639–42. 36. Tunkel AR, Glaser CA, Bloch KC, et al. The management of encephalitis: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis. 2008;47:303–27. 37. Dauphin G, Zientara S. West Nile virus: recent trends in diagnosis and vaccine development. Vaccine. 2007;25:5563–76. 38. Martin DA, Biggerstaff BJ, Allen B, Johnson AJ, Lanciotti RS, Roehrig JT. Use of immunoglobulin M cross-reactions in differential diagnosis of human flaviviral encephalitis infections in the United States. Clin Diagn Lab Immunol. 2002;9:544–9. 39. Lindsey HS, Calisher CH, Mathews JH. Serum dilution neutralization test for California group virus identification and serology. J Clin Microbiol. 1976;4:503–10. 40. Herrmann S, Leshem B, Landes S, Rager-Zisman B, Marks RS. Chemiluminescent optical fiber immunosensor for the detection of anti-West Nile virus IgG. Talanta. 2005;66:6–14. 41. Jackson JB, Parsons JS, Nichols LS, Knoble N, Kennedy S, Piwowar EM. Detection of human immunodeficiency virus type 1 (HIV-1) antibody by western blotting and HIV-1 DNA by PCR in patients with AIDS. J Clin Microbiol. 1997;35:1118–21. 42. Fordan S, Bennett B, Lee M, Crowe S. Comparative performance of the Geenius™ HIV-1/HIV-2 supplemental test in Florida’s public health testing population. J Clin Virol. 2017;91:79–83. 43. Ghany MG, Strader DB, Thomas DL, Seeff LB. Diagnosis, management, and treatment of hepatitis C: an update. Hepatology. 2009;49:1335–74. 44. Delaney KP, Branson BM, Uniyal A, et al. Evaluation of the performance characteristics of 6 rapid HIV antibody tests. Clin Infect Dis. 2011;52:257–63. 45. Lee SR, Yearwood GD, Guillon GB, et al. Evaluation of a rapid, point-of-care test device for the diagnosis of hepatitis C infection. J Clin Virol. 2010;48:15–7. 46. Livant E, Heaps A, Kelly C, Maharaj R, Samsunder N, Nhlangulela L, Karugaba P, Panchia R, Marrazzo J, Chirenje ZM, Parikh UM; VOICE Study Team. The fourth generation AlereTM HIV Combo rapid test improves detection of acute infection in MTN-003 (VOICE) samples. J Clin Virol. 2017;94:15–21. 47. Masciotra S, Luo W, Westheimer E, Cohen SE, Gayd CL, Hall L, Pan Y, Peters PJ, Owen SM. Performance evaluation of the FDA-approved DetermineTM HIV-1/2Ag/Ab Combo assay using plasma and whole blood specimens. J Clin Virol. 2017;91:95–100.

146

Y. F. (Wayne) Wang

48. Adams S, Luo W, Wesolowski L, Cohen SE, Peters PJ, Owen SM, Masciotra S. Performance evaluation of the point-of-care INSTITMHIV-1/2 antibody test in early and established HIV infections. J Clin Virol. 2017;91:90–4. 49. Maple PA, Gray J, Breuer J, Kafatos G, Parker S, Brown D. Performance of a time-resolved fluorescence immunoassay for measuring varicella-zoster virus immunoglobulin G levels in adults and comparison with commercial enzyme immunoassays and Merck glycoprotein enzyme immunoassay. Clin Vaccine Immunol. 2006;13:214–8. 50. Chris Maple PA, Gray J, Brown K, Brown D. Performance characteristics of a quantitative, standardised varicella zoster IgG time resolved fluorescence immunoassay (VZV TRFIA) for measuring antibody following natural infection. J Virol Methods. 2009;157:90–2. 51. McDonald SL, Maple PA, Andrews N, et al. Evaluation of the time resolved fluorescence immunoassay (TRFIA) for the detection of varicella zoster virus (VZV) antibodies following vaccination of healthcare workers. J Virol Methods. 2011;172:60–5. 52. Gomez E, Jespersen DJ, Harring JA, Binnicker MJ. Evaluation of the Bio-Rad BioPlex 2200 syphilis multiplex flow immunoassay for the detection of IgM- and IgG-class antitreponemal antibodies. Clin Vaccine Immunol. 2010;17:966–8. 53. Berth M, Bosmans E. Comparison of three automated immunoassay methods for the determination of Epstein-Barr virus-specific immunoglobulin M. Clin Vaccine Immunol. 2010;17:559–63. 54. Binnicker MJ, Jespersen DJ, Rollins LO. Evaluation of the Bio-Rad BioPlex measles, mumps, rubella, and varicella-zoster virus IgG multiplex bead immunoassay. Clin Vaccine Immunol. 2011;18:1524–6. 55. Hatchette TF, Scholz H, Bolotin S, Crowcroft NS, Jackson C, McLachlan E, Severini A. Calibration and evaluation of quantitative antibody titers for measles virus by using the BioPlex 2200. Clin Vaccine Immunol. 2017;24pii:e00269–16. 56. Fonseca BP, Marques CF, Nascimento LD, et al. Development of a multiplex bead-based assay for detection of hepatitis C virus. Clin Vaccine Immunol. 2011;18:802–6. 57. Kellar KL, Iannone MA. Multiplexed microsphere-based flow cytometric assays. Exp Hematol. 2002;30:1227–37. 58. Martins TB. Development of internal controls for the Luminex instrument as part of a multiplex seven-analyte viral respiratory antibody profile. Clin Diagn Lab Immunol. 2002;9:41–5. 59. Binnicker MJ, Jespersen DJ, Rollins LO. Treponema-specific tests for Serodiagnosis of syphilis: comparative evaluation of seven assays. J Clin Microbiol. 2011;49:1313–7. 60. Quinn CP, Semenova VA, Elie CM, et al. Specific, sensitive, and quantitative enzyme-linked immunosorbent assay for human immunoglobulin G antibodies to anthrax toxin protective antigen. Emerg Infect Dis. 2002;8:1103–10. 61. Biagini RE, Schlottmann SA, Sammons DL, et al. Method for simultaneous measurement of antibodies to 23 pneumococcal capsular polysaccharides. Clin Diagn Lab Immunol. 2003;10:744–50. 62. Robin L, Mboumba Bouassa RS, Nodjikouambaye ZA, Charmant L, Matta M, Simon S, Filali M, Mboup S, Bélec L. Analytical performances of simultaneous detection of HIV-1, HIV-2 and hepatitis C- specific antibodies and hepatitis B surface antigen (HBsAg) by multiplex immunochromatographic rapid test with serum samples: a cross-sectional study. J Virol Methods. 2018;253:1–4. 63. Bellisario R, Colinas RJ, Pass KA. Simultaneous measurement of antibodies to three HIV-1 antigens in newborn dried blood-spot specimens using a multiplexed microsphere-based immunoassay. Early Hum Dev. 2001;64:21–5. 64. Faucher S, Martel A, Sherring A, et al. Protein bead array for the detection of HIV-1 antibodies from fresh plasma and dried-blood-spot specimens. Clin Chem. 2004;50:1250–3. 65. Lukacs Z, Dietrich A, Ganschow R, Kohlschutter A, Kruithof R. Simultaneous determination of HIV antibodies, hepatitis C antibodies, and hepatitis B antigens in dried blood spots—a feasibility study using a multi-analyte immunoassay. Clin Chem Lab Med. 2005;43:141–5.


147

66. Pickering JW, Martins TB, Greer RW, et al. A multiplexed fluorescent microsphere immunoassay for antibodies to pneumococcal capsular polysaccharides. Am J Clin Pathol. 2002;117:589–96. 67. Shoma S, Verkaik NJ, de Vogel CP, et al. Development of a multiplexed bead-based immunoassay for the simultaneous detection of antibodies to 17 pneumococcal proteins. Eur J Clin Microbiol Infect Dis. 2011;30:521–6. 68. Binnicker MJ, Jespersen DJ, Harring JA, Rollins LO, Beito EM. Evaluation of a multiplex flow immunoassay for detection of epstein-barr virus-specific antibodies. Clin Vaccine Immunol. 2008;15:1410–3. 69. Fiore M, Mitchell J, Doan T, et al. The Abbott IMx automated benchtop immunochemistry analyzer system. Clin Chem. 1988;34:1726–32. 70. Hennig H, Schlenke P, Kirchner H, Bauer I, Schulte-Kellinghaus B, Bludau H. Evaluation of newly developed microparticle enzyme immunoassays for the detection of HCV antibodies. J Virol Methods. 2000;84:181–90. 71. Lazzarotto T, Galli C, Pulvirenti R, et al. Evaluation of the Abbott AxSYM cytomegalovirus (CMV) immunoglobulin M (IgM) assay in conjunction with other CMV IgM tests and a CMV IgG avidity assay. Clin Diagn Lab Immunol. 2001;8:196–8. 72. Li TM, Chuang T, Tse S, Hovanec-Burns D, El Shami AS. Development and validation of a third generation allergen-specific IgE assay on the continuous random access IMMULITE 2000 analyzer. Ann Clin Lab Sci. 2004;34:67–74. 73. MMWR. Discordant results from reverse sequence syphilis screening—five laboratories, United States, 2006–2010. MMWR Morb Mortal Wkly Rep. 2011;60:133–7. 74. Marangoni A, Sambri V, Storni E, D’Antuono A, Negosanti M, Cevenini R. Treponema pallidum surface immunofluorescence assay for serologic diagnosis of syphilis. Clin Diagn Lab Immunol. 2000;7:417–21.

Procalcitonin and Other Host-Response-Based Biomarkers for Evaluation of Infection and Guidance of Antimicrobial Treatment Philipp Schuetz, Ramon Sager, Yannick Wirz, and Beat Mueller

Host-Response Markers in Systemic Infections and Sepsis In systemic bacterial infections, pathogens and their antigens stimulate pro- and anti-inflammatory immune and host-response mediators that constitute the host defences and coordinate leukocyte recruitment to the site of acute infection [1]. These molecules have multiple adaptive and maladaptive functions including regulation of the osmotic and volume status, appetite, blood circulation, and food intake among others [2]. Precursors, mature forms, and degradation products of these various mediators leave the original site of action and enter the circulation, where, theoretically, they can all be measured. As surrogate host-response biomarkers, these substances mirror the extent and severity of an infection, with their levels falling upon resolution of the infection [3]. Significant attempts have been made to correlate the levels of different mediators with the presence of infection as potential diagnostic markers and to use these markers for risk stratification of patients [4].

Host-Response Markers and the Definition of Sepsis The recognition over 25 years ago that the host response plays an integral role in sepsis led to the definition of sepsis that has recently been updated to also include evidence of organ dysfunction as this has been shown to drive mortality in this patient population [5]. Unfortunately, the systemic inflammatory response syndrome (SIRS) variables (i.e. body temperature, heart rate, tachypnoea, and white P. Schuetz (*) · R. Sager · Y. Wirz · B. Mueller Medical University Department, Division of General Internal and Emergency Medicine, Kantonsspital Aarau, Aarau, Switzerland Medical Faculty, University of Basel, Basel, Switzerland © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_7

149

150

P. Schuetz et al.

blood count) used in the original definition from the early 1990s turned out to be less useful than anticipated, lacking sensitivity, specificity, and ease of clinical application [6]. Had a newer host-response biomarker been available when the SIRS criteria were initially put forth, it would arguably have been preferable to use white blood cell counts as a laboratory-based SIRS criterion. Indeed, specific immune and host-response markers (e.g. procalcitonin (PCT), presepsin, and lactate) measured on admission and during follow-up have been suggested to improve early sepsis recognition, severity assessment, and therapeutic decision-making for individual patients and may thereby allow transformation of bundled sepsis care based on prespecified protocols to more individualized patient management. Of these, PCT has generated much interest as it has a high diagnostic ability, helps in risk stratification of patients, and has been successfully used as a guide to antibiotic treatment decisions [4]. As for diagnostic accuracy, a meta-analysis from 2013 that included 3244 critically ill patients classified as experiencing sepsis or SIRS of noninfectious origin pooled the diagnostic power of PCT [7]. Studies between 1996 and 2011 were included and showed a good high discriminatory ability of PCT (area under the curve [AUC] of 0.85), with pooled sensitivity and specificity of 0.77 and 0.79, respectively [7].

Regulation of Procalcitonin During Infections Procalcitonin is released ubiquitously in response to endotoxin or mediators released in response to bacterial infections (e.g. interleukin [IL]-1β, tumour necrosis factor [TNF]-α, and IL-6) and correlates with the extent and severity of bacterial infections [8]. Because upregulation of PCT is attenuated by interferon gamma (INFγ), a cytokine released in response to viral infections, PCT is more specific for bacterial infections and may help to distinguish bacterial infections from viral illnesses. It shows a favourable kinetic profile for use as a clinical marker, with its levels promptly increasing within 6–12 h following infection and falling by half daily when the infection is controlled by the host immune system or through antibiotic therapy. Its levels correlate well with bacterial load and severity of infection, and its course predicts fatal outcome in patients with systemic infections and critically ill patients with sepsis, thus suggesting prognostic implications [9].

Procalcitonin as a Diagnostic Guide in Patients with Infections Procalcitonin has been investigated as a diagnostic guide for patients with infections in several observational studies focusing on different clinical situations and different types and sites of infections.

Procalcitonin and Other Host-Response-Based Biomarkers for Evaluation of Infection…

151

Blood Stream Infections Procalcitonin has been demonstrated to be very useful in the diagnosis of blood stream infections and bacteraemia [10–12]. It was superior to white blood counts and C-reactive protein (CRP) in its ability to distinguish blood contamination from true blood stream infection in patients with growth of coagulase-negative staphylococci in their blood cultures [10]. At a cut-off of 0.1ug/L, PCT had a very high sensitivity to correctly exclude true bacterial infection. Two other studies that focused on the use of PCT to predict bacteraemia in patients with urinary tract infections (UTI) [11] and pneumonia [12] revealed a PCT cut-off of 0.25ug/L to be most helpful to exclude bacteraemic disease, with a high negative predictive value in both settings.

Urinary Tract Infections Evidence for the utility of PCT in urinary tract infections comes primarily from the paediatric literature, where it has a similar sensitivity but superior specificity as compared to CRP for the prediction of pyelonephritis in children with febrile UTIs [13]. It correlates well with both the extent of renal involvement and with renal scarring. The utility of PCT in this setting was also investigated in a randomized trial [14].The study showed an antibiotic usage reduction of 30% with PCT-guided treatment compared to the standard treatment. The elaborated algorithm combined serum PCT concentration and quantitative pyuria measurement [14]. Patients were grouped on the basis of uncomplicated or complicated UTI and on the basis of the treatment setting (outpatient versus inpatient) resulting in differences in antibiotics administered and lengths of treatment or were subjected to a monitored approach with measurement of PCT and the degree of pyuria. There were no negative effects. The authors concluded that a PCT/pyuria-based approach is safe in terms of outcome and has the potential to reduce antibiotic consumption.

Endocarditis In patients with infectious endocarditis, two independent studies revealed that circulating PCT levels were elevated compared to noninfected patients [15, 16]. Unfortunately, a reliable PCT threshold for diagnosing or excluding infective endocarditis has not been proposed or tested in interventional studies. Importantly, subacute forms of endocarditis or prosthetic valve endocarditis may show different characteristics compared to acute forms due to their low inflammatory nature and possible biofilm production [17].

152

P. Schuetz et al.

Meningitis Several older studies that evaluated PCT-guided therapy in meningitis revealed that PCT-guided therapy reduces antimicrobial consumption during viral outbreaks. Two recent meta-analyses confirmed its accuracy in differentiating viral from bacterial meningitis [18, 19]. The most recent meta-analysis from 2016 included 2058 subjects and showed a sensitivity of 0.95 (95% CI, 0.89–0.97), a specificity of 0.97 (95% CI, 0.89–0.99), a positive likelihood ratio of 31.7 (95% CI, 8.0–124.8), and a negative likelihood ratio of 0.06 (95% CI, 0.03–0.11). The diagnostic performance was even better when PCT levels were combined with those of cerebrospinal fluid lactate. Serum PCT was found to be more sensitive and specific compared to cerebrospinal fluid PCT. Furthermore, PCT was useful for prognostication of poor outcome, for follow-up of treatment, and for differentiating other bacterial meningitis from tuberculous meningitis.

Intra-abdominal Infections Few studies have investigated the utility of PCT in intra-abdominal infections. While PCT as a marker promised to exclude perforation and ischaemia in obstructive bowel syndrome, its utility in acute appendicitis and pancreatitis was limited, and it was more helpful as a prognostic marker for severe disease and adverse outcome. While localized infections may not induce a massive upregulation of PCT, studies found PCT to be of diagnostic utility in patients with arthritis [20] and osteomyelitis, particularly when subtle increases were considered and a low PCT cut-off (0.1ug/L) was employed.

Febrile Neutropenia Multiple studies have evaluated the utility of PCT in patients with febrile neutropenia. A systematic review of 30 studies on the topic concluded that PCT has value as a diagnostic and prognostic tool in patients with febrile neutropenia but that due to differences in patient populations and study quality, further research is needed [21]. Importantly, it is worth noting that the production of PCT seems not to be attenuated by corticosteroids and that it does not rely on white blood cells. A study that included 102 critically ill patients with systemic infections in a medical ICU found significantly lower CRP and IL-6 levels, but similar PCT levels, in patients treated with systemic corticosteroids (20–1500 mg per day of prednisone parenterally) compared to untreated patients [22]. These observations were confirmed in healthy male volunteers who received different doses of prednisolone up to 30 mg before a

Procalcitonin and Other Host-Response-Based Biomarkers for Evaluation of Infection…

153

sepsis-like syndrome was induced with Escherichia coli lipopolysaccharide (LPS) injections [23]. While other biomarkers were significantly inhibited in a dose- dependent manner, levels of PCT showed no inhibition during the study. Whether this is also true for other corticosteroid doses, however, remains unknown. The value of PCT in febrile neutropenia may be as part of a combination with other biomarkers of bacterial infection such as IL-6 and IL-8, as shown in a small study of paediatric febrile neutropenia.

Antibiotic Stewardship Aided by the Use of Procalcitonin The clinical implications of the above-mentioned observational studies may be limited by differences in disease definitions and patient populations, the use of insensitive (semi-quantitative) PCT assays, methodological issues such as observer bias and selection bias, and issues with sample availability, co-infection, and colonization. To overcome these limitations, several randomized-controlled studies have investigated the utility of PCT in assisting with decisions regarding initiation and/or duration of antibiotic therapy (antibiotic stewardship). The benefit of PCT was measured through clinical outcomes, assuming that if the patient recovers without antibiotics, there was no relevant bacterial illness in need of chemotherapy.

Antibiotic Stewardship Algorithms All published studies on antibiotic stewardship have used similar clinical algorithms with recommendations for or against antibiotic treatment based on PCT cut-off ranges. For moderate-risk patients with respiratory tract infections seen in the emergency department (Fig. 1), algorithms recommended initiation and discontinuation of antibiotic therapy based on four different cut-off ranges ( 0.5 ng/mL). Initial antibiotics were withheld mostly in patients with low risk for systemic infection with acute bronchitis or exacerbation of COPD [ECOPD]). Clinical re-evaluation and repeat measurement of PCT were recommended after 6–24 h if the clinical condition did not improve spontaneously. If PCT values were higher and antibiotic therapy was initiated, repeated PCT measurements every 1–2 days, depending on the clinical severity of disease, were recommended, and antibiotics were discontinued using the same cut-off ranges or if a marked drop in PCT levels by 80–90% was seen if initial levels had been high (e.g. >5 μg/L). To ensure safety, specific criteria where this algorithm could be overruled, such as life- threatening disease or immediate need for ICU admission, were predefined. For high-risk patients in the ICU setting (Fig. 2), algorithms focused on discontinuation of antibiotic therapy if a patient showed a clinical recovery and PCT levels decreased to ‘normal’ levels, or by at least 80–90%.

154

P. Schuetz et al.

Procalcitonin algorithm for patients with low risk infections (eg, respiratory tract infections) PCT (ug/L)

Bacterial infection?

Recommendation regarding antibiotics (AB)

Important considerations and criteria to overrule

10

2 1

Very likely

0.5 Likely

AB YES!

AB Yes

0.25 0.1

Unlikely Very unlikely

AB No AB NO!

- Consider the course of PCT - If antibiotics are initiated: - Repeat PCT every other day, 7; stop antibiotics using the same cutoffs - If peak PCT levels are very high, sto when decreased 80-90% from peak level - If PCT remains high, consider case to be treatment failure - If antibiotics are withheld, repeat PCT after 6-24 hours - Initial antibiotics (overruling) can be considered in case of: - Respiratory or hemodynamic instability, severest comorbidities, and ICU admission - PCT < 0.1 ug/L: CAP with PSI V or CURB >3, or COPD with GOLD IV - PCT < 0.25 ug/L: CAP with PSI IV and V or CURB >2, or COPD with GOLD III and IV

0.01

Fig. 1 Procalcitonin algorithm for patients with low-risk infections (e.g. respiratory tract infections) Procalcitonin algorithm for stopping antibiotics in high risk ICU patients with sepsis PCT (ug/L)

Ongoing infection?

Recommendation for stopping antibiotics

Important considerations

10

2 1

Very likely

AB CONTINUE!

Likely

AB continue

Unlikely

AB stop

0.5 0.25 0.1

Very unlikely

AB STOP!

- Consider the course of PCT - If antibiotics are continued : - Measure PCT daily; discontinue antibiotics when PCT decreases by >80% of the peak level or an absolute PCT value $100.00 per test for many molecular assays. Although the list of consumables needed for sample preparation and chromatographic analysis is quite lengthy, they are all stored at room temperature, and though the initial investment for the instrument itself is not trivial, it is comparable to that of other identification systems on the market. CFA analysis requires growth in pure culture using standardized conditions negating the possibility of direct testing of patient specimens. In addition, numerous steps are required for sample preparation including harvesting of the bacterial

Cellular Fatty Acid-Based Microbial Identification and Antimicrobial Susceptibility…

207

unknown which can result in unintentional laboratory exposures. Laboratory directors must consider all aspects of their particular testing needs when considering implementation of not only existing technologies but emerging ones as well.

merging and Future Uses of Fatty Acid-Based Identification E Methods Technological advances continue to move the field of microbiology forward with improvements in both detection and identification of organisms. Molecular-based methods such as DNA sequencing of various targets and matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI TOF MS) analysis of microbial peptides are now competing with other, previously established techniques for detection and identification of microorganisms. However, a single, perfect diagnostic method for genus−/species-level identification of microorganisms remains elusive, and more often than not, final confirmation requires a combination of methods. For this reason, many clinical laboratories continue to use CFA analysis in conjunction with other methods such as biochemical testing, DNA sequencing, and/or MALDI TOF MS for identification of bacterial unknowns. Many laboratories find CFA analysis especially helpful in identification of gram-negative, non- fermentative organisms for which biochemical testing is not routinely performed. New applications for CFA analysis are emerging in the field of environmental microbiology. For example, changes in the relative abundance of a particular fatty acid may be indicative of specific ratios of microorganisms from an environmental sample. Such an approach has enabled the identification of shifting ratios between particular fungi and bacteria in soil samples which could serve as a marker of soil health which is a useful indicator during reclamation [32]. Likewise, the use of CFA analysis has also been harnessed for analysis of biological wastewater treatment plants to estimate activated sludge microbial communities as a means to evaluate changes in bacterial communities during plant operations [33].

References 1. Abel K, Deschmertzing H, Peterson JI. Classification of microorganisms by analysis of chemical composition. I Feasibility of utilizing gas chromatography. J Bacteriol. 1963;85:1039–44. 2. Kaneda T. Biosynthesis of branched chain fatty acids. I. Isolation and identification of fatty acids from Bacillus subtilis (ATCC 7059). J Biol Chem. 1963;238:1222–8. 3. VanDamme P. Taxonomy and classification of bacteria. In: Baron EJ, Jorgensen JH, Landry ML, Pfaller MA, Murray P, editors. Manual of clinical microbiology. Washington, DC: ASM; 2007. p. 543–72. 4. Lim DV, Simpson JM, Kearns EA, Kramer MF. Current and developing technologies for monitoring agents of bioterrorism and biowarfare. Clin Microbiol Rev. 2005;18:583–607.

208

N. Parrish and S. Riedel

5. Kunitsky C, Osterhout J, Sasser M. Identification of microorganisms using fatty acid methyl ester (FAME) analysis and the MIDI Sherlock® microbial Identification System. In: Miller MJ, editor. Encyclopedia of rapid microbiological methods, vol. 3. River Grove: DHI Publishing, LLC; 2006. p. 1–56. 6. Drucker DB. Chemotaxonomic fatty-acid fingerprints of some streptococci with subsequent statistical analysis. Can J Microbiol. 1974;20:1723–8. 7. Knivett VA, Cullen J. Some factors affecting cyclopropane acid formation in Escherichia coli. Biochem J. 1965;96:771–6. 8. Marr AG, Ingraham JL. Effect of temperature on the composition of fatty acids in Escherichia coli. J Bacteriol. 1990;84:1260–7. 9. Welch DF. Applications of cellular fatty acid analysis. Clin Microbiol Rev. 1991;4:422–38. 10. www.midi-inc.com. 11. O’Hara C. Manual and automated instrumentation for identification of Enterobacteriaceae and other aerobic gram-negative bacilli. Clin Microbiol Rev. 2005;18:147–62. 12. Jantzen E, Berdal BP, Omland T. Cellular fatty acid composition of Francisella tularensis. J Clin Microbiol. 1979;10:928–30. 13. Clarridge J, Raich T, Sjosted A, et al. Characterization of two unusual clinically significant Francisella tularensis strains. J Clin Microbiol. 1996;34:1995–2000. 14. Leclerq A, Guiyoule A, El Lioui M, Carniel E, Decallone J. High homogeneity of the Yersinia pestis fatty acid composition. J Clin Microbiol. 2000;38:1545–51. 15. AOAC International. Initiative yields effective methods for anthrax detection; RAMP and MIDI, Inc., methods approved. Inside Lab Manage 10:3, 2004. 16. Song Y, Yang R, Guo Z, Zhang M, Wang X, Zhou F. Distinctness of spore and vegetative cellular fatty acid profiles of some aerobic endospore-forming bacilli. J Microbiol Methods. 2000;39:225–41. 17. Srinivasan A, Kraus CN, DeShazer D, et al. Glanders in a military research biologist. N Engl J Med. 1999;345:256–8. 18. Timothy J, Inglis J, Aravena-Roman M, et al. Cellular fatty acid profile distinguishes Burkholderia pseudomallei from avirulent Burkholderia thailandensis. J Clin Microbiol. 2003;41:4812–4. 19. Khan AS, Ashford DA. Ready or not-preparedness for bioterrorism. N Engl J Med. 2001;345:287–9. 20. Tan Y, Wu M, Liu H, et al. Cellular fatty acids as chemical markers for differentiation of Yersinia pestis and Yersinia pseudotuberculosis. Lett Appl Microbiol. 2010;50:104–11. 21. Whittaker P, Fry FS, Curtis SK, et al. Use of fatty acid profiles to identify food-borne bacterial pathogens and aerobic endospore-forming bacilli. J Agric Food Chem. 2005;53:3735–42. 22. Butler WR, Jost KC Jr, Kilburn JO. Identification of mycobacteria by high-performance liquid chromatography. J Clin Microbiol. 1991;29:2468–72. 23. Centers for Disease Control. Standardized method for HPLC identification of mycobacteria. Atlanta: CDC; 1996. 24. Mycobacteria Identification System Operating Manual Version 1.0. MIDI incorporated, Newark, 2003. 25. Pfyffer G. Mycobacterium: general characteristics, laboratory detection, and staining procedures. In: Murray P, Baron EJ, Jorgensen JH, Landry ML, Pfaller MA, editors. Manual of clinical microbiology. Washington, DC: ASM; 2007. p. 543–72. 26. Flauta V, Osterhout G, Ellis B, et al. Use of the “RAM” susceptibility testing method for rapid detection of clarithromycin resistance in the Mycobacterium avium complex. Diagn Microbiol Infect Dis. 2010;67(1):47–51. 27. Garza-Gonzales E, Guerrero-Olazaran M, Tijerina-Menchaca R, Viader-Salvado J. Determination of drug susceptibility of Mycobacterium tuberculosis through mycolic acid analysis. J Clin Microbiol. 1997;35:1287–9. 28. Parrish N, Osterhout G, Dionne K, et al. A rapid, standardized, susceptibility method for Mycobacterium tuberculosis using mycolic acid analysis. J Clin Microbiol. 2007;45:3915–20.

Cellular Fatty Acid-Based Microbial Identification and Antimicrobial Susceptibility…

209

29. Viader-Salvado JM, Garza-Gonzalez E, Valdez-Leal R, Bosque-Moncaya M, Tijerina- Menchaca R, Guerrero-Olazaran M. Mycolic acid index susceptibility method for Mycobacterium tuberculosis. J Clin Microbiol. 2001;2001(39):2642–5. 30. Clinical and Laboratory Standards Institute (CLSI). Laboratory detection and identification of mycobacteria. Approved guideline, 2008. 31. Dionne K, Carroll K, Parrish N. “RAM”-based rapid identification and determination of resistance in Mycobacteria direct from sputum. An abstract presented at the 111th ASM General Meeting, New Orleans, 2011. 32. Mummey DL, Stahl PD, Buyer JS. Microbial biomarkers as an indicator of ecosystem recovery following surface mine reclamation. Appl Soil Ecol. 2002;21:251–9. 33. Sreenivasulu B, Paramageetham C, Sreenivasulu D, Suman B, Umamahesh K, Babu GP. Analysis of chemical signatures of alkaliphiles using fatty acid methyl ester analysis. J Pharm Bioallied Sci. 2017;9(2):106–14.

MALDI-TOF Mass Spectrometry-Based Microbial Identification and Beyond Alexander Mellmann and Johannes Müthing

Introduction Rapid and accurate species identification of bacteria, fungi, and viruses is a fundamental requirement in clinical and food microbiology and other fields of diagnostic microbiology. Whereas virus recognition is usually achieved within a few hours by either serological tests or genotyping approaches using various nucleic acid detection systems, the conventional identification of bacteria and fungi still mainly relies on methods that include laborious and time-consuming initial cultivation and ensuing isolation of the microorganism. This approach is therefore dependent on the generation time (growth) of the particular microorganism, resulting in assay durations of several hours minimum, e.g., in the case of Enterobacteriaceae or other fast-growing prokaryotes, and up to several days or weeks in the case of slowgrowing mycobacteria and some fungi. Though species identification of a pure culture is achievable within 4–48 h with various (semi)automated systems, additional isolation steps are frequently necessary, which can extend the time until diagnosis by days, e.g., if the potential pathogen must be separated from the physiological background flora. Realistically species assignment of a putative pathogen from a non-sterile specimen takes at least 2–3 days. In many areas of patient care, elapsed time until diagnosis may considerably reduce the therapeutic quality of care due to a lack of information about the infecting pathogen. Therefore, a rapid species diagnosis is of high priority as a focused therapy might be lifesaving for the patient [1, 2]. Similarly, a timely diagnosis is imperative for surveillance studies or screenings with particular demands during outbreak situations of foodborne pathogens or pre-admission screening to detect multiresistant bacteria in the hospital setting [3, 4]. Both species identification and resistance testing are of equal importance; however, this chapter will focus primarily on species identification but A. Mellmann (*) · J. Müthing Institute of Hygiene, University Hospital Münster, Münster, Germany e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_10

211

212

A. Mellmann and J. Müthing

will also give an outlook upon what will be possible in the future with respect to resistance testing and subtyping. In addition to the time required to identify an unknown species, some bacterial species or groups are still difficult to differentiate. During the last decade, molecular studies have raised doubts about traditional genus and species assignments, resulting in profound reclassifications of numerous bacterial genera and species as well as the discovery of a large number of novel species. Furthermore, these investigations demonstrated substantial limitations of previously employed methods and the urgent need for the development of more reliable techniques [5, 6]. Finally, in some bacterial species, such as within the diverse group of Gram-negative, non-fermenting rods, extensive reclassification and their reactive biochemical behavior and different colony morphologies pose further challenges in unequivocal species identification [7]. Great efforts have been made to enhance the accuracy and the speed of species identification. In addition to genotypic methods that rely on DNA sequencing of discriminatory regions, for example, 16S rRNA encoding genes in prokaryotes [8, 9], or nowadays even on whole genome sequencing [10], various phenotype-related procedures have been developed such as cell wall analysis or determination of fatty acid and protein profiles [11–13], enabling a robust species identification that is independent from the bacterial metabolism or regulatory phenomena. In addition to the ameliorated species identification, the expense per assay is a key issue and has to be considered. The applicability to automation plays a pivotal role in modern clinical laboratories and must be taken into account in addition to the hands-on/turnaround time and assay costs. Finally, ease and robustness of procedures are prerequisites for their implementation in the clinical laboratory. In this context, reproducibility of results and acceptance by both the client and regulatory authority are essential for the establishment in a clinical laboratory. Here, the power of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) is demonstrated as a redevelopment that has evolved to revolutionize the identification of prokaryotic and eukaryotic pathogens in microbial laboratories during recent years. This chapter starts with a short technical overview about the principles of mass spectrometry, in particular MALDI- TOF MS. Speed, accuracy, and reproducibility of MS techniques will be compared with customary methods, and current limitations of MALDI-TOF MS-based approaches will be discussed. Additionally, an alternative MS strategy based on electrospray ionization (ESI), also suitable for species identification, will be introduced.

General Remarks on Mass Spectrometry Mass spectrometry is an emerging technique that has been developed into a very useful tool to structurally analyze biomolecules of various substance classes, such as nucleobases, nucleosides, and nucleotides as components of nucleic acids [14, 15], (glyco)proteins including proteolytic digested glycopeptides and released

MALDI-TOF Mass Spectrometry-Based Microbial Identification and Beyond

213

glycans as well as quantitative mass spectrometric analysis of glycoproteins [16– 18], phospholipids and (glyco)lipids [19, 20], and others. The primary objective of MS lies in identification of the exact molecular mass of an isolated biomolecule or individual molecules such as proteins in mixture. The applicability of mass spectrometry to the analysis of complex biomolecules has been greatly improved by introducing two soft ionization techniques – MALDI and ESI MS. Furthermore, MALDI restrictions by limited ionization have been overcome by laser-induced postionization as an excellent tool for MS imaging of biomolecules in tissue slices [21]. Both methods ionize large molecules, which tend to be fragile and fragment when more conventional ionization methods are applied. MALDI and ESI MS can be easily implemented in a straightforward diagnostic procedure to reliably identify the genus, species, and, in some cases, subspecies of bacteria. Generally, a typical mass spectrometer is built up from three components: an ion source, a mass analyzer, and a detector. The ion source produces ions from the sample, the mass analyzer separates ions with different mass-to-charge ratios (m/z), and the numbers of different ions are detected by the detector. The resulting output is a mass spectrum which is displayed as a graph of the ion intensities versus m/z values and consists of a number of mass spectral peaks, forming a unique pattern. The majority of ions generated by MALDI contain only one charge, and only one peak appears for each individual compound in the spectrum, facilitating data interpretation. Ions produced by ESI may be multiply charged resulting in considerably more complex mass spectra than MALDI. Notably, signal intensities do not reflect the quantities of different sample molecules.

MALDI and ESI MS In the past decades, several different MALDI- and ESI-based methods were developed, a few of which are nowadays applied in systems used for microbial identification purposes. Both methods are highly advantageous as the analyte structure is preserved due to the use of soft ionization. MALDI, invented in the 1980s [22, 23], is based on UV laser ionization of the analyte (any substance up to whole bacterial or fungal cells), which is embedded in an appropriate matrix on a target plate. Matrix molecules fulfill several requirements that are crucial for ionization of the investigated biomolecules. They are of low molecular weight and low volatility preventing vaporization during sample preparation. Acidic matrices are useful as they act as proton donors that are essential for ionization of the analyte. Furthermore, they possess not only polar groups that enable use in aqueous solutions but also exhibit strong optical absorption in the UV range, so that they efficiently absorb the energy from laser irradiation. In most cases small organic molecules, such as 2,5-dihydroxybenzoic acid (DHB) or α-cyano-4-hydroxycinnamic acid (CHCA), are used for UV MALDI and mixed in a 1000–10,000-fold excess to the analyte [22, 24]. Co-crystallization of the analyte with the matrix is another key issue in selecting a proper matrix to obtain a good-quality mass spectrum of the analytes of interest. A pulsed laser beam, usually a nitrogen UV laser (λ = 337 nm), is fired at the

214


matrix/analyte crystals of the dried-droplet spot on the target plate within a vacuum. The laser energy is absorbed by the matrix, which in turn is desorbed in an expanding plume and ionized by addition of a proton [25]. Charging of the analyte occurs through the transfer of protons or sodium ions to the sample molecules, and quasimolecular singly charged ions are formed, e.g., [M + H]+ or [M + Na]+, respectively (Fig. 1). After the sample and matrix molecules have entered the gas phase of the vacuum environment, the newly generated ions are accelerated in an electric field of known strength in the time-of-flight (TOF) analyzer. The TOF analyzer is a field-free flight tube, where the ions can travel in a strait and linear direction to the detector (Fig. 1). A linear TOF spectrum is limited in resolution and cannot distinguish ions with similar m/z values. This can be partially corrected in a reflectron TOF analyzer in which ions are reflected by an “iron mirror” using an electric field, thereby doubling the ion flight path and increasing the resolution (not shown here). The velocity of the ions depends on the mass and the degree of ionization, i.e., the m/z ratio, and is measured as the time it takes for the ions to fly from one end of the analyzer to the other and strike the detector. The flying speeds of ions are proportional to their m/z ratio and the m/z values versus signal intensity are then finally drawn as the mass spectrum where the x-axis depicts the m/z value and the y-axis the intensity [24]. Figure 1 displays a schematic graphical representation of a linear TOF MALDI mass spectrometer and an example mass spectrum of a sample composed of three different mass components.

Fig. 1 Schematic illustration of a linear UV laser MALDI-TOF mass spectrometer. The matrix- analyte mixture is shot with an UV laser leading to desorption of the analyte and matrix and to a transfer of protons from the matrix to the analyte molecules. Ions are subsequently accelerated in an electric field, separated during their travel in a field-free flight tube according to their mass-to- charge (m/z) ratio, and finally detected with the detector. The resulting mass spectrum is displayed as a graph of the ion intensities versus m/z values


215

The mathematical basis of the mass determination is the following equation:

= Ekin ½= mv 2 zeU

with Ekin as the kinetic energy of the ions after acceleration within the electric field with the voltage U. The speed v is calculated via the TOF and the covered distance, and e is the elementary electric charge. Altogether, this enables the exact calculation of the m/z values. For further characterization of biomolecules, mass spectrometers can also be equipped with a collision chamber filled with an inert gas, e.g., argon. Collision with these molecules leads to fragmentation of the analyte ions and assists in structural characterization of the unknown analyte molecules. In contrast to MALDI, in ESI the sample is solubilized in volatile organic solvent (e.g., methanol, acetonitrile) prior to ionization. The analyte is then nebulized, together with the solvent, as a fine spray through a very small, charged, and usually metal or glass capillary equipped with a stainless steel needle into the electric field at atmospheric pressure. The resulting charged droplets of the analyte are then subjected to a TOF analyzer for generation of a mass spectrum. During the spraying process, the solvent continuously evaporates leading to an increase of the charge density. At the Rayleigh limit, the electrostatic repulsive forces exceed the surface tension, and the droplets divide into smaller subunits (Coulomb explosion). The smaller droplets continue to evaporate, and the process is repeated again ultimately resulting in charged analyte molecules which enter the mass analyzer. As whole bacterial or fungal cells cannot be solubilized sufficiently, ESI is mainly used for analysis of cellular components or other soluble analytes such as polymerase chain reaction (PCR)-amplified microbial nucleic acids providing an alternative biomarker analysis [26, 27]. As mentioned above, a general advantage of MALDI-TOF MS is the soft ionization of the embedded analyte without destruction of the analyte, which makes it especially suitable in the structural characterization of intact biomolecules. Moreover, only a few microliters of the matrix-analyte mixture are required for placement onto the target plate. Preparation of this mixture is usually very simple and requires only a few minutes to complete [24].

Species Identification Using MALDI-TOF MS Species identification of intact microorganisms, taken directly from culture, by means of MALDI-TOF MS has been firstly described for various Gram-positive and Gram-negative bacteria by Claydon et al. in 1996 [28], and the general applicability has been reviewed, including intact viruses, spores, and fungi, by Fenselau and Demirev in 2001 [29]. The spectra obtained allowed identification of microorganisms from different genera, different species, and from different strains of the same species [28]. Assignment was realized with whole cell extracts by the exact mass determination of desorbed peptides and small proteins of the cell wall resulting in a unique mass spectral fingerprint of the microorganism under

216


investigation. At that time it was assumed that the measured masses were unique and representative for individual microorganisms forming the basis of applications of mass spectrometry in microbiology without knowing the detailed characterization of each component [29–31]. Figure 2 illustrates the workflow for species identification using MALDI-TOF MS.

Fig. 2 Principal workflow of MALDI-TOF MS-based species identification. (1) Bacterial or fungal colonies on an agar plate. (2) Mixing of matrix and analyte for sample preparation and subsequent co-crystallization of the matrix-analyte mix by gentle air-drying. (3) Laser desorption, ionization, and mass analysis of the analyte molecules. (4) Generation of the mass spectrum. (5) Computer-assisted data analysis and comparison of acquired spectra with reference database entries


217

Shortly after the first description [28], it became apparent that the poor reproducibility of measurements was mainly due to (i) shortcomings in sample preparation, especially in Gram-positive bacteria [32], (ii) inconsistencies among different MALDI-TOF mass spectrometers employed, and (iii) variability in bacterial culture conditions [33]. Moreover, until approximately 2005, there was still a lack of an urgently needed sophisticated software and data analysis tools that were essential for the integration of MALDI-TOF MS outcomes into the clinical laboratory. A valid identification of bacterial and fungal species was further hampered by the deficit in central spectra databases that comprised fingerprint libraries derived from well-characterized reference strains required for comparison with newly generated mass spectra of unknown composition. For these reasons microbial MALDI-TOF MS at its infancy was neither well accepted nor implemented for routine application into the microbial and/or clinical laboratory. To address poor reproducibility and serious inconsistency that could be attributed to different culture conditions and resulting alteration of the microorganisms’ metabolism, the investigation of proteins was shifted to a higher m/z range. Whereas the first publications reported on measured mass ranges from 550 to 2000 daltons to identify a species [28], current systems encompass a mass range of 2000–20,000 daltons. This mass range mainly measures ribosomal proteins [34, 35], conserved proteins that are highly abundant in any type of prokaryotic and eukaryotic cells. This approach warrants a relatively high robustness against variability of metabolic products and fluctuation of other cell components that may occur by varying culture conditions. Additionally, ribosomal proteins are positively charged, which facilitates MALDI-TOF MS. [31] This improved stability and reproducibility has been demonstrated shortly afterward for some prokaryotic microorganisms, e.g., nonfermenting bacteria or staphylococci [36, 37]. For other groups of pathogens, a strict adherence to certain culture conditions is strongly recommended as accumulation of metabolites [38], sporulation (e.g., Bacillus spp.), or autolytic processes due to long-term storage (e.g., Streptococcus spp.) might cause sizeable changes in the mass pattern. Technical advancements in the last decade have eliminated the production of diverging spectra due to operation on individual instruments (Fig. 3) [37]. Additionally, optimization and standardization of sample preparation protocols now allow for rapid processing of nearly any bacterial species within minutes [32, 39]. To further enhance the quality of MALDI-TOF mass spectra for use in species identification, new algorithms have been developed which aim at better and more valid comparison of spectra from unknown microorganisms against a database harboring spectra from well-characterized reference strains [40–44]. In combination with continuously growing reference databases, today two commercially available platforms, Bruker MALDI Biotyper® (Bruker Daltonik, Bremen, Germany) and VITEK MS® (bioMérieux, Marcy-l’Etoile, France), produce a correct species identification in most instances. Only a few discrepancies were detected in comparative studies that applied both systems; common reasons for these discrepancies were the presence or absence of individual reference spectra of the investigated organisms in the reference databases of the respective platforms [45–57].

218


Fig. 3 Reproducibility of MALDI-TOF mass spectra. Spectra were directly obtained from bacterial colonies for species identification using three different MALDI-TOF mass spectrometers: autoflex, microflex, and ultraflex (all from Bruker Daltonics). Masses are depicted in the m/z range from 4000 to 12,000 daltons. Intens. [a.u.], intensity (in arbitrary units). (Copyright © Mellmann et al. [37])

To reduce the likelihood of a misidentification due to the database content, the following two prerequisites should be fulfilled for a reference database: First, the database should contain spectra from well-characterized culture collection strains that have been cultured under optimal and standardized conditions. These strains should be relevant for the specific diagnostic questions that may differ significantly between various disciplines (e.g., medical microbiology, food microbiology, or environmental monitoring). Second, the reference database should include not only the prototypical strain of a certain species but also as many other strains as possible from the same species to both determine and compensate for the naturally occurring intra-species variability. In general, these two core conditions are essential for any diagnostic procedure that is based on comparisons against a reference [8, 58]. Similar to other identification strategies that rely on phenotype-related fingerprint libraries, the assignment is usually supplemented by calculation of an error probability or result validity extrapolated from the quality and quantity of matched peaks within the newly generated and the reference MALDI pattern [35]. Tables 1 and 2 provide an overview about the growing number of studies focusing on mass spectrometry-derived identification of bacterial and fungal species. In these studies, comparisons between MALDI-TOF MS and classical genotypic and


219

Table 1 Synopsis of selected MALDI-TOF MS studies focusing on bacterial identification (as of February 2018) Organism(s) Acinetobacter baumannii complex Aeromonas spp. Anaerobes

Arthrobacter spp. Bacillus spp. Bacteroides spp. Bartonella spp. Brucella spp. Campylobacter / Helicobacter spp. Cardiobacterium hominis Clavibacter spp. Corynebacterium spp. Coxiella burnetii Enterobacteriaceae Enterococcus spp. Gallibacterium spp. HACEK group Haemophilus spp. Foodborne pathogens Legionella spp. Listeria spp. Mycobacteria spp. Nocardia spp. Nonfermenters Pantoea spp. Plesiomonas spp. Salmonella spp. Staphylococcus spp.

Stenotrophomonas spp. Streptococcus spp.

Vibrio spp. Yersinia spp.

References Pailhories et al. [59] Donohue et al. [60] Coltella et al. [61], Fedorko et al. [62], Grosse-.Herrenthey et al. [63], Justensen et al. [52], La Scola et al. [64], Nagy et al. [65], Shah et al. [66], Stingu et al. [67], Veloo et al. [45, 68, 69] Vargha et al. [38] Hotta et al. [70] Culebras et al. [71] Fournier et al. [72] Ferreira et al. [73], Lista et al. [74] Alispahic et al. [75], Bessède et al. [76], Mandrell et al. [77], Winkler et al. [78] Wallet et al. [79] Zaluga et al. [80] Konrad et al. [81] Hernychova et al. [82] Conway et al. [83], Lynn et al. [84], Saffert et al. [85] Fang et al. [54] Alispahic et al. [86] Couturier et al. [87], Schulthess et al. [88] Haag et al. [89] Mazzeo et al. [90] Gaia et al. [91], He et al. [92], Moliner et al. [93] Barbuddhe et al. [94] Bouakaze et al. [95], Hettick et al. [96], Lotz et al. [97], Pignone et al. [98], Saleeb et al. [99], Tseng et al. [100] Verroken et al. [101] Degand et al. [102], Jacquier et al. [103], Marko et al. [104], Mellmann et al. [37], Vanlaere et al. [105] Rezzonico et al. [106] Kolinska et al. [107] Dieckmann et al. [108], Dieckmann and Malorny [109], Sparbier et al. [110] Bergeron et al. [111], Bernardo et al. [112], Carbonnelle et al. [36], Carpaij et al. [113], Decristophoris et al. [114], Dubois et al. [115], Dupont et al. [116], Elbehiry et al. [117], Rajakaruna et al. [118] Vasileuskaya-Schulz et al. [119] Agergaard et al. [57], Cherkaoui et al. [120], Friedrichs et al. [121], Hinse et al. [122], Karpanoja et al. [51], Lartigue et al. [123], Rupf et al. [124] Dieckmann et al. [125] Ayyadurai et al. [126], Lasch et al. [127], Stephan et al. [128]

220


Table 2 Synopsis of selected MALDI-TOF MS studies focusing on fungal identification (as of February 2018) Organism(s) Aspergillus spp. Candida spp. Cryptococcus spp. Fusarium spp. Dermatophytes Mucorales spp. Penicillium spp. Pseudallescheria/Scedos porium complex spp. Various yeasts/fungi

References Alanio et al. [129], De Carolis et al. [130], Hettick et al. [131] Quiles-Merelo et al. [132] McTaggart et al. [133] De Carolis et al. [130], Kemptner et al. [134] Erhard et al. [135] De Carolis et al. [130] Hettick et al. [131] Coulibaly et al. [136] Bader et al. [137], Dhiman et al. [138], Gosh et al. [139], Kaleta et al. [140], Marklein et al. [141], Pence et al. [47], Putignani et al. [142], Qian et al. [143], Stevenson et al. [144], Van Herendal et al. [145]

phenotypic methods for species identification demonstrate not only the capability but also comparability of MALDI to provide reliable results [121, 146–148]. Furthermore, in many studies, MALDI-TOF MS has impressively shown its potential to further differentiate species to the subspecies or even clonal level. For example, it is possible to differentiate the Burkholderia cepacia complex into its species by mass spectrometry [37, 102, 105], while in the past this issue required great efforts using conventional phenotypic or even genotypic methods. This aspect is of high relevance in medical microbiology due to the distinct outcome and infection control measures related to the different species. Other impressive examples of the enhanced discriminatory power (see also Tables 1 and 2) are the differentiation of Salmonella spp. [108, 109] or Listeria spp. [94]. After the major obstacles impending reproducibility and accuracy were resolved, the use of MALDI-TOF MS exhibits its outstanding advantage over all other identification methods: its rapidness of identification. Starting from a single colony, it has now become an easy task to identify an unknown species within only a few minutes [37, 149, 150]. In contrast to the majority of the conventional procedures, a pure culture is no longer needed for MALDI-TOF MS, allowing to skip tedious, laborious, and time-consuming isolation steps. First studies have even demonstrated the proof of concept to identify species within a mixed culture [150]. While the computer-assisted spectral analysis works within a time span of a few seconds or minutes, the sample preparation prior to the MALDI-TOF MS analysis represents the most time-consuming step. However, sample extraction is not required for all organisms. For example, in Gram-negative bacteria analysis, the whole bacterial cell suspension can be directly mixed with the matrix. Still, for organisms with a more refractory cell wall, such as Gram-positive bacteria or fungi, the extraction step is required or at least beneficial for the diagnostic accuracy before mixing the sample with the matrix. Regardless of the analyte, a dedicated extraction step prior to MALDI-TOF MS increases the quality of spectra [151, 152]. Ultimately, the operator must decide if an analyte pretreatment is always necessary or whether


221

preceding extraction should be performed only in situations where the analysis of a directly spotted sample failed. Depending on the specification of the mass spectrometer, high-throughput analyses of more than 100 identifications per hour are achievable. Further advantages of the MALDI-TOF MS-based strategy of microbial identification are the relatively low consumable costs which include the expenses for the matrix and chemicals needed for sample extraction. Still, the financial investment for installation of a mass spectrometer and the running costs of the reference database and maintenance of the system are not irrelevant and should also be taken into account, when comparing novel MALDI-TOF MS with traditional methods.

imitations and Future Challenges of MALDI-TOF MS L in Clinical Microbiology Despite the impressive improvements of MALDI-TOF MS, some limitations still exist regarding species identification of microorganisms. The crucial point for a valid and successful species identification is, as mentioned above, the quality of entries in the reference database. During the last several years, the MALDI-TOF MS manufacturers have constantly updated and extended their reference databases to overcome this problem. Several evaluation studies (Tables 1 and 2) have addressed this issue; however, it seems that even comprehensive databases are not able to differentiate all species with the necessary precision underlining the high variability of pathogens in clinical routine samples. Moreover, the lack of species-specific molecules that could be detected using MALDI-TOF MS interferes with a successful species identification. Well-known examples are the differentiation of pneumococci and members of the Streptococcus mitis/oralis group [153, 154] or the delineation of some Enterobacter species [155], where their close relationship likely hampers the valid species identification [156, 157]. Therefore, like the users of other customary laboratory methods, MALDI-TOF mass spectrometer users need to be aware of the boundaries of the system to warrant an accurate species identification of microorganisms. A challenge during the early days of MALDI-TOF MS was the lack of standardization of laboratory procedures, which decreased interlaboratory reproducibility and therefore the broader use in clinical routine. This issue was technically solved more than a decade ago; however, it took several more years until various multicenter ring trials approved the nowadays broadly accepted high reproducibility of MALDI-TOF MS-based identification [158–168]. Although MALDI-TOF MS is a rapid method that is both similar to and in competition with molecular methods like real-time PCR assays, it is still classified as a phenotypic method as whole cell extracts are analyzed that do not require any template-driven amplification steps of molecular techniques. To date, either a single colony on solid medium or an aliquot

222


of a pure liquid culture with a minimal number of 1000–5000 cells is necessary for a valid identification [169, 170]. This detection limit, i.e., the minimal amount of cellular material that is needed for a sufficient MALDI-TOF MS, poses a major challenge when the direct investigation of material from primarily sterile body sites, e.g., urine or cerebrospinal fluid, or from a positive blood culture or other broth culture systems, is wished. In the past, malfunctions using such materials were due to the interference of the MS-based approach by mass signals of proteins or other molecules derived from human cells or body fluids. Great efforts were made in the past to overcome these obstacles as the timely and correct diagnosis of pathogens has a great impact on the success of therapy, especially in critically ill patients [4, 171]. Different approaches, mainly based on differential centrifugation and absorption of disturbing proteins or other components, have demonstrated the general applicability for direct use of such samples in MALDI-TOF MS [152, 153, 172–184]; however, none of these approaches is widely used yet. In addition to accurate species identification, antimicrobial susceptibility testing is the second major task in clinical microbiology. Due to the inability of MALDI- TOF MS to provide susceptibility testing, matching of resistance testing data – determined by means of different, mostly phenotypic approaches, such as agar diffusion or using automated systems – faces the ambitious challenge to validate species identification and the resistance profile of the species, e.g., to reflect intrinsic resistance. To enable resistance testing, early studies have demonstrated the feasibility of MALDI-TOF MS to detect certain resistance determinants [185–187]; however, most assays took place on an experimental level or suffered from insufficient robustness for routine applications. Afterward, the detection of antibiotics, e.g., carbapenems, and – after a short incubation time with the resistant organisms – their degradation products, was able to validly determine specific resistance determinants such as extended-spectrum beta-lactamases by measuring the presence of the intact or degraded antibiotic agent [188, 189]. Although this approach was quite promising, it is limited due to the fact that only selected antibiotics and resistance mechanisms could be investigated. Another example, which followed a similar principle but measured changes of the protein spectral profile of a pathogen after incubation with an antimicrobial agent, rather than degradation products of antibiotics, was published. Here, the spectra of fluconazole- susceptible and fluconazole-resistant Candida albicans were compared after growth in the presence of increasing concentrations of fluconazole and successfully correlated with the minimal inhibitory concentration (MIC) determined by microdilution [190]. In 2017, a novel direct-on-target microdroplet growth assay was published [191], where the bacterial growth was determined in the presence of increasing antimicrobial agents directly on a MALDI-TOF MS target plate. In case of a successful species identification using the recorded MALDI-TOF MS spectra, the organism was rated as “non-susceptible”; if no growth, i.e., no spectra could be recorded for a valid species identification, was detected, the organism was rated as “susceptible.” Although this approach is not yet commercially available for routine testing, it is the most promising approach for antimicrobial susceptibility testing


223

using MALDI-TOF MS as it works independent of the tested microbial species, the antimicrobial agent in question, and underlying resistance mechanism [191]. The high discrimination on species level motivated researchers also to investigate the possibility to use MALDI-TOF MS also for further differentiation on strain level, i.e., subtyping. Here, early studies demonstrated that a differentiation of diverse clonal lineages within the species Staphylococcus aureus could be attained. Attempts, however, to discriminate methicillin-resistant S. aureus (MRSA) from methicillin-susceptible S. aureus were unsuccessful [112]. Further attempts to apply MALDI-TOF MS for subtyping were nicely reviewed by Spinali et al. and Lartigue [192, 193], where in selected species a discrimination level similar to previous methods like Rep-PCR was achieved [194]. Due to the rapid advances in next- generation sequencing technologies and its application for strain typing, it is unlikely that MALDI-TOF MS will play a major role in this field. However, as the spectra are already available from the species identification efforts, they might be at least used for a rapid screening to rule out a clonal relationship in case of diverse spectra.

Conclusion Collectively, MALDI-TOF mass spectrometry has successfully entered the microbiological laboratory and has initiated an ongoing revolution for prokaryotic and eukaryotic species identification. Whereas MALDI-TOF MS-based species identification is already well accepted and has replaced the classical methods in many clinical laboratories as a fast, accurate, and robust method for reliable species identification, it is yet unknown to which extent MALDI-TOF MS might also replace current methods for antimicrobial susceptibility testing. Clients will at least be easily convinced of a fast species result, which allows for an empiric antibiotic therapy based on local resistance data in clinical microbiology.

References 1. O’Riordan P, Schwab U, Logan S, et al. Rapid molecular detection of rifampicin resistance facilitates early diagnosis and treatment of multi-drug resistant tuberculosis: case control study. PLoS One. 2008;3:e3173. 2. Park BS, Park YJ, Kim YJ, et al. A case of disseminated Nocardia farcinica diagnosed through DNA sequencing in a kidney transplantation patient. Clin Nephrol. 2008;70:542–5. 3. Harbarth S, Masuet-Aumatell C, Schrenzel J, et al. Evaluation of rapid screening and preemptive contact isolation for detecting and controlling methicillin-resistant Staphylococcus aureus in critical care: an interventional cohort study. Crit Care. 2006;10:R25. 4. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006;34:1589–96.

224


5. O’Hara CM. Manual and automated instrumentation for identification of Enterobacteriaceae and other aerobic gram-negative bacilli. Clin Microbiol Rev. 2005;18:147–62. 6. Bergey DHG, G.M. Volume 2: the Proteobacteria. Berlin: Springer; 2005. 7. Gilligan PH. Microbiology of airway disease in patients with cystic fibrosis. Clin Microbiol Rev. 1991;4:35–51. 8. Clarridge JE 3rd. Impact of 16S rRNA gene sequence analysis for identification of bacteria on clinical microbiology and infectious diseases. Clin Microbiol Rev. 2004;17:840–62. table of contents 9. Harmsen DK, H. 16S rDNA for diagnosing pathogens: a living tree. ASM News. 2004;70:19–24. 10. Didelot X, Bowden R, Wilson DJ, Peto TEA, Crook DW. Transforming clinical microbiology with bacterial genome sequencing. Nat Rev Genet. 2012;13:601–12. 11. Groth I, Schumann P, Martin K, et al. Ornithinicoccus hortensis gen. Nov., sp. nov., a soil actinomycete which contains L-ornithine. Int J Syst Bacteriol. 1999;49(Pt 4):1717–24. 12. Stackebrandt E, Koch C, Gvozdiak O, Schumann P. Taxonomic dissection of the genus Micrococcus: Kocuria gen. nov., Nesterenkonia gen. nov., Kytococcus gen. nov., Dermacoccus gen. nov., and Micrococcus Cohn 1872 gen. emend. Int J Syst Bacteriol. 1995;45:682–92. 13. Miller LT. Single derivatization method for routine analysis of bacterial whole-cell fatty acid methyl esters, including hydroxy acids. J Clin Microbiol. 1982;16:584–6. 14. Dudley E, Bond L. Mass spectrometry analysis of nucleosides and nucleotides. Mass Spectrom Rev. 2014;33:302–31. 15. Huber CG, Oberacher H. Analysis of nucleic acids by on-line liquid chromatography-mass spectrometry. Mass Spectrom Rev. 2001;20:310–43. 16. Ahn YH, Kim JY, Yoo JS. Quantitative mass spectrometric analysis of glycoproteins combined with enrichment methods. Mass Spectrom Rev. 2015;34:148–65. 17. Wang H, Hanash S. Intact-protein based sample preparation strategies for proteome analysis in combination with mass spectrometry. Mass Spectrom Rev. 2005;24:413–26. 18. Zhang Q, Li Z, Wang Y, Zheng Q, Li J. Mass spectrometry for protein sialoglycosylation. Mass Spectrom Rev. 2017. (https://doi.org/10.1002/mas.21555. [Epub ahead of print]). 19. Müthing J, Distler U. Advances on the compositional analysis of glycosphingolipids combining thin-layer chromatography with mass spectrometry. Mass Spectrom Rev. 2010;29:425–79. 20. Fuchs B. Analysis of phospholipids and glycolipids by thin-layer chromatography-matrix- assisted laser desorption and ionization mass spectrometry. J Chromatogr A. 2012;1259:62–73. 21. Soltwisch J, Kettling H, Vens-Cappell S, Wiegelmann M, Müthing J, Dreisewerd K. Mass spectrometry imaging with laser-induced postionization. Science. 2015;348:211–5. 22. Karas M, Hillenkamp F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal Chem. 1988;60:2299–301. 23. Tanaka K, Waki H, Ido Y, Akita Y, Yoshida Y, Yoshida T, Matsuo T. Protein and polymer analyses up to m/z 100 000 by laser ionization time-of-flight mass spectrometry. Rapid Commun Mass Spectrom. 1988;2:151–3. 24. Hillenkamp F, Peter-Katalinic J. MALDI MS. A practical guide to instrumentation, methods and applications. Wiley-VCH: Weinheim; 2007. 25. Dreisewerd K. The desorption process in MALDI. Chem Rev. 2003;103:395–426. 26. Hall TA, Budowle B, Jiang Y, et al. Base composition analysis of human mitochondrial DNA using electrospray ionization mass spectrometry: a novel tool for the identification and differentiation of humans. Anal Biochem. 2005;344:53–69. 27. Muddiman DC, Anderson GA, Hofstadler SA, Smith RD. Length and base composition of PCR-amplified nucleic acids using mass measurements from electrospray ionization mass spectrometry. Anal Chem. 1997;69:1543–9. 28. Claydon MA, Davey SN, Edwards-Jones V, Gordon DB. The rapid identification of intact microorganisms using mass spectrometry. Nat Biotechnol. 1996;14:1584–6. 29. Fenselau C, Demirev PA. Characterization of intact microorganisms by MALDI mass spectrometry. Mass Spectrom Rev. 2001;20:157–71.


225

30. Holland RD, Wilkes JG, Rafii F, et al. Rapid identification of intact whole bacteria based on spectral patterns using matrix-assisted laser desorption/ionization with time-of-flight mass spectrometry. Rapid Commun Mass Spectrom. 1996;10:1227–32. 31. Krishnamurthy T, Ross PL, Rajamani U. Detection of pathogenic and non-pathogenic bacteria by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun Mass Spectrom. 1996;10:883–8. 32. Smole SC, King LA, Leopold PE, Arbeit RD. Sample preparation of gram-positive bacteria for identification by matrix assisted laser desorption/ionization time-of-flight. J Microbiol Methods. 2002;48:107–15. 33. Wang Z, Russon L, Li L, Roser DC, Long SR. Investigation of spectral reproducibility in direct analysis of bacteria proteins by matrix-assisted laser desorption/ionization time-of- flight mass spectrometry. Rapid Commun Mass Spectrom. 1998;12:456–64. 34. Suh MJ, Hamburg DM, Gregory ST, Dahlberg AE, Limbach PA. Extending ribosomal protein identifications to unsequenced bacterial strains using matrix-assisted laser desorption/ ionization mass spectrometry. Proteomics. 2005;5:4818–31. 35. Maier AK, M. Fast and reliable MALDI-TOF MS-based microorganism identification. Chem Today. 2007;25:68–71. 36. Carbonnelle E, Beretti JL, Cottyn S, et al. Rapid identification of Staphylococci isolated in clinical microbiology laboratories by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2007;45:2156–61. 37. Mellmann A, Cloud J, Maier T, et al. Evaluation of matrix-assisted laser desorption ionization- time-of-flight mass spectrometry in comparison to 16S rRNA gene sequencing for species identification of nonfermenting bacteria. J Clin Microbiol. 2008;46:1946–54. 38. Vargha M, Takats Z, Konopka A, Nakatsu CH. Optimization of MALDI-TOF MS for strain level differentiation of Arthrobacter isolates. J Microbiol Methods. 2006;66:399–409. 39. Liu H, Du Z, Wang J, Yang R. Universal sample preparation method for characterization of bacteria by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl Environ Microbiol. 2007;73:1899–907. 40. Arnold RJ, Reilly JP. Fingerprint matching of E. coli strains with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry of whole cells using a modified correlation approach. Rapid Commun Mass Spectrom. 1998;12:630–6. 41. Jarman KH, Daly DS, Petersen CE, Saenz AJ, Valentine NB, Wahl KL. Extracting and visualizing matrix-assisted laser desorption/ionization time-of-flight mass spectral fingerprints. Rapid Commun Mass Spectrom. 1999;13:1586–94. 42. Jarman KH, Cebula ST, Saenz AJ, et al. An algorithm for automated bacterial identification using matrix-assisted laser desorption/ionization mass spectrometry. Anal Chem. 2000;72:1217–23. 43. Pineda FJ, Lin JS, Fenselau C, Demirev PA. Testing the significance of microorganism identification by mass spectrometry and proteome database search. Anal Chem. 2000;72:3739–44. 44. Bright JJ, Claydon MA, Soufian M, Gordon DB. Rapid typing of bacteria using matrix- assisted laser desorption ionisation time-of-flight mass spectrometry and pattern recognition software. J Microbiol Methods. 2002;48:127–38. 45. Veloo AC, Knoester M, Degener JE, Kuijper EJ. Comparison of two matrix-assisted laser desorption ionisation-time of flight mass spectrometry methods for the identification of clinically relevant anaerobic bacteria. Clin Microbiol Infect. 2011;17:1501–6. 46. Porte L, Garcia P, Braun S, et al. Head-to-head comparison of Microflex LT and Vitek MS systems for routine identification of microorganisms by MALDI-TOF mass spectrometry in Chile. PLoS One. 2017;12:e0177929. 47. Pence MA, McElvania TeKippe E, Wallace MA, Burnham CA. Comparison and optimization of two MALDI-TOF MS platforms for the identification of medically relevant yeast species. Eur J Clin Microbiol Infect Dis. 2014;33:1703–12.

226


48. McElvania TeKippe E, Burnham CA. Evaluation of the Bruker Biotyper and VITEK MS MALDI-TOF MS systems for the identification of unusual and/or difficult-to-identify microorganisms isolated from clinical specimens. Eur J Clin Microbiol Infect Dis. 2014;33:2163–71. 49. Martiny D, Busson L, Wybo I, El Haj RA, Dediste A, Vandenberg O. Comparison of the Microflex LT and Vitek MS systems for routine identification of bacteria by matrix- assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2012;50:1313–25. 50. Levesque S, Dufresne PJ, Soualhine H, et al. A side by side comparison of Bruker Biotyper and VITEK MS: utility of MALDI-TOF MS technology for microorganism identification in a public health reference laboratory. PLoS One. 2015;10:e0144878. 51. Karpanoja P, Harju I, Rantakokko-Jalava K, Haanpera M, Sarkkinen H. Evaluation of two matrix-assisted laser desorption ionization-time of flight mass spectrometry systems for identification of viridans group streptococci. Eur J Clin Microbiol Infect Dis. 2014; 33:779–88. 52. Justesen US, Holm A, Knudsen E, et al. Species identification of clinical isolates of anaerobic bacteria: a comparison of two matrix-assisted laser desorption ionization-time of flight mass spectrometry systems. J Clin Microbiol. 2011;49:4314–8. 53. Jamal W, Albert MJ, Rotimi VO. Real-time comparative evaluation of bioMerieux VITEK MS versus Bruker Microflex MS, two matrix-assisted laser desorption-ionization time-of- flight mass spectrometry systems, for identification of clinically significant bacteria. BMC Microbiol. 2014;14:289. 54. Fang H, Ohlsson AK, Ullberg M, Ozenci V. Evaluation of species-specific PCR, Bruker MS, VITEK MS and the VITEK 2 system for the identification of clinical Enterococcus isolates. Eur J Clin Microbiol Infect Dis. 2012;31:3073–7. 55. Deak E, Charlton CL, Bobenchik AM, et al. Comparison of the Vitek MS and Bruker Microflex LT MALDI-TOF MS platforms for routine identification of commonly isolated bacteria and yeast in the clinical microbiology laboratory. Diagn Microbiol Infect Dis. 2015;81:27–33. 56. Cherkaoui A, Hibbs J, Emonet S, et al. Comparison of two matrix-assisted laser desorption ionization-time of flight mass spectrometry methods with conventional phenotypic identification for routine identification of bacteria to the species level. J Clin Microbiol. 2010;48:1169–75. 57. Agergaard CN, Knudsen E, Dargis R, Nielsen XC, Christensen JJ, Justesen US. Species identification of Streptococcus bovis group isolates causing bacteremia: a comparison of two MALDI-TOF MS systems. Diagn Microbiol Infect Dis. 2017;88:23–5. 58. Harmsen D, Rothganger J, Frosch M, Albert J. RIDOM: ribosomal differentiation of medical micro-organisms database. Nucleic Acids Res. 2002;30:416–7. 59. Pailhories H, Daure S, Eveillard M, Joly-Guillou ML, Kempf M. Using Vitek MALDI- TOF mass spectrometry to identify species belonging to the Acinetobacter calcoaceticus- Acinetobacter baumannii complex: a relevant alternative to molecular biology? Diagn Microbiol Infect Dis. 2015;83:99–104. 60. Donohue MJ, Smallwood AW, Pfaller S, Rodgers M, Shoemaker JA. The development of a matrix-assisted laser desorption/ionization mass spectrometry-based method for the protein fingerprinting and identification of Aeromonas species using whole cells. J Microbiol Methods. 2006;65:380–9. 61. Coltella L, Mancinelli L, Onori M, et al. Advancement in the routine identification of anaerobic bacteria by MALDI-TOF mass spectrometry. Eur J Clin Microbiol Infect Dis. 2013;32:1183–92. 62. Fedorko DP, Drake SK, Stock F, Murray PR. Identification of clinical isolates of anaerobic bacteria using matrix-assisted laser desorption ionization-time of flight mass spectrometry. Eur J Clin Microbiol Infect Dis. 2012;31:2257–62.


227

63. Grosse-Herrenthey A, Maier T, Gessler F, et al. Challenging the problem of clostridial identification with matrix-assisted laser desorption and ionization-time-of-flight mass spectrometry (MALDI-TOF MS). Anaerobe. 2008;14:242–9. 64. La Scola B, Fournier PE, Raoult D. Burden of emerging anaerobes in the MALDI-TOF and 16S rRNA gene sequencing era. Anaerobe. 2011;17:106–12. 65. Nagy E, Maier T, Urban E, Terhes G, Kostrzewa M, Bacteria ESGoARiA. Species identification of clinical isolates of Bacteroides by matrix-assisted laser-desorption/ionization time-offlight mass spectrometry. Clin Microbiol Infect. 2009;15:796–802. 66. Shah HN, Keys CJ, Schmid O, Gharbia SE. Matrix-assisted laser desorption/ionization time- of-flight mass spectrometry and proteomics: a new era in anaerobic microbiology. Clin Infect Dis. 2002;35:S58–64. 67. Stingu CS, Rodloff AC, Jentsch H, Schaumann R, Eschrich K. Rapid identification of oral anaerobic bacteria cultivated from subgingival biofilm by MALDI-TOF-MS. Oral Microbiol Immunol. 2008;23:372–6. 68. Veloo AC, Erhard M, Welker M, Welling GW, Degener JE. Identification of Gram-positive anaerobic cocci by MALDI-TOF mass spectrometry. Syst Appl Microbiol. 2011;34:58–62. 69. Veloo AC, Welling GW, Degener JE. The identification of anaerobic bacteria using MALDI- TOF MS. Anaerobe. 2011;17:211–2. 70. Hotta Y, Sato J, Sato H, Hosoda A, Tamura H. Classification of the genus Bacillus based on MALDI-TOF MS analysis of ribosomal proteins coded in S10 and spc operons. J Agric Food Chem. 2011;59:5222–30. 71. Culebras E, Rodriguez-Avial I, Betriu C, Gomez M, Picazo JJ. Rapid identification of clinical isolates of Bacteroides species by matrix-assisted laser-desorption/ionization time-of-flight mass spectrometry. Anaerobe. 2012;18:163–5. 72. Fournier PE, Couderc C, Buffet S, Flaudrops C, Raoult D. Rapid and cost-effective identification of Bartonella species using mass spectrometry. J Med Microbiol. 2009;58:1154–9. 73. Ferreira L, Vega Castano S, Sanchez-Juanes F, et al. Identification of Brucella by MALDI- TOF mass spectrometry. Fast and reliable identification from agar plates and blood cultures. PLoS ONE. 2010;5:e14235. 74. Lista F, Reubsaet FA, De Santis R, et al. Reliable identification at the species level of Brucella isolates with MALDI-TOF-MS. BMC Microbiol. 2011;11:267. 75. Alispahic M, Hummel K, Jandreski-Cvetkovic D, et al. Species-specific identification and differentiation of Arcobacter, Helicobacter and Campylobacter by full-spectral matrix- associated laser desorption/ionization time of flight mass spectrometry analysis. J Med Microbiol. 2010;59:295–301. 76. Bessede E, Solecki O, Sifre E, Labadi L, Megraud F. Identification of Campylobacter species and related organisms by matrix assisted laser desorption ionization-time of flight (MALDI- TOF) mass spectrometry. Clin Microbiol Infect. 2011;17:1735–9. 77. Mandrell RE, Harden LA, Bates A, Miller WG, Haddon WF, Fagerquist CK. Speciation of Campylobacter coli, C. jejuni, C. helveticus, C. lari, C. sputorum, and C. upsaliensis by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl Environ Microbiol. 2005;71:6292–307. 78. Winkler MA, Uher J, Cepa S. Direct analysis and identification of Helicobacter and Campylobacter species by MALDI-TOF mass spectrometry. Anal Chem. 1999;71:3416–9. 79. Wallet F, Loiez C, Decoene C, Courcol R. Rapid identification of Cardiobacterium hominis by MALDI-TOF mass spectrometry during infective endocarditis. Jpn J Infect Dis. 2011;64:327–9. 80. Zaluga J, Heylen K, Van Hoorde K, et al. GyrB sequence analysis and MALDI-TOF MS as identification tools for plant pathogenic Clavibacter. Syst Appl Microbiol. 2011;34:400–7. 81. Konrad R, Berger A, Huber I, et al. Matrix-assisted laser desorption/ionisation time-of-flight (MALDI-TOF) mass spectrometry as a tool for rapid diagnosis of potentially toxigenic Corynebacterium species in the laboratory management of diphtheria-associated bacteria. Euro Surveill. 2010;15:19699

228


82. Hernychova L, Toman R, Ciampor F, et al. Detection and identification of Coxiella burnetii based on the mass spectrometric analyses of the extracted proteins. Anal Chem. 2008;80:7097–104. 83. Conway GC, Smole SC, Sarracino DA, Arbeit RD, Leopold PE. Phyloproteomics: species identification of Enterobacteriaceae using matrix-assisted laser desorption/ionization time- of-flight mass spectrometry. J Mol Microbiol Biotechnol. 2001;3:103–12. 84. Lynn EC, Chung MC, Tsai WC, Han CC. Identification of Enterobacteriaceae bacteria by direct matrix-assisted laser desorption/ionization mass spectrometric analysis of whole cells. Rapid Commun Mass Spectrom. 1999;13:2022–7. 85. Saffert RT, Cunningham SA, Ihde SM, Jobe KE, Mandrekar J, Patel R. Comparison of Bruker Biotyper matrix-assisted laser desorption ionization-time of flight mass spectrometer to BD Phoenix automated microbiology system for identification of gram-negative bacilli. J Clin Microbiol. 2011;49:887–92. 86. Alispahic M, Christensen H, Hess C, Razzazi-Fazeli E, Bisgaard M, Hess M. Identification of Gallibacterium species by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry evaluated by multilocus sequence analysis. Int J Med Microbiol. 2011;301:513–22. 87. Couturier MR, Mehinovic E, Croft AC, Fisher MA. Identification of HACEK clinical isolates by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2011;49:1104–6. 88. Schulthess B, Bloemberg GV, Zbinden A, et al. Evaluation of the Bruker MALDI Biotyper for identification of fastidious Gram-negative rods. J Clin Microbiol. 2016;54:543–8. 89. Haag AM, Taylor SN, Johnston KH, Cole RB. Rapid identification and speciation of Haemophilus bacteria by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. J Mass Spectrom. 1998;33:750–6. 90. Mazzeo MF, Sorrentino A, Gaita M, et al. Matrix-assisted laser desorption ionization-time of flight mass spectrometry for the discrimination of food-borne microorganisms. Appl Environ Microbiol. 2006;72:1180–9. 91. Gaia V, Casati S, Tonolla M. Rapid identification of Legionella spp. by MALDI-TOF MS based protein mass fingerprinting. Syst Appl Microbiol. 2011;34:40–4. 92. He Y, Chang TC, Li H, Shi G, Tang YW. Matrix-assisted laser desorption ionization time- of- flight mass spectrometry and database for identification of Legionella species. Can J Microbiol. 2011;57:533–8. 93. Moliner C, Ginevra C, Jarraud S, et al. Rapid identification of Legionella species by mass spectrometry. J Med Microbiol. 2010;59:273–84. 94. Barbuddhe SB, Maier T, Schwarz G, et al. Rapid identification and typing of listeria species by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl Environ Microbiol. 2008;74:5402–7. 95. Bouakaze C, Keyser C, Gonzalez A, et al. Matrix-assisted laser desorption ionization-time of flight mass spectrometry-based single nucleotide polymorphism genotyping assay using iPLEX gold technology for identification of Mycobacterium tuberculosis complex species and lineages. J Clin Microbiol. 2011;49:3292–9. 96. Hettick JM, Kashon ML, Slaven JE, et al. Discrimination of intact mycobacteria at the strain level: a combined MALDI-TOF MS and biostatistical analysis. Proteomics. 2006;6:6416–25. 97. Lotz A, Ferroni A, Beretti JL, et al. Rapid identification of mycobacterial whole cells in solid and liquid culture media by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2010;48:4481–6. 98. Pignone M, Greth KM, Cooper J, Emerson D, Tang J. Identification of mycobacteria by matrix-assisted laser desorption ionization-time-of-flight mass spectrometry. J Clin Microbiol. 2006;44:1963–70. 99. Saleeb PG, Drake SK, Murray PR, Zelazny AM. Identification of mycobacteria in solid- culture media by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2011;49:1790–4.


229

100. Tseng SP, Teng SH, Lee PS, Wang CF, Yu JS, Lu PL. Rapid identification of M. abscessus and M. massiliense by MALDI-TOF mass spectrometry with a comparison to sequencing methods and antimicrobial susceptibility patterns. Future Microbiol. 2013;8:1381–9. 101. Verroken A, Janssens M, Berhin C, et al. Evaluation of matrix-assisted laser desorption ionization-time of flight mass spectrometry for identification of nocardia species. J Clin Microbiol. 2010;48:4015–21. 102. Degand N, Carbonnelle E, Dauphin B, et al. Matrix-assisted laser desorption ionization-time of flight mass spectrometry for identification of nonfermenting gram-negative bacilli isolated from cystic fibrosis patients. J Clin Microbiol. 2008;46:3361–7. 103. Jacquier H, Carbonnelle E, Corvec S, et al. Revisited distribution of nonfermenting Gram- negative bacilli clinical isolates. Eur J Clin Microbiol Infect Dis. 2011;30:1579–86. 104. Marko DC, Saffert RT, Cunningham SA, et al. Evaluation of the Bruker Biotyper and Vitek MS matrix-assisted laser desorption ionization-time of flight mass spectrometry systems for identification of nonfermenting gram-negative bacilli isolated from cultures from cystic fibrosis patients. J Clin Microbiol. 2012;50:2034–9. 105. Vanlaere E, Sergeant K, Dawyndt P, et al. Matrix-assisted laser desorption ionisation-time- of-flight mass spectrometry of intact cells allows rapid identification of Burkholderia cepacia complex. J Microbiol Methods. 2008;75:279–86. 106. Rezzonico F, Vogel G, Duffy B, Tonolla M. Application of whole-cell matrix-assisted laser desorption ionization-time of flight mass spectrometry for rapid identification and clustering analysis of pantoea species. Appl Environ Microbiol. 2010;76:4497–509. 107. Kolinska R, Drevinek M, Aldova E, Zemlickova H. Identification of Plesiomonas spp.: serological and MALDI-TOF MS methods. Folia Microbiol (Praha). 2010;55:669–72. 108. Dieckmann R, Helmuth R, Erhard M, Malorny B. Rapid classification and identification of salmonellae at the species and subspecies levels by whole-cell matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl Environ Microbiol. 2008;74:7767–78. 109. Dieckmann R, Malorny B. Rapid screening of epidemiologically important Salmonella enterica subsp. enterica serovars by whole-cell matrix-assisted laser desorption ionization- time of flight mass spectrometry. Appl Environ Microbiol. 2011;77:4136–46. 110. Sparbier K, Weller U, Boogen C, Kostrzewa M. Rapid detection of Salmonella sp. by means of a combination of selective enrichment broth and MALDI-TOF MS. Eur J Clin Microbiol Infect Dis. 2012;31:767–73. 111. Bergeron M, Dauwalder O, Gouy M, et al. Species identification of staphylococci by amplification and sequencing of the tuf gene compared to the gap gene and by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Eur J Clin Microbiol Infect Dis. 2011;30:343–54. 112. Bernardo K, Pakulat N, Macht M, et al. Identification and discrimination of Staphylococcus aureus strains using matrix-assisted laser desorption/ionization-time of flight mass spectrometry. Proteomics. 2002;2:747–53. 113. Carpaij N, Willems RJ, Bonten MJ, Fluit AC. Comparison of the identification of coagulase- negative staphylococci by matrix-assisted laser desorption ionization time-of-flight mass spectrometry and tuf sequencing. Eur J Clin Microbiol Infect Dis. 2011;30:1169–72. 114. Decristophoris P, Fasola A, Benagli C, Tonolla M, Petrini O. Identification of Staphylococcus intermedius group by MALDI-TOF MS. Syst Appl Microbiol. 2011;34:45–51. 115. Dubois D, Leyssene D, Chacornac JP, et al. Identification of a variety of Staphylococcus species by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2010;48:941–5. 116. Dupont C, Sivadon-Tardy V, Bille E, et al. Identification of clinical coagulase-negative staphylococci, isolated in microbiology laboratories, by matrix-assisted laser desorption/ionization-time of flight mass spectrometry and two automated systems. Clin Microbiol Infect. 2010;16:998–1004.

230


117. Elbehiry A, Al-Dubaib M, Marzouk E, Osman S, Edrees H. Performance of MALDI biotyper compared with Vitek 2 compact system for fast identification and discrimination of Staphylococcus species isolated from bovine mastitis. Microbiology. 2016;5:1061–70. 118. Rajakaruna L, Hallas G, Molenaar L, et al. High throughput identification of clinical isolates of Staphylococcus aureus using MALDI-TOF-MS of intact cells. Infect Genet Evol. 2009;9:507–13. 119. Vasileuskaya-Schulz Z, Kaiser S, Maier T, Kostrzewa M, Jonas D. Delineation of Stenotrophomonas spp. by multi-locus sequence analysis and MALDI-TOF mass spectrometry. Syst Appl Microbiol. 2011;34:35–9. 120. Cherkaoui A, Emonet S, Fernandez J, Schorderet D, Schrenzel J. Evaluation of matrix- assisted laser desorption ionization-time of flight mass spectrometry for rapid identification of Beta-hemolytic streptococci. J Clin Microbiol. 2011;49:3004–5. 121. Friedrichs C, Rodloff AC, Chhatwal GS, Schellenberger W, Eschrich K. Rapid identification of viridans streptococci by mass spectrometric discrimination. J Clin Microbiol. 2007;45:2392–7. 122. Hinse D, Vollmer T, Erhard M, et al. Differentiation of species of the Streptococcus bovis/equinus-complex by MALDI-TOF mass spectrometry in comparison to sodA sequence analyses. Syst Appl Microbiol. 2011;34:52–7. 123. Lartigue MF, Kostrzewa M, Salloum M, et al. Rapid detection of “highly virulent” Group B Streptococcus ST-17 and emerging ST-1 clones by MALDI-TOF mass spectrometry. J Microbiol Methods. 2011;86:262–5. 124. Rupf S, Breitung K, Schellenberger W, Merte K, Kneist S, Eschrich K. Differentiation of mutans streptococci by intact cell matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Oral Microbiol Immunol. 2005;20:267–73. 125. Dieckmann R, Strauch E, Alter T. Rapid identification and characterization of Vibrio species using whole-cell MALDI-TOF mass spectrometry. J Appl Microbiol. 2010;109:199–211. 126. Ayyadurai S, Flaudrops C, Raoult D, Drancourt M. Rapid identification and typing of Yersinia pestis and other Yersinia species by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry. BMC Microbiol. 2010;10:285. 127. Lasch P, Drevinek M, Nattermann H, et al. Characterization of Yersinia using MALDI-TOF mass spectrometry and chemometrics. Anal Chem. 2010;82:8464–75. 128. Stephan R, Cernela N, Ziegler D, et al. Rapid species specific identification and subtyping of Yersinia enterocolitica by MALDI-TOF mass spectrometry. J Microbiol Methods. 2011;87:150–3. 129. Alanio A, Beretti JL, Dauphin B, et al. Matrix-assisted laser desorption ionization time-of- flight mass spectrometry for fast and accurate identification of clinically relevant Aspergillus species. Clin Microbiol Infect. 2011;17:750–5. 130. De Carolis E, Posteraro B, Lass-Florl C, et al. Species identification of Aspergillus, Fusarium and Mucorales with direct surface analysis by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Clin Microbiol Infect. 2012;18:475–84. 131. Hettick JM, Green BJ, Buskirk AD, et al. Discrimination of Penicillium isolates by matrix- assisted laser desorption/ionization time-of-flight mass spectrometry fingerprinting. Rapid Commun Mass Spectrom. 2008;22:2555–60. 132. Quiles-Melero I, Garcia-Rodriguez J, Gomez-Lopez A, Mingorance J. Evaluation of matrix- assisted laser desorption/ionisation time-of-flight (MALDI-TOF) mass spectrometry for identification of Candida parapsilosis, C. orthopsilosis and C. metapsilosis. Eur J Clin Microbiol Infect Dis. 2012;31:67–71. 133. McTaggart LR, Lei E, Richardson SE, Hoang L, Fothergill A, Zhang SX. Rapid identification of Cryptococcus neoformans and Cryptococcus gattii by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2011;49:3050–3. 134. Kemptner J, Marchetti-Deschmann M, Mach R, Druzhinina IS, Kubicek CP, Allmaier G. Evaluation of matrix-assisted laser desorption/ionization (MALDI) preparation tech-


231

niques for surface characterization of intact Fusarium spores by MALDI linear time-of-flight mass spectrometry. Rapid Commun Mass Spectrom. 2009;23:877–84. 135. Erhard M, Hipler UC, Burmester A, Brakhage AA, Wostemeyer J. Identification of dermatophyte species causing onychomycosis and tinea pedis by MALDI-TOF mass spectrometry. Exp Dermatol. 2008;17:356–61. 136. Coulibaly O, Marinach-Patrice C, Cassagne C, Piarroux R, Mazier D, Ranque S. Pseudallescheria/Scedosporium complex species identification by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Med Mycol. 2011;49:621–6. 137. Bader O, Weig M, Taverne-Ghadwal L, Lugert R, Gross U, Kuhns M. Improved clinical laboratory identification of human pathogenic yeasts by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Clin Microbiol Infect. 2011;17:1359–65. 138. Dhiman N, Hall L, Wohlfiel SL, Buckwalter SP, Wengenack NL. Performance and cost analysis of matrix-assisted laser desorption ionization-time of flight mass spectrometry for routine identification of yeast. J Clin Microbiol. 2011;49:1614–6. 139. Ghosh AK, Paul S, Sood P, et al. Matrix-assisted laser desorption ionization time-of-flight mass spectrometry for the rapid identification of yeasts causing bloodstream infections. Clin Microbiol Infect. 2015;21:372–8. 140. Kaleta EJ, Clark AE, Cherkaoui A, et al. Comparative analysis of PCR-electrospray ionization/mass spectrometry (MS) and MALDI-TOF/MS for the identification of bacteria and yeast from positive blood culture bottles. Clin Chem. 2011;57:1057–67. 141. Marklein G, Josten M, Klanke U, et al. Matrix-assisted laser desorption ionization-time of flight mass spectrometry for fast and reliable identification of clinical yeast isolates. J Clin Microbiol. 2009;47:2912–7. 142. Putignani L, Del Chierico F, Onori M, et al. MALDI-TOF mass spectrometry proteomic phenotyping of clinically relevant fungi. Mol BioSyst. 2011;7:620–9. 143. Qian J, Cutler JE, Cole RB, Cai Y. MALDI-TOF mass signatures for differentiation of yeast species, strain grouping and monitoring of morphogenesis markers. Anal Bioanal Chem. 2008;392:439–49. 144. Stevenson LG, Drake SK, Shea YR, Zelazny AM, Murray PR. Evaluation of matrix-assisted laser desorption ionization-time of flight mass spectrometry for identification of clinically important yeast species. J Clin Microbiol. 2010;48:3482–6. 145. Van Herendael BH, Bruynseels P, Bensaid M, et al. Validation of a modified algorithm for the identification of yeast isolates using matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF MS). Eur J Clin Microbiol Infect Dis. 2012;31:841–8. 146. Seng P, Rolain JM, Fournier PE, La Scola B, Drancourt M, Raoult D. MALDI-TOF-mass spectrometry applications in clinical microbiology. Future Microbiol. 2010;5:1733–54. 147. Benagli C, Rossi V, Dolina M, Tonolla M, Petrini O. Matrix-assisted laser desorption ionization-time of flight mass spectrometry for the identification of clinically relevant bacteria. PLoS One. 2011;6:e16424. 148. Neville SA, Lecordier A, Ziochos H, et al. Utility of matrix-assisted laser desorption ionization-time of flight mass spectrometry following introduction for routine laboratory bacterial identification. J Clin Microbiol. 2011;49:2980–4. 149. Ruelle V, El Moualij B, Zorzi W, Ledent P, Pauw ED. Rapid identification of environmental bacterial strains by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun Mass Spectrom. 2004;18:2013–9. 150. Warscheid B, Fenselau C. A targeted proteomics approach to the rapid identification of bacterial cell mixtures by matrix-assisted laser desorption/ionization mass spectrometry. Proteomics. 2004;4:2877–92. 151. Alatoom AA, Cunningham SA, Ihde SM, Mandrekar J, Patel R. Comparison of direct colony method versus extraction method for identification of gram-positive cocci by use of Bruker Biotyper matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2011;49:2868–73.

232


152. Ferreira L, Sanchez-Juanes F, Munoz-Bellido JL, Gonzalez-Buitrago JM. Rapid method for direct identification of bacteria in urine and blood culture samples by matrix-assisted laser desorption ionization time-of-flight mass spectrometry: intact cell vs. extraction method. Clin Microbiol Infect. 2011;17:1007–12. 153. Drancourt M. Detection of microorganisms in blood specimens using matrix-assisted laser desorption ionization time-of-flight mass spectrometry: a review. Clin Microbiol Infect. 2010;16:1620–5. 154. van Veen SQ, Claas EC, Kuijper EJ. High-throughput identification of bacteria and yeast by matrix-assisted laser desorption ionization-time of flight mass spectrometry in conventional medical microbiology laboratories. J Clin Microbiol. 2010;48:900–7. 155. Risch M, Radjenovic D, Han JN, Wydler M, Nydegger U, Risch L. Comparison of MALDI TOF with conventional identification of clinically relevant bacteria. Swiss Med Wkly. 2010;140:w13095. 156. Hoffmann H, Stindl S, Ludwig W, et al. Enterobacter hormaechei subsp. oharae subsp. nov., E. hormaechei subsp. hormaechei comb. nov., and E. hormaechei subsp. steigerwaltii subsp. nov., three new subspecies of clinical importance. J Clin Microbiol. 2005;43:3297–303. 157. Paauw A, Caspers MP, Schuren FH, et al. Genomic diversity within the Enterobacter cloacae complex. PLoS One. 2008;3:e3018. 158. Blosser SJ, Drake SK, Andrasko JL, et al. Multicenter matrix-assisted laser desorption ionization-time of flight mass spectrometry study for identification of clinically relevant Nocardia spp. J Clin Microbiol. 2016;54:1251–8. 159. Branda JA, Rychert J, Burnham CA, et al. Multicenter validation of the VITEK MS v2.0 MALDI-TOF mass spectrometry system for the identification of fastidious gram-negative bacteria. Diagn Microbiol Infect Dis. 2014;78:129–31. 160. Garner O, Mochon A, Branda J, et al. Multi-centre evaluation of mass spectrometric identification of anaerobic bacteria using the VITEK(R) MS system. Clin Microbiol Infect. 2014;20:335–9. 161. Lasch P, Wahab T, Weil S, et al. Identification of highly pathogenic microorganisms by matrix-assisted laser desorption ionization-time of flight mass spectrometry: results of an interlaboratory ring trial. J Clin Microbiol. 2015;53:2632–40. 162. Manji R, Bythrow M, Branda JA, et al. Multi-center evaluation of the VITEK(R) MS system for mass spectrometric identification of non-Enterobacteriaceae Gram-negative bacilli. Eur J Clin Microbiol Infect Dis. 2014;33:337–46. 163. Mellmann A, Bimet F, Bizet C, et al. High interlaboratory reproducibility of matrix-assisted laser desorption ionization-time of flight mass spectrometry-based species identification of nonfermenting bacteria. J Clin Microbiol. 2009;47:3732–4. 164. Rychert J, Burnham CA, Bythrow M, et al. Multicenter evaluation of the Vitek MS matrix- assisted laser desorption ionization-time of flight mass spectrometry system for identification of Gram-positive aerobic bacteria. J Clin Microbiol. 2013;51:2225–31. 165. Rychert J, Slechta ES, Barker AP, et al. Multicenter evaluation of the Vitek MS v3.0 system for the identification of filamentous fungi. J Clin Microbiol. 2018;56:e01353–17. 166. Schieffer KM, Tan KE, Stamper PD, et al. Multicenter evaluation of the Sepsityper extraction kit and MALDI-TOF MS for direct identification of positive blood culture isolates using the BD BACTEC FX and VersaTREK((R)) diagnostic blood culture systems. J Appl Microbiol. 2014;116:934–41. 167. Veloo ACM, Jean-Pierre H, Justesen US, et al. A multi-center ring trial for the identification of anaerobic bacteria using MALDI-TOF MS. Anaerobe. 2017;48:94–7. 168. Vlek A, Kolecka A, Khayhan K, et al. Interlaboratory comparison of sample preparation methods, database expansions, and cutoff values for identification of yeasts by matrix-assisted laser desorption ionization-time of flight mass spectrometry using a yeast test panel. J Clin Microbiol. 2014;52:3023–9. 169. Hathout Y, Demirev PA, Ho YP, et al. Identification of Bacillus spores by matrix-assisted laser desorption ionization-mass spectrometry. Appl Environ Microbiol. 1999;65:4313–9. 170. Tao L, Yu X, Snyder AP, Li L. Bacterial identification by protein mass mapping combined with an experimentally derived protein mass database. Anal Chem. 2004;76:6609–17.


233

171. Kumar A, Ellis P, Arabi Y, et al. Initiation of inappropriate antimicrobial therapy results in a fivefold reduction of survival in human septic shock. Chest. 2009;136:1237–48. 172. Yan Y, He Y, Maier T, et al. Improved identification of yeast species directly from positive blood culture media by combining Sepsityper specimen processing and Microflex analysis with the matrix-assisted laser desorption ionization Biotyper system. J Clin Microbiol. 2011;49:2528–32. 173. Schmidt V, Jarosch A, Marz P, Sander C, Vacata V, Kalka-Moll W. Rapid identification of bacteria in positive blood culture by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Eur J Clin Microbiol Infect Dis. 2012;31:311–7. 174. Mauri C, Principe L, Bracco S, et al. Identification by mass spectrometry and automated susceptibility testing from positive bottles: a simple, rapid, and standardized approach to reduce the turnaround time in the management of blood cultures. BMC Infect Dis. 2017;17:749. 175. Huang TS, Lee CC, Tu HZ, Lee SS. Rapid identification of mycobacteria from positive MGIT broths of primary cultures by MALDI-TOF mass spectrometry. PLoS One. 2018;13:e0192291. 176. Fothergill A, Kasinathan V, Hyman J, Walsh J, Drake T, Wang YF. Rapid identification of bacteria and yeasts from positive-blood-culture bottles by using a lysis-filtration method and matrix-assisted laser desorption ionization-time of flight mass spectrum analysis with the SARAMIS database. J Clin Microbiol. 2013;51:805–9. 177. Burillo A, Rodriguez-Sanchez B, Ramiro A, Cercenado E, Rodriguez-Creixems M, Bouza E. Gram-stain plus MALDI-TOF MS (matrix-assisted laser desorption ionization-time of flight mass spectrometry) for a rapid diagnosis of urinary tract infection. PLoS One. 2014;9:e86915. 178. Barnini S, Ghelardi E, Brucculeri V, Morici P, Lupetti A. Rapid and reliable identification of Gram-negative bacteria and Gram-positive cocci by deposition of bacteria harvested from blood cultures onto the MALDI-TOF plate. BMC Microbiol. 2015;15:124. 179. Ferreira L, Sanchez-Juanes F, Porras-Guerra I, et al. Microorganisms direct identification from blood culture by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Clin Microbiol Infect. 2011;17:546–51. 180. Moussaoui W, Jaulhac B, Hoffmann AM, et al. Matrix-assisted laser desorption ionization time-of-flight mass spectrometry identifies 90% of bacteria directly from blood culture vials. Clin Microbiol Infect. 2010;16:1631–8. 181. Romero-Gomez MP, Mingorance J. The effect of the blood culture bottle type in the rate of direct identification from positive cultures by matrix-assisted laser desorption/ionisation time-of-flight (MALDI-TOF) mass spectrometry. J Infect. 2011;62:251–3. 182. Kok J, Thomas LC, Olma T, Chen SC, Iredell JR. Identification of bacteria in blood culture broths using matrix-assisted laser desorption-ionization Sepsityper and time of flight mass spectrometry. PLoS One. 2011;6:e23285. 183. Schubert S, Weinert K, Wagner C, et al. Novel, improved sample preparation for rapid, direct identification from positive blood cultures using matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry. J Mol Diagn. 2011;13:701–6. 184. Kroumova V, Gobbato E, Basso E, Mucedola L, Giani T, Fortina G. Direct identification of bacteria in blood culture by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry: a new methodological approach. Rapid Commun Mass Spectrom. 2011;25:2247–9. 185. Edwards-Jones V, Claydon MA, Evason DJ, Walker J, Fox AJ, Gordon DB. Rapid discrimination between methicillin-sensitive and methicillin-resistant Staphylococcus aureus by intact cell mass spectrometry. J Med Microbiol. 2000;49:295–300. 186. Du Z, Yang R, Guo Z, Song Y, Wang J. Identification of Staphylococcus aureus and determination of its methicillin resistance by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal Chem. 2002;74:5487–91.

234


187. Camara JE, Hays FA. Discrimination between wild-type and ampicillin-resistant Escherichia coli by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal Bioanal Chem. 2007;389:1633–8. 188. Burckhardt I, Zimmermann S. Using matrix-assisted laser desorption ionization-time of flight mass spectrometry to detect carbapenem resistance within 1 to 2.5 hours. J Clin Microbiol. 2011;49:3321–4. 189. Hrabak J, Walkova R, Studentova V, Chudackova E, Bergerova T. Carbapenemase activity detection by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2011;49:3222–7. 190. Marinach C, Alanio A, Palous M, et al. MALDI-TOF MS-based drug susceptibility testing of pathogens: the example of Candida albicans and fluconazole. Proteomics. 2009;9:4627–31. 191. Idelevich EA, Sparbier K, Kostrzewa M, Becker K. Rapid detection of antibiotic resistance by MALDI-TOF mass spectrometry using a novel direct-on-target microdroplet growth assay. Clin Microbiol Infect. 2018;24(7):738–43. https://doi.org/10.1016/j.cmi.2017.10.016. Epub 2017 Oct 24. 192. Spinali S, van Belkum A, Goering RV, et al. Microbial typing by matrix-assisted laser desorption ionization-time of flight mass spectrometry: do we need guidance for data interpretation? J Clin Microbiol. 2015;53:760–5. 193. Lartigue MF. Matrix-assisted laser desorption ionization time-of-flight mass spectrometry for bacterial strain characterization. Infect Genet Evol. 2013;13:230–5. 194. Mencacci A, Monari C, Leli C, et al. Typing of nosocomial outbreaks of Acinetobacter baumannii by use of matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol. 2013;51:603–6.

Transcriptomic Techniques in Diagnostic Microbiology Zachary E. Holcomb and Ephraim L. Tsalik

Introduction Within the realm of infectious diseases, rapid and accurate diagnosis is crucial for appropriately guiding triage and treatment decisions. Historically, diagnosis of infectious diseases has relied on pathogen detection, with microbial culture considered the gold standard. However, pathogen detection-based diagnostics often have limited sensitivity and specificity, require a clinical suspicion for the pathogen of interest, display prolonged time to result, or do not reliably distinguish between infection and colonization. Given the rise in multidrug resistance resulting from inappropriate antibacterial use, the threats of emerging pandemic infections, and increased morbidity and mortality from diagnostic delay, a need exists for improved diagnostics in the field of infectious diseases. An emerging alternative to pathogen detection-based modalities involves analysis of the host immune response to infection. This is not a new concept, as macroscopic observation of the host immune response through qualifying the syndromic manifestation of disease (e.g., fever, purulence, erythema) has been used throughout history to diagnose infections. Furthermore, host biomarkers, which are quantifiable indicators of a biological state produced by the host, have been used for nearly a

Z. E. Holcomb (*) Duke University School of Medicine, Durham, NC, USA e-mail: [email protected] E. L. Tsalik Emergency Medicine Service, Durham VAMC, Durham, NC, USA Center for Applied Genomics & Precision Medicine, Duke University Medical Center, Durham, NC, USA Division of Infectious Diseases, Department of Medicine, Duke University Medical Center, Durham, NC, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_11

235

236

Z. E. Holcomb and E. L. Tsalik

century to quantify the host response to infection and inflammation. The first of these focused on measuring single biomarkers, such as the erythrocyte sedimentation rate or C-reactive protein as general markers of inflammation [1, 2]. More recently, the single-analyte biomarker procalcitonin (PCT) has been utilized to indicate the presence or absence of bacterial infection in both sepsis and lower respiratory tract infections [3, 4]. However, all of these single-analyte biomarkers display limited sensitivity and specificity and therefore demonstrate efficacy only within highly focused clinical scenarios. These pitfalls can largely be overcome by using multi-analyte biomarker panels to quantify the host response to infectious perturbations. However, this approach quickly becomes computationally challenging, demanding new techniques for derivation and interpretation of biomarker panels. Within the past decade, our approach to human disease has dramatically changed. Sequencing of the human genome and development of computational approaches to analyze large datasets have allowed for new possibilities in terms of analysis and understanding the host immune response to infection. These revolutions have allowed for the analysis of multi-analyte biomarkers in the host, including tests measuring gene expression, protein panels, metabolite panels, cytokines, and more. We will focus here on transcriptomics, the analysis of host gene expression through quantification of the RNA transcripts produced. The advantages of transcriptomics, as compared to traditional pathogen detection methods, include the capacity to distinguish among infectious pathogen classes (such as distinguishing bacterial from viral infection), the capability to differentiate between active infection and colonization, and the ability to prognosticate disease course and predict disease severity [5–7]. Moreover, RNA biomarkers are likely substantially better able to characterize a disease state than peptide biomarkers given the vastly greater amount of DNA transcribed (>80%) compared to the amount of RNA translated (~1.5%) [8–10]. This chapter will focus on the use of transcriptomics for the diagnosis of infectious diseases. It starts with an overview of the discovery and development of a transcriptomic-based disease classifier, including a description of analyzing large datasets and reducing dimensionality. Disease states that have been studied with transcriptomic analysis will then be reviewed, followed by a discussion of current and future clinical applications of these transcriptomic techniques.

Classifier Design Disease classifiers are patterns of biomarker perturbations capable of categorizing individuals into defined clinical groups. A transcriptomic disease classifier consists of multiple genes that are reproducibly up- or downregulated in a certain disease state. Development of a disease-specific “gene signature” requires several broad steps, including a discovery phase to evaluate gene expression changes in the disease state of interest, a disease classifier generation step, and a validation step that evaluates the performance characteristics of the disease signature in an appropriate population [11]. This workflow is summarized in Fig. 1.

237

Transcriptomic Techniques in Diagnostic Microbiology

Phenotype A Phenotype B

Sample collection from cohorts

Peripheral blood

mRNA

Generation of High-Dimensional Gene Expression Data Dimensionality Reduction and Statistical Analysis

Classifier Development

?

?

?

?

Validation of classifier in a separate population Fig. 1 Workflow for development of a gene expression diagnostic classifier

Study Design The first step in this process requires a discovery phase of experimental analysis that defines host gene expression levels due to the disease in question. For a disease classifier to be broadly applicable in a general population, appropriate experimental design is crucial. This means that the study population should include an infected cohort and controls that share similar phenotypic features of syndromic illness. This is particularly important since a disease classifier defined using a sick cohort versus completely healthy controls will not be as useful for clinical practice. Healthy

238


individuals do not typically require a test to differentiate them from the sick. Therefore, clinically relevant similarities should exist between the infected and control groups. For example, a signature of bacterial respiratory infection should use as controls people with non-bacterial causes of an acute respiratory illness syndrome (e.g., viral infection, pulmonary edema, asthma, acute exacerbation of chronic obstructive pulmonary disease). Gene expression differences in these phenotypically similar cohorts will then be more useful diagnostically.

Techniques for Signature Discovery Once experimental cohorts have been identified based on the characteristics of interest, samples are collected for transcriptomic analysis. For most studies, peripheral blood is an ideal source, as it is easily accessible and commonly acquired in most clinical settings. Furthermore, circulating peripheral white blood cells are often directly responding to immune signals cascading from localized infection sites, and therefore these cells provide unique and specific responses to the particular pathogen of interest. Transcriptomic analysis of these circulating immune cells involves measurement of RNA transcripts, which quantifies gene expression alterations across the genome. For most gene expression applications, microarray technology has been utilized due to its relatively low cost, ease of data generation, established standardized methods for analysis, and good quantitative accuracy. Microarrays consist of large collections of thousands of oligonucleotide probes bound to a substrate (either a chip or beads). These probes can detect the presence of complementary sequences in test samples, thereby detailing the amount of RNA present in a sample and thus the levels of gene expression in a semiquantitative manner, since measurements are reported as hybridization signal intensity in relation to control samples [12]. However, a notable limitation of microarrays is that they are restricted to detection of sequences that are complementary to the probes included on the array, thus precluding the detection of many splice and sequence variants. Newer technology using next-generation RNA sequencing (RNA-Seq), initially impeded by cost and analytic complexity, is now rapidly replacing microarray technology due to decreasing prices and improved data management and analysis capabilities. RNA-Seq converts mRNA in the sample to a complementary DNA library and then sequences the library and maps it against a reference genome [12]. The determination of differentially expressed genes in a sample is based on how frequently a given mRNA sequence is detected in the sequencing reaction, so transcriptomic analysis using RNA-Seq depends on counts, while microarray depends on hybridization signal intensity. RNA-Seq provides a snapshot view of the entire transcriptome at the time of sample acquisition and is not limited by the probes present on a microarray [13]. RNA-Seq offers a number of additional advantages, including greater sensitivity and a less-biased view of the transcriptome while simultaneously having the capability to detect expressed sequence variants and splice variants.


239

Classifier Validation Once a disease classifier has been developed, the classifier must be validated in a separate population from which it was derived in order to determine diagnostic accuracy. This step is crucial in determining the performance characteristics of the gene expression test of interest. Finally, the gene expression signature must be hosted on a technological platform to allow it to be translated to the clinical environment. Currently available and future platforms are discussed in greater detail below.

Data Analysis and Interpretation Advances in our diagnostic approach to infectious diseases have led to the recent rise in popularity of genomics, transcriptomics, proteomics, metabolomics, and many other “omics” fields of study as tools for understanding the host response to disease. These approaches have been met with a concomitant need for new, advanced mathematical approaches to analyzing these vast amounts of data. Many of these relatively recent computational developments may be less familiar to microbiologists.

Analytical Hurdles Since measurement and quantification of alterations in gene expression involves collection of vast amounts of complex, high-dimensional data, traditional analytic methods fail in interpretation of this data. New techniques and development of de novo mathematical models are required to explore data on such a large scale, whereby analysis can move beyond the significance of single genes or lists of genes and focus on biologically relevant pathways. The technical details of these computational methods are beyond the scope of this text, but here we will briefly summarize the goals behind this methodology and highlight one specific example of a commonly used approach to gene expression data analysis. Additional details regarding these analytical methods can be found elsewhere in published literature [14–16]. An additional analytical challenge provided by gene expression data is the “large p, small n” problem. The essence of this conundrum is that gene expression studies often involve large numbers of variables (e.g., genes) measured from a small number of samples (e.g., humans or other hosts). One resulting consequence of measuring a large number of genes in a small number of samples is over-fitting of the data [17], meaning that when evaluating the predictive capabilities of tens of thousands of genes, chance alone dictates that some non-differentially expressed genes will still classify disease phenotypes well. This inevitable noise (i.e., falsepositive gene expression correlations) must be accounted for in the determination of statistically significant gene expression patterns.

240


Methods for Dimensionality Reduction With tens of thousands of data points in the form of gene expression levels, the process of accurately and reproducibly analyzing this information requires a reduction in the dimensionality of the data. Put simply, dimensionality reduction is the process of decreasing the number of random variables being considered in the analysis. For example, the Affymetrix U133 Plus 2.0 microarray contains probes for over 47,000 transcripts [18]. Advanced statistical algorithms simplify those 47,000 data points to hundreds or even dozens of composite variables. Within the past decade, many approaches to reducing the dimensionality of gene expression data have been developed, allowing for the prediction of disease phenotypes based on transcriptomic or proteomic changes observed in the host. Examples of such statistical algorithms include sparse factor modeling [19], Bayesian constructions of the elastic net [20], sparse principal component analysis [21], penalized matrix decomposition [22], modular transcriptional analysis [23], and the molecular distance to health [24]. Here we briefly discuss sparse latent factor modeling. In the context of genomics, sparse latent factor regression modeling, also known as Bayesian factor regression modeling (BFRM), involves the identification of “factors.” Factors are sets of genes whose expression is correlated and are thought to represent underlying biological pathways. Following the creation of these factors, which may contain up to hundreds of gene transcripts, the factors themselves serve as variables in linear regression models to predict the outcome variable such as diagnosis or prognosis [25]. In order to simplify analyses of complex high- dimensional data, an assumption is made that there are relatively few associations between variables [26]. This process reduces the number of variables from tens of thousands of gene expression variations (as measured by mRNA transcripts) to dozens or hundreds of factors, which are much more amenable to conventional statistical analysis. Factors that can successfully predict disease phenotype are identified. Subsequently, the individual genes represented by these predictive factors are evaluated for significance and biological plausibility. While sparse latent factor regression modeling is capable of analyzing data with hundreds or even thousands of variables, the process of reducing data dimensionality concomitantly risks obscuring the significant effect of any single gene if it falls within a factor that itself is not associated with the phenotype of interest. Additional details regarding sparse latent factor regression modeling and its usage in genomic datasets are published in the literature [27]. An example application of sparse latent factor regression modeling to gene expression data involved the identification of a signature for the diagnosis of acute respiratory viral infection [28]. This project involved a human challenge study, whereby 57 individuals grouped into three separate cohorts were inoculated intranasally with one of three respiratory viruses (rhinovirus, respiratory syncytial virus, or influenza A). Approximately half of the inoculated volunteers became symptomatic. Peripheral blood was collected for measurement of gene expression, which was used for sparse latent factor regression analysis. This analysis yielded multiple


241

factors capable of classifying subject phenotypes into symptomatic versus asymptomatic infection. The most predictive factor produced in this analysis consisted of 30 unique genes, most of which were previously known to be involved in the host immune response to viral infection. This 30-gene classifier was found to have >95% accuracy in identifying infected subjects from this volunteer population. Subsequently, in a separate validation group, the 30-gene classifier demonstrated 93% accuracy in distinguishing individuals infected with influenza A from individuals with pneumococcal respiratory infections [28]. This represents one of multiple examples of sparse latent factor regression analysis of gene expression data reported in the literature.

Analytical Biases When measuring the up- and downregulation of tens of thousands of genes across the genome, many differences in transcription levels will exist at various points in time, even within the same host, depending on the current biological state (e.g., infected versus healthy, young versus old, etc.). Therefore, gene expression-based signatures are subject to analytical biases if they are not carefully defined in the appropriate populations. In the infectious disease diagnostics realm, this means gene expression changes should be defined in the same populations as those in which the intended diagnostic test will ultimately be used. For example, the comparison of patients with bacterial sepsis to completely healthy controls will undoubtedly generate a multitude of gene expression alterations across the genome, with a heavy focus on up- and downregulation of immune response genes. However, the diagnostic signature derived from these gene expression changes will likely not prove useful in distinguishing bacterial sepsis from noninfectious systemic inflammatory response syndrome (SIRS), as both are inflammatory conditions and their distinction will likely require analysis of a completely different set of genes [11]. Although comparing gene expression changes in the infected host to the healthy control provides invaluable information on disease pathobiology, this comparison is not ideal for diagnostic assay development. Since the clinical distinction between “sick” and “healthy” is not subtle, diagnostic targets defined by the host response should be developed and evaluated in patients with differing phenotypes but similar syndromic presentations. Therefore, in the above example of developing a gene expression-based disease classifier to identify bacterial sepsis, the “control” population should have similar syndromic features of sepsis, such as patients with SIRS. In general, if a disease classifier is used in a population that differs significantly from the population in which it was derived, this classifier would likely yield erroneous and misleading diagnostic information. Although there are many potential biases involved with analysis of gene expression-based disease classifiers, there are also several methods that have been proposed to reduce or avoid biases [29]. One observation has been that many genes

242


from several distinct biological pathways may be up- or downregulated in the diseased state, meaning multiple predictive gene signatures may exist. Since these different signatures represent different groups of genes, isolated focus on only one predictive signature may provide an incomplete picture of the underlying pathobiology, which is, after all, represented by the entire collection of gene expression changes. Also, there will likely be “redundant” biomarkers. These may occur when multiple genes, such as from the same biological pathway, are all highly discriminating of the desired phenotype but are highly correlated with each other. In this scenario, perhaps only one representative gene is necessary for classification. Eliminating redundancy in a classifier helps limit the size of the signature, which is an important consideration for the eventual translation to a useable clinical platform. Finally, in order to agnostically arrive at the genes selected in a signature, any data preprocessing procedures should be designed so as not to bias biomarker selection.

Disease States Host-based gene expression classifiers have been developed in a variety of disease states. Classifiers capable of identifying the class of the infectious pathogen (bacterial, viral, fungal, parasitic) have been more successful in showing clinical utility than classifiers attempting to define the specific pathogen. This type of classification scheme is useful to facilitate appropriate antibiotic usage or guidance of proper empiric antimicrobial therapy in a patient with suspected infection. Here we will briefly discuss examples of host-based gene expression classifiers that have been published across the four major pathogen classes, ending with a brief consideration of bioterrorism implications.

Bacterial Pathogens Although most infectious disease diagnostics focus on the identification of bacterial pathogens, simply knowing whether a bacterial pathogen is present or absent is enough to initiate or withhold antibiotic treatment. Gene expression alterations are capable of quickly informing the clinician of the presence of bacterial infection, as well as providing useful prognostic information. Sepsis Among acute infectious processes, sepsis is a leading cause of morbidity and mortality in hospitalized patients [30]. Due to the wide spectrum of heterogeneous pathophysiologies that underlie sepsis and the difficulties in identifying the


243

etiological agent, current clinical practice focuses on empiric treatment rather than individualized therapy. Since many deaths related to sepsis arise from initially mild clinical disease that progresses, improved diagnostics are needed that can guide treatment options and predict prognosis. The first primary clinical decision point in a patient meeting criteria for systemic inflammatory response syndrome (SIRS) or with suspected sepsis is to determine if the process is due to an infectious or noninfectious etiology. Many examples discussing the diagnostic utility of gene expression analysis in septic patients can be found in the literature, ranging from a four-gene classifier [31] to a 138-gene classifier [32] capable of distinguishing infectious sepsis from SIRS. Since the term sepsis encompasses clinical manifestations in response to a broad range of underlying etiological agents, the next important task for clinicians is to subclassify septic patients across this spectrum into groups of likely prognostic outcomes in order to individualize treatment decisions. Transcriptomic analyses of critically ill patients with sepsis have revealed that gene expression signatures are capable of “clustering” patients into groups with respect to degree of organ failure, length of ICU stay, and overall mortality [33–36]. These studies, along with many others in the literature, indicate the utility of gene expression analysis not only in recognizing and diagnosing sepsis but also in enhancing our understanding of the underlying pathophysiological differences in the presentation of sepsis and improving our predictions of prognostic outcomes and response to therapy for septic patients. Gram-Positive Versus Gram-Negative Many antibiotic therapies in the clinician’s repertoire preferentially target bacteria based on Gram stain characteristics. Therefore, a Gram-positive bacterial infection often requires different pharmacological management than a Gram-negative bacterial infection. In reality, many patients are treated with broad-spectrum antibiotics covering both Gram-positive and Gram-negative bacteria until a positive microbiological identification is made. Previously, gene expression-based classifiers have demonstrated the ability to distinguish Gram-positive and Gram-negative bacterial infections, potentially allowing for earlier institution of targeted antibiotic therapy and quicker cessation of broad-spectrum antibiotics (and their associated side effects). In a study of pediatric patients with either Gram-negative Escherichia coli (E. coli) or Gram-positive Staphylococcus aureus (S. aureus), the two infection groups demonstrated significantly different gene expression profiles [6]. A 30-gene classifier based on these gene expression changes highlights the specificity of the immune response, as it was able to distinguish E. coli from S. aureus infection with 95% accuracy. This result, along with other similar findings [37, 38], suggests that gene expression modalities have the potential to guide appropriate empiric antibiotic therapy and narrow antibiotic regimens early in the disease course by subclassifying bacterial invaders into more narrow categories.

244


Tuberculosis In contrast to sepsis, tuberculosis (TB) presents a different challenge due to its latency and often indolent progression. TB, caused by the bacterium Mycobacterium tuberculosis, is among the most common infections worldwide. The diseases caused by this bacterium cover a wide spectrum of clinical syndromes, with differing clinical decisions and treatment options for those with subclinical (latent) infection and patients with varied types of active disease. There are many manifestations associated with infection by M. tuberculosis, which broadly include isolated pulmonary infection, extrapulmonary infection, and disseminated disease. The current gold standard for diagnosis of active TB involves culturing the organism, which can take up to 6 weeks [39]. Newer PCR-based pathogen detection technologies have allowed for more rapid diagnosis, but these tests can be less sensitive than traditional culture and their clinical utility remains a topic of active study [40, 41]. In order to address the diagnostic challenges of M. tuberculosis infection, the host immune response to infection has been examined on the transcriptomic level. Gene expression-based classifiers have been developed that are not only capable of distinguishing active from latent TB, but these classifiers can also distinguish patients based on extent of disease and predict response to treatment [42, 43]. More significantly, a host gene expression-based TB classifier can distinguish TB from other similar pulmonary diseases such as lung cancer and community-acquired pneumonia [44]. These results are promising for areas of the world with a heavy burden of tuberculosis.

Viral Pathogens Many early studies in the field of host-based gene expression diagnostics focused on distinguishing viral from bacterial infection. Several of the earliest published disease classifiers are therefore signatures for acute viral infection. The interest in separating bacterial from viral infection is driven by clinical need, since inappropriate administration of antibiotics drives up healthcare costs and increases the incidence of adverse effects and development of antibacterial resistance. Although the host response to many viral agents has been studied on the transcriptomic level, here we will focus mainly on diagnosis of acute viral respiratory infection, with a brief mention of gene expression diagnostics in dengue virus as well. Acute Respiratory Infection One of the richest areas for host-based gene expression research revolves around pathogens causing acute respiratory infections. Although the clinical presentation of acute respiratory infection is relatively easily recognized based on symptomatology, the etiologic agent driving the clinical disease is often much more difficult to


245

identify. This distinction between viral and bacterial etiologies is critical in directing appropriate antimicrobial therapy, considering that 72% of ambulatory care patients with acute respiratory illnesses are treated with antibiotic therapies, even though most of these infections are caused by viruses [45]. Diagnostic needs in this patient population can be addressed by measuring host gene expression alterations in response to viral or bacterial infection. The initial paradigm-altering study in this area involved a human challenge study, whereby healthy volunteers were experimentally infected with either live rhinovirus (HRV), respiratory syncytial virus (RSV), or influenza A [28]. An “acute respiratory viral infection” gene signature was developed that not only distinguished virally infected from uninfected volunteers but also distinguished viral from bacterial causes of acute respiratory infection in a separate cohort of patients. However, this signature was only to discriminate viral from non-viral infection. For example, it could not differentiate bacterial infection from healthy individuals – it could only identify them both as non-viral. This prompted the development of a gene expression classifier that can partition respiratory illness into bacterial, viral, and noninfectious etiologies, as shown in Fig. 2 [46, 47]. These results confirmed that infectious pathogens thought to be limited to the respiratory tract elicit a robust and reproducible immune response in the host that is specific to the invading pathogen, thus potentially offering clinicians a new diagnostic strategy to guide appropriate antimicrobial prescribing. Additional gene expression work in patients with acute respiratory infections has even demonstrated the ability to distinguish between specific viral causes (separating RSV from HRV and influenza) [48], further highlighting the specificity of the host immune response. In addition to distinguishing between viral and bacterial etiologies of infection at the time of peak illness, host gene expression analysis also lends insight into the presymptomatic state. Since most symptoms of infection are due to the host response, it was hypothesized that gene expression changes could precede the development of symptomatic disease. Indeed, in an experimental human challenge model, volunteers were inoculated with influenza A, and their peripheral blood transcriptome was measured every 8 h for up to 7 days [7]. A gene signature was found that not only differentiated symptomatic from asymptomatic subjects, but was able to do so as early as 29 h postexposure. Furthermore, the signature achieved maximal efficacy approximately 40 h before patients experienced peak clinical symptoms, indicating that gene expression alterations occur early in the immune response. In certain circumstances, such as in an outbreak scenario, presymptomatic detection of disease could enable earlier treatment and isolation to limit contagion. Dengue Analysis of host-based gene expression alterations has also found potential clinical utility in other viral infections, such as identifying the mosquito-borne tropical illness caused by dengue virus. Dengue infections range from a mild dengue fever to more severe dengue hemorrhagic fever or even dengue shock syndrome. Given the

246


Fig. 2 Predicted probability ternary plot for acute respiratory illness etiology generated by leave- one-out cross-validation. Each subject has three separate probabilities calculated: probability of bacterial ARI, probability of viral ARI, and probability of SIRS. The highest probability determines class assignment. Each corner represents a class having probability 1. Since there are three possible classifications, the probabilities for all three classes (i.e., distances from the subject to the corner) must sum to 1. The probability of a class at a given point is proportional to the distance along a line extending from the vertex to its opposing side. Dashed lines represent the edge of the classification boundary for each of the three classes. (Reproduced with permission [46])

wide variability in the spectrum of disease caused by dengue virus, there is a need for a diagnostic test capable of grouping patients early in the infectious process based on symptoms and prognosis. Gene signatures capable of separating severe dengue cases from non-severe cases with >90% accuracy have been identified [49]. Moreover, these signatures can separate the “early acute phase” of infection characterized by innate immune response genes from a “late acute phase” characterized by genes associated with the cell cycle. This transcriptomic shift from an early “immunity and inflammation” phase to a later “repair and recovery” phenotype has also been observed by others [50], highlighting the potential for these diagnostic gene expression changes in the host immune response to dengue virus.


247

Fungal Pathogens Invasive fungal organisms are capable of producing a wide array of clinical syndromes and often present diagnostic challenges, including differentiating active fungal infection from colonization. In fact, studies have shown that despite the identification of fungi using culture in up to 64% of ICU patients, the actual rate of disease-causing invasive mycosis is only around 2% in this patient population [5]. Most published research in the realm of host-based gene expression diagnostics has focused on bacterial and viral pathogens. However, focus is shifting to include the host response to invasive fungi as well, particularly given its relevance to the growing population of immunosuppressed hosts. Candidiasis The majority of host-based diagnostic research in the fungal arena focuses on Candida species, particularly Candida albicans. Candida is the most prevalent cause of fungal bloodstream infection in hospitalized patients, particularly those who are immunocompromised or have indwelling vascular catheters. Candida bloodstream infections are the fourth leading cause of nosocomial bloodstream infections and can provide diagnostic challenges, with delays in diagnosis contributing to increased morbidity and mortality [51]. Invasive candidiasis has been shown to increase a patient’s risk of mortality by 10–14% [52]. Since early interventions improve outcomes in candidemia [53, 54], improved diagnostic technologies are imperative for this patient population where empiric antifungal therapy is rarely administered. In order to address this diagnostic gap, a murine challenge study was performed to develop a gene expression signature for candidemia. This work yielded a 67-gene classifier found to be 98% sensitive and 96% specific for distinguishing Candida albicans infection from S. aureus bacteremia and uninfected controls [55]. Many of the genes in the signature have known roles in the host defense against Candida and other invasive fungal organisms. Furthermore, the evolution of the host response to fungal infection could be observed by tracing the upregulated genes to underlying biological pathways through the process of gene set enrichment analysis, as shown in Fig. 3. These pathways included myeloid cell activation and CD40 signaling pathways [55]. Despite this study and others like it, a gene expression-based disease classifier has yet to be developed and validated in human patients with candidemia.

248


Gene Expression Data

Database of Biological Pathways Gene 1

Gene 2 Gene 4

Gene 3

Gene 14

Gene 6

Gene 13

Gene 18

Gene 10

Correlation with Phenotype

Gene 9

Gene 15

Gene 16

Gene 8 Gene 7

Gene 17

Gene 11

Gene 12

Gene 19 Gene 22

Gene 25

Gene 20 Gene 23

Gene 24

Gene 26

Gene 21

Enrichment Score

Gene 5

Gene Set Enrichment Analysis Genes in previously defined pathway

Gene List S

Fig. 3 Workflow for gene set enrichment analysis. Alterations in gene expression levels are quantified for each experimental cohort, and statistically significant differences in gene expression are compiled into a list (we will call this set S), which is ranked based on the magnitude of change in expression level between the two phenotypes (left panel). The set of genes S is compared to genes involved in known biological pathways (center panel). Gene set enrichment analysis then individually examines each known pathway, calculating an “enrichment score” that represents the difference in expression of genes in the pathway between two phenotypes (e.g., infected and not infected; right panel). The enrichment score reflects the degree to which genes in the defined pathway (represented at the top by vertical bars) are significantly up- or downregulated in gene set S. When a gene in the defined pathway is also in S, the enrichment score increases, as shown by the green line. As shown under the “correlation with phenotype” section, some genes in the known pathway were upregulated in S (left side), while others were downregulated in S (right side). The significance of the enrichment score can be correlated with a p-value, allowing for determination of statistically significant expression levels of biological pathways between phenotypes [56]

Aspergillosis and Other Fungal Infections Aside from candidiasis, transcriptomic analysis of the host immune response to invasive fungal infections has been limited. Aspergillus, an opportunistic invasive fungal infection with a predilection for immunocompromised patients, is a common cause of morbidity and mortality in the hospital setting, with Aspergillus fumigatus as the most commonly implicated species [57–60]. Aspergillus species are responsible for myriad clinically significant syndromic illnesses, including respiratory, gastrointestinal, cutaneous, and disseminated disease processes. Although components of the human immune response to invasive aspergillosis have been studied and defined, no transcriptomic disease classifiers based on gene expression changes have been published. However, gene expression changes have been analyzed in response to Aspergillus infection, including one study of human monocytes exposed to Aspergillus fumigatus conidia that resulted in statistically significant differential expression of 1827 genes, many known to be involved in the host immune defense to aspergillosis [61]. This and other published studies in the literature have focused mainly on the underlying pathobiology of infection, rather than attempting to develop a gene expression-based classifier.


249

There are many other clinically relevant fungi that cause infectious diseases. However, published research in regard to the host biomarker immune response to these fungi is lacking. As the field of host-based diagnostics continues to grow, pathogenic fungi such as Cryptococcus, Mucorales, and endemic dimorphic fungi, all of which drive diagnostically challenging clinical syndromes, will likely be prime targets for the development of gene expression signatures of disease.

Parasitic Pathogens Parasitic infections pose a unique diagnostic challenge, particularly in resource- limited settings where many laboratory diagnostic techniques are unavailable. For example, the differential diagnosis for a febrile patient in a malaria-endemic area is extremely broad, including not only parasitic infection but many potentially lethal bacterial and viral pathogens as well. Given the overlapping syndromic presentations and broad differential, pathogen detection in these settings can be quite difficult. There is much room for diagnostic improvement by utilizing biomarkers of the host response to narrow the differential. Unfortunately, there is currently a paucity of published research regarding host gene expression alterations in response to parasitic infections. Perhaps the most extensively studied parasitic infection, at least using transcriptomic techniques, is malaria. In 2006, there were an estimated 247 million cases of malaria and approximately 3.3 billion people at risk, resulting in approximately 881,000 deaths [62]. Given the global burden of malarial disease and its prevalence in resource-limited settings, there is motivation to develop improved and cost-efficient diagnostic techniques that can aid in the recognition of this infection. To date, there have been no gene expression-based classifiers developed capable of diagnosing malaria. The only analysis of host gene expression alterations in this parasitic infection focused on the underlying pathobiology rather than diagnostics. In this study, a longitudinal gene profiling analysis of five children who presented with severe malaria and again 1 month later with mild malaria was performed [63]. Sixty-eight genes, mostly reflecting interferon and T cell function, demonstrated significantly different levels of expression between the severe and milder malarial episodes. Although this enhances our understanding of the pathophysiology of malaria, this work did not produce a gene signature capable of diagnosing the infection itself. There are indications such gene expression-based signatures likely exist. For example, host-based proteomic analyses have been performed for both malaria and Leishmania, including a study that identified differentially expressed proteins between Plasmodium falciparum malaria, Plasmodium vivax malaria, healthy controls, and leptospirosis [64]. If proteomic biomarkers can differentiate these conditions, then gene expression is likely to be a successful modality as well. Further work is necessary to develop and validate such a gene expression-based classifier. As host-based diagnostics evolve and diagnostic assays become more established, gene signatures and gene expression-derived diagnostics for parasitic pathogens will likely follow.

250


Bioterrorism Implications Another interesting diagnostic area with the potential to be impacted by host-based gene expression diagnostics is the field of biological threat agents. Bioterrorism requires prompt prophylaxis and treatment for infections and therefore relies on early diagnosis of the causative pathogen. Currently, most biological threat agents are identified by pathogen detection in tissue and body fluid, with assays relying primarily on antibodies or pathogen culture [65]. These techniques require pathogen proliferation to become detectable and necessitate a priori knowledge of the offending agent. Gene expression could play a role in earlier detection of bioterrorism before pathogens proliferate to dangerous levels and before the onset of clinical symptoms. Multiple gene expression-based biomarker studies of bioterrorism-related pathogens have been published. Differences in gene expression have been identified in a nonhuman primate model of infection with Bacillus anthracis, Yersinia pestis, and Brucella melitensis, viruses such as Venezuelan equine encephalitis virus and dengue virus, and toxins such as Vibrio cholera toxin, Clostridium botulinum toxin A, and Staphylococcal enterotoxin B [66]. These bioterrorism agents induced certain gene expression patterns that were unique to each pathogen and detectible early postexposure. In addition, the biological response to tularemia infection has been studied from the host gene expression perspective, identifying six differentially expressed genes in human peripheral blood mononuclear cells in response to Francisella tularensis inoculation [67]. The capacity of gene expression diagnostics to provide rapid detection of bioterrorism pathogens and the subsequent ability to initiate early therapeutic intervention can mitigate the deadly impact of biological threat agents. The major current limitation is the time required to collect blood samples and process the diagnostic assay, although newer technologies are rapidly addressing this issue.

Clinical Applications The development of transcriptomic-based tests begins with signature discovery in a cohort that is representative of the eventual intended use. Then, the signature should be validated in a similar but independent cohort. Following validation, the development of an actual clinical test depends most heavily on technical considerations. Whereas several host gene expression-based tests have been cleared for use by the FDA, none (as of yet) utilize a platform that enables rapid, simple, point-of-need (or point-of-care) testing. Here, we will discuss technical requirements for such a platform, examples of host gene expression-based tests available for clinical use, and where progress is currently being made.


251

Technical Requirements for Gene Expression-Based Tests In the absence of an existing simple and rapid diagnostic platform for host gene expression analysis, it is challenging to define its technical characteristics. Moreover, there are likely to be many solutions to solve the demands of such an assay. However, there are some general principles that apply to the most readily attainable solutions. This will undoubtedly change as new technologies and solutions emerge. However, there are four major components for successful test development. These include sample processing, transcript quantification, detection, and mathematical transformation. Sample processing, by some metrics, is the most challenging aspect of this process. Most host signatures have been developed based on peripheral blood gene expression. Whereas the respiratory tract is an attractive source of diagnostic information for respiratory tract infection, particularly given its noninvasive nature, there are concerns about adequate and reliable specimen collections. This sampling inconsistency can be avoided by using whole blood. However, blood is a complex medium and includes a number of elements that may interfere with PCR amplification. To maximize the efficiency of PCR, the RNA should be extracted and purified to some degree. Numerous off-the-shelf kits for RNA extraction are available, although nearly all are independent of the subsequent PCR and detection steps. Sample processing that is integrated with subsequent steps is more challenging but certainly possible. Moreover, this integration is essential to develop a rapid and simple test that motivates clinical adoption. Once RNA is available, the next major step is amplification. This is because currently available technologies are not sufficiently sensitive to directly detect the levels of transcripts in a sample, particularly when the sample volume is small (e.g., 600 copies) throughout a typical 2–3 Mb bacterial chromosome. This is illustrated in Fig. 1 with a comparison of EcoRI REA for Staphylococcus aureus strains COL and NCTC8325. Thus, the resulting challenge is to accurately compare electrophoretic patterns comprised of hundreds of restriction fragments, often co-migrating in clusters of similar size and potentially including resident plasmid DNA [9]. Consequently, at the present time, this method continues to be used only with Clostridium difficile [10]. Since the mid-1970s, Southern hybridization [11] has been a staple of molecular biology, and its power to probe for specific DNA sequences soon began to find clinical application. For diagnostic purposes, tests to detect the presence or absence of clinically relevant sequences (e.g., related to organism identification, antibiotic resistance) began to be developed. For epidemiological analysis probes specific for sequences found at multiple chromosomal locations can be hybridized against chromosomal restriction enzyme fragments which have been electrophoretically

Fig. 1 Diagrammatic representation of REA with chromosomal DNA from S. aureus strains COL and NCTC8325 digested with the restriction enzyme EcoRI. (Data were originally generated using the Comprehensive Microbial Resource of the J. Craig Venter Institute website: http://cmr.jcvi.org/ cgi-bin/CMR/shared/Menu.cgi?menu=genome)

308

R. V. Goering

separated. The resulting hybridization patterns (restriction fragment length polymorphisms [RFLPs]) provide an indication of chromosomal relatedness between different bacterial isolates. However, at present this approach continues to find use only with probes for the insertion sequence IS6110 in the RFLP analysis of Mycobacterium tuberculososis [12].

Pulsed-Field Gel Electrophoresis (PFGE) In contrast to conventional REA, rare-cutting restriction enzymes cleave the bacterial chromosome into a relatively small number of fragments (e.g., 10–30) due to the length and/or DNA base composition of their recognition sequences. However, electrophoretic analysis of the megabase-size restriction fragments generated is complicated by their size-independent migration during conventional agarose gel electrophoresis [13, 14]. In the 1980s alternative electrophoretic approaches were developed based on the principal of periodic reorientation of the electric field (and DNA migration) relative to the direction of the gel. The pulsed electric field separates DNA fragments over a wide range of sizes from kb to Mb (Fig. 2) thus allowing a more manageable comparison of isolate patterns. The usefulness of PFGE for molecular typing has been extensively reviewed [15, 16]. However, it is important to note that while the method is far from new, PFGE has exhibited enormous staying power as a valuable method of genomic analysis and comparison. This is especially true for molecular typing where for the majority of bacterial pathogens, it has long been considered the “gold standard” for assessing isolate interrelationships. The reason for this longevity is multifold. Overall, the method for chromosomal DNA isolation (i.e., the in situ lysis of bacterial cells encased in agarose plugs) requires only minor variation with different bacterial species. A wide range of bacterial pathogens can be analyzed using a small number of different restriction enzymes (commonly SmaI and XbaI for gram-positive and gram-negative isolates, respectively). Despite the fact that PFGE obviously does not detect every genetic change and macro-restriction fragment, for most organisms analyzed, the sum of the visible fragment sizes represents greater than 90% of the chromosome. This visual sense of global chromosomal monitoring can be highly informative, not only for isolate comparisons but also in associating characteristic PFGE patterns with specific (e.g., internationally recognized) bacterial strains [17]. In addition, the chromosomal overview provided by PFGE allows visualization of genomic rearrangements as in the case with S. aureus strain USA300 where changes in PFGE patterns can be specifically associated with loss of the staphylococcal chromosomal cassette encoding methicillin resistance (SCCmec) or the adjacent arginine catabolic mobile element (ACME) [18] (Fig. 3). However, despite its historic utility, PFGE is now seeing decreased use in favor of whole-genome sequencing which provides more discriminating and comprehensive analysis (see below).

Molecular Typing Techniques: State of the Art

309

Fig. 2 Illustration of PFGE workflow moving from chromosomal digestion with rare-cutting restriction enzymes to macro-restriction fragment separation by PFGE to final analysis of fragment patterns from different (patient) sources

PCR-Based Amplified Fragment Length Polymorphism (AFLP) AFLP remains in limited use as an interesting approach that combines the use of restriction enzymes and PCR to potentially analyze a wide range of bacterial pathogens [19]. The process involves creation of typing patterns based on PCR amplification of a subset of chromosomal restriction fragments (Fig. 4). This is accomplished by digesting isolated DNA with two different restriction endonucleases, usually chosen so that one cuts more frequently than the other (e.g., EcoRI and MseI). While a large group of restriction fragments are initially created, only specific subsets are utilized for isolate comparison. Adapters specific for the cleaved restriction sites are ligated to the fragment ends, thus extending the length

310

R. V. Goering

Fig. 3 SmaI-digested chromosomal DNA from USA300 S. aureus isolates which are (lane 1) methicillin resistant and PCR positive for the adjacent ACME arcA gene, (lane 2) methicillin susceptible due to loss of SCCmec but arcA positive, or (lane 3) negative for both SCCmec and arcA. (Modified from [18])

of the known end sequences and serving as primer binding sites for PCR. The adapter design includes extra nucleotides beyond the restriction-site sequence allowing only a subset of fragments to be amplified. Using labeled primers the specificity of the process may be further controlled, ultimately leading to an electrophoretic pattern of amplified products that becomes the basis for assessing isolate interrelationships. AFLP improvements have included multiple enzyme-adapter combinations and either fluorescent or radioactively labeled primers, allowing high-throughput analysis to be achieved using an automated DNA sequencer, phosphorimager, etc. [19, 20] However, issues regarding data analysis and interlaboratory sharing and the specialized equipment required for electrophoresis have limited the use of this method in the clinical setting.


311

Fig. 4 AFLP protocol. (a) Genomic DNA is restricted using two different enzymes to yield fragments (b) with a mixture of restriction sequence ends. (c) Restriction-site-specific adapters are ligated to the fragment ends. (d) PCR primers complementary to the adapters with additional bases at their 3′ ends restrict amplification to a subset of fragments (e) the sizes of which are then analyzed by electrophoresis (f). (Modified from [78])

Repetitive-Sequence-Based PCR (rep-PCR) Well before our current level of technology and understanding regarding bacterial genomics, specific DNA sequences were known to be repeated at multiple chromosomal sites in a variety of clinically important pathogens. Enterobacteria were found to contain several hundred copies of repetitive extragenic palindromic (REP) elements and enterobacterial repetitive intergenic consensus (ERIC) sequences [21]. Repeated BOX element sequences were observed in the chromosome of Streptococcus pneumoniae [22]. Multiple copies of IS256 were found in staphylococcal genomes [23]. These and other repeat elements represent genomic landmarks of known sequence to which PCR primers may be specifically anchored in an outwardly oriented direction. The resulting amplicons represent inter-repeat distances that do not exceed the capability of the Taq polymerase (Fig. 5).

312

R. V. Goering

Fig. 5 Illustration of rep-PCR. (a) Repetitive sequences in the bacterial chromosome are recognized by outwardly directed primers (b) allowing PCR amplification of the inter-repetitive regions (c) which are then analyzed by electrophoresis (d)

Thus, strain typing by rep-PCR is accomplished by comparing the chromosomal distribution of such repeated sequences as reflected by the resulting pattern of amplicon sizes. Performed under relatively stringent conditions, rep-PCR is much more reproducible than other more generic PCR approaches such as randomly amplified polymorphic DNA (RAPD) and arbitrarily primed PCR (AP-PCR) which are not considered here [1, 4]. Initial “home brew” efforts at rep-PCR encountered issues such as appropriate primer combinations, PCR conditions, and optimum


313

visualization of amplicon fragment patterns by agarose gel electrophoresis [24]. However, the process has become highly reproducible via commercial automation. The DiversiLab System (bioMérieux, Marcy-l’Étoile, France) employs optimized protocols, separation of PCR products in a charged microfluidic field (i.e., on a chip) rather than by conventional agarose gel electrophoresis, and software for data analysis. While often less discriminatory than PFGE [25–27], rep-PCR remains an interesting typing method although its use has diminished with issues regarding database libraries, interlaboratory data sharing [3], and costs associated with the commercial approach.

PCR Ribotyping Bacteria typically contain multiple chromosomal copies of rRNA genes. Historical ribotyping exploited the fact that strain-to-strain differences in the chromosomal regions flanking rRNA genes affect restriction enzyme recognition sites producing different RFLP hybridization patterns with rRNA probes [28]. However, this approach has been replaced by PCR ribotyping. PCR primers amplifying polymorphisms in the 16S–23S intergenic spacer regions of rRNA gene copies continue to be used as an important tool in the epidemiological monitoring of Clostridium difficile [29]. However, it is important to note that the amplicons generated typically include a variety of similar sizes which can be a challenge to analyze and are thus best separated by capillary (rather than agarose gel) electrophoresis. Nevertheless, the patterns obtained are amenable to databasing and interlaboratory comparison especially with regard to highly toxigenic strains such as C. difficile ribotype 027 [30–32].

Staphylococcal Cassette Chromosome mec (SCCmec) Typing Staphylococci resistant to the antibiotic methicillin, especially S. aureus (MRSA), represent an infectious disease problem of global concern. Central to this issue is the mobile genetic element SCCmec encoding the altered penicillin-binding protein (known as PBP2a or PBP2’) responsible for resistance [33]. Increased understanding of staphylococcal genomics has revealed SCCmec variations (termed SCCmec types) which differ with regard to their internal organization and total size (60 kb) [34]. A variety of multiplex PCR approaches have been developed with primers positioned to detect type-specific differences reflected by amplicon banding patterns in agarose gels [35–37]. However, SCCmec represents one of the most highly recombinogenic regions in the staphylococcal genome. This is reflected in the increasing complexity associated with newly described SCCmec types and subtypes and the multiplex PCR protocols required for their detection (http://www.sccmec.org/Pages/SCC_TypesEN.html) [34].

314

R. V. Goering

Nevertheless, SCCmec typing represents an important means of studying the element’s organization, persistence, and movement in staphylococcal populations. In this context, SCCmec type has become a landmark trait in the definition of specific staphylococcal epidemic strains (especially MRSA). However, the method is not discriminating enough to stand alone as an approach to epidemiological monitoring, and SCCmec differences do not significantly impact antistaphylococcal chemotherapy [38].

Multiple-Locus VNTR Analysis (MLVA) Similar to the repetitive sequences discussed earlier (i.e., rep-PCR), advances in bacterial genomics have revealed the presence of chromosomal regions consisting of tandemly repeated sequence “units” varying both in the number and sequence of the individual repeats (Fig. 6). These occur by slipped strand mispairing during chromosomal replication resulting in the insertion or deletion of repeat units [39– 41]. Bacterial genomes may contain different variable number tandem repeats (VNTR) at multiple chromosomal sites. Properly designed multiplexed PCR primers thus produce MLVA banding patterns by electrophoresis with potential application for strain typing [42]. Finding and validating the epidemiological usefulness of specific MLVA approaches is a deliberative process which varies depending on a number of factors including the degree of VNTR polymorphisms, the organism being analyzed, etc. [42, 43] Nevertheless, MLVA strain typing has been described for a variety of clinically important bacterial pathogens including Bacillus anthracis, Brucella spp., Escherichia coli, Legionella pneumophila, Leptospira interrogans, Mycobacterium tuberculosis, Pseudomonas aeruginosa, Yersinia pestis, Shigella spp., S. aureus, and S. pneumoniae (see [42, 43] for reviews). This trend has been facilitated by a number of advances including digitized MLVA pattern nomenclature based on VNTR repeat numbers; improved accuracy with pattern visualization by capillary, rather than agarose gel, electrophoresis; and proper molecular size standards. Overall, with some exceptions such as PFGE, electrophoretic-based typing methods tend to be relatively simple to perform and also benefit from the potential for decreased cost when agarose gels are used for analysis. However, it is important to emphasize that strain typing based on electrophoretic banding patterns is a comparison of DNA fragment sizes rather than specific genomic content. Thus, interpretation of isolate interrelationships must be tempered by the caveat that equivalent-sized DNA fragments or amplicons in pattern comparisons may or may not represent the same chromosomal sequence. Electrophoresis-based typing approaches are also challenged regarding issues of typing pattern nomenclature, databasing, and interlaboratory sharing. Nevertheless, as noted earlier, in the context of locally available economic and scientific resources, these methods continue to remain of value as options for the epidemiological evaluation of problem bacterial pathogens.


315

Fig. 6 Diagram of a chromosomal VNTR where (a) a sequence unit of “X” base pairs is (b) tandemly repeated “Y” number of times during chromosomal replication. PCR primers anchored to chromosomal regions adjacent to the VNTR (c) allow amplification with subsequent electrophoretic analysis to determine the VNTR “Y” repeat number

DNA Sequence-Based Methods Since the bacterial chromosome is the most fundamental molecule of identity in the cell, strain typing based on DNA sequence analysis is the most direct approach to assessing isolate relatedness. Sequence-based approaches have a number of additional advantages over electrophoresis-based typing methods including: 1. Simplicity and reproducibility. Older molecular methods for epidemiological analysis involve numerous experimental variables including types of equipment, reagents, experimental protocols, etc., all of which affect inter- and intra-laboratory reproducibility. With enough time and effort, any epidemiological method can be standardized as evidenced by classical bacteriophage typing of staphylococci [44] or the success of the nationwide PFGE Pulse-Net System for the investigation of foodborne outbreaks designed by the United States Centers for Disease Control [45]. However, DNA sequence analysis is a more straightforward process that can be performed in a more controlled, uniform, and reproducible manner with specific known chromosomal loci. 2. Data sharing and storage. Electronic storage and sharing of data from electrophoresis-based typing methods is accomplished using bitmapped (e.g., .tiff, .jpeg) computer images. However, the larger the number of isolates the more unwieldy the process can become. In addition, some form of nomenclature must be used to identify and interrelate isolate banding patterns. With large data sets, the use of appropriate computer software is essential to accomplish this process. However, the framework for data sharing, storage, and retrieval is necessarily based on visual images and the limits that format imposes. Conversely, nucleotide sequences represent simple, highly portable, quaternary data that can much more easily be shared, stored, and retrieved.

316

R. V. Goering

3. Data interpretation and detection of significant differences. As discussed below, the most crucial aspect of any typing method is its ability to detect significant (epidemiologically relevant) differences between isolates. While the goal of molecular typing is a comparison of chromosomal similarity, electrophoretic banding patterns only indirectly address this issue. Despite computer programs which can assist the process, there is always an element of end-user judgment that can affect the final evaluation. In contrast, nucleotide sequence data allows direct and unambiguous genomic comparison. Advances in DNA sequencing and the rapidly expanding database of sequenced microbial genomes have served as the foundation for a variety of typing approaches which can generally be categorized as single-locus, multiple-locus, or whole- genome sequence typing.

Single-Locus Sequence Typing Since the genome of bacterial pathogens is mega-base in size, it is remarkable to think that a single locus of ca. 1000 bases could contain sufficient information to be epidemiologically relevant. Nevertheless, three instances where this is the case are detailed below. S. aureus Protein A (Spa) Typing The production of protein A is a hallmark characteristic of S. aureus. Thus, the gene for protein A (spa) is found in all S. aureus strains. The 3′ end of the spa locus (i.e., the polymorphix “X” region) contains a 24-bp VNTR which can be amplified with appropriate primers (e.g., see Fig. 6) and sequenced to determine the specific spa type. Software packages such as StaphType (Ridom GmbH, Münster, Germany) and BioNumerics (Applied Maths NV, Sint-Martens-Latem, Belgium) are available to assist with the sequence analysis process. Numerous studies have shown that comparisons of S. aureus spa types, facilitated by an Internet-based spa server (http://spaserver.ridom.de), provide epidemiologically relevant information that correlates well with other typing methods such as PFGE [38, 46, 47]. In Europe this has led to the formally organized use of spa typing in the epidemiological monitoring of specific S. aureus strains (i.e., SeqNet; http://www.seqnet.org) involving 60 laboratories from 29 countries. Streptococcus pyogenes M protein (emm) Typing The cell surface M protein is an important virulence factor in S. pyogenes [48]. Genomic analysis has revealed that the M protein locus (emm) is variable and can encode at least 100 different M protein types which were initially detected and


317

cataloged serologically. However, PCR primers flanking the hypervariable region of the emm gene allow direct sequencing to determine specific isolate emm types. As a result, sequence-based emm typing remains a widely used approach to group A streptococcal epidemiology [48–50]. As with S. aureus spa typing, emm typing is facilitated by an Internet-based server (hosted by the US Centers for Disease Control) which houses the S. pyogenes emm sequence database (https://www2a.cdc. gov/ncidod/biotech/strepblast.asp). This resource has allowed the CDC to follow specific S. pyogenes epidemiological trends such as the proportion of emm types contributing to specific disease in different global regions (e.g., Africa, Asia, Latin America, Middle East, Australia, Pacific Island) (https://www.cdc.gov/streplab/ emmtype-proportions.html). mec-Associated Direct Repeat Unit (dru) Typing In 1991, Ryffell et al. [51] identified a cluster of repeated imperfect 40-bp sequences (i.e., direct repeat units; dru) adjacent to IS431 within the SCCmec element of S. aureus isolates. While the dru VNTR is absent in a minority of MRSA isolates [52], its constant location in different SCCmec types of both coagulasepositive and coagulase-negative staphylococcal species represents a valuable and stable internal SCCmec characteristic [53]. As with staphylococcal spa typing, properly positioned PCR primers allow amplification and sequencing of the dru region. Software such as BioNumerics (Applied Maths NV, Sint-Martens-Latem, Belgium) is available to assist with assignment of dru types the central repository for which is an Internet- based server (http://www.dru-typing.org). As with SCCmec typing, dru typing has become an increasingly important means of characterizing the persistence and movement of SCCmec in staphylococcal populations. While not discriminating enough to serve as a standalone approach to epidemiological monitoring, dru typing has proven helpful in assessing movement of SCCmec in staphylococcal populations and in subtyping highly clonal (i.e., difficult to differentiate) staphylococcal strains [52, 54, 55]. In addition, a combination of dru typing and analysis of SCCmec (ccr) recombinase genes has proved highly informative with regard to the phylogeny of specific S. aureus MRSA strains [56, 57].

Multi-locus Sequence Typing (MLST) Since its initial description in 1998 [58], MLST has become one of the most popular approaches to microbial strain typing with demonstrated utility for a wide range of clinically relevant pathogens (http://www.mlst.net/databases/default. asp). The method is based on PCR amplification and subsequent sequencing of the internal regions (450–500 bp) of multiple essential (i.e., housekeeping) genes. Seven genes are typically employed, the sequences of which are assigned numeric allelic designations (Fig. 7a). Individual strains are thus characterized by a

318

R. V. Goering

Fig. 7 Illustration of MLST with hypothetical S. aureus strains a and b depicting the seven chromosomal housekeeping genes with an example of allelic differences (e.g., in yqiL) constituting different STs (a). An eBURST example of a clonal complex with central founding ST and associated SLVs is also shown (b)

seven-digit MLST sequence type (ST). For a given organism, individual STs are interrelated based on an algorithm that identifies a parent or “founding” ST as that which has the greatest number of single-locus variants (SLV). Using online graphic tools (eBURST; http://saureus.mlst.net/eburst/), the STs can be further grouped into clonal complexes (CC) where members of the group share a minimum of five or six of the seven allellic designations (Fig. 7b). The highly portable nature of such data and availability of online databases (https://pubmlst.org) has facilitated the use of MLST for global epidemiological analysis [59, 60] and


319

long-term (i.e., phylogenetic) investigation of bacterial lineages [61–63]. However, the method has not found routine epidemiological application since MLST housekeeping gene sequences are too conserved to reliably differentiate the closely related isolates typically encountered during short-term outbreaks (e.g., MRSA and MSSA could both have the same ST). The time and cost associated with multiple-gene sequencing (a total of ca. 3–4 kb for seven loci) has also been a disincentive to routine use.

Other Multi-locus Approaches Hybridization-Based Typing As noted earlier, only a small number of loci may be simultaneously queried using DNA hybridization with restriction fragment-based typing. However, this is not the case with array-based methods where thousands of specific oligonucleotide probes (e.g., representing species-specific, antimicrobial resistance, and virulence-associated genes) can be anchored to solid surfaces such as glass, plastic, or silicone chips. The hybridization pattern of labeled genomic DNA from isolates to be analyzed thus has the potential to provide a wealth of information regarding genomic content (e.g., the presence or absence of specific genes). Depending on the length of the anchored array sequences, even minor sequence variations including insertions, deletions, or changes in a single base of a sequence (single nucleotide polymorphism; SNP) can be detected. The power of this approach has been applied to the characterization of a wide variety of clinically relevant organisms [2, 64–66]. However, while large-scale microarrays have the potential for high-throughput genomic analysis, their use is not cost-effective for routine clinical use. In addition, a high level of technical expertise is required especially for data analysis which can be complicated by “background” noise due to partial hybridization, etc. The issues associated with large-scale arrays have largely been solved by a more streamlined approach provided by Alere Technologies, GmbH (Jena, Germany). The company provides data helpful for epidemiological investigation and isolate comparison via a microarray-based genotyping service surveying hundreds of genomic loci in a variety clinically relevant organisms [67, 68]. Spacer oligonucleotide genotyping (spoligotyping) represents another variation on hybridization-based bacterial typing. The method targets a region of highly conserved DNA direct repeat sequences found in the genome of Mycobacterium tuberculosis complex (MTC) strains. The direct repeats are separated by a variety of interspersed spacer sequences, and the order of the different spacers is highly conserved. PCR amplification and hybridization of the direct repeat region to an ordered spacer sequence template will reveal the presence or absence of specific spacers in different isolates which constitutes the spoligotype [69].

320

R. V. Goering

Whole-Genome Sequence (WGS) Typing As noted earlier, the goal of molecular strain typing is epidemiological assessment based on the most fundamental molecule of identity in the cell – the bacterial chromosome. Thus, the ability to compare whole-genome sequences represents the ultimate molecular typing approach with great promise as the new “gold standard” for epidemiological analysis. While this was impossible with older dideoxy/chain termination sequencing technology [70], newer (i.e., next-generation sequencing, NGS) methods have made this goal a reality. NGS technology, different “generations” of approaches to sequencing and application are more fully discussed elsewhere (e.g., this book and also see [6, 8, 71, 72]) and will not be considered further here. However, from a strain typing standpoint, it is important to note that revolutionary developments in NGS have made WGS possible with benchtop platforms such as the MiSeq and HiSeq (Illumina, San Diego) and MiniON (Oxford Nanopore Technologies, Oxford, UK). Such instrumentation now allows WGS to be completed in only a few hours with extensive multifold coverage allowing isolates to be compared down to the level of SNPs. However, for NGS, as for previous sequencing iterations, the critical issues are throughput, quality, read length, and cost. All of these are currently in a state of flux as commercial technology improves and positions itself in the scientific marketplace. For all WGS applications, especially including bacterial strain typing, one of the greatest needs is bioinformatic data interpretation and analysis. Important progress toward addressing these issues is being made through an increasing number of online tools [72] as well as commercially available software solutions such as BioNumerics (Applied Maths NV, Sint- Martens-Latem, Belgium) and SeqSphere+ (Ridom GmbH, Münster, Germany). However, concerns remain regarding standardization, accuracy, accessibility, ease of use, cost, analysis time, etc. Nevertheless, these are exciting “problems” to have which ultimately will be solved setting the scientific stage for additional remarkable developments in this most fundamental approach to determine isolate epidemiological interrelationships.

Non-sequence-Based Whole Cell Typing While strain typing is firmly directed toward sequence-based analysis, MALDI- TOF mass spectrometry is finding renewed emphasis in applications for strain typing. MALDI-TOF mass spectrometry is considered in detail elsewhere in this book. The method has generated intense interest as a means of rapid microbial identification via the detection of unique cellular protein biomarkers. As a related issue, MALDI-TOF is also being investigated as an approach to bacterial strain typing [73–77]. However, experimental parameters (e.g., loading of the target plate, matrix composition) must be carefully controlled with optimized post-processing and analysis of the mass spectra. Nevertheless, as an adjunct to microbial identification, strain typing is a logical goal for MALDI-TOF technology which will most certainly see additional refinement and application in the future.


321

Strain Typing in the Context of the Epidemiological Window In the final analysis, regardless of the quantity or quality of strain typing data, the issue ultimately comes down to data interpretation. In this context it is important to note that while the term “molecular” epidemiology implies a precise process, this is not always the case regardless of the method employed since such investigations have an unavoidable context-driven component. A variety of environmental factors as well as interaction between the host and infectious agent may all influence the course of disease transmission. In addition, infectious disease issues benefiting from epidemiological evaluation do not typically give advance warning. Thus, identification of the index patient in a particular outbreak is a common problem in epidemiological analysis. Nevertheless, the analysis must be conducted in the context of the available isolate data (i.e., the epidemiological window [8]) which, unfortunately, does not always include the outbreak source. Thus, the analytical process is commonly one of attempting to work backward in time which, depending on the available data, may necessitate conclusions based on probabilities rather than hard data. For this reason, regardless of the sophistication of the typing approach, epidemiological analysis commonly contains an element of educated guess. Nevertheless, for most clinical scenarios (i.e., outbreak investigation), the key issue is whether or not a series of bacterial isolates are the result of personto-person transfer. At the heart of this question is the concept of significant difference which for chromosomal comparison relates to epidemiologically relevant genomic clock speed. In the absence of an index case or isolate, strain typing methods are challenged as the opportunity for chromosomal change over time increases the potential for genetic distance between epidemiologically related isolates. As illustrated in Fig. 8, if one considers a simple reference genome of six characteristics (“x”) evolving through two generations with sequential genetic events of unknown complexity (x → y), the resulting second-generation genomes would vary from each other by four differences. The potential complexity of the scenario obviously increases further over time. These issues underscore the potential difficulties one may encounter in attempting to discern lineages of infectious agent transmission, regardless of the typing method employed. Thus, for an

Fig. 8 Depiction of the interrelationships between a reference genome containing six characteristics (each designated “x”) with two subsequent generations each experiencing sequential single genetic events (x → y). Within each generation, the number of resulting genetic differences is indicated in parenthesis

322

R. V. Goering

optimum outcome (e.g., in an outbreak setting) analysis of strain typing data and its epidemiological relevance requires knowledge of (i) the limitations of the typing method, (ii) the etiological agent (e.g., genomic clock speed), and (iii) the clinical setting within which the issue is being studied [6]. For the future, it is exciting to consider the advances in strain typing that will continue to be made. The persistence and spread of problem pathogens in patient populations will obviously continue to occur. Thus, perhaps the most important point of all is to emphasize that, more than ever before, strain typing and epidemiological analysis benefit from communication. It is when all interested parties participate (e.g., physician, nurse, infection control specialist, laboratory) that the epidemiological educated guess is most likely to be correct, and that most certainly is a key goal for the treatment of infectious disease both now and in the future.

References 1. Tenover FC, Arbeit RD, Goering RV. How to select and interpret molecular strain typing methods for epidemiological studies of bacterial infections: a review for healthcare epidemiologists. Infect Control Hosp Epidemiol. 1997;18:426–39. 2. Chen Y, Brown E, Knabel SJ. Molecular epidemiology of foodborne pathogens. In: Wiedmann M, Zhang W, editors. Genomics of foodborne bacterial pathogens. New York: Springer; 2011. p. 403–53. 3. Van Belkum A, Tassios PT, Dijkshoorn L, et al. Guidelines for the validation and application of typing methods for use in bacterial epidemiology. Clin Microbiol Infect. 2007;13 (Suppl 3):1–46. 4. Goering RV. The molecular epidemiology of nosocomial infection: past, present, and future. Rev Med Microbiol. 2000;11:145–52. 5. Goering RV. Molecular strain typing for the clinical laboratory: current application and future direction. Clin Microbiol Newsl. 2000;22:169–73. 6. Schurch AC, Arredondo-Alonso S, Willems RJL, Goering RV. Whole genome sequencing options for bacterial strain typing and epidemiologic analysis based on single nucleotide polymorphism versus gene-by-gene-based approaches. Clin Microbiol Infect. 2018;24:342–349. 7. Lindsay JA. Genomic variation and evolution of Staphylococcus aureus. Int J Med Microbiol. 2010;300:98–103. 8. Goering RV, Kock R, Grundmann H, Werner G, Friedrich AW. From theory to practice: molecular strain typing for the clinical and public health setting. Euro Surveill. 2013;18:20383. 9. Bialkowska-Hobrzanska H, Jaskot D, Hammerberg O. Evaluation of restriction endonuclease fingerprinting of chromosomal DNA and plasmid profile analysis for characterization of multiresistant coagulase-negative staphylococci in bacteremic neonates. J Clin Microbiol. 1990;28:269–75. 10. Huber CA, Foster NF, Riley TV, Paterson DL. Clostridium difficile typing methods: challenges for standardization. J Clin Microbiol. 2013;51:2810–4. 11. Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol. 1975;98:503–17. 12. Thorne N, Borrell S, Evans J, et al. IS6110-based global phylogeny of Mycobacterium tuberculosis. Infect Genet Evol. 2011;11:132–8. 13. Schwartz DC, Saffran W, Welsh J, Haas R, Goldenberg M, Cantor CR. New techniques for purifying large DNA’s and studying their properties and packaging. Cold Spring Harb Symp Quant Biol. 1983;47:189–95.


323

14. Schwartz DC, Koval M. Conformational dynamics of individual DNA molecules during gel electrophoresis. Nature. 1989;338:520–2. 15. Goering RV, Ribot EM, Gerner-Smidt P. Pulsed-field gel electrophoresis: laboratory and epidemiologic considerations for interpretation of data. In: Persing DH, Tenover FC, Tang YW, Nolte FS, Hayden RT, et al., editors. Molecular microbiology. 2nd ed. Washington, DC: ASM Press; 2011. p. 167–77. 16. Goering RV. Pulsed field gel electrophoresis: a review of application and interpretation in the molecular epidemiology of infectious disease. Infect Genet Evol. 2010;10:866–75. 17. McDougal LK, Steward CD, Killgore GE, Chaitram JM, McAllister SK, Tenover FC. Pulsed- field gel electrophoresis typing of oxacillin-resistant Staphylococcus aureus isolates from the United States: establishing a national database. J Clin Microbiol. 2003;41:5113–20. 18. Goering RV, McDougal LK, Fosheim GE, Bonnstetter KK, Wolter DJ, Tenover FC. Epidemiologic distribution of the arginine catabolic mobile element among selected methicillin- resistant and methicillin-susceptible Staphylococcus aureus isolates. J Clin Microbiol. 2007;45:1981–4. 19. Goering RV, Stemper ME, Shukla SK, Foley SL. Restriction analysis techniques. In: Foley SL, Chen AY, Simjee S, Zervos MJ, editors. Molecular techniques for the study of hospital acquired infection. Hoboken: Wiley-Blackwell; 2011. p. 135–44. 20. Melles DC, Schouls L, Francois P, et al. High-throughput typing of Staphylococcus aureus by amplified fragment length polymorphism (AFLP) or multi-locus variable number of tandem repeat analysis (MLVA) reveals consistent strain relatedness. Eur J Clin Microbiol Infect Dis. 2009;28:39–45. 21. Versalovic J, Koeuth T, Lupski JR. Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Res. 1991;19:6823–31. 22. Van Belkum A, Sluijter M, De Groot R, Verbrugh H, Hermans PWM. Novel BOX repeat PCR assay for high-resolution typing of Streptococcus pneumoniae strains. J Clin Microbiol. 1996;34:1176–9. 23. Deplano A, Vaneechoutte M, Verschraegen G, Struelens MJ. Typing of Staphylococcus aureus and Staphylococcus epidermidis strains by PCR analysis of inter-IS256 spacer length polymorphisms. J Clin Microbiol. 1997;35:2580–7. 24. Frye SR, Healy M. Repetitive sequence-based PCR typing of bacteria and fungi. In: Persing DH, Tenover FC, Tang YW, Nolte FS, Hayden RT, Van Belkum A, editors. Molecular microbiology: diagnostic principles and practice. Washington, DC: ASM Press; 2011. p. 199–212. 25. Ross TL, Merz WG, Farkosh M, Carroll KC. Comparison of an automated repetitive sequence- based PCR microbial typing system to pulsed-field gel electrophoresis for analysis of outbreaks of methicillin-resistant Staphylococcus aureus. J Clin Microbiol. 2005;43:5642–7. 26. Roussel S, Felix B, Colaneri C, et al. Semi-automated repetitive-sequence-based polymerase chain reaction compared to pulsed-field gel electrophoresis for Listeria monocytogenes subtyping. Foodborne Pathog Dis. 2010;7:1005–12. 27. Aguadero V, Gonzalez VC, Vindel A, Gonzalez VM, Moreno JJ. Evaluation of rep-PCR/ DiversiLab versus PFGE and spa typing in genotyping methicillin-resistant Staphylococcus aureus (MRSA). Br J Biomed Sci. 2015;72:120–7. 28. Bouchet V, Huot H, Goldstein R. Molecular genetic basis of ribotyping. Clin Microbiol Rev. 2008;21:262–73. 29. Tenover FC, Novak-Weekley S, Woods CW, et al. Impact of strain types on detection of toxigenic Clostridium difficile: comparison of molecular diagnostic and enzyme immunoassay approaches. J Clin Microbiol. 2010;48:3719–24. 30. Valiente E, Dawson LF, Cairns MD, Stabler RA, Wren BW. Emergence of new PCR-ribotypes from the hypervirulent Clostridium difficile 027 lineage. J Med Microbiol. 2012;61:49–56. 31. Solomon K, Fanning S, McDermott S, et al. PCR ribotype prevalence and molecular basis of macrolide-lincosamide-streptogramin B (MLSB) and fluoroquinolone resistance in Irish clinical Clostridium difficile isolates. J Antimicrob Chemother. 2011;66:1976–82. 32. Gerding DN. Global epidemiology of Clostridium difficile infection in 2010. Infect Control Hosp Epidemiol. 2010;31(Suppl 1):S32–4.

324

R. V. Goering

33. Katayama Y, Ito T, Hiramatsu K. A new class of genetic element, staphylococcus cassette chromosome mec, encodes methicillin resistance in Staphylococcus aureus. Antimicrob Agents Chemother. 2000;44:1549–55. 34. IWG-SCC. Classification of staphylococcal cassette chromosome mec (SCCmec): guidelines for reporting novel SCCmec elements. Antimicrob Agents Chemother. 2009;53:4961–7. 35. Kondo Y, Ito T, Ma XX, et al. Combination of multiplex PCRs for staphylococcal cassette chromosome mec type assignment: rapid identification system for mec, ccr, and major differences in junkyard regions. Antimicrob Agents Chemother. 2007;51:264–74. 36. Milheirico C, Oliveira DC, De Lencastre H. Multiplex PCR strategy for subtyping the staphylococcal cassette chromosome mec type IV in methicillin-resistant Staphylococcus aureus: ‘SCCmec IV multiplex’. J Antimicrob Chemother. 2007;60:42–8. 37. Oliveira DC, De Lencastre H. Multiplex PCR strategy for rapid identification of structural types and variants of the mec element in methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother. 2002;46:2155–61. 38. Deurenberg RH, Stobberingh EE. The evolution of Staphylococcus aureus. Infect Genet Evol. 2008;8:747–63. 39. Van Belkum A, Scherer S, van Alphen L, Verbrugh H. Short-sequence DNA repeats in prokaryotic genomes. Microbiol Mol Biol Rev. 1998;62:275–93. 40. Lindstedt BA. Multiple-locus variable number tandem repeats analysis for genetic fingerprinting of pathogenic bacteria. Electrophoresis. 2005;26:2567–82. 41. Hammerschmidt S, Muller A, Sillmann H, et al. Capsule phase variation in Neisseria meningitidis serogroup B by slipped-strand mispairing in the polysialyltransferase gene (siaD): correlation with bacterial invasion and the outbreak of meningococcal disease. Mol Microbiol. 1996;20:1211–20. 42. Pourcel C, Vergnaud G. Strain typing using multiple “variable number of tandem repeat” analysis and genetic element CRISPR. In: Persing DH, Tenover FC, Tang YW, Nolte FS, Hayden RT, Van Belkum A, editors. Molecular microbiology: diagnostic principles and practive. 2nd ed. Washington, DC: ASM Press; 2011. p. 179–97. 43. Nadon CA, Trees E, Ng LK, et al. Development and application of MLVA methods as a tool for inter-laboratory surveillance. Euro Surveill. 2013;18(35):20565. 44. Bannerman TL, Hancock GA, Tenover FC, Miller JM. Pulsed-field gel electrophore sis as a replacement for bacteriophage typing of Staphylococcus aureus. J Clin Microbiol. 1995;33:551–5. 45. Ribot EM, Hise KB. Future challenges for tracking foodborne diseases: PulseNet, a 20-year- old US surveillance system for foodborne diseases, is expanding both globally and technologically. EMBO Rep. 2016;17:1499–505. 46. Harmsen D, Claus H, Witte W, et al. Typing of methicillin-resistant Staphylococcus aureus in a university hospital setting by using novel software for spa repeat determination and database management. J Clin Microbiol. 2003;41:5442–8. 47. Church DL, Chow BL, Lloyd T, Gregson DB. Comparison of automated repetitive-sequence- based polymerase chain reaction and spa typing versus pulsed-field gel electrophoresis for molecular typing of methicillin-resistant Staphylococcus aureus. Diagn Microbiol Infect Dis. 2011;69:30–7. 48. Bessen DE. Population biology of the human restricted pathogen, Streptococcus pyogenes. Infect Genet Evol. 2009;9:581–93. 49. Sanderson-Smith M, De Oliveira DM, Guglielmini J, et al. A systematic and functional classification of Streptococcus pyogenes that serves as a new tool for molecular typing and vaccine development. J Infect Dis. 2014;210:1325–38. 50. Steer AC, Law I, Matatolu L, Beall BW, Carapetis JR. Global emm type distribution of group a streptococci: systematic review and implications for vaccine development. Lancet Infect Dis. 2009;9:611–6. 51. Ryffel C, Bucher R, Kayser FH, Berger-Bächi B. The Staphylococcus aureus mec determinant comprises an unusual cluster of direct repeats and codes for a gene product similar to the Escherichia coli sn-glycerophosphoryl diester phosphodiesterase. J Bacteriol. 1991;173:7416–22.


325

52. Ionescu R, Mediavilla JR, Chen L, et al. Molecular characterization and antibiotic susceptibility of Staphylococcus aureus from a multidisciplinary hospital in Romania. Microb Drug Resist. 2010;16:263–72. 53. Goering RV, Morrison D, Al-Doori Z, Edwards GF, Gemmell CG. Usefulness of mec-associated direct repeat unit (dru) typing in the epidemiological analysis of highly clonal methicillin- resistant Staphylococcus aureus in Scotland. Clin Microbiol Infect. 2008;14:964–9. 54. Fessler A, Scott C, Kadlec K, Ehricht R, Monecke S, Schwarz S. Characterization of methicillin-resistant Staphylococcus aureus ST398 from cases of bovine mastitis. J Antimicrob Chemother. 2010;65:619–25. 55. Shore AC, Rossney AS, Kinnevey PM, et al. Enhanced discrimination of highly clonal ST22- methicillin-resistant Staphylococcus aureus IV isolates achieved by combining spa, dru, and pulsed-field gel electrophoresis typing data. J Clin Microbiol. 2010;48:1839–52. 56. Smyth DS, Wong A, Robinson DA. Cross-species spread of SCCmec IV subtypes in staphylococci. Infect Genet Evol. 2011;11:446–53. 57. Smyth DS, McDougal LK, Gran FW, et al. Population structure of a hybrid clonal group of methicillin-resistant Staphylococcus aureus, ST239-MRSA-III. PLoS One. 2010;5:e8582. 58. Maiden MCJ, Bygraves JA, Feil E, et al. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc Natl Acad Sci U S A. 1998;95:3140–5145. 59. Feil EJ, Spratt BG. Recombination and the population structures of bacterial pathogens. Annu Rev Microbiol. 2001;55:561–90. 60. Aanensen DM, Spratt BG. The multilocus sequence typing network: mlst.net. Nucleic Acids Res. 2005;33(Web Server issue):W728–33. 61. Yan Y, Cui Y, Han H, et al. Extended MLST-based population genetics and phylog eny of Vibrio parahaemolyticus with high levels of recombination. Int J Food Microbiol. 2011;145:106–12. 62. Ch'ng SL, Octavia S, Xia Q, et al. Population structure and evolution of pathogenicity of Yersinia pseudotuberculosis. Appl Environ Microbiol. 2011;77:768–75. 63. Litrup E, Torpdahl M, Malorny B, Huehn S, Christensen H, Nielsen EM. Association between phylogeny, virulence potential and serovars of Salmonella enterica. Infect Genet Evol. 2010;10:1132–9. 64. Miller MB, Tang YW. Basic concepts of microarrays and potential applications in clinical microbiology. Clin Microbiol Rev. 2009;22:611–33. 65. Musser JM, Shelburne SA III. A decade of molecular pathogenomic analysis of group a Streptococcus. J Clin Invest. 2009;119:2455–63. 66. Li W, Raoult D, Fournier PE. Bacterial strain typing in the genomic era. FEMS Microbiol Rev. 2009;33:892–916. 67. Matussek A, Jernberg C, Einemo IM, et al. Genetic makeup of Shiga toxin-producing Escherichia coli in relation to clinical symptoms and duration of shedding: a microarray analysis of isolates from Swedish children. Eur J Clin Microbiol Infect Dis. 2017;36:1433–41. 68. Blomfeldt A, Aamot HV, Eskesen AN, et al. DNA microarray analysis of Staphylococcus aureus causing bloodstream infection: bacterial genes associated with mortality? Eur J Clin Microbiol Infect Dis. 2016;35:1285–95. 69. Driscoll JR. Spoligotyping for molecular epidemiology of the Mycobacterium tuberculosis complex. Methods Mol Biol. 2009;551:117–28. 70. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A. 1977;74:5463–7. 71. Heather JM, Chain B. The sequence of sequencers: the history of sequencing DNA. Genomics. 2016;107:1–8. 72. Carrico JA, Rossi M, Moran-Gilad J, Van DG, Ramirez M. A primer on microbial bioinformatics for nonbioinformaticians. Clin Microbiol Infect. 2018;24:342–49. 73. Spinali S, van BA, Goering RV, et al. Microbial typing by matrix-assisted laser desorption ionization-time of flight mass spectrometry: do we need guidance for data interpretation? J Clin Microbiol. 2015;53:760–5.

326

R. V. Goering

74. Moura H, Woolfitt AR, Carvalho MG, et al. MALDI-TOF mass spectrometry as a tool for differentiation of invasive and noninvasive Streptococcus pyogenes isolates. FEMS Immunol Med Microbiol. 2008;53:333–42. 75. Wolters M, Rohde H, Maier T, et al. MALDI-TOF MS fingerprinting allows for discrimination of major methicillin-resistant Staphylococcus aureus lineages. Int J Med Microbiol. 2011;301:64–8. 76. Williamson YM, Moura H, Woolfitt AR, et al. Differentiation of Streptococcus pneumoniae conjunctivitis outbreak isolates by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl Environ Microbiol. 2008;74:5891–7. 77. Dieckmann R, Malorny B. Rapid screening of epidemiologically important Salmonella enterica subsp. enterica serovars by whole-cell matrix-assisted laser desorption ionization- time of flight mass spectrometry. Appl Environ Microbiol. 2011;77:4136–46. 78. Rademaker JL, Savelkoul P. PCR amplification-based microbial typing. In: Persing DH, Tenover FC, Versalovic J, Tang YW, Unger ER, Relman DA, et al., editors. Molecular microbiology: diagnostic principles and practice. Washington, DC: ASM Press; 2004. p. 197–221.

PCR and Its Variations Eleanor A. Powell and Michael Loeffelholz

PCR: The Quintessential Nucleic Acid Amplification Method The polymerase chain reaction (PCR) is an in vitro technique used to replicate, or amplify, a specific region of DNA billions-fold in a few hours or less [1–3]. The amplification is primer directed; oligonucleotide primers anneal to and flank the DNA region to be amplified. PCR is utilized in diagnostic and research laboratories to generate sufficient quantities of DNA to be adequately tested, analyzed, or manipulated. Because of the exquisite sensitivity it offers, PCR has rapidly become a standard method in diagnostic microbiology. More recently, reagent kits and various instrument platforms have added speed, flexibility, and simplicity [4–10]. How significant is the contribution of PCR to the field of biomedicine? A PubMed search in 2005 for the first edition of this book using the keyword “PCR” produced 214,352 hits [11]. A search conducted in April 2017 using the same keyword produced 413,113 hits. PCR was conceived in 1983 by Kary Mullis, an achievement that earned him the Nobel Prize in chemistry in 1993 [12]. The first practical application of PCR was described by Saiki and colleagues in 1985 [2], and less than 10 years later, the US Food and Drug Administration (FDA) cleared the first PCR-based test for diagnosis of an infectious disease [9]. The 1990s saw the birth of a number of alternative nucleic acid amplification methods, including Qβ replicase, ligase chain reaction, strand displacement amplification, transcription-mediated amplification, and others. Some of these methods are used extensively in FDA-approved E. A. Powell (*) Department of Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA e-mail: [email protected] M. Loeffelholz Department of Pathology, University of Texas Medical Branch, Galveston, TX, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_16

327

328

E. A. Powell and M. Loeffelholz

diagnostic methods and are discussed elsewhere in this text. Research and diagnostic applications of PCR continued to be developed during the 1990s. In an incredibly short period of time, PCR revolutionized the field and became a staple on the clinical microbiologist’s menu of tests. Indeed, molecular diagnostics is now a recognized subspecialty within clinical microbiology. The 2000s could be characterized as the decade of PCR instrument platforms, with the availability of numerous commercial instruments for real-time and multiplexed PCR that provided significant advances in speed and automation [13–16]. The subsequent years have seen a dramatic expansion in the number of FDA-approved PCR tests for individual analytes and syndromic multiplex panels. While the menu of FDAapproved PCR assays is rapidly expanding, the menu is still relatively limited. Clinical microbiologists still rely heavily on laboratory-developed PCR assays for the detection of many important infectious agents.

Principles of PCR nzymatic Amplification of DNA: Components of the PCR E Reaction Two early innovations responsible for making PCR a practical research and diagnostic tool are thermal stable DNA polymerase and the thermal cycler. The thermal cycler will be discussed later in this chapter. PCR was first performed using heat- labile DNA polymerase. This necessitated manual replenishment of enzyme that was destroyed by the high temperatures of every cycle. Heat-stable DNA polymerase was isolated from the bacterium Thermus aquaticus, which inhabits hot springs where temperatures exceed 90 °C. This enzyme, called “Taq” DNA polymerase, remains active despite repeated heating during many cycles of amplification. In addition to DNA polymerase, the essential components of the PCR reaction include oligonucleotide primers, deoxynucleoside triphosphates (dNTPs), a divalent cation such as magnesium chloride, template or target DNA, and buffer (usually Tris). Primers are oligonucleotides, generally 20–25 bases long. They are designed to recognize specific sequences of the intended target and define the amplified region. At temperatures appropriate for annealing, the two primers bind to opposite ends of this region, each to a complementary strand of target DNA. Primers must be designed carefully to avoid self-annealing or dimerization. The length and sequence of the primer determine its melting temperature and, hence, annealing temperature. Once annealed to target DNA, primers create a binding site for DNA polymerase, which requires a double-stranded DNA template. This short double-stranded section primes the DNA replication or amplification process. As stated at the beginning of the chapter, PCR is a primer-directed amplification of DNA. Taq DNA polymerase is the enzyme responsible for synthesizing or extending the new DNA strand. Complementary base pairing creates a new strand, which is in essence the mirror image of the template strand. dNTPs are the building blocks for the new DNA strands, or amplicons. The dNTP mixture includes

329

PCR and Its Variations Fig. 1 PCR cycling steps

Double-stranded 5’ target DNA 3’

3’ 5’

5’

3’

3’

5’

Denaturation ~94-95 C

5’

Primer annealing ~50-60 C

3’ 5’ Primer

Primer 5’

3’

5’

3’

5’ 5’

Extension ~72 C

5’ 3’

5’

dATP, dCTP, dGTP, and dTTP, generally at equimolar concentrations. If the enzyme uracil N-glycosylase (UNG) is used in the PCR reaction to prevent carry-over contamination, dUTP is added in place of or in combination with dTTP. Magnesium is the cofactor most commonly used in PCR reactions and is required for Taq DNA polymerase activity. Magnesium concentration must be carefully optimized, as the window of optimal activity is rather narrow.

The PCR Cycle Traditionally, PCR consists of three steps: denaturation, primer annealing, and extension (Fig. 1). One round of these three steps is referred to as a PCR cycle. These steps require different temperatures. This is accomplished using an automated thermal cycler, which can heat and cool tubes rapidly. While most PCR protocols use three different temperatures for each step, two-temperature PCR cycles, where primer annealing and extension occur at the same temperature, are frequently used for faster cycling. Generally, 30–40 rounds of temperature cycling are required to generate a sufficient amount of PCR product (amplicons).

Denaturation At a temperature generally between 94°C and 95°C, the two strands of the DNA target are separated or denatured. At this temperature, all enzymatic reactions, such as the extension from a previous cycle, stop.

330


Annealing Following denaturation, the temperature of the reaction is reduced to allow strands of DNA with complementary sequence to anneal. The annealing temperature varies, depending on the sequence, and, hence, melting temperature of the oligonucleotide primers, but is often between 50°C and 60°C. At annealing temperature, the primers are in movement, caused by Brownian motion. Ionic bonds are constantly formed and broken between the single stranded primer and DNA target. When primers come in contact with a perfectly complimentary target sequence, the bond that forms is sufficiently stable to allow DNA polymerase to sit and initiate DNA synthesis at the 3′ end of each primer.

Extension Extension of the primers generally occurs at 72°C, the temperature at which Taq DNA polymerase is most active. As bases are added to the 3′ end of the primer and the double-stranded section lengthens, the resulting ionic bond is greater than the forces that break these attractions. Each round of temperature cycling theoretically doubles the amount of DNA. After several rounds of temperature cycling, the amount of short double- stranded DNA product (flanked by sequence complementary to the primers) vastly exceeds the amount of the original target DNA. As a result, amplicons accumulate geometrically. After the first PCR cycle, a single starting piece of double-stranded DNA becomes two, after two cycles, there are four copies, after three cycles, there are eight copies, and so on (Fig. 2). After n cycles, a single starting template has theoretically become 2n copies of the original template. As stated, 30–40 rounds of PCR are generally required to produce detectable amounts of amplicon. Due to the presence of inhibitory substances in the PCR reaction and other factors, amplification efficiency probably never reaches 100%. While the analytical sensitivity of PCR is theoretically at the single-copy level [4, 17], sampling variability and lower amplification efficiency generally prevent reliable detection of less than 10–20 target copies per PCR reaction. The entire procedure is carried out in a programmable thermal cycler – a computer-controlled cycling system with heating and cooling parameters. Many techniques for thermoregulation are used in the designs of thermal cyclers. These include the Peltier effect [18], heated and chilled airstreams [19, 20], and a continuous flow manner [21]. In this last design, heat from one side of a semiconductor is transferred to another, heating or cooling the overall temperature of the system. This design is much more effective than traditional designs of thermoregulation, which requires the use of refrigerants and compressors. Other approaches for thermoregulation include the use of continually circulating airstreams, water baths, or a combination of Peltier and convective technologies. Recently, thermal cyclers designed to

PCR and Its Variations

331

5’ 3’

3’ 5’ 1st cycle 5’

3’ 5’ 5’

3’

5’ 2nd cycle

3’

5’

5’ 3’

3’ 5’

5’

3rd cycle

3’

Fig. 2 Exponential amplification of DNA by PCR. Primers are represented by bold lines, newly replicated DNA strands are represented by long dashes, and template DNA strands by solid lines

decrease the time required for PCR have become available. These instruments typically have a faster ramp rate, which results in faster transitions between temperatures and a faster overall PCR run. When choosing a thermal cycler, functions that should be considered (depending on a particular laboratory’s needs) include temperature accuracy and consistency across all PCR positions, maintenance costs, throughput, menu of FDA-cleared or approved assays, and the ability to add user- defined cycling protocols for laboratory-developed assays.

Detection and Analysis of the PCR Product The PCR product should be a fragment or fragments of DNA of defined length. Before the PCR product is used in further applications, it should be analyzed. First, reactions should be examined to ensure product is actually formed. This seems intuitive, but when amplicon is detected with a probe, unexpectedly negative results could be due to either lack of amplification or failure of the probe to hybridize and produce a detectable signal. Causes of unsuccessful PCR amplification include poor quality of target DNA, too little (or too much) target DNA, lack of sequence homology between primers and the intended target, the presence of PCR inhibitors, and failure to optimize PCR conditions. PCR product must also be the expected length

332


and sequence. Unexpected amplicon length or melting curve indicates that the target region itself is different than expected, that the target sequence is shared, or that amplification conditions are suboptimal and allow nonspecific annealing. PCR product should also be evaluated to ensure that the correct number of distinct products is produced. In most diagnostic applications, a single amplicon is generated by one primer pair. Additional, unintended product is usually produced as a result of suboptimal amplification conditions (poor primer design, Taq or MgCl2 concentration too high, annealing temperature not optimized). In conventional PCR, endpoint product is analyzed after PCR is completed, by electrophoresis in agarose gels and visualization with ethidium bromide. DNA fragment size is determined by comparison with known molecular weight markers. Agarose gel electrophoresis is not recommended as a stand-alone method, as amplicon sequence cannot be confirmed. In real-time PCR, product is detected by the production of fluorescence, usually as cycling is occurring (hence, in real time). Amplicons are detected by the intercalation of nonspecific fluorescent dyes or by hybridization of sequence-specific probes that are labeled with a fluorescent reporter dye. Depending on the detection method used, amplicons can be further analyzed by subjecting PCR reactions to a melting curve analysis, which determines the temperature at which the double- stranded product disassociates. Sequencing of the resulting products using either Sanger sequencing or next- generation sequencing (both techniques are discussed elsewhere in this text) can provide the sequence of the amplicon. This sequence can be used to confirm that the products are from amplification of the expected target. It can also be further analyzed to provide information about that target sequence and its functions in the original organism. PCR detection formats are discussed in detail in another chapter of this text.

PCR Variations Allele-Specific PCR Allele-specific PCR (AS-PCR) is used as a diagnostic technique to identify or utilize single nucleotide polymorphisms (SNPs), which in turn can distinguish closely related bacterial species [22]. The 3′-end of the primer is essential in the extension of the primer in a PCR reaction. Selective amplification is usually achieved by designing a primer such that the primer will match/mismatch one of the target sequences at the 3′-end of the primer. Therefore, AS-PCR does require prior knowledge of the target sequences. PCR amplification under stringent conditions is much less efficient in the presence of a mismatch between template and primer, so successful amplification with a SNP-specific primer signals presence of the specific SNP in a sequence. Sometimes, an additional mismatch is deliberately


333

introduced close to the targeted SNP to further decrease the melting temperature of the double-stranded DNA product and enhance sequence discrimination. Several detection methods can be used in AS-PCR. The most commonly used detection method involves fluorescence labeled, dideoxynucleotide terminators that stop the chain extension. Alternatively, the primer can be labeled and extension products separated by electrophoresis. The single base extension product can also be broken down into smaller pieces and measured by mass spectrometry.

Hot-Start PCR Hot-start PCR was first described in the literature in 1991 [23], and practical applications were demonstrated in 1992 [24]. Hot-start PCR techniques focus on the inhibition of DNA polymerase activity during reaction setup. By limiting polymerase activity prior to the elevated temperatures of PCR, nonspecific amplification is reduced, and target yield is increased. This is accomplished by physically separating or chemically inactivating one or more of the reaction components until high temperature triggers mixing or reactivation to give a complete reaction mixture. Many PCR kits that incorporate a hot-start mechanism are now commercially available. In manual hot-start PCR, reactions lacking one essential component (usually DNA polymerase) are prepared and held at a temperature above the threshold of nonspecific binding of primer to template. Just prior to cycling, the missing component is added to allow the reaction to take place at the higher temperature. This procedure limits nonspecific annealing of the primers and generally improved yield of the desired amplicon. This manual method is tedious and ungainly, as the tubes must be kept at 95–100°C. At this temperature, tubes are uncomfortable to handle. Opening the tubes to add the final reagent increases the chance of introducing contamination. This procedure has generally been replaced by chemical or physical barriers within the reaction mixture. Hot-start PCR is also accomplished by creating a physical barrier between the essential components, such as primers and template, or enzyme and magnesium chloride. This barrier can be created by adding wax over an incomplete PCR reaction mixture in a tube [25–27]. The wax can be preformulated for PCR reactions or in bulk form, such as paraffin. The remaining PCR component(s) is placed on top of the wax layer. During the first denaturation step, the wax barrier melts, and convection currents mix the essential PCR components. Additional hot-start methods include chemically modified Taq DNA polymerase and an antibody-inhibited Taq DNA polymerase [28, 29]. The antibody is directed against the active site of the enzyme, preventing DNA replication until the high temperature of the denaturation step disassociates the antibody [30]. These modified enzyme preparations require a longer initial denaturation step than standard Taq DNA polymerase.

334


Similar to controlled temperature and physical separation of PCR reaction components, cosolvents and enzymes have also been used to reduce or eliminate nonspecific annealing of primers [31]. Cosolvents such as dimethyl sulfoxide (DMSO) and formamide increase stringency by changing the melting temperature of the primer-template hybrid. Glycerol is believed to function similar to cosolvents. Cosolvents have various effects on the polymerase enzyme. Glycerol increases the temperature stability of Taq DNA polymerase, while formamide lowers it. The enzyme uracil N-glycosylase (UNG) is used in PCR as part of a system to degrade dUTP-containing product carried over from previous PCR reactions [32, 33]. Another benefit of UNG is that it degrades PCR product formed during the PCR setup process, prior to the high temperatures of cycling that provide specificity. In this role, UNG essentially provides an enzymatic hot-start PCR.

Touchdown PCR Unlike a standard PCR program that utilizes a constant annealing temperature, touchdown PCR incorporates a range of annealing temperatures. The earliest cycles of touchdown PCR have high annealing temperatures. In subsequent cycles, the annealing temperature is decreased by small increments (usually 1°C) every several cycles to a final “touchdown” annealing temp which is then used for the remaining ten or so cycles. This gradual decrease in annealing temperature selects for the most complementary primer/target binding in early cycles. This is most likely the sequence of interest. As the annealing temperature decreases, primers will anneal to nonspecific sequences; however, amplification of these products will lag behind that of the specific product. This favors synthesis of intended product over any nonspecific products [34]. Touchdown PCR was originally utilized to simplify the process of determining optimal PCR annealing temperatures. Touchdown PCR has been utilized in laboratory-developed multiplex tests for stool parasites, Vibrio cholerae serotypes, antimicrobial resistance genes, and Propionibacterium acnes phylogroups [35–38].

Degenerate PCR Degenerate PCR is a procedure that intentionally lowers analytical specificity, to allow divergent sequences to be detected in spite of sequence variation in the primer- binding region. Rather than using a single primer pair with a specific sequence, degenerate primer sets may contain several primers that vary at one or more nucleotide positions or a primer containing a nonspecific base, such as inosine, at a divergent position. There are circumstances in diagnostic microbiology when greater inclusivity is useful. For example, the genus Norovirus (family Caliciviridae) is comprised of dozens of distinct strains with relatively high genetic diversity [39]. A single, standard primer pair would lack sufficient complementarity to detect most


335

strains and have little diagnostic value. Degenerate reverse transcription PCR has been used successfully to detect a broad range of noroviruses [39].

Nested and Heminested PCR Nested and heminested PCR are designed to increase the sensitivity of PCR by directly reamplifying the product from a primary PCR with a second PCR. Nested PCR uses two sets of amplification primers and two separate stages of PCR [40, 41]. The second (nested) set of primers anneal to a sequence internal to the region flanked by the first set. In heminested PCR, the second stage of PCR utilizes one of the first-stage primers and one new, internal primer. The advantage of nested PCR is increased sensitivity and specificity of the reaction, since the internal primers anneal only if the amplicon has the corresponding expected sequence. The second-stage PCR is usually free of inhibitory substances that can reduce the efficiency and sensitivity of the first-stage PCR. Disadvantages of nested PCR include additional time and cost associated with two stages of PCR and the increased risk of contamination incurred during transfer of first-stage amplification products to a second tube. The physical separation of amplification mixtures with wax or oil [42], using an integrated PCR system in a cartridge or pouch [5, 6], and designing the second primer set with a higher annealing temperature are variations used to reduce the potential for contamination.

Multiplex PCR In multiplex PCR, two or more unique DNA sequences in the same specimen are amplified simultaneously [10, 43]. Primers used in multiplex reactions must be designed carefully to have similar annealing temperatures and to lack complementarity, to avoid dimerization. Primers are designed such that each amplification product has a unique size, melting temperature, or probe-binding sequence. This allows the detection and identification of different microorganisms in the same specimen. Multiplex PCR requires careful optimization of annealing conditions for maximal amplification efficiency. Commercial kits have been shown to efficiently amplify different sequences under single annealing conditions due to careful design of primer sequences and the use of buffer conditions that widen the annealing temperature window [44]. Multiplex PCR assays frequently incorporate an internal control to monitor every step of the procedure. Multiplex PCR has been well demonstrated to accurately and reliably detect multiple targets in a single PCR reaction [6, 13, 15, 45–47]. Highly multiplexed PCR assays (generally detect more than six targets) are based on conventional PCR, as multiplexing by real-time PCR is limited by the emission spectra and overlapping absorption (“bleedthrough”) of reporter dyes.

336


Multiplex PCR panels are now utilized in many clinical microbiology labs. These panels are configured based on a specific clinical syndrome (acute gastroenteritis, upper respiratory disease, meningitis, etc.) and test for many of the pathogens that commonly cause these syndromes. PCR methodologies vary from system to system. One system uses a nested PCR approach where the first stage of PCR occurs in a single mixture and the second stage occurs in individual reactions specific to each target [48]. Nucleic acid extraction is performed in the same pouch as the PCR, and melt curve analysis is used for detection. A second system requires separate DNA extraction and uses a bead-based microarray for detection [49]. The use of multiplex panels and sample-to-answer systems is likely to continue to expand in clinical microbiology laboratories.

Reverse Transcription PCR Reverse transcription (RT)-PCR is a technique used to amplify RNA targets. Because DNA polymerase requires a double-stranded DNA template, RNA must be transcribed into complementary (c) DNA prior to PCR by the enzyme reverse transcriptase. The cDNA then serves as the template for the first PCR temperature cycle (Fig. 3). The combined use of RT and PCR with thermostable DNA polymerase to amplify RNA targets was first described in 1988 [3]. Reviews describing the numerous applications of RT-PCR are available [50]. RT-PCR is an important technique in the diagnosis of infectious diseases, given the large number of clinically significant RNA viruses. Two reverse transcriptase enzymes commonly used are Moloney murine leukemia virus (M-MuLV) reverse transcriptase and avian myeloblastosis virus (AMV) reverse transcriptase. Both enzymes have the same fundamental activities but differ in some characteristics, including temperature and pH optima. In addition to M-MuLV and AMV, other variants of this enzyme are available for use in the molecular diagnostic laboratory. These enzymes are available in pre-optimized RT-PCR kits. In vitro reverse transcription is primer directed: a single primer is used to generate cDNA and can be one of the primers used in the subsequent PCR reaction (sequence-specific), or a random oligonucleotide. Specificity is not required of reverse transcription. Random oligonucleotides are convenient in that one RT kit or reaction can be used for all RNA targets. When initially described, the RT step was performed in a separate tube containing only components necessary for reverse transcription. Following RT, an aliquot is removed, added to a PCR reaction tube, and subjected to amplification. Drawbacks of the separate tube method include inconvenience and cross-contamination risk. Currently, single tube RT-PCR assays, either two enzyme or single enzyme, are the norm. A DNA polymerase isolated from Thermus thermophilus (Tth) is a thermostable enzyme that has both reverse transcriptase and DNA polymerase activity. This enzyme replicates DNA at approximately 74 °C in the presence of magnesium and makes cDNA from an RNA


337 mRNA or total RNA Down stream primer

Reverse transcription

RNA degradation or denature First strand complementary (c) DNA PCR cycle 1 Upstream primer

PCR cycle 2

Fig. 3 Amplification of RNA by reverse transcription-PCR. Primers are represented by bold lines, newly replicated DNA strands are represented by long dashes, and template DNA strands by short dashes

template in the presence of manganese. This allows reverse transcriptase and PCR to occur in the same reaction mixture without manipulation. This enzyme is used extensively for PCR amplification of RNA targets [51].

Quantitative PCR A variety of quantitative PCR assays have been developed to accurately quantify nucleic acid targets in clinical specimens [8, 52–54]. In addition to PCR, other molecular techniques such as branched (b) DNA provide accurate quantification of nucleic acids. While these methods determine the amount of DNA or RNA template in a clinical specimen, the results can be easily extrapolated to organism equivalents, hence the use of terms bacterial load, viral load, and so forth. The clinical value of quantitative PCR has led to commercialization of tests for such viruses as human immunodeficiency virus (HIV), cytomegalovirus, hepatitis C virus, and hepatitis B virus. Quantitative PCR results have become a valuable tool for guiding antiviral therapy, monitoring clinical course, and predicting outcome from a variety of infectious diseases [8, 55–57].

338


Nucleic acids can be quantified using an absolute standard in order to generate concrete numbers or a relative standard to give comparative data. Absolute standards can be used whenever definite numbers are needed. Relative standards are useful when absolute quantities are less important than knowing how a sample differs from a control. Fundamental PCR quantification strategies are relative, competitive, and comparative. Relative quantitative PCR compares nucleic acid amount across a number of serial dilutions of a sample, using a co-amplified internal control for sample normalization. Results are expressed as ratios of the sequence-specific signal to the internal control signal. This yields a corrected relative value for the sequencespecific product in each sample. Relative PCR uses primers for an internal control that are multiplexed in the same PCR reaction with the target-specific primers. Internal control and target-specific primers must be compatible – that is, they must not produce additional bands or hybridize each other. The signal from the internal control is used to normalize sample data to account for variation in RT or amplification efficiency. Common internal controls include the housekeeping genes (or their mRNAs) β-actin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and 18S rDNA. Competitive PCR provides absolute quantification of a nucleic acid target in a sample. An internal control or quantification standard is added at a known concentration to samples and co-amplified with the target sequence. Addition of the internal control to the sample prior to processing monitors for nucleic acid recovery during this step. The internal control is often a synthetic RNA or DNA with the same primer-binding sequence (hence the term competitive PCR) but designed to produce an amplicon slightly different in size than the target amplicon or with a unique internal sequence allowing detection with a different probe. Following amplification and detection, the amount of product or signal generated by the internal control is equated to its known input copy number. This relationship is then used to determine the copy number of the target sequence.

Real-Time PCR Real-time PCR, also known as kinetic PCR, detects and measures amplification as it occurs [58], compared to conventional PCR where the product of amplification is detected after the reaction is complete. Real-time PCR monitors the fluorescence emitted during the reaction as an indicator of amplicon production during each PCR cycle. During amplification, fluorescence increases in direct proportion to the amount of PCR product (Fig. 4). During the exponential phase of amplification, the increase in the amount of PCR product correlates to the initial amount of target template. Hence, real-time PCR is well suited for quantitative PCR. The availability of real-time PCR has allowed comparative quantification comparing PCR results to an external standard curve to determine target copy number.

339

PCR and Its Variations Sample A

Fluorescence

Positive control

Log-linear phase

Threshold Negative control 10

CT A

20

CTPC

40

50

Cycles

Fig. 4 Real-time PCR amplification plot

A separate internal control of an unrelated sequence should be incorporated into each PCR reaction to monitor all steps of the procedure. Because measurements are taken at each cycle, during the exponential phase of PCR, efficiency is consistent between samples. Conventional PCR measures product only at the endpoint, when the effects of inhibitors are significant. Because PCR product is measured during the exponential cycles, quantification by real-time PCR is more accurate and precise over a greater range than conventional PCR. Real-time PCR offers dynamic range of up to 107-fold, compared to 1000-fold in conventional PCR. External standards are used to create a standard curve across the dynamic range of the PCR assay. Real-time PCR generates a crossing threshold (CT) or crossing point (CP) cycle for each sample. This is the point where product (fluorescence) crosses a predetermined threshold. The higher the amount of starting target, the lower the CT. The CT for an unknown patient sample is analyzed against the standard curve to yield a target DNA or RNA copy number. The two common methods for detection of real-time PCR products are (1) DNA-binding agents (nonspecific fluorescent dyes such as SYBR Green) that intercalate nonspecifically with double-stranded DNA and (2) sequence-specific DNA probes labeled with a reporter which fluoresces only after hybridization of the probe with its complementary DNA target. Among the sequence-specific probes, there are hybridization probes and hydrolysis probe [59]. All methods require a thermal cycler equipped with optics that monitor the fluorescence in each PCR reaction at frequent intervals. Since SYBR green binds nonspecifically to double-stranded DNA, fluorescence will be produced by any amplification product, including primer dimers. This necessitates the post-amplification analysis of PCR product-melting curve analysis. Dissociation or “melting” of double-stranded DNA during heating is measurable by the large reduction in fluorescence that results. The melting temperature of DNA is very exact and depends on the base composition (and length

340


Fig. 5 Real-time PCR melting curve

if it is very short). In addition to its use with SYBR green real-time PCR, melting curve analysis is often performed when using hybridization probes. The temperature at which the probes dissociate from amplicons can be used to distinguish single nucleotide polymorphisms and other minor differences such as those of amplicons generated from the glycoprotein D gene of Herpes simplex virus types 1 and 2 [60]. In real-time PCR, amplification and amplicon analysis are performed in the same tube and do not require the transfer or opening of PCR reaction vessels. The melting curve analysis is performed automatically as per a predefined program, immediately upon completion of amplification. A graph of the negative first derivative of the temperature of dissociation (defined as 50% dissociation) produces distinct peaks (Fig. 5).

Digital PCR Finally, droplet digital PCR (ddPCR) is a new approach to absolute nucleic acid quantification that directly counts the number of target molecules, rather than relying on reference standards or endogenous controls [61]. In ddPCR, sample is combined with master mix, and the resulting solution is partitioned either in emulsion droplets or wells on a nanofluidics chip. Each droplet or well is a separate PCR reaction that ideally contains either one copy of the template or no template. If template is present, amplification occurs and the dye fluoresces. Each droplet or well is read at the end of the PCR, and the number of fluorescent partitions is counted. ddPCR has been used to verify the viral load in international standards of BK virus and CMV [62, 63]. ddPCR has also been used to more accurately quantify viruses with high-sequence diversity, as the results are not dependent on equal primer binding [64].


341

PCR-Based Strain Typing Techniques PCR-based strain typing techniques are designed to generate multiple bands that provide a unique fingerprint for a particular species or strain of microorganism [65, 66]. Unlike diagnostic tests that determine presence or absence of a microorganism (or its nucleic acid) in a specimen, these procedures are used to differentiate epidemiologically unrelated organisms at the species or subspecies level. They must generally produce multiple DNA bands to provide sufficient discrimination power, and these banding patterns must be reproducible run-to-run and lab-to-lab and among isolates of the same predefined group while clearly distinguishing isolates that epidemiologically or phenotypically fall outside of that group.

AP-PCR and RAPD Arbitrarily primed PCR (AP-PCR) and random amplified polymorphic DNA (RAPD) are methods of creating genomic fingerprints from species, even if little is known about the target sequence to be amplified [67–72]. Strain-specific arrays of amplicons (fingerprints) are generated by PCR amplification using arbitrary, or random sequence oligonucleotides that are often less than ten nucleotides in length, and low-temperature annealing. A single primer is often used, since it will anneal in both orientations. Detectable PCR product is generated when the primers anneal at the proper orientation and within a reasonable distance of one another. In spite of the arbitrary nature of the assay and amplification conditions that are relatively nonspecific, these methods have been shown to generate reproducible DNA banding patterns. These same characteristics make these methods suitable for a wide range of bacteria.

AFLP Amplified fragment-length polymorphism (AFLP) involves the restriction of genomic DNA, followed by ligation of adapters or linkers containing the restriction sites to the ends of the DNA fragments. The linkers and the adjacent restriction site serve as primer binding sites for subsequent amplification of the restriction fragments by PCR. Selective nucleotides extending into the restriction fragments are added to the 3’ ends of the PCR primers such that only a subset of the restriction fragments are recognized. Only restriction fragments in which the nucleotides flanking the restriction site match the selective nucleotides will be amplified. The amplified fragments are visualized by means of autoradiography, phospho-imaging, or other methods. Like AP-PCR and RAPD, AFLP can be applied to organisms without previous knowledge of genomic sequence.

342


ERIC-, Rep-, BOX-, IS-, and VNTR-PCR Enterobacterial repetitive intergenic consensus (ERIC)-PCR, repetitive element (Rep)-PCR, insertion sequence (IS)-PCR, and variable number tandem repeat (VNTR)-PCR are examples of PCR-based typing methods that target repetitive, conserved sequences found in bacteria and, in some cases, fungi. In a seminal 1991 paper, Versalovic et al. described the presence of repetitive sequences in a wide range of bacterial species and demonstrated their use to directly fingerprint bacterial genomes (Fig. 6) [73]. Specific repetitive sequences include the 124–127 base pair ERIC sequence, the 154 base pair BOX sequence, and the 35–40 base pair repetitive extragenic palindromic sequence. These sequences are located intergenically throughout the chromosome. Some repetitive sequences translocate to new locations in the genome and are called transposons or insertion sequences. Some ISs are species-specific, while others have no species restriction. VNTRs are repeated sequences of noncoding DNA. Whether ERIC, IS, VNTR, or other repetitive element or sequence, the basis of the strain typing is the same. The ability of repetitive element-based PCR methods to distinguish unrelated strains or species is based on the random distribution of elements within the genome and the time required for these to become established. That is, bacteria associated with a common source outbreak are highly unlikely to have any differences in the number or location of repetitive elements, while bacteria that are geographically, temporally, and epidemiologically unrelated are more likely to have experienced mutational events. Repetitive element-based PCR assays are designed so that primers anneal to the specific sequence in an outward orientation, so that DNA between the repeated elements is amplified. Variability between unrelated organisms is due to the random

Bacterial genome with conserved elements (shaded areas)

PCR amplification

PCR amplicons of varying lengths, determined by number and spacing of repetitive elements

Separate and visualize in agarose gel

Fig. 6 Repetitive element (Rep)-PCR


343

number and location of the elements on the genome. The utility of this PCR-based fingerprinting method is demonstrated by its successful commercialization. A kit includes reagents to perform Rep-PCR and electrophoresis, a fingerprint library, and software for data analysis for a variety of bacteria and fungi [74].

Appendix: Primer Design Resources There are a number of primer design programs and related resources available for free on the World Wide Web. The following is a brief sampling of primer design programs. (URLs accessed on May 21, 2017). For information on Primer design BLAST search for primer design Real-time PCR (Taqman) primer design Degenerate primer design General resource site PCR and qPCR primer and probe design

URL http://biotools.umassmed.edu/bioapps/primer3_www.cgi http://www.ncbi.nlm.nih.gov/tools/primer-blast/index. cgi?LINK_LOC=NcbiHomeAd https://www.genscript.com/ssl-bin/app/primer http://bibiserv.techfak.uni-bielefeld.de/genefisher2/ http://molbiol-tools.ca/PCR.htm https://www.idtdna.com/Primerquest/Home/Index

References 1. Mullis KB, Faloona FA. Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Methods Enzymol. 1987;155:335–50. 2. Saiki RK, Scharf S, Faloona F, et al. Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science. 1985;230:1350–4. 3. Saiki RK, Gelfand DH, Stoffel S, et al. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science. 1988;239:487–91. 4. Fredricks DN, Relman DA. Application of polymerase chain reaction to the diagnosis of infectious diseases. Clin Infect Dis. 1999;29:475–86. 5. Marlowe EM, Novak-Weekley SM, Cumpio J, et al. Evaluation of the Cepheid Xpert MTB/ RIF assay for direct detection of Mycobacterium tuberculosis complex in respiratory specimens. J Clin Microbiol. 2011;49:1621–3. 6. Rand KH, Rampersaud H, Houck HJ. A comparison of two multiplex methods for the detection of respiratory viruses: FilmArray RP and xTAG RVP. J Clin Microbiol. 2011;49:24494–2453. 7. Selvaraju SB, Wurst M, Horvat RT, Selvarangan R. Evaluation of three analyte-specific reagents for detection and typing of herpes simplex virus in cerebrospinal fluid. Diagn Microbiol Infect Dis. 2009;63:286–91. 8. Sire JM, Vray M, Merzouk M, et al. Comparative RNA quantification of HIV-1 group M and non-M with the Roche Cobas AmpliPrep/Cobas TaqMan HIV-1 v2.0 and Abbott real-time HIV-1 PCR assays. J Acquir Immune Defic Syndr. 2011;56:239–43.

344


9. Tang YW, Procop GW, Persing DH. Molecular diagnostics of infectious diseases. Clin Chem. 1997;43:2021–38. 10. Tang Y, Persing DH. Molecular detection and identification of microorganisms. In: Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH, editors. Manual of clinical microbiology. 7th ed. Washington, DC: ASM Press; 1999. 11. Loeffelholz M, Deng H. PCR and its variations. In: Tang YW, Stratton CW, editors. Advanced techniques in diagnostic microbiology. 1st ed. New York: Springer; 2006. 12. Mullis KB. The unusual origin of the polymerase chain reaction. Sci Am. 1990;262:56–61. 64–55 13. Balada-Llasat JM, LaRue H, Kelly C, Rigali L, Pancholi P. Evaluation of commercial ResPlex II v2.0, MultiCode-PLx, and xTAG respiratory viral panels for the diagnosis of respiratory viral infections in adults. J Clin Virol. 2011;50:42–5. 14. Kost CB, Rogers B, Oberste MS, et al. Multicenter beta trial of the GeneXpert enterovirus assay. J Clin Microbiol. 2007;45:1081–6. 15. Selvaraju SB, Selvarangan R. Evaluation of three influenza A and B real-time reverse transcription-PCR assays and a new 2009 H1N1 assay for detection of influenza viruses. J Clin Microbiol. 2010;48:3870–5. 16. Stevenson JB, Hymas WC, Hillyard DR. A novel capillary electrophoresis-based multiplex PCR assay for detection of respiratory pathogens. Ann Clin Lab Sci. 2011;41:33–8. 17. White TJ, Madej R, Persing DH. The polymerase chain reaction: clinical applications. Adv Clin Chem. 1992;29:161–96. 18. Collasius M, Falk H, Ciesler C, Valet G. How to build an inexpensive cyclotherm instrument for automated polymerase chain reaction. Anal Biochem. 1989;181:163–6. 19. Wittwer CT, Fillmore GC, Hillyard DR. Automated polymerase chain reaction in capillary tubes with hot air. Nucleic Acids Res. 1989;17:4353–7. 20. Wittwer CT, Hillyard DR, Ririe KM. US patent 5455175. 1995. 21. Martin UK, de Mello AJ, Andreas M. Chemical amplification: continuous-flow PCR on a chip. Science. 1998;280:1046–8. 22. de los Monteros LE E, Galan JC, Gutierrez M, et al. Allele-specific PCR method based on pncA and oxyR sequences for distinguishing Mycobacterium bovis from Mycobacterium tuberculosis: intraspecific M. bovis pncA sequence polymorphism. J Clin Microbiol. 1998;36:39–242. 23. Mullis KB. The polymerase chain reaction in an anemic mode: how to avoid cold oligodeoxyribonuclear fusion. PCR Methods Appl. 1991;1:1–4. 24. Chou Q, Russell M, Birch DE, Raymond J, Bloch W. Prevention of pre-PCR mis-priming and primer dimerization improves low-copy-number amplifications. Nucleic Acids Res. 1992;20:1717–23. 25. Bassam BJ, Caetano-Anolles G. Automated “hot start” PCR using mineral oil and paraffin wax. BioTechniques. 1993;14:30–4. 26. Horton RM, Hoppe BL, Conti-Tronconi BM. AmpliGrease: “hot start” PCR using petroleum jelly. BioTechniques. 1994;16:42–3. 27. Riol H, Levesque G, Murthy MR. A method of using heavy mineral oil for performing “hot- start” amplification of rare nucleic acids. Anal Biochem. 1994;221:210–2. 28. Birch DE. Simplified hot start PCR. Nature. 1996;381:445–6. 29. Kebelmann-Betzing C, Seeger K, Dragon S, et al. Advantages of a new Taq DNA polymerase in multiplex PCR and time-release PCR. BioTechniques. 1998;24:154–8. 30. Kellogg DE, Rybalkin I, Chen S, et al. TaqStart Antibody’: “hot start” PCR facilitated by a neutralizing monoclonal antibody directed against Taq DNA polymerase. BioTechniques. 1994;16:1134–8. 31. Rasmussen HN, Rasmussen OF, Andersen JK, Olsen JE. Specific detection of pathogenic Yersinia enterocolitica by two-step PCR using hot-start and DMSO. Mol Cell Probes. 1994;8:99–108. 32. Pang J, Modlin J, Yolken R. Use of modified nucleotides and uracil-DNA glycosylase (UNG) for the control of contamination in the PCR-based amplification of RNA. Mol Cell Probes. 1992;6:251–6.


345

33. Udaykumar EJS, Hewlett IK. A novel method employing UNG to avoid carry-over contamination in RNA-PCR. Nucleic Acids Res. 1993;21:3917–8. 34. Don RH, Cox PT, Wainwright BJ, Baker K, Mattick JS. ‘Touchdown’ PCR to circumvent spurious priming during gene amplification. Nucleic Acids Res. 1991;19:4008. 35. Shin J-H, Lee S-E, Kim TS, Ma D-W, Chai J-Y, Shin E-H. Multiplex-touchdown PCR to simultaneously detect Cryptosporidium parvum, Giardia lamblia, and Cyclospora cayetanensis, the major causes of Traveler’s diarrhea. Korean J Parasitol. 2016;54:631–6. 36. Wang MY, Geng JL, Chen YJ, Song Y, Sun M, Liu HZ, Hu CJ. Direct detection of mecA, blaSHV , blaCTX-M , blaTEM and blaOXA genes from positive blood culture bottles by multiplex-touchdown PCR assay. Lett Appl Microbiol. 2017;64:138–43. 37. Bhumiratana A, Siriphap A, Khamsuwan N, Borthong J, Chonsin K, Sutheinkul O. O serogroup-specific touchdown-multiplex polymerase chain reaction for detection and identification of Vibrio cholerae O1, O139, and Non-O1/Non-O139. Biochem Res Int. 2014;2014:295421. 38. Barnard E, Nagy I, Hunyadkürti J, Patrick S, McDowell A. Multiplex touchdown PCR for rapid typing of the opportunistic pathogen Propionibacterium acnes. J Clin Microbiol. 2015;53:1149–55. 39. Moe CL, Gentsch J, Ando T, et al. Application of PCR to detect Norwalk virus in fecal specimens from outbreaks of gastroenteritis. J Clin Microbiol. 1994;32:642–8. 40. Haqqi TM, Sarkar G, David CS, Sommer SS. Specific amplification with PCR of a refractory segment of genomic DNA. Nucleic Acids Res. 1988;16:11844. 41. Schmidt B, Muellegger RR, Stockenhuber C, et al. Detection of Borrelia burgdorferi-specific DNA in urine specimens from patients with erythema migrans before and after antibiotic therapy. J Clin Microbiol. 1996;34:1359–63. 42. Whelen AC, Felmlee TA, Hunt JM, et al. Direct genotypic detection of Mycobacterium tuberculosis rifampin resistance in clinical specimens by using single-tube heminested PCR. J Clin Microbiol. 1995;33:556–61. 43. Chamberlain JS, Gibbs RA, Ranier JE, Nguyen PN, Caskey CT. Deletion screening of the Duchenne muscular dystrophy locus via multiplex DNA amplification. Nucleic Acids Res. 1988;16:11141–56. 44. Rossister BJF, Grompe M, Caskey CT. In: MJ MP, Quirke P, Taylor GR, editors. PCR. A practical approach. New York: Oxford University Press; 1991. 45. Bej AK, Mahbubani MH, Miller R, DiCesare JL, Haff L, Atlas RM. Multiplex PCR amplification and immobilized capture probes for detection of bacterial pathogens and indicators in water. Mol Cell Probes. 1990;4:353–65. 46. Geha DJ, Uhl JR, Gustaferro CA, Persing DH. Multiplex PCR for identification of methicillin- resistant staphylococci in the clinical laboratory. J Clin Microbiol. 1994;32:1768–72. 47. Roberts TC, Storch GA. Multiplex PCR for diagnosis of AIDS-related central nervous system lymphoma and toxoplasmosis. J Clin Microbiol. 1997;35:268–9. 48. Poritz MA, Blaschke AJ, Byington CL, et al. FilmArray, an automated nested multiplex PCR system for multi-pathogen detection: development and application to respiratory tract infection. PLoS One. 2011;6:e26047. 49. Dunbar SA. Applications of Luminex xMAP technology for rapid, high-throughput multiplexed nucleic acid detection. Clin Chim Acta. 2006;363:71–82. 50. Larrick JW. Message amplification phenotyping (MAPPing) – principles, practice and potential. Trends Biotechnol. 1992;10:146–52. 51. Myers TW, Gelfand DH. Reverse transcription and DNA amplification by a Thermus thermophilus DNA polymerase. Biochemistry. 1991;30:7661–6. 52. Boyer N, Marcellin P. Pathogenesis, diagnosis and management of hepatitis C. J Hepatol. 2000;32:98–112. 53. Clarke JR, McClure MO. HIV-1 viral load testing. J Infect. 1999;38:141–6. 54. Mylonakis E, Paliou M, Rich JD. Plasma viral load testing in the management of HIV infection. Am Fam Physician 2001;63:483–490, 495–486. 55. Hodinka RL. The clinical utility of viral quantitation using molecular methods. Clin Diagn Virol. 1998;10:25–47.

346


5 6. Jung R, Soondrum K, Neumaier M. Quantitative PCR. Clin Chem Lab Med. 2000;38:833–6. 57. Orlando C, Pinzani P, Pazzagli M. Developments in quantitative PCR. Clin Chem Lab Med. 1998;36:255–69. 58. Higuchi R, Dollinger G, Walsh PS, Griffith R. Simultaneous amplification and detection of specific DNA sequences. Biotechnology (NY). 1992;10:413–7. 59. Nolte FS, Caliendo AM. Molecular microbiology. In: Versalovic J, Carroll KC, Funke G, Jorgensen JH, Landry ML, Warnock DW, editors. Manual of clinical microbiology. 10th ed. Washington, D C: ASM Press; 2011. p. 27–59. 60. Whiley DM, Mackay IM, Syrmis MW, Witt MJ, Sloots TP. Detection and differentiation of herpes simplex virus types 1 and 2 by a duplex LightCyler PCR that incorporates an internal control PCR reaction. J Clin Virol. 2004;30:32–8. 61. Zhong Q, Bhattacharya S, Kotsopoulos S, et al. Multiplex digital PCR: breaking the one target per color barrier of quantitative PCR. Lab Chip. 2011;11:2167–74. 62. Bateman AC, Greninger AL, Atienza EE, Limaye AP, Jerome KR, Cook L. Quantification of BK virus standards by quantitative real-time PCR and droplet digital PCR is confounded by multiple virus populations in the WHO BKV international standard. Clin Chem. 2017;63:761–9. 63. Hayden RT, Gu Z, Sam SS, Sun Y, Tang L, Pounds S, Caliendo AM. Comparative evaluation of three commercial quantitative cytomegalovirus standards by use of digital and real-time PCR. J Clin Microbiol. 2015;53:1500–5. 64. Sedlak RH, Nguyen T, Palileo I, Jerome KR, Kuypers J. Superiority of digital reverse transcription-PCR (RT-PCR) over real-time RT-PCR for quantitation of highly divergent human rhinoviruses. J Clin Microbiol. 2017;55:442–9. 65. Fernandez-Cuenca F. Applications of PCR techniques for molecular epidemiology of infectious diseases. Enferm Infecc Microbiol Clin. 2004;22:355–60. 66. Olive DM, Bean P. Principles and applications of methods for DNA-based typing of microbial organisms. J Clin Microbiol. 1999;37:1661–9. 67. Grattard F, Berthelot P, Reyrolle M, Ros A, Etienne J, Pozzetto B. Molecular typing of nosocomial strains of Legionella pneumophila by arbitrarily primed PCR. J Clin Microbiol. 1996;34:1595–8. 68. MacGowan AP, O'Donaghue K, Nicholls S, McLauchlin J, Bennett PM, Reeves DS. Typing of Listeria spp. by random amplified polymorphic DNA (RAPD) analysis. J Med Microbiol. 1993;38:322–7. 69. Matsui C, Pereira P, Wang CK, et al. Extent of laminin-5 assembly and secretion effect junctional epidermolysis bullosa phenotype. J Exp Med. 1998;187:1273–83. 70. van Belkum A, Kluytmans J, van Leeuwen W, et al. Multicenter evaluation of arbitrarily primed PCR for typing of Staphylococcus aureus strains. J Clin Microbiol. 1995;33:1537–47. 71. Welsh EA, Clark HH, Epstein SZ, Reveille JD, Duvic M. Human leukocyte antigen-DQB1*03 alleles are associated with alopecia areata. J Invest Dermatol. 1994;103:758–63. 72. Woods JP, Kersulyte D, Tolan RW Jr, Berg CM, Berg DE. Use of arbitrarily primed polymerase chain reaction analysis to type disease and carrier strains of Neisseria meningitidis isolated during a university outbreak. J Infect Dis. 1994;169:1384–9. 73. Versalovic J, Koeuth T, Lupski JR. Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Res. 1991;19:6823–31. 74. Wise MG, Siragusa GR, Plumbee J, Healy M, Cary PJ, Seal BS. Predicting Salmonella enterica serotypes by repetitive sequence-based PCR. J Microbiol Methods. 2009;76:18–24.

Non-PCR Amplification Techniques Rosemary C. She, Ted E. Schutzbank, and Elizabeth M. Marlowe

Introduction In 1983 while driving up a mountain road, Dr. Kary Mullis envisioned the concept of the polymerase chain reaction (PCR). These scientific “driving” thoughts completely revolutionized biology and have created an entire biotechnology industry, resulting in new biotechnology methods, instruments, companies, and jobs. As with any good idea, PCR was quickly patented by Cetus and then sold to Hoffman La Roche for $300 million [1]. In the world of clinical diagnostics, the ability to compete with PCR also has fostered new technology. Since the advent of PCR, several other patents have been filed relating to non-PCR-mediated signal, probe, or target amplification techniques. Today, numerous companies have developed commercially available diagnostic methods and instruments that are based on non-PCR mediated signal or target amplification techniques (Table 1). Signal amplification technologies have one major advantage over target amplification in that contamination of an assay run with previously amplified material from a previous run is less of an issue. Also, signal amplification technologies are isothermal, meaning that unlike PCR, thermocycling instrumentation is not required. Lastly, assays can be designed to detect and/or quantify specific DNA or RNA targets and, in the case of RNA, without the need to first convert the RNA target to R. C. She (*) Keck School of Medicine of the University of Southern California, Los Angeles, CA, USA e-mail: [email protected] T. E. Schutzbank Ascension – St. John Providence, Grosse Pointe, MI, USA e-mail: [email protected] E. M. Marlowe Roche Molecular Systems, Inc., Pleasanton, CA, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_17

347

348

R. C. She et al.

Table 1 Commercially available non-PCR target amplification assays for the detection of microorganisms Manufacturer Hologic (Gen-Probe)

Grifols (Gen-Probe) bioMerieux Becton Dickinson Meridian Biosciences

Quidel

Technology Targets TMA Diagnostic Mycobacterium tuberculosis C. trachomatis/N. gonorrhoeae Trichomonas vaginalis Mycoplasma gentialium (CE-IVD only) HIV-1, HBV (CE-IVD only), HCV, HSV-1 and HSV-2, Zika virus, HPV and HPV 16 and 18/45 genotyping TMA Blood Screening HIV-1, HIV-2, HCV, HBV, WNV, HAV/ parvo, HEV, dengue and Zika viruses NASBA HIV-1 v. 2.0a SDA C. trachomatis/N. gonorrhoeae HSV- 1 & 2b Trichomonas vaginalisc LAMP C. difficile Group A Streptococcus Group B Streptococcus Mycoplasma pneumoniae Bordetella pertussis HSV- 1 and HSV-2 HDA HSV-1 and HSV-2 (AmpliVue) Bordetella C. difficile Group B Streptococcus HDA HSV-1 and HSV-2 and VZV (Solana) C. difficile Group A Streptococcus Groups A, C, and G Streptococcus Trichomonas vaginalis Influenza A and B RSV + hMPV

URL www.hologic.com

www.diagnostics. grifols.com www.biomerieux.com www.bd.com

www. meridianbiosciences. com

www.quidel.com

Research use only On BD Protec Tec ET and Viper XTR only c On BD Viper XTR only a

b

DNA via reverse transcription. Care must still be taken, however, to minimize cross- contamination between samples being tested due to the enhanced analytical sensitivity inherent in assays using signal amplification technologies. An inherent concern with signal amplification methods is that great care must be taken during the design of the assay to ensure that carryover of the different assay components, from one step to the next, is minimized to reduce background noise, which will degrade the analytical sensitivity of the test in question. Non-PCR-mediated target amplification techniques are also usually based on isothermal amplification. While various names of isothermal amplification have appeared in the literature throughout the years, this chapter will focus on the

Non-PCR Amplification Techniques

349

c ommercially available non-PCR techniques. For a historical perspective and a more in-depth technical review of non-PCR-mediated target amplification techniques, see this chapter in the first edition of this book. The basis of these techniques is mimicking cellular or viral replication of DNA or RNA. For example, such techniques may consist of transcription-based amplification, which generates an RNA product, or strand displacement-based amplification, which generates a DNA product [2]. The advantage of using non-PCR-mediated target amplification techniques is that they are very robust due to the amount of target that is generated in a very short amount of time. They also avoid the royalty costs associated with PCR. However, like any amplification technology, they also are subject to contamination. Thus, adherence to good molecular practices still applies when using non-PCR methods. Given the ease of designing and ordering PCR primers and probes, PCR still remains a common practice for many research and clinical laboratories. Both signal and non-PCR target amplification methods will be discussed in this chapter.

Signal Amplification Technologies Hybrid Capture Hybrid capture (HC) is a term that describes a signal amplification methodology incorporating synthetic RNA probes complementary to a specific target sequence (e.g., a viral or bacterial DNA target sequence), followed by capture and detection of the RNA/DNA hybrid molecules using an antibody specific for such RNA/DNA heteroduplexes. The production of anti-RNA/DNA antibodies actually dates back to the 1970s [3, 4]. These were polyclonal antibodies elicited by injection of poly (A) • poly (dT) hybrids into rabbits [5]. The development of monoclonal antibodies against heteropolymer duplexes against the bacteriophage ΦX174 single-stranded DNA genome hybridized with RNA transcribed in vitro from this template [6]. This same study described the use of such antibodies in the immunodetection of RNA/DNA hybrid molecules. The first report for the use of anti-RNA/DNA antibodies in a clinical assay was in 1993 by Carpenter et al. [7]. This paper described the development of novel oligonucleotide hairpin probe encoding a T7 RNA polymerase promoter. The hairpin probe and an adjacently hybridizing biotinylated capture probe were hybridized to target DNA (purified Chlamydia trachomatis DNA), and the duplex was captured onto streptavidin-coated magnetic particles. After ligation of the immobilized probes, which served to maintain specificity, the hairpin probe was transcribed by T7 RNA polymerase. The transcribed RNA product was hybridized to a biotinylated capture probe and bound to streptavidin-coated magnetic particles. The immobilized heteroduplex was detected with a RNA/DNA antibody-alkaline phosphatase conjugate. After a wash step to remove unbound antibody-enzyme conjugate, the chemiluminescent substrate adamantyl-1,2- dioxetane phenyl

350

R. C. She et al.

Fig. 1 Hybrid Capture technology (a, b, c). See text for details

phosphate was added and chemiluminescence measured in a luminometer. This method was shown to detect down to 10 attomoles of C. trachomatis DNA in a background of 5 μg of background DNA. Signal amplification using this procedure was achieved both by the linear amplification of the ligatable probes through the RNA transcription process, followed by binding of the enzyme conjugated anti-DNA/RNA antibodies. This methodology, however, was never developed for commercial use. The first use of anti-DNA/RNA antibodies in a commercial assay was the Hybrid Capture 1 (HC1) human papillomavirus test manufactured by Digene Corporation (now Qiagen Inc.) [8]. Since then, improvements to the methodology have been made with the introduction of Hybrid Capture 2 (HC2). The methodology can use either a DNA probe to detect an RNA target or an RNA probe to detect a DNA target. The first step in an HC2 test is lysis of the target organism, to release the target nucleic acid. If the target is DNA, a denaturation step is required. Therefore, the first step of an HC2 assay for the detection of a DNAcontaining target organism is the addition of a lysis solution containing sodium hydroxide, which disrupts the virus or bacteria in the sample, releases the target DNA, and denatures the DNA into single strands, which are accessible for hybridization with a target-specific RNA probe. After hybridization to the probes (illustrated in Fig. 1a), the next step is to transfer the sample to a microwell containing an immobilized anti-DNA/RNA monoclonal antibody (Fig. 1b). During this step, the RNA/DNA duplexes are captured by the immobilized antibodies onto the surface of the microwell. A wash cycle removes any unbound material. The third step is the addition of an alkaline phosphatase-conjugated (AP) RNA/ DNA monoclonal antibody. Several AP molecules are conjugated to each antibody, and multiple conjugated antibodies bind to each captured hybrid, which in turn results in signal amplification (Fig. 1c). The fourth step is detection of amplified chemiluminescent signal. The container is washed to remove all of the unbound components, while the RNA/DNA hybrids and the labeled antibody remain bound to the container. Dioxetane, a chemiluminescent substrate, is added


351

to the reaction well, which is cleaved by the bound alkaline phosphatase to produce light [9], which in turn is measured using a luminometer. Hybrid capture technology has been used to develop nucleic acid detection assays for human papillomavirus (HPV), cytomegalovirus (CMV), Chlamydia trachomatis (CT), Neisseria gonorrhoeae (GC), hepatitis B virus (HBV), and herpes simplex virus types 1 and 2 (HSV-1 and HSV-2). From a public health perspective, cervical cancer is one of the most important cancers due to the fact that the causative agent of cervical cancer, HPV [10–13], is sexually transmitted [14, 15]. HPV is also implicated in other types of cancer, such as anal cancer, vulvar cancer, penile cancer, and oral and laryngeal cancers. There are over 100 different types of HPV, with only a subset implicated in male and female genital infections. These can be differentiated into two groups, high risk and low risk, based on the predilection of infection by these viruses to progress to cancer. Digene Inc. (Beltsville, MD, now Qiagen Inc.) developed the first FDA-approved diagnostic assays for the detection of HPV in cervical scrapings. This test is available in two different configurations. The Digene HC2 HPV DNA Test employs probe cocktails to detect 5 low-risk HPV types, 6, 11, 42, 43, and 44 and 13 high-risk HPV types, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68, which are associated with cervical cancer. The second configuration, the Digene HC2 High-Risk HPV DNA Test, detects the 13 high-risk HPV types. The clinical utility of the Digene HC2 HPV DNA Test is questionable; guidelines for HPV testing, published by the American Society for Colposcopy and Cervical Pathology, state that there is no clinical utility in testing for non-oncogenic HPV types [16]. Samples for testing are collected with a proprietary collection device, available from Qiagen. Alternatively, cells collected in the liquid cytology media, Cytyc PreservCyt Solution (Hologic Inc., Bedford, MA), or from cervical biopsies are also approved for testing. Testing may be performed manually, or with the Rapid Capture System, a semiautomated pipetting and microwell plate handling system for high- volume sample-throughput testing. The analytical sensitivity of the HC2 assays is ~5000 HPV genome copies. On the surface, this may appear to be insensitive from a purely analytical perspective; however, this level of analytical sensitivity is ideal with respect to clinical sensitivity. Clinical utility of HPV screening is based on the association of detecting high- risk HPV in a clinical sample and the association with the presence of cervical cancer determined by a PAP smear, not just simply the presence of the virus. Therefore, detection of very low levels of HPV by ultra-sensitive methods, such as PCR, may have no clinical significance [17]. PCR methods require that a threshold of detection representing a clinically significant result be determined. The Hybrid Capture assays are not cleared to be used independently as a screening assay for the general population [18]. As just mentioned, FDA approval of HPV detection assays such as the Digene HC2 High-Risk HPV DNA Test is based not on sensitivity and specificity of detection of HPV but the sensitivity and specificity in conjunction with Papanicolaou smear (PAP smear) results, for risk of developing cervical cancer.

352

R. C. She et al.

The indications for using this test are: • To aid in the diagnosis of sexually transmitted HPV infections and to differentiate between high- and low-risk HPV-type infections. • To screen patients with a PAP smear interpretation of atypical squamous cells of undetermined significance (ASCUS) to determine the need for referral to colposcopy. • As a supplement to the PAP smear in women with low-grade squamous intraepithelial lesion (LSIL) or high-grade squamous intraepithelial lesion (HSIL) results to assess the risk for developing cervical cancer. The HC2 High-Risk HPV DNA Test is not intended for use as a screening test for PAP normal women under age 30 [16] and is not meant to be a substitute for regular PAP screening. More recently, PCR-based HPV tests have become available with a primary screening claim for HPV in women 25 years and older (https://www. sciencedaily.com/releases/2015/01/150108083704.htm). There are some limitations to the HC2 assay, one of which is that cross-reactivity of the high-risk HPV probes with low-risk HPV DNA sequences has been observed, which may lead to false-positive results [19, 20]. Neither of the HC-based assays have an internal positive control to differentiate true from false-negative results; the latter have been reported with the HC2 HPV high-risk test [20, 21]. HC2 technology has also been used to develop assays for the detection of two other sexually transmitted organisms, CT and GC. These are the most common causes of bacterial infections of the lower genital tract. The FDA-approved HC2 CT-ID DNA Test Version 2.0 Test is an in vitro diagnostic (IVD) assay for the qualitative detection of Chlamydia trachomatis DNA in cervical specimens collected using the HC2 DNA Collection Device (cervical brush and Specimen Transport Medium™ [STM]) and in cervical specimens collected using the Female Swab Specimen Collection Kit™ (Dacron swab and STM). The CT Probe Cocktail supplied with the HC2 CT-ID DNA Test is complementary to approximately 39,300 bp or 4% of the Chlamydia genomic DNA (1 × 106 bp). One probe is complementary to 100% of the cryptic plasmid of 7500 bp. In studies comparing the sensitivity and specificity of this test to culture, the HC2 CT-ID DNA Test demonstrated very comparable results in symptomatic and asymptomatic women [22, 23]. When compared to PCR, the HC2 method also gave very similar results, with PCR showing slightly higher sensitivity, as would be expected by a target amplification procedure [24]. The Digene CT/GC Dual ID HC2 DNA Test, which is also FDA-approved, is designed to screen patient populations for CT and GC. The clinical performance of the GC component of this test also compares very favorably to culture [23]. Hybrid Capture technology has also been applied to the detection of CMV [25– 27] and HBV [28, 29]. However, when compared with PCR-based methods, the low analytical sensitivities for the HC2 assays put them at a distinct disadvantage and are no longer being manufactured.


353

Invader Technology The Invader® assay, originally developed by Third Wave Technologies (now Hologic Inc.) is a homogeneous, isothermal DNA probe-based system with broad utility, ranging from the detection of single base pair mutations to the qualitative and quantitative detection of specific DNA sequences. The Invader methodology relies on the linear amplification of a target-specific signal, but not the actual target itself. Specificity is achieved through a combination of sequence-specific oligonucleotide hybridization steps and structure-specific enzymatic cleavage using the Cleavase® enzyme. A diagram of this process is illustrated in Fig. 2. Two synthetic oligonucleotides, an upstream Invader oligonucleotide (Invader oligo) and a downstream probe, both hybridize to the single-stranded target DNA to create the structure shown in Fig. 2. This structure contains a single base overlap precisely at the nucleotide being interrogated [30]. The 3′ sequence of the probe, the targetspecific region (TSR), is complementary to the target. The nonspecific 5′ domain remains unhybridized and forms a flap upon hybridization. The tripartite structure formed by the hybridization of the Invader oligo and probe to the target is the substrate for the Cleavase® enzyme which recognizes this structure and specifically cleaves the probe, releasing the “flap.” If, in the case of the structure shown in Fig. 2, the highlighted base was other than a “C,” the base at the site of this mismatch on the TSR becomes part of the “flap.” Cleavage does not occur because this structure is a poor substrate for Cleavase® [31]. How this is all configured to create a signal amplification system is shown in Fig. 3. The signal-generating reaction mixture contains all of the components just described. The reactions are performed at temperatures very near the melting temperature (Tm) of the probe, which is present in molar excess. Each time an intact probe molecule binds to the target in the presence of the Invader oligo, a cleavage can occur. Therefore, multiple probes are cleaved per target molecule since the probes cycle rapidly on and off the target molecule. The signal-generating component of the reaction is the fluorescence energy transfer (FRET) cassette composed of synthetic hairpin oligonucleotides containing a fluorescent dye (F) and a nonfluorescent quencher (Q) that acts as a FRET pair [31, 32]. The FRET cassette is comprised of two elements. The first is a region that is complementary to the flap sequence described above. The second is a self-complementary region that forms a hairpin structure that mimics the binding of an Invader probe to a target DNA molecule. Signal amplification occurs when the released flap from the first reaction hybridizes to the FRET cassette, forming the complex recognized by Cleavase®. Cleavage occurs between the fluorescent dye and the quencher, releasing the dye from the quencher, permitting the formation of a fluorescent signal when excited by light at the appropriate wavelength in a fluorimeter. Each 5’flap generates between 103 and 104 cleaved FRET cassettes per hour in a linear signal amplification reaction [33, 34].

354

R. C. She et al.

Fig. 2 Invader technology. See text for details

One of the earliest applications of Invader® technology was the development of assays for the detection of mutations in genes involved in hypercoagulability disorders, specifically factor V Leiden, factor II (prothrombin), and the methylenetetrahydrofolate reductase (MTHFR) genes [35]. These assays were available initially as analyte specific reagents but have since been approved by the FDA. All three tests are based on Invader Plus®. This process involves a PCR reaction, followed by an Invader reaction, all of which occurs in a single reaction, in a closed, single-tube format. This sequential process combines PCR-based target amplification with subsequent signal amplification through Invader® chemistry.


355

Fig. 3 Invader Biplex technology. See text for details

Another commercial application of Invader® technology was for the detection of mutations in the UGT1A1 gene that codes for the enzyme UDP- glucuronosyltransferase. This gene is responsible for the metabolism of certain drugs, including irinotecan, a chemotherapy agent commonly used to treat colorectal and lung cancer. Detecting specific variations in the UGT1A1 predicts which patients are at an increased risk of toxicity from irinotecan [36, 37]. The Invader® UGT1A1 Molecular Assay (Hologic Inc., Bedford, MA) targets the *28 allele, which is two base pair insertion (TA) in the UGT1A1 promoter region. Individuals either homozygous or heterozygous for this allele are seven times more likely to demonstrate severe toxicity to irinotecan than patients homozygous for the wild-type allele [36]. This test received FDA approval in 2005 but is no longer in production. Invader technology has also been successfully applied to the molecular diagnosis of cystic fibrosis (CF). The InPlex® CF Molecular Test (Hologic Inc., Bedford, MA) was an IVD assay that tests for 23 separate mutations in the cystic fibrosis transmembrane receptor (CFTR) gene. The IVS8-5 T/7 T/9 T markers were automatically reflexed as part of the test. All mutations contained in the assay are recommended for testing by the American College of Obstetricians and Gynecologists (ACOG) and the American College of Medical Genetics (ACMG). For the InPlex® technology, the appropriate genomic region was amplified by PCR using a limiting number of cycles. For the CF test, this was 13 cycles. The amplification products were injected into an assay-specific microfluidics card that has multiple chambers for analyzing each of the CFTR mutations. FDA clearance for this test was granted in 2008; however, this test is no longer marketed by Hologic. The Cervista® HPV HR test is an FDA-approved assay that tests for 13 high-risk HPV types listed above for the HC2 method but also includes HPV type 66. In addition, this test also includes an internal positive control to confirm adequacy of

356

R. C. She et al.

sample collection and to indicate if inhibitory substances are present. Instrumentation choices available from Hologic to perform this test range from semiautomated to fully automated platforms, depending on the needs of the laboratory. Cells collected in Cytyc PreservCyt Solution (Hologic Inc., Bedford, MA) are the indicated sample type for this test, although cervical cells collected in SurePath preservative-fixative fluid (BD TriPath Imaging, Burlington, SC) have been shown to work with this test as well [38]. However, use of any collection medium other than Cytyc PreservCyt Solution would be considered an off-label use of the assay. Several studies have been published comparing the Invader® HPV assay with the Digene HC2 HPV DNA Test [20, 39–41]. Performance characteristics for the two methods are comparable in terms of clinical sensitivity. All of the three studies also concurred that the Invader® assay had improved specificity in terms of fewer false-positive results. In addition, the Cervista® HPV HR test can detect infection by multiple HPV types, which is important since recent studies have suggested that the risk of progression toward cervical cancer can be as high for simultaneous infections with multiple HPV types as it is for infection with just one of the highest risk types, HPV 16 or HPV 18. A companion test for the Cervista® HPV HR assay is the Cervista® HPV 16/18 test. HPV types 16 and 18 are recognized as highly oncogenic and persistent and are associated with 60% and 10% of cervical cancers, respectively. The test is intended to be used adjunctively with the Cervista® HPV HR test in combination with cervical cytology to assess the presence or absence of high-risk HPV types 16 and 18. In addition to the commercially produced assays described above, Invader® technology has been applied widely in the development of laboratory-developed tests for a broad range of clinical and nonclinical applications. The reason for this is the availability of Universal Invader® reagents and Universal Invader™ software for designing Invader®-based assays. Several Invader-based methods have been published for a variety of infectious disease-related applications, including detection assays for varicella zoster virus [42] and rifampin and isoniazid resistance of Mycobacterium tuberculosis [43], quantification of bacteria involved in periodontitis [44], and HBV genotyping and detection of drug-resistant mutations [45, 46]. A study published by Xie et al. describes an Invader®-based immunoassay [47]. Two different antibodies are employed in this method, one of which captures an antigen to a solid surface, and the other, which is biotinylated, is used as the detector. After removal of unbound antibody via a wash step, streptavidin and a biotinylated oligonucleotide are added to the reaction mixture. After washing away the unbound oligonucleotide, the remaining components required for the Invader assay are added. The authors report that this method could detect 0.1 pg/mL of the target antigen (tumor necrosis factor-α) versus 3.5 pg/mL for a commercially available standard enzyme immunoassay method using colorimetric detection. It is not known if Hologic will continue commercializing applications for Invader® technology. Hologic has licensed the use of Invader® technology to other in vitro diagnostic and biotechnology companies for new test development. Such an example is Exact Sciences (Madison WI), which licensed Invader® technology for the development of cancer diagnostic and screening assays. Their commercially available Cologuard® colon cancer screening test is based on Invader technology


357

[48]. In addition, Hologic has an AgBIO services division dedicated to working with the agriculture industry for custom assay development using Invader® technology, specifically in the areas of SNP genotyping analysis, copy number determination, marker-assisted selection, and animal diagnostics.

Target Amplification Technologies Isothermal Transcription-Based Amplification Isothermal transcription-based amplification (ITA) is available under several commercial names. These names include transcription-mediated amplification (TMA) and nucleic acid sequence-based amplification (NASBA). The basic concept of ITA is derived from retroviral amplification that relies on three key enzymatic reactions. Once RNA is present in the reaction, the first primer attaches to its complementary site at the 3′ end of the template. The reverse transcriptase (RT) enzyme is used to make the complementary DNA (cDNA) of the RNA target. RNAse H then degrades the initial RNA template only with RNA-DNA hybrids, but not single-stranded RNA. The second primer attaches to the 5′ end of the DNA strand and is extended to form a double-stranded DNA molecule using reverse transcriptase’s DNA- dependent DNA polymerase activity. The T7 RNA polymerase promoter is incorporated in the first primer enabling T7 RNA polymerase to subsequently produce a complementary negative sense RNA “amplicon” which can be used again in the initial step of the reaction [49]. The process repeats automatically, resulting in an exponential amplification of the original target that can produce over a billion copies of amplicon in less than 30 min. The crux of the commercial Aptima (Hologic, San Diego, California) amplification technology is TMA. Like PCR, this isothermal technology relies on target amplification with complementary oligonucleotide primers to specifically anneal and allow enzymatic amplification of the target nucleic acid strands (Fig. 4). End- point TMA detection is achieved by nucleic acid hybridization using single-stranded chemiluminescent DNA probes, which are complementary to a region of each target amplicon, labeled with different acridinium ester molecules. Through the hybridization protection assay (HPA), selection reagent is added which differentiates hybridized from unhybridized probe, eliminating the generation of signal from unhybridized probe. During the detection step, light emitted from the labeled RNA/DNA hybrids is measured as photon signals in a luminometer and is reported as relative light units (RLU). Real-time TMA (rtTMA) uses single-stranded nucleic acid torches that are present during the amplification and hybridize specifically to the amplicon in real time. Each torch has a fluorophore and a quencher, which perform similarly to a molecular beacon, emitting signal when bound to the complementary amplicon. As more torches hybridize to amplicon, a higher fluorescent signal is generated, and the time for the fluorescent signal is recorded that can be correlated to a specific threshold for detection.

358

R. C. She et al.

RT

RT

RT

copies

Amplicon

RT

Fig. 4 Transcription-mediated amplification (TMA). (Image courtesy of Gen-Probe, Incorporated)

The TMA bacterial assays replicate specific regions of ribosomal RNA. Thus the assays are already targeting a template that is in abundant quantity, making the TMA assays very robust [49]. TMA is also used to detect viral targets [50, 51]. TMA performed manually remains labor intensive and requires considerable handson time. Hologic has overcome this by automating the process. The Tigris® DTS® system and the Panther™ system are fully automated systems for TMA assays. The Panther Fusion® allows for automation of TMA, rtTMA, and real-time PCR assays. Automation is particularly useful in high-volume laboratories and laboratories looking to consolidate testing onto a single platform. For a list of available TMA assays, see Table 1. NASBA is also an isothermal reaction that uses three enzymes (avian myeloblastosis virus-RT (AMV-RT), RNase H, and T7 RNA polymerase) and target-specific oligonucleotides (Fig. 5). NASBA, like TMA, has been introduced into medical


359

Target specific molecular beacons sense RNA

RNase H oligo P1

oligo P1 Reverse Transcriptase

RT

RT

RNase H & oligo P2 Reverse Transcriptase

oligo P2

anti-sense RNA

T7 RNAP

T7 RNA polymerase Molecular beacon hybridization

Fig. 5 Real-time nucleic acid sequence-based amplification (NASBA). (Image courtesy of bioMerieux)

diagnostics, where it has been shown to have a faster turn-around time than PCR, and can also be more sensitive [52]. After its introduction, NASBA was used for quantification of HIV-1 in human sera [53]. However, the HIV-1 assay was never FDA-approved and is only available as a research use only (RUO) product from bioMerieux (Marcy l’Etoile, France). bioMerieux has marketed NASBA under the NucliSENS® name. Currently available NASBA assays use real-time-based detection with molecular beacons. TMA and NASBA are very similar technologies. The differences between them are the specific enzymes used in the reactions and their detection systems. For reverse transcription, TMA utilizes Moloney murine leukemia virus-RT (MMLV-RT), while NASBA utilizes AMV-RT [54]. For a list of the targets for which there are commercially available products, see Table 1.

Strand Displacement Amplification Strand displacement amplification (SDA) is an isothermal amplification technique that was first described in 1992 [55]. Rather than using thermocycling to facilitate repeating rounds of DNA synthesis as in PCR, SDA relies upon the activities of a restriction endonuclease and an exonuclease-deficient DNA polymerase to continually and isothermally amplify a DNA target. Inner primers are designed to have a restriction enzyme recognition sequence within their 5′ overhang. Flanking outer primers bind upstream of the inner primer targets. Extension of all four primers by DNA polymerase results in amplicons incorporating restriction enzyme

360

R. C. She et al.

a

S1

(1)

B1

b

S1

(1) S2

S1

(2)

B1

(2)

(3) B2

(3)

S2

B2

(4)

Complete BsoBI nick site Partial BsoBI nick site

S2 S1

(5) S2 S1

(6) S2

(7)

Fig. 6 Strand displacement amplification (SDA) (a, b). (Image courtesy of Beckton Dickinson)

sites (Fig. 6a). A restriction endonuclease added to the reaction creates a singlestranded nick at these sites, where DNA polymerase can then initiate replication and displace the annealed strand (Fig. 6b). The substitution of a dNTP with a modified form containing a 5′-[alpha-thio]-triphosphate, typically dATPαS, allows for incorporation of a hemiphosphorothioate linkage at the restriction site during amplification. This permits single-stranded nicking by the restriction enzyme. The DNA polymerase that is used in SDA must be able to initiate polymerization at a nick site and must lack exonuclease activity. Such polymerases include exo- Klenow, exo-Bst, and exo-Bca polymerases [55, 56]. The restriction endonuclease must be able to create a single-stranded nick at a hemiphosphorothioated recognition site, dissociate quickly to allow the polymerase to act, and repeatedly perform rounds of nicking and dissociation. Restriction enzymes which meet these criteria include HincII, BsoBI, AvaI, NciI, and Fnu4HI [55, 57]. The maximum target length is influenced by the stringency of the reaction conditions, the processivity of the DNA polymerase, and the frequency of recognition sites of the restriction enzyme surrounding the target sequence. Amplicons are generally 50–120 bp long, and SDA is not well suited for target sequences longer


361

than 1000 bp [57]. At 37 °C, exo-Klenow polymerase demonstrates decreased amplification efficiency with target lengths greater than 50 base pairs. At this temperature, the stringency of the reaction is relatively low with increased background amplification. BsoBI restriction endonuclease and exo-Bst polymerase, derived from the thermophilic bacterium Bacillus stearothermophilus, function well at relatively high temperatures (50–60 °C). This higher temperature enables increased stringency of primer hybridization and improved reaction kinetics [58]. Doubling time is approximately 30 s, and up to 1010 amplicons can be generated within 15 min [56]. In SDA, target detection may be accomplished in real time with the use of fluorescently labeled probes (Fig. 7) and, if desired, produce quantitative results [59, 60]. Use of dUTP instead of dTTP to enable contamination control with uracil-DNA glycosylase has been described [56]. Initial studies focused on the application of SDA to the detection of M. tuberculosis [56, 60, 61]. The BDProbeTec™ ET assay (Becton Dickinson, Franklin Lakes, New Jersey) uses real-time SDA to identify M. tuberculosis complex from colony growth or from respiratory specimens and has comparable performance to other amplified detection assays [62–64]. It is only commercially available outside of the USA. The BD ProbeTec™ ET system is FDA-cleared for real-time SDA-based detection of Chlamydia trachomatis and Neisseria gonorrhoeae (CT/GC) and HSV-1 and HSV-2. Newer generation systems employing SDA, the BD Viper XTR and BD Viper LT, have added sample-to-answer capability. The Viper LT is a benchtop system which accommodates both SDA- and PCR-based assays. FDA-cleared assays using SDA at this time include C. trachomatis and N. gonorrhoeae (CT/GC Qx) on both Viper XTR and Viper LT systems and additionally HSV-1 and HSV-2 and Trichomonas vaginalis on the Viper XTR. The performance of the CT/GC Qx assay is similar to other FDA-cleared amplified assays [65–67]. Of note, the CT/GC Qx is FDA-cleared for endocervical specimens collected in SurePath media for cytological screening. The HSV-1 and HSV-2 assay is FDA-cleared for testing from anogenital lesions and the T. vaginalis assay for vaginal swabs; both assays have performed comparably to other molecular methods [68–70].

Loop-Mediated Isothermal Amplification Loop-mediated isothermal amplification (LAMP) was developed by Eiken Chemical Company (Tokyo, Japan), and the method was first published in 2000 [71]. Like SDA, LAMP requires a DNA polymerase with high strand displacement activity, e.g., BstI polymerase. Also like SDA, two inner and two outer primers are used in an isothermal reaction (Fig. 8a). The inner primers incorporate a sequence in their 5′ overhang that is complementary to the region that will be amplified just downstream of the primer. After the outer primers anneal upstream to the inner primers and initiate strand displacement, the complementary regions of the displaced strand can then anneal and form a loop structure at its 3′ end (Fig. 8b). Subsequent amplification may occur as self-primed DNA synthesis off of the 3′ end of the loop or as

362

R. C. She et al.

SDA Primer

Linear Detector Probe (1)

(2)

Continued Exponential SDA

(3)

(4)

(5)

Complete BsoBI nick site Partial BsoBI nick site Complete BsoBI cleavage Partial BsoBI cleavage site Quencher Fluorophore

Fig. 7 Real-time strand displacement amplification (SDA). (Image courtesy of Beckton Dickinson)


363

Fig. 8 Loop-mediated isothermal amplification (LAMP) (a, b, c). (Image courtesy of Eiken Chemical Co.)

primers anneal to the single-stranded area of the loop. A variety of stem-loop structures of varying sizes are formed in the process (Fig. 8c). Amplicons may be detected by measuring change in turbidity from the formation of magnesium pyrophosphate salt [72]. This change is turbidity can also be directly visualized. Other methods of detection, such as fluorescent probes and bioluminescence, have been described [73, 74]. While detection of precipitated magnesium phosphate is not specific for the target of interest, specificity is achieved by the use of four primers targeting six regions of DNA during the reaction process.

364

R. C. She et al.

Reverse-transcription LAMP has also been developed for detection of RNA targets and can be accomplished in a one-step, isothermal reaction with the use of AMV-RT [75]. Use of dUTP instead of dTTP to enable contamination control with uracil-DNA glycosylase has also been shown to be feasible [76]. One study demonstrated that LAMP is less inhibited by urine and stool than PCR, suggesting that purification of nucleic acids may not be necessary prior to LAMP reactions on certain human specimens [77]. The LAMP technology has been licensed by Meridian Biosciences, Inc. (Cincinnati, Ohio). The first FDA-cleared test utilizing LAMP for clinical diagnostics was the illumigene® Clostridium difficile assay. Specimens undergo a heat lysis rather than purified nucleic acid extraction, thus decreasing the complexity and turn-around time of the assay. Inhibition control is assayed in a separate reaction tube. The reactions occur at 60–65 °C and can be completed in less than 1 h within a closed system. Detection of product is by turbidimetric readings on a luminometer. According to the manufacturers’ data, the analytical sensitivity of this assay (4–64 colony-forming units/reactions) is roughly the same as for other FDA-cleared C. difficile assays that are PCR-based. Studies have shown this LAMP-based method to have comparable performance compared to PCR-based assays using clinical specimens [78–80]. Other FDA-cleared tests that are now available on the illumigene® LAMP platform include Group B Streptococcus for prenatal screening, Group A Streptococcus from throat swabs, Mycoplasma pneumoniae from throat and nasopharyngeal swabs, Bordetella pertussis from nasopharyngeal swabs, and HSV-1 and HSV-2 from cutaneous and mucocutaneous lesions. The Group B Streptococcus assay detects organism after broth enrichment and has sensitivity similar to other commercial NAAT for this target [81, 82]. Detection of group A streptococci from throat swabs by the illumigene® has been reported to be sensitive by most studies, with a range of sensitivity of 87–100% [83–85]. The illumigene® LAMP assay for M. pneumoniae has limited performance data available, with one study reporting a sensitivity of 100% out of 22 positive specimens. This study included non-FDA-cleared respiratory specimen types such as bronchoalveolar lavage fluid and sputum [86]. The HSV-1 and HSV-2 assay has been reported to perform well compared to ELVIS® HSV culture test system [87]. Published data are limited for the B. pertussis assay. Although not FDA-cleared, the illumigene® malaria and malaria Plus assays have reportedly high sensitivity and negative predictive value which would be useful in ruling out malaria in low prevalence populations [88]. The simple format could be easily adopted for use in field laboratories in endemic areas [89]. Currently the illumigene® assay can test up to ten samples per instrument, and the pipetting steps are manually performed. This format would seem to be most beneficial for small- to medium-sized laboratories that desire to perform molecular testing using a simple protocol with minimal investment in capital equipment. Laboratory-developed LAMP-based tests have also been validated for detection of a variety of pathogens from clinical specimens and may be particularly useful in resource-poor settings [90–93].


365

Helicase-Dependent Amplification Helicase-dependent amplification (HDA) was developed by BioHelix (Beverly, Massachusetts) and first described in 2004 [94]. Unlike PCR which relies upon heat to denature double-stranded DNA, HDA relies on helicase activity to separate double-stranded DNA or RNA-DNA hybrid strands isothermally. Similar to PCR, oligonucleotide primers flanking a target region result in an amplicon of expected base pair length after extension by DNA polymerase (Fig. 9a). The ease of primer design provides an advantage over other isothermal methods already discussed. Use of a thermophilic helicase, such as Tte-UvrD, allows the reaction to occur at a higher temperature. This in turn results in greater assay specificity and sensitivity and also may obviate the need for accessory proteins in the reaction [95]. A variety of strand-displacing DNA polymerases may be used and have included Bst exo- polymerase, exo-Klenow fragment, and GST large fragment polymerase [94–96]. HDA is compatible with many methods of amplicon detection. Colorimetric detection, intercalating DNA dyes, hybrid capture with fluorescence end-point detection, and real-time detection by TaqMan probes have all been performed successfully [96–99]. One-step reverse-transcription HDA has also been proven to be feasible, highly efficient, and amenable to real-time detection of human pathogens [98]. In 2011, the FDA cleared an HDA assay developed by BioHelix for the detection of HSV-1 and HSV-2 from oral and genital lesions [100]. Quidel acquired Biohelix in 2013 and has made commercially available several HDA-based assays under the AmpliVue and Solana line of products. The AmpliVue system requires several manual steps. FDA-cleared assays on the AmpliVue are Bordetella for nasopharyngeal swab specimens, C. difficile and GBS for prenatal screening, and HSV-1 and HSV-2 for cutaneous and mucocutaneous lesions. The basic steps involve an initial specimen preparation followed by 45–60 min incubation at 64 °C. The HDA reaction is designed to occur asymmetrically and generate excess single-stranded biotinylated products that are hybridized to capture probes tagged with a specific label, e.g., 6-carboxyfluorescein (FAM) or dinitrophenyl (DNP). This differential labeling enables detection of multiple targets including an internal control. HDA product is then applied to a cassette with a vertical-flow test strip where the biotinylated products bind to streptavidin-labeled color beads. Control and test lines are coated with antibodies to the molecular label of the capture probes, and the resulting lines are read visually (Fig. 9b). It is suitable for smallvolume laboratories because it is simple, self-contained, low throughput, and low cost in terms of capital equipment. This format has been shown to have comparable performance to PCR and ELVIS culture systems for detection of HSV from clinical specimens [101, 102]. For C. difficile, the AmpliVue HDA method is performed similarly to other molecular methods [103]. On the Solana platform, specimens are loaded onto an instrument where HDA and detection occur, and results are electronically displayed after about 25 min. Up to 12 specimens can be run at once. Detection of target depends on emission of fluorescence from probes tagged with a fluorophore and quencher. An internal control

366

R. C. She et al.

P2

P1

Fig. 9 (a) Helicase-dependent amplification (HDA): Step 1, Helicase enzyme unwinds the DNA at 64 °C. Step 2, Biotin-labeled primer and polymerase extend the copy. Step 3, Replication completed. Helicase replicates procedure again. FITC-labeled capture probe attaches to biotin-labeled DNA. (b) HDA with chromatographic detection. (Image courtesy of Quidel Corp)


Fig. 9 (continued)

367

368

R. C. She et al.

is included. At this time, FDA-cleared assays include the Solana C. difficile, Group A Streptococcus from throat swabs, Strep Complete (Group C/G in addition to Group A Streptococcus) from throat swabs, HSV-1 and HSV-2/VZV from cutaneous or mucocutaneous lesion, Trichomonas vaginalis from vaginal swabs and female urine, influenza A + B from nasopharyngeal swabs, and RSV + hMPV from nasopharyngeal swabs. The Trichomonas assay performed comparably with wet preparation, culture, and TMA [104]. The Group A Streptococcus assay performed with high sensitivity compared to culture [105]. The Strep Complete assay appears to also perform well compared to culture, but published studies are limited [106]. At this time, published studies are not yet available for RSV + hMPV, influenza, or C. difficile on the Solana platform. Great Basin Scientific uses HDA in their portrait platform which is FDA-cleared for C. difficile and Group B Streptococcus [107]. Laboratory-developed HDA tests have been successfully developed and shown to be a feasible low-tech alternative that may be used in resource-limited settings [108].

Conclusions In recent years, the progressive expansion of the isothermal amplification test menu in the clinical microbiology marketplace has been seen. Non-PCR amplification technologies that are currently commercially available for infectious disease testing offer numerous advantages over PCR. These non-PCR molecular assays also have become more user-friendly for smaller laboratories. Isothermal amplification techniques offer flexibility in terms of inexpensive capital equipment, rapid turnaround time, and competitive pricing against PCR-based assays. Laboratories now have more FDA-cleared options to choose from, and this menu promises to expand with time. Automation has also enhanced the process of integrating these assays into the larger clinical laboratory. Contained, automated systems obviate the need for separate workspaces for individual steps in performing the assay. However, for quantitative nucleic acid testing, real-time PCR is still the best-studied and most highly utilized method. Competition among quantitative assay platforms may soon increase as more non-PCR real-time assays are introduced. As with any technology introduced for patient testing, each laboratory will have to consider its patient population, test volume, and resources to decide which molecular platform best fits its needs.

References 1. Mullis K. Dancing naked in the mind field. 1st ed. New York: Vintage Books; 2000. 2. Pendrak ML, Yan SS. Non-polymerase chain reaction mediated target amplification techniques. In: Advanced techniques in diagnostic microbiology. 1st ed. New York: Springer; 2006.


369

3. Colby C, Stollar BD, Simon MI. Interferon induction: DNA-RNA hybrid or double stranded RNA? Nat New Biol. 1971;229(6):172–4. 4. Rudkin GT, Stollar BD. High resolution detection of DNA-RNA hybrids in situ by indirect immunofluorescence. Nature. 1977;265(5593):472–3. 5. Stollar BD. Doubls-helical polynucleotides: immunochemical recognition of differing conformations. Science. 1970;169(3945):609–11. 6. Boguslawski SJ, Smith DE, Michalak MA, Mickelson KE, Yehle CO, Patterson WL, et al. Characterization of monoclonal antibody to DNA.RNA and its application to immunodetection of hybrids. J Immunol Methods. 1986;89(1):123–30. 7. Carpenter WR, Schutzbank TE, Tevere VJ, Tocyloski KR, Dattagupta N, Yeung KK. A transcriptionally amplified DNA probe assay with ligatable probes and immunochemical detection. Clin Chem. 1993;39(9):1934–8. 8. Schiffman MH, Kiviat NB, Burk RD, Shah KV, Daniel RW, Lewis R, et al. Accuracy and interlaboratory reliability of human papillomavirus DNA testing by hybrid capture. J Clin Microbiol. 1995;33(3):545–50. 9. Beck S, Koster H. Applications of dioxetane chemiluminescent probes to molecular biology. Anal Chem. 1990;62(21):2258–70. 10. Munger K, Baldwin A, Edwards KM, Hayakawa H, Nguyen CL, Owens M, et al. Mechanisms of human papillomavirus-induced oncogenesis. J Virol. 2004;78(21):11451–60. https://doi. org/10.1128/JVI.78.21.-60.2004. 11. Snijders PJ, Steenbergen RD, Heideman DA, Meijer CJ. HPV-mediated cervical carcinogenesis: concepts and clinical implications. J Pathol. 2006;208(2):152–64. https://doi. org/10.1002/path.866. 12. zur Hausen H. Condylomata acuminata and human genital cancer. Cancer Res. 1976;36 (2 pt 2):794. 13. zur Hausen H. Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer. 2002;2(5):342–50. https://doi.org/10.1038/nrc798. 14. Schiffman MH. Epidemiology of cervical human papillomaviruses. In: zur Hausen H, editor. Human pathogenic papillomaviruses. Heidelberg: Springer; 1994. 15. Munoz N, Bosch FX, de Sanjose S, Herrero R, Castellsague X, Shah KV, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med. 2003;348(6):518–27. https://doi.org/10.1056/NEJMoa021641. 16. Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam SL, Cain J, et al. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer. CA Cancer J Clin. 2012;62(3):147–72. https://doi.org/10.3322/caac.21139. Epub 2012 Mar 14. 17. Arney A, Bennett K. Molecular diagnostics of human papillomavirus. Lab Med. 2010;41:523–30. 18. Burd EM. Human papillomavirus and cervical cancer. Clin Microbiol Rev. 2003;16(1):1–17. 19. Castle PE, Solomon D, Wheeler CM, Gravitt PE, Wacholder S, Schiffman M. Human papillomavirus genotype specificity of hybrid capture 2. J Clin Microbiol. 2008;46(8):2595–604. https://doi.org/10.1128/JCM.00824-08. Epub 2008 Jun 25. 20. Schutzbank TE, Jarvis C, Kahmann N, Lopez K, Weimer M, Yount A. Detection of high-risk papillomavirus DNA with commercial invader-technology-based analyte-specific reagents following automated extraction of DNA from cervical brushings in ThinPrep media. J Clin Microbiol. 2007;45(12):4067–9. https://doi.org/10.1128/JCM.01833-07. Epub 2007 Oct 24. 21. Munson E, Du Chateau BK, Bellerose B, Czarnecka J, Griep J. Clinical laboratory evaluation of invader(R) chemistry and hybrid capture for detection of high-risk human papillomavirus in liquid-based cytology specimens. Diagn Microbiol Infect Dis. 2011;71(3):230–5. https:// doi.org/10.1016/j.diagmicrobio.2011.07.004. Epub Sep 6. 22. Schachter J, Hook EW 3rd, McCormack WM, Quinn TC, Chernesky M, Chong S, et al. Ability of the digene hybrid capture II test to identify chlamydia trachomatis and Neisseria gonorrhoeae in cervical specimens. J Clin Microbiol. 1999;37(11):3668–71.

370

R. C. She et al.

23. Darwin LH, Cullen AP, Crowe SR, Modarress KJ, Willis DE, Payne WJ. Evaluation of the hybrid capture 2 CT/GC DNA tests and the GenProbe PACE 2 tests from the same male urethral swab specimens. Sex Transm Dis. 2002;29(10):576–80. 24. Quint K, Porras C, Safaeian M, Gonzalez P, Hildesheim A, Quint W, et al. Evaluation of a novel PCR-based assay for detection and identification of chlamydia trachomatis serovars in cervical specimens. J Clin Microbiol. 2007;45(12):3986–91. https://doi.org/10.1128/ JCM.01155-07. Epub 2007 Oct 24. 25. Mazzulli T, Drew LW, Yen-Lieberman B, Jekic-McMullen D, Kohn DJ, Isada C, et al. Multicenter comparison of the digene hybrid capture CMV DNA assay (version 2.0), the pp65 antigenemia assay, and cell culture for detection of cytomegalovirus viremia. J Clin Microbiol. 1999;37(4):958–63. 26. Walmsley S, O’Rourke K, Mortimer C, Rachlis A, Fong I, Mazzulli T. Predictive value of cytomegalovirus (CMV) antigenemia and digene hybrid capture DNA assays for CMV disease in human immunodeficiency virus-infected patients. Clin Infect Dis. 1998;27(3):573–81. 27. Hanson KE, Reller LB, Kurtzberg J, Horwitz M, Long G, Alexander BD. Comparison of the Digene Hybrid Capture System Cytomegalovirus (CMV) DNA (version 2.0), Roche CMV UL54 analyte-specific reagent, and QIAGEN RealArt CMV LightCycler PCR reagent tests using AcroMetrix OptiQuant CMV DNA quantification panels and specimens from allogeneic-stem-cell transplant recipients. J Clin Microbiol. 2007;45(6):1972–3. https://doi. org/10.1128/JCM.02515-06. Epub 2007 Feb 21. 28. Yuan HJ, Yuen MF, Wong DK, Sum SS, Lai CL. Clinical evaluation of the digene hybrid capture II test and the COBAS AMPLICOR monitor test for determination of hepatitis B virus DNA levels. J Clin Microbiol. 2004;42(8):3513–7. https://doi.org/10.1128/ JCM.42.8.3513-7.2004. 29. Konnick EQ, Erali M, Ashwood ER, Hillyard DR. Evaluation of the COBAS amplicor HBV monitor assay and comparison with the ultrasensitive HBV hybrid capture 2 assay for quantification of hepatitis B virus DNA. J Clin Microbiol. 2005;43(2):596–603. https://doi. org/10.1128/JCM.43.2.596-603.2005. 30. Lyamichev V, Mast AL, Hall JG, Prudent JR, Kaiser MW, Takova T, et al. Polymorphism identification and quantitative detection of genomic DNA by invasive cleavage of oligonucleotide probes. Nat Biotechnol. 1999;17(3):292–6. https://doi.org/10.1038/7044. 31. Lyamichev VI, Kaiser MW, Lyamicheva NE, Vologodskii AV, Hall JG, Ma WP, et al. Experimental and theoretical analysis of the invasive signal amplification reaction. Biochemistry. 2000;39(31):9523–32. 32. Stryer L. Fluorescence energy transfer as a spectroscopic ruler. Annu Rev Biochem. 1978;47:819–46. https://doi.org/10.1146/annurev.bi.47.070178.04131. 33. Kwiatkowski RW, Lyamichev V, de Arruda M, Neri B. Clinical, genetic, and pharmacogenetic applications of the Invader assay. Mol Diagn. 1999;4(4):353–64. https://doi.org/10.154/ MODI00400353. 34. Hall JG, Eis PS, Law SM, Reynaldo LP, Prudent JR, Marshall DJ, et al. Sensitive detection of DNA polymorphisms by the serial invasive signal amplification reaction. Proc Natl Acad Sci U S A. 2000;97(15):8272–7. https://doi.org/10.1073/pnas.140225597. 35. Ledford M, Friedman KD, Hessner MJ, Moehlenkamp C, Williams TM, Larson RS. A multi- site study for detection of the factor V (Leiden) mutation from genomic DNA using a homogeneous invader microtiter plate fluorescence resonance energy transfer (FRET) assay. J Mol Diagn. 2000;2(2):97–104. https://doi.org/10.1016/S525-578(10)60623-X. 36. Ando Y, Saka H, Ando M, Sawa T, Muro K, Ueoka H, et al. Polymorphisms of UDP- glucuronosyltransferase gene and irinotecan toxicity: a pharmacogenetic analysis. Cancer Res. 2000;60(24):6921–6. 37. Innocenti F, Undevia SD, Iyer L, Chen PX, Das S, Kocherginsky M, et al. Genetic variants in the UDP-glucuronosyltransferase 1A1 gene predict the risk of severe neutropenia of irinotecan. J Clin Oncol. 2004;22(8):1382–8. https://doi.org/10.1200/JCO.2004.07.173. Epub Mar 8.


371

38. Harvey M, Stout S, Starkey CR, Hendren R, Holt S, Miller GC. The clinical performance of invader technology and SurePath when detecting the presence of high-risk HPV cervical infection. J Clin Virol. 2009;45(Suppl 1):S79–83. https://doi.org/10.1016/ S386-6532(09)70012-5. 39. Ginocchio CC, Barth D, Zhang F. Comparison of the Third Wave Invader human papillomavirus (HPV) assay and the digene HPV hybrid capture 2 assay for detection of high-risk HPV DNA. J Clin Microbiol. 2008;46(5):1641–6. https://doi.org/10.1128/JCM.01824-07. Epub 2008 Mar 26. 40. Johnson LR, Starkey CR, Palmer J, Taylor J, Stout S, Holt S, et al. A comparison of two methods to determine the presence of high-risk HPV cervical infections. Am J Clin Pathol. 2008;130(3):401–8. https://doi.org/10.1309/4DXEAFG2JXYF34N3. 41. Wong AK, Chan RC, Nichols WS, Bose S. Human papillomavirus (HPV) in atypical squamous cervical cytology: the invader HPV test as a new screening assay. J Clin Microbiol. 2008;46(3):869–75. https://doi.org/10.1128/JCM.01424-07. Epub 2008 Jan 3. 42. Tang YW, Allawi HT, DeLeon-Carnes M, Li H, Day SP, Schmid D. Detection and differentiation of wild-type and vaccine mutant varicella-zoster viruses using an invader Plus method. J Clin Virol. 2007;40(2):129–34. https://doi.org/10.1016/j.jcv.2007.07.007. Epub Aug 28. 43. Cooksey RC, Holloway BP, Oldenburg MC, Listenbee S, Miller CW. Evaluation of the invader assay, a linear signal amplification method, for identification of mutations associated with resistance to rifampin and isoniazid in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 2000;44(5):1296–301. 44. Tadokoro K, Yamaguchi T, Kawamura K, Shimizu H, Egashira T, Minabe M, et al. Rapid quantification of periodontitis-related bacteria using a novel modification of invader PLUS technologies. Microbiol Res. 2010;165(1):43–9. https://doi.org/10.1016/j.micres.2008.06.001. Epub Aug 20. 45. Tadokoro K, Kobayashi M, Yamaguchi T, Suzuki F, Miyauchi S, Egashira T, et al. Classification of hepatitis B virus genotypes by the PCR-invader method with genotype-specific probes. J Virol Methods. 2006;138(1–2):30–9. https://doi.org/10.1016/j.jviromet.2006.07.014. Epub Aug 24. 46. Tadokoro K, Suzuki F, Kobayashi M, Yamaguchi T, Nagano M, Egashira T, et al. Rapid detection of drug-resistant mutations in hepatitis B virus by the PCR-invader assay. J Virol Methods. 2011;171(1):67–73. https://doi.org/10.1016/j.jviromet.2010.10.001. Epub Oct 13. 47. Xie MJ, Fukui K, Horie M, Sakihama Y, Hashino K, Kimura H, et al. A novel sensitive immunoassay method based on the invader technique. Anal Biochem. 2008;374(2):278–84. https:// doi.org/10.1016/j.ab.2007.11.031. Epub Nov 29. 48. Zou H, Allawi H, Cao X, Domanico M, Harrington J, Taylor WR, et al. Quantification of methylated markers with a multiplex methylation-specific technology. Clin Chem. 2012;58(2):375–83. https://doi.org/10.1373/clinchem.2011.171264. Epub 2011 Dec 22. 49. Hill CS. Molecular diagnostic testing for infectious diseases using TMA technology. Expert Rev Mol Diagn. 2001;1(4):445–55. 50. McAuley JD, Caglioti S, Williams RC, Robertson GF, Morgan L, Tobler LH, et al. Clinical significance of nondiscriminated reactive results with the Chiron Procleix HIV-1 and HCV assay. Transfusion. 2004;44(1):91–6. 51. Linnen JM, Deras ML, Cline J, Wu W, Broulik AS, Cory RE, et al. Performance evaluation of the PROCLEIX West Nile virus assay on semi-automated and automated systems. J Med Virol. 2007;79(9):1422–30. 52. Schneider P, Wolters L, Schoone G, Schallig H, Sillekens P, Hermsen R, et al. Real-time nucleic acid sequence-based amplification is more convenient than real-time PCR for quantification of Plasmodium falciparum. J Clin Microbiol. 2005;43(1):402–5. 53. Kievits T, van Gemen B, van Strijp D, Schukkink R, Dircks M, Adriaanse H, et al. NASBA isothermal enzymatic in vitro nucleic acid amplification optimized for the diagnosis of HIV-1 infection. J Virol Methods. 1991;35(3):273–86. 54. Compton J. Nucleic acid sequence-based amplification. Nature. 1991;350(6313):91–2. 55. Walker GT, Little MC, Nadeau JG, Shank DD. Isothermal in vitro amplification of DNA by a restriction enzyme/DNA polymerase system. Proc Natl Acad Sci U S A. 1992;89(1):392–6.

372

R. C. She et al.

56. Spargo CA, Fraiser MS, Van Cleve M, Wright DJ, Nycz CM, Spears PA, et al. Detection of M. Tuberculosis DNA using thermophilic strand displacement amplification. Mol Cell Probes. 1996;10(4):247–56. 57. Hellyer TJ, Nadeau JG. Strand displacement amplification: a versatile tool for molecular diagnostics. Expert Rev Mol Diagn. 2004;4(2):251–61. 58. Little MC, Andrews J, Moore R, Bustos S, Jones L, Embres C, et al. Strand displacement amplification and homogeneous real-time detection incorporated in a second-generation DNA probe system, BDProbeTecET. Clin Chem. 1999;45(6 Pt 1):777–84. 59. Nadeau JG, Pitner JB, Linn CP, Schram JL, Dean CH, Nycz CM. Real-time, sequence- specific detection of nucleic acids during strand displacement amplification. Anal Biochem. 1999;276(2):177–87. 60. Walker GT, Linn CP. Detection of Mycobacterium tuberculosis DNA with thermophilic strand displacement amplification and fluorescence polarization. Clin Chem. 1996;42(10):1604–8. 61. Walker GT, Nadeau JG, Spears PA, Schram JL, Nycz CM, Shank DD. Multiplex strand displacement amplification (SDA) and detection of DNA sequences from Mycobacterium tuberculosis and other mycobacteria. Nucleic Acids Res. 1994;22(13):2670–7. 62. Mazzarelli G, Rindi L, Piccoli P, Scarparo C, Garzelli C, Tortoli E. Evaluation of the BDProbeTec ET system for direct detection of Mycobacterium tuberculosis in pulmonary and extrapulmonary samples: a multicenter study. J Clin Microbiol. 2003;41(4):1779–82. 63. Iinuma Y, Senda K, Fujihara N, Saito T, Takakura S, Shimojima M, et al. Comparison of the BDProbeTec ET system with the Cobas Amplicor PCR for direct detection of Mycobacterium tuberculosis in respiratory samples. Eur J Clin Microbiol Infect Dis. 2003;22(6):368–71. Epub 2003 May 16. 64. Goessens WH, de Man P, Koeleman JG, Luijendijk A, te Witt R, Endtz HP, et al. Comparison of the COBAS AMPLICOR MTB and BDProbeTec ET assays for detection of Mycobacterium tuberculosis in respiratory specimens. J Clin Microbiol. 2005;43(6):2563–6. 65. Mushanski LM, Brandt K, Coffin N, Levett PN, Horsman GB, Rank EL. Comparison of the BD viper system with XTR technology to the gen-probe APTIMA COMBO 2 assay using the TIGRIS DTS system for the detection of chlamydia trachomatis and Neisseria gonorrhoeae in urine specimens. Sex Transm Dis. 2012;39(7):514–7. https://doi.org/10.1097/ OLQ.0b013e31824f2f5b. 66. Chernesky M, Jang D, Gilchrist J, Hatchette T, Poirier A, Flandin JF, et al. Head-to-head comparison of second-generation nucleic acid amplification tests for detection of chlamydia trachomatis and Neisseria gonorrhoeae on urine samples from female subjects and self-collected vaginal swabs. J Clin Microbiol. 2014;52(7):2305–10. https://doi.org/10.1128/ JCM.03552-13. Epub 2014 Apr 2. 67. Van Der Pol B, Liesenfeld O, Williams JA, Taylor SN, Lillis RA, Body BA, et al. Performance of the cobas CT/NG test compared to the Aptima AC2 and viper CTQ/GCQ assays for detection of chlamydia trachomatis and Neisseria gonorrhoeae. J Clin Microbiol. 2012;50(7):2244–9. https://doi.org/10.1128/JCM.06481-11. Epub 2012 Apr 18. 68. Van Der Pol B, Williams JA, Taylor SN, Cammarata CL, Rivers CA, Body BA, et al. Detection of trichomonas vaginalis DNA by use of self-obtained vaginal swabs with the BD ProbeTec Qx assay on the BD viper system. J Clin Microbiol. 2014;52(3):885–9. https://doi. org/10.1128/JCM.02966-13. Epub 2014 Jan 3. 69. Lang AL, Roberts C, Mazzulli T, Hatchette TF, LeBlanc JJ. Detection and differentiation of herpes simplex viruses by use of the viper platform: advantages, limitations, and concerns. J Clin Microbiol. 2014;52(6):2186–8. https://doi.org/10.1128/JCM.03636-13. Epub 2014 Apr 2. 70. Van Der Pol B, Warren T, Taylor SN, Martens M, Jerome KR, Mena L, et al. Type-specific identification of anogenital herpes simplex virus infections by use of a commercially available nucleic acid amplification test. J Clin Microbiol. 2012;50(11):3466–71. https://doi. org/10.1128/JCM.01685-12. Epub 2012 Aug 8. 71. Notomi T, Okayama H, Masubuchi H, Yonekawa T, Watanabe K, Amino N, et al. Loop-mediated isothermal amplification of DNA. Nucleic Acids Res. 2000;28(12):E63.


373

72. Mori Y, Nagamine K, Tomita N, Notomi T. Detection of loop-mediated isothermal amplification reaction by turbidity derived from magnesium pyrophosphate formation. Biochem Biophys Res Commun. 2001;289(1):150–4. 73. Aoi Y, Hosogai M, Tsuneda S. Real-time quantitative LAMP (loop-mediated isothermal amplification of DNA) as a simple method for monitoring ammonia-oxidizing bacteria. J Biotechnol. 2006;125(4):484–91. 74. Gandelman OA, Church VL, Moore CA, Kiddle G, Carne CA, Parmar S, et al. Novel bioluminescent quantitative detection of nucleic acid amplification in real-time. PLoS One. 2010;5(11):e14155. https://doi.org/10.1371/journal.pone.0014155. 75. Francois P, Tangomo M, Hibbs J, Bonetti EJ, Boehme CC, Notomi T, et al. Robustness of a loop-mediated isothermal amplification reaction for diagnostic applications. FEMS Immunol Med Microbiol. 2011;62(1):41–8. https://doi.org/10.1111/j.574-695X.2011.00785.x. Epub 2011 Mar 16. 76. He L, Xu HS, Wang MZ, Rong HN. Development of rapid detection of infectious hypodermal and hematopoietic necrosis virus by loop-mediated isothermal amplification. Bing Du Xue Bao. 2010;26(6):490–5. 77. Ren X, Li P. Development of reverse transcription loop-mediated isothermal amplification for rapid detection of porcine epidemic diarrhea virus. Virus Genes. 2011;42(2):229–35. 78. Lalande V, Barrault L, Wadel S, Eckert C, Petit JC, Barbut F. Evaluation of a loop-mediated isothermal amplification assay for diagnosis of Clostridium difficile infections. J Clin Microbiol. 2011;49(7):2714–6. 79. Noren T, Alriksson I, Andersson J, Akerlund T, Unemo M. Rapid and sensitive loop-mediated isothermal amplification test for Clostridium difficile detection challenges cytotoxin B cell test and culture as gold standard. J Clin Microbiol. 2011;49(2):710–1. 80. Lloyd A, Pasupuleti V, Thota P, Pant C, Rolston DD, Hernandez AV, et al. Accuracy of loop- mediated isothermal amplification for the diagnosis of Clostridium difficile infection: a systematic review. Diagn Microbiol Infect Dis. 2015;82(1):4–10. https://doi.org/10.1016/j. diagmicrobio.2015.02.007. Epub Feb 23. 81. Miller SA, Deak E, Humphries R. Comparison of the AmpliVue, BD Max System, and illumigene Molecular Assays for Detection of Group B Streptococcus in Antenatal Screening Specimens. J Clin Microbiol. 2015;53(6):1938–41. https://doi.org/10.1128/JCM.00261-15. Epub 2015 Mar 18. 82. Couturier BA, Weight T, Elmer H, Schlaberg R. Antepartum screening for group B Streptococcus by three FDA-cleared molecular tests and effect of shortened enrichment culture on molecular detection rates. J Clin Microbiol. 2014;52(9):3429–32. https://doi. org/10.1128/JCM.01081-14. Epub 2014 Jul 9. 83. Upton A, Bissessor L, Farrell E, Shulman ST, Zheng X, Lennon D. Comparison of illumigene group a streptococcus assay with culture of throat swabs from children with sore throats in the New Zealand school-based rheumatic fever prevention program. J Clin Microbiol. 2016;54(1):153–6. https://doi.org/10.1128/JCM.02440-15. Epub 2015 Nov 11. 84. Anderson NW, Buchan BW, Mayne D, Mortensen JE, Mackey TL, Ledeboer NA. Multicenter clinical evaluation of the illumigene group A Streptococcus DNA amplification assay for detection of group A Streptococcus from pharyngeal swabs. J Clin Microbiol. 2013;51(5):1474–7. https://doi.org/10.1128/JCM.00176-13. Epub 2013 Feb 27. 85. Henson AM, Carter D, Todd K, Shulman ST, Zheng X. Detection of Streptococcus pyogenes by use of illumigene group a streptococcus assay. J Clin Microbiol. 2013;51(12):4207–9. https://doi.org/10.1128/JCM.01892-13. Epub 2013 Sep 18. 86. Ratliff AE, Duffy LB, Waites KB. Comparison of the illumigene Mycoplasma DNA amplification assay and culture for detection of Mycoplasma pneumoniae. J Clin Microbiol. 2014;52(4):1060–3. https://doi.org/10.1128/JCM.02913-13. Epub 2014 Jan 15. 87. Faron ML, Ledeboer NA, Patel A, Beqa SH, Yen-Lieberman B, Kohn D, et al. Multicenter evaluation of meridian bioscience HSV 1&2 molecular assay for detection of herpes simplex virus 1 and 2 from clinical cutaneous and mucocutaneous specimens. J Clin Microbiol. 2016;54(8):2008–13. https://doi.org/10.1128/JCM.00483-16. Epub 2016 May 18.

374

R. C. She et al.

88. Rypien C, Chow B, Chan WW, Church DL, Pillai DR. Detection of plasmodium infection by the illumigene malaria assay compared to reference microscopy and real-time PCR. J Clin Microbiol. 2017;55(10):3037–45. https://doi.org/10.1128/JCM.00806-17. Epub 2017 Aug 2. 89. Lucchi NW, Gaye M, Diallo MA, Goldman IF, Ljolje D, Deme AB, et al. Evaluation of the illumigene Malaria LAMP: a robust molecular diagnostic tool for Malaria parasites. Sci Rep. 2016;6:36808. https://doi.org/10.1038/srep36808. 90. Piera KA, Aziz A, William T, Bell D, Gonzalez IJ, Barber BE, et al. Detection of plasmodium knowlesi, plasmodium falciparum and Plasmodium vivax using loop-mediated isothermal amplification (LAMP) in a co-endemic area in Malaysia. Malar J. 2017;16(1):29. https://doi. org/10.1186/s12936-016-1676-9. 91. Petrone BL, Wolff BJ, DeLaney AA, Diaz MH, Winchell JM. Isothermal detection of mycoplasma pneumoniae directly from respiratory clinical specimens. J Clin Microbiol. 2015;53(9):2970–6. https://doi.org/10.1128/JCM.01431-15. Epub 2015 Jul 15. 92. Reddy S, Ntoyanto S, Sakadavan Y, Reddy T, Mahomed S, Dlamini M, et al. Detecting mycobacterium tuberculosis using the loop-mediated isothermal amplification test in South Africa. Int J Tuberc Lung Dis. 2017;21(10):1154–60. https://doi.org/10.5588/ijtld.16.0863. 93. Kamachi K, Toyoizumi-Ajisaka H, Toda K, Soeung SC, Sarath S, Nareth Y, et al. Development and evaluation of a loop-mediated isothermal amplification method for rapid diagnosis of Bordetella pertussis infection. J Clin Microbiol. 2006;44(5):1899–902. 94. Vincent M, Xu Y, Kong H. Helicase-dependent isothermal DNA amplification. EMBO Rep. 2004;5(8):795–800. 95. An L, Tang W, Ranalli TA, Kim HJ, Wytiaz J, Kong H. Characterization of a thermostable UvrD helicase and its participation in helicase-dependent amplification. J Biol Chem. 2005;280(32):28952–8. 96. Doseeva V, Forbes T, Wolff J, Khripin Y, O'Neil D, Rothmann T, et al. Multiplex isothermal helicase-dependent amplification assay for detection of chlamydia trachomatis and Neisseria gonorrhoeae. Diagn Microbiol Infect Dis. 2011;71(4):354–65. https://doi.org/10.1016/j.diagmicrobio.2011.08.021. Epub Oct 14. 97. Gill P, Alvandi AH, Abdul-Tehrani H, Sadeghizadeh M. Colorimetric detection of helicobacter pylori DNA using isothermal helicase-dependent amplification and gold nanoparticle probes. Diagn Microbiol Infect Dis. 2008;62(2):119–24. 98. Goldmeyer J, Kong H, Tang W. Development of a novel one-tube isothermal reverse transcription thermophilic helicase-dependent amplification platform for rapid RNA detection. J Mol Diagn. 2007;9(5):639–44. 99. Tong Y, Tang W, Kim HJ, Pan X, Ranalli T, Kong H. Development of isothermal TaqMan assays for detection of biothreat organisms. BioTechniques. 2008;45(5):543–57. 100. Lemieux B, Li Y, Kong H, Tang YW. Near instrument-free, simple molecular device for rapid detection of herpes simplex viruses. Expert Rev Mol Diagn. 2012;12(5):437–43. https://doi. org/10.1586/erm.12.34. 101. Kim HJ, Tong Y, Tang W, Quimson L, Cope VA, Pan X, et al. A rapid and simple isothermal nucleic acid amplification test for detection of herpes simplex virus types 1 and 2. J Clin Virol. 2011;50(1):26–30. https://doi.org/10.1016/j.jcv.2010.09.006. Epub Oct 14. 102. Granato PA, Alkins BR, Yen-Lieberman B, Greene WH, Connolly J, Buchan BW, et al. Comparative evaluation of AmpliVue HSV 1+2 assay with ELVIS culture for detecting herpes simplex virus 1 (HSV-1) and HSV-2 in clinical specimens. J Clin Microbiol. 2015;53(12):3922–5. https://doi.org/10.1128/JCM.01905-15. Epub 2015 Oct 14. 103. Paitan Y, Miller-Roll T, Adler A. Comparative performance study of six commercial molecular assays for rapid detection of toxigenic Clostridium difficile. Clin Microbiol Infect. 2017;23(8):567–72. https://doi.org/10.1016/j.cmi.2017.02.016. Epub Feb 20. 104. Gaydos CA, Schwebke J, Dombrowski J, Marrazzo J, Coleman J, Silver B, et al. Clinical performance of the Solana(R) point-of-care trichomonas assay from clinician-collected vaginal swabs and urine specimens from symptomatic and asymptomatic women. Expert Rev Mol Diagn. 2017;17(3):303–6. https://doi.org/10.1080/14737159.2017.1282823. Epub 2017 Jan 29.


375

105. Uphoff TS, Buchan BW, Ledeboer NA, Granato PA, Daly JA, Marti TN. Multicenter evaluation of the Solana Group a Streptococcus assay: comparison with culture. J Clin Microbiol. 2016;54(9):2388–90. https://doi.org/10.1128/JCM.01268-16. Epub 2016 Jun 29. 106. Pierce K, Hopper A, Holt S, Blaschke A, Ampofo K, Korgenski K, et al. Evaluation of the Solana Strep complete assay for the detection of groups A, C and G strep from pediatric throat swabs. J Mol Diagn. 2016;18(6):977–8. 107. Denys GA. Portrait toxigenic Clostridium difficile assay, an isothermal amplification assay detects toxigenic C. difficile in clinical stool specimens. Expert Rev Mol Diagn. 2014;14(1):17–26. https://doi.org/10.1586/14737159.2014.864239. Epub 2013 Nov 28. 108. Huang S, Do J, Mahalanabis M, Fan A, Zhao L, Jepeal L, et al. Low cost extraction and isothermal amplification of DNA for infectious diarrhea diagnosis. PLoS One. 2013;8(3):e60059. https://doi.org/10.1371/journal.pone.0060059. Epub 2013 Mar 28.

Real-Time and Digital PCR for Nucleic Acid Quantification Alexander J. McAdam

Introduction It is difficult to overstate the role of polymerase chain reaction (PCR) and its derivatives in clinical microbiology. The ease of adapting PCR, combined with the availability of complete genomic sequences for most pathogens, has permitted the rapid development of commercial as well as laboratory-developed molecular techniques for diagnostic purposes. As its name implies, real-time PCR, a common PCR variant, has furthered clinical utility by permitting the detection and quantification of PCR products during the amplification process. In real-time PCR, the presence of amplified target DNA is monitored continuously through the detection of a fluorescent signal that increases in intensity with the concentration of amplified DNA. Digital PCR (dPCR) is an alternative method for quantitative detection of DNA [1]. In dPCR, the DNA and reagents needed for amplification and detection of target DNA are separated into a large number of partitions, amplification occurs and the number of DNA templates is determined by scoring each partition as positive or negative for amplified DNA. Real-time PCR and dPCR each have advantages and disadvantages; however, the former is much more widely used than the latter. In this chapter, pathogen detection and quantitative measurement using real-time PCR will be discussed in more detail, as it is the more common method, while dPCR will be briefly discussed.

A. J. McAdam (*) Infectious Disease Diagnostics Laboratory, Department of Laboratory Medicine, Boston Children’s Hospital and Harvard Medical School, Children’s Hospital Boston, Boston, MA, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_18

377

378

A. J. McAdam

Quantitative Real-Time PCR If one could monitor a PCR reaction during every cycle of amplification, the quantity of amplified DNA (product) produced would approximate a logistic function, which follows a sigmoidal curve (Fig. 1). After an initial lag phase, the rate of product formation is nearly exponential and depends primarily on the starting template concentration. Under ideal conditions, the product is initially doubled after every cycle, until the reaction is inhibited by excessive product and/or limiting PCR reactants. Prior to this inhibition, the exponential phase of the PCR reaction may be described by the following equation: P = T (1 + E )

n

where P is the PCR product concentration, T is the initial template concentration, E is the efficiency of the PCR reaction, and n is the number of amplification cycles. The efficiency E ranges between 0 and 1.0, with 1.0 (100% efficiency) representing an ideal amplification reaction where the target molecule is doubled after every reaction cycle. Under such conditions, it would take 3.32 amplification cycles to increase the target PCR product tenfold (Fig. 2a). Plotting the relationship between the number of amplification cycles necessary to reach a specified concentration of product (as determined by a set level of fluorescence) and the log of the starting concentration of the nucleic acid template gives a negative linear relationship (Fig. 2b). In theory, the product threshold set point can be any amount during the exponential phase of the reaction; however, this threshold should be set at the lowest reliably detectable level of fluorescence above background, since this is where the

Product (arbitrary units)

1400 1200 1000 800 600 400 200 0

0

0

2

4

4

8

6

12

8

16

10

20

12

24

Cycle Number Fig. 1 A hypothetical PCR amplification curve represents a logistic function and follows a sigmoidal curve (solid red line). A purely exponential amplification is also shown for comparison (dashed blue line). Notice that during early cycles, the two curves are nearly identical

Real-Time and Digital PCR for Nucleic Acid Quantification

379

a 19.52 22.74 26.05 29.67 33.1336.25

800

Fluorescence

600 400 200 0

30.0

10

b

20 Cycles

30

40

Log Concentration

10 8 6 4 2 0

20

40

30 Threshold Cycle

Fig. 2 Example of a real-time PCR standard curve. (a) Serial tenfold dilutions (ranging from 5 to 5 ×105 copies per reaction) of a plasmid bearing the hexon gene for adenovirus were used as template for detection in real-time PCR. Numbers on top indicate the threshold cycle (Ct) for each dilution. (b) Plot of the threshold cycle versus log template concentration demonstrates an inverse linear curve

reaction behaves most likely to the exponential ideal (Fig. 1). The latter is defined as the crossing threshold (Ct) of the reaction. For any given PCR reaction, plotting Ct values of a serial dilution of template nucleic acid serves two purposes. First, one can empirically determine the slope of the Ct versus log product curve, which may be used to determine the percent efficiency of the PCR reaction:

(

)

E = 10 − slope – 1 × 100

Suboptimal PCR conditions require, on average, more than one cycle to double the product, thus flattening the slope of the curve. As a general rule, a high PCR efficiency is desirable, primarily because the lower limit of detection will be otherwise compromised. The second purpose in plotting such a curve is to create calibration or standard curves, which are used to calculate the concentration of the

380

A. J. McAdam

target sequence in any given sample. The Ct value determined from the unknown sample may be applied using linear regression methods to the standard curve to determine the concentration of the target sequence. The use of such a method assumes that the amplification reaction in the unknown sample behaved identically as the amplification reactions used to create the standard curve. The presence of PCR inhibitors in a sample would, for example, negate this assumption and result in underestimation of the target DNA concentration. PCR inhibitors will be considered later in this chapter.

Dynamic Range and Detection Limits In theory, the upper limit of quantification in real-time PCR exceeds 1010 copies per reaction, and the lower limit of detection is one molecule of template. Generally, the former is true in practice but is seldom approached in the clinical laboratory due to a lack of need for such a large dynamic range. The lower limit of detection is far more critical, especially for analysis of viral loads such as for human immunodeficiency virus, and cannot reliably reach one copy of template per reaction. Because of Poisson’s law of distribution, it is impossible to be certain that one copy of template is present in every given reaction at such low concentrations. For clinical purposes, the lower limit of detection is usually defined as the lowest quantity of analyte (nucleic acid in this case) with a 95% probability of detection. For PCR, this is theoretically possible with three copies of template. In reality, this is not typically achievable due to losses during sample extraction and preparation, as well as the imperfect efficiency of the reverse transcription step if RT-PCR is being performed for RNA quantification. In our laboratory, we have been successful in the validation of assays that detect as few as five to ten copies of template per reaction, in agreement with other published validations [2–4]. In cases where it is desirable to further improve the lower limit of detection, clinical samples may be concentrated during the extraction step and the appropriate correction factor be implemented for quantification purposes.

Quantification Methodologies Differences in the techniques will be discussed as they are relevant to quantitative testing here. In the vast majority of cases, a fluorescent marker that detects amplified product is included in the reaction, and the real-time PCR machine contains a fluorimeter calibrated to excite and detect the marker. Some years after the discovery of PCR, the addition of ethidium bromide (EtBr) to the PCR reaction was described as a method to detect and quantify the amplification product in real time. EtBr is an intercalating agent that preferentially binds to double-stranded DNA and has since been replaced with the minor groove-binding dye SYBR Green I. SYBR Green I


381

and its derivatives distinguish double-stranded DNA from single-stranded DNA better than EtBr and are widely used in research laboratories for quantitative PCR and RT-PCR. The use of DNA-binding dyes in real-time PCR in the clinical laboratory is uncommon, in part because of concerns about the accuracy of quantitative results obtained with these dyes. At higher concentrations, SYBR Green I can inhibit the PCR reaction, resulting in inaccurate quantitative results [5], although this can be overcome by using a lower concentration of the dye [6]. A greater concern is that the binding (and hence detection and quantification) of such dyes is not sequence specific, and thus specific and nonspecific amplification products, as well as primer dimers, will all contribute to detection of reaction signal. If such an approach is used clinically, it is imperative that signals due to these potential nonspecific products be recognized prior to test implementation and rectified wherever possible to avoid potential over-quantification and false-positive results. With the use of such dyes, melting curve analysis will help to discriminate specific products from other contaminants. An alternative method involves coupling of a fluorophore to one of the primers (e.g., light upon extension or LUX technology) [6]. The primer is designed to form a hairpin structure in the absence of product. In the hairpin conformation, the fluorophore is quenched, but it fluoresces upon incorporation in the PCR product, which results in extension of the hairpin structure. Like DNA-binding dyes, a primer-based detection method depends largely on the specificity of the primers, in contrast to probe-based detection methods, discussed below. In place of a dye-conjugated primer or DNA-binding dyes such SYBR Green I, a sequence-specific fluorescent probe may be used for detection and quantification of PCR products. The main advantage of using any of these probe methods instead of dyes such as SYBR Green I or a primer-coupled fluorophore is that there is increased specificity due to independent confirmation of specific product formation, since the probe hybridization sequence does not overlap with the primers used for amplification. As a result of their sequence-specific binding, probes are unlikely to anneal to nonspecific amplification products and primer dimers and so are predicted to give more accurate quantitative results.

ariability of Real-Time PCR Results and External V Quantification Standards There is significant variability in quantitative real-time PCR results, both within and between laboratories. The variability of quantitative results within a laboratory should be measured, and clinical staff should be educated to understand how large a change can be taken as significant. Although values vary, most laboratories find quantitative results between PCR runs fall within a tenfold or smaller range [7–10]. Interlaboratory variability of quantitative results is a significant and long-standing problem, although there has been recent progress toward understanding and

382

A. J. McAdam

addressing this issue. Interlaboratory variation DNA quantification means that patient results cannot be meaningfully compared if samples are tested in different laboratories, and it also reduces the possibilities for research collaborations at multiple institutions. The range of interlaboratory variation is at least 100- and more than 10,000-fold for various analytes measured using laboratory-developed or commercial tests, including Epstein-Barr virus, cytomegalovirus, BK virus, adenovirus, and human herpesvirus 6 [11–13]. In contrast, exclusive use of FDA-approved commercial assays, as for human immunodeficiency virus, reduces interlaboratory variation in results [14]. A number of factors contribute to variability in assay results, including the calibrator used, use of commercial primers and probes, and selection of amplification target [14]. An important consideration in reducing variability in quantitative real-time PCR tests is what nucleic acid material will be used as the calibrator for the standard curve. The calibrator can be a synthetic oligonucleotide, a purified PCR product, recombinant plasmid, or quantified pathogen nucleic acid. All of these are reasonable choices for laboratory-developed tests, although quantified pathogen nucleic acid is preferred if it is available, since it most closely resembles clinical specimens. One potential disadvantage of using a quantified pathogen DNA or RNA (such as extracted viral nucleic acid) is that they may only be available in relatively low concentrations, thus limiting the upper limit of quantification. Because of the large variability in assay results associated with the use of diverse calibrators, the WHO has created calibrators which can be used as reference standards for commercial calibrator reagents. The WHO calibrators consist of intact, lyophilized virus, not purified nucleic acid. There are WHO calibrators for quantitative measurement of cytomegalovirus, Epstein-Barr virus, JC virus, and BK virus [12, 15–17]. Use of these calibrators or derived secondary calibrators reduces bias in assay results [18] and also reduces the variation of interlaboratory results, although the size of the reduction varies considerably [12, 19]. The WHO calibrators for JC virus and BK virus each have heterogeneity in the number of copies of genes present due to deletions within specific genes in a subpopulation of the viruses in the calibrators, suggesting that changes to these may be needed [17, 20]. If these standards are used, it is important to select amplification targets for real-time PCR assays with care so that quantitative results will be accurate.

Amplicon Selection A number of other factors need to be considered when selecting the amplicon region. These include the primer and probe length, their melting temperatures (Tm), and the overall size of the amplicon. Longer amplicons are associated with lower PCR efficiencies and thus greater challenges in quantification, and therefore the amplicon size should be kept between 50 and 150 base pairs. The Tm of the primers should be between 56 °C and 62 °C, but more importantly, the Tm of each of the


383

primers should be as similar as possible. If a Taqman probe is used, it is important for the Tm of the probe to be approximately 10 °C higher than the primers. This is because the probe needs to remain annealed during polymerase extension, and incomplete annealing during extension may result in under-quantification. The amplification target selected influences the consistency of quantitative real- time PCR results [14]. Sequence variation in the amplification target due to heterogeneity in the pathogen’s genome can lead to large differences in results obtained in different assays and laboratories [7, 21, 22]. For certain pathogens, such as adenovirus or BK virus, choosing the target sequence for the assay amplification (the amplicon) is particularly challenging, because numerous viral subtypes and common sequence variants abound, respectively [4, 23]. Careful sequence comparison to all known variants of a given pathogen deposited in genome databases is essential in choosing the appropriate amplicon sequence to avoid regions of high sequence variability. A single nucleotide mismatch in the primer or probe sequence can be sufficient to falsely reduce quantification several orders of magnitude or give a false-negative result. A number of computer programs exist which can help the user in selection of the amplicon. On occasion, it may be impossible to find a specific pair of primers and a single probe that will work for all known sequence variants of a given pathogen. In these cases, it will be necessary to synthesize a small pool of primers or probes that will target a reasonably well-conserved region and cover all potential variants. It is important to realize that in the cases where combinations of primers and probes are pooled, colinearity of quantification between the variants may not always be the norm, although this has been reported for at least one quantitative PCR assay for adenovirus [23]. Another factor related to selection of the amplification target which can have a marked effect upon quantitative results of real-time PCR is the size of the target amplicon. This has been demonstrated for cytomegalovirus testing using plasma, and it remains to be seen whether it is more generally applicable [13]. Shorter amplification targets are associated with detection of higher levels of DNA in plasma for cytomegalovirus [13]. This suggests that the viral DNA in plasma consists of variable lengths, some of which are short enough such that longer amplification targets are less frequently intact than shorter amplification targets. Consistent with this, it has been shown that cytomegalovirus DNA in plasma and serum from renal transplant recipients is free DNA, not associated with viral particles, and highly fragmented [24, 25].

Internal Controls and PCR Inhibitors In the research environment, real-time PCR methods are often used to quantify a gene product of interest relative to an internal reference, such as actin. In the clinical lab, such an internal reference gene is often not present in the clinical sample (e.g., in urine) or, even if present, may be widely variable due to a particular state of the patient (e.g., severe pancytopenia in the context of cancer chemotherapy). For these

384

A. J. McAdam

reasons, absolute quantification is preferred to relative quantification in the clinical setting. Nevertheless, use of an internal reference for amplification is important in the clinical context. As mentioned earlier, the presence of PCR inhibitors or amplification failure for other reasons must be ruled out for any negative result. An internal reference, usually an unrelated plasmid or RNA for RT-PCR, is added into the amplification reaction to verify the amplification conditions. Internal references are available from many commercial sources and can be amplified separately using the identical reaction conditions as the target of interest, or can be implemented in a multiplex platform using a probe that is detected at a channel distinct from the target. The latter is preferred because the same reaction is used for quantification of the clinical sample and for control purposes. Care must be taken, however, that the presence of the internal control does not interfere with the amplification of the target itself. This is because certain components of the two simultaneous amplification reactions are shared (e.g., DNA polymerase) and under certain conditions may be present in relatively limiting quantities. For this reason, the internal control is added at a relatively low copy number to avoid potential competition for PCR reactants. The internal reference can be added at two different steps prior to amplification: (1) directly to the clinical sample prior to sample extraction or (2) to the amplification reaction itself. In the latter case, any loss of amplification of the internal reference can be attributed to a PCR inhibitor or general amplification failure. In the former case, reduced extraction efficiency can also contribute to any observed reduction of internal reference amplification. Thus, the addition of an internal reference standard is an important consideration in setting up a quantitative PCR assay.

Quantitative Digital PCR dPCR uses the same basic chemistry as real-time PCR; however, the method of quantitative measurement of target DNA is entirely different [26]. In dPCR, the sample DNA and PCR reagents, including primers and probes, are mixed and then separated into many separate reactions or partitions. Various commercial platforms are available for creating the partitions, and these are paired with amplification and detection platforms. The partitions can be created using mechanical separation, for example, using an array or by using water-oil emulsions. Due to dilution into a large number of partitions, each PCR reaction should contain one or no target DNA molecules. The PCR amplification is then performed on all the partitions. The total number of template DNA molecules is determined by measuring the total number of partitions which are positive for amplified DNA following completion of the PCR reaction. No quantified reference samples or standard curve are necessary in dPCR. The process is “digital” in that each partition is scored as either positive or negative, rather than along a gradient of positivity as in real-time PCR. It is important that the DNA be adequately diluted such that most partitions contain one or no target DNA molecules or to calculate the required correction if this is not the case [1, 26].


385

The advantages and disadvantages of dPCR relative to real-time PCR should be understood if one is considering use of dPCR. dPCR has several advantages over real-time PCR. First, as mentioned, dPCR does not require a standard curve for quantitative measurement of DNA, although some have stressed that this method does require careful assessment of linearity, accuracy, and precision [27]. Second, most but not all studies find that dPCR provides less variable quantitative results than real-time PCR [10, 28, 29]. Third, the presence of PCR inhibitors has little effect of dPCR [30–32]. This may be because of the dilution of the sample during partitioning or because it does not matter whether amplification is logarithmic or efficient in dPCR, as long as the threshold for a positive result is met in the partitions. Finally, dPCR performs better with divergent target sequences, again probably because the amplification efficiency is less critical in dPCR than in real-time PCR [33]. The disadvantages of dPCR are significant and have prevented widespread use of this method in clinical microbiology testing. First, with the devices that are currently available, dPCR requires more labor and has lower throughput than can be achieved with real-time PCR [1]. Second, dPCR is generally less sensitive than real-time PCR because the sample volumes used are usually lower [28, 29]. Third, dPCR has a smaller dynamic range than real-time PCR, and so samples might need to be diluted following an initial quantitative measurement to provide accurate results. Fourth, currently available systems do not support large multiplex assays, limiting the number of analytes which can be included in an assay [1]. It should be acknowledged that clinical microbiologists are less familiar with dPCR than they are with long-established real-time PCR, and this contributes to the limited use of dPCR as well.

Conclusions Real-time PCR has made PCR-based nucleic acid quantification readily accessible to the clinical microbiology laboratory. The laboratory director must consider several factors when designing or validating a quantitative real-time PCR assay. The sensitivity of real-time PCR quantification depends on the efficiency of the PCR reaction as well as the efficiency of nucleic acid purification and, if used, reverse transcription. Accurate quantification depends on several factors that are important in assay design. These include the selection of an appropriate method for generation of fluorescence as the PCR product accumulates. Although dyes for double-stranded DNA and primer-based fluorescence are commonly used in research, the additional specificity provided by probe-based methods of amplicon detection makes these appropriate for clinical use. The selection of an external quantification control is an important step in designing a quantitative real-time PCR test, but this is an area in which further development is needed. Although it is desirable that the quantification control resemble (or be) quantified pathogen, this is not always practical, and purified nucleic acids are a reasonable alternative. When national or international

386

A. J. McAdam

quantification standards exist, these should be used if possible. dPCR holds potential for application to quantitative measurement of nucleic acid in clinical microbiology primarily due to increased precision of this method. The disadvantages of dPCR, most importantly the reduced sensitivity of this method and low throughput, have limited the use of this method.

References 1. Kuypers J, Jerome KR. Applications of digital PCR for clinical microbiology. J Clin Microbiol. 2017;55:1621–8. 2. Forman M, Wilson A, Valsamakis A. Cytomegalovirus DNA quantification using an automated platform for nucleic acid extraction and real-time PCR assay setup. J Clin Microbiol. 2011;49:2703–5. 3. Huang ML, Nguy L, Ferrenberg J, Boeckh M, Cent A, Corey L. Development of multiplexed real-time quantitative polymerase chain reaction assay for detecting human adenoviruses. Diagn Microbiol Infect Dis. 2008;62:263–71. 4. Iwaki KK, Qazi SH, Garcia-Gomez J, et al. Development of a real-time quantitative PCR assay for detection of a stable genomic region of BK virus. Virol J. 2010;7:295. 5. Karsai A, Muller S, Platz S, Hauser MT. Evaluation of a homemade SYBR green I reaction mixture for real-time PCR quantification of gene expression. BioTechniques. 2002;32:790–2, 4–6. 6. Buh Gasparic M, Tengs T, La Paz JL, et al. Comparison of nine different real-time PCR chemistries for qualitative and quantitative applications in GMO detection. Anal Bioanal Chem. 2010;396:2023–9. 7. Solis M, Meddeb M, Sueur C, et al. Sequence variation in amplification target genes and standards influences interlaboratory comparison of BK virus DNA load measurement. J Clin Microbiol. 2015;53:3842–52. 8. Germi R, Lupo J, Semenova T, et al. Comparison of commercial extraction systems and PCR assays for quantification of Epstein-Barr virus DNA load in whole blood. J Clin Microbiol. 2012;50:1384–9. 9. Descamps V, Martin E, Morel V, et al. Comparative evaluation of three nucleic acid-based assays for BK virus quantification. J Clin Microbiol. 2015;53:3822–7. 10. Hayden RT, Gu Z, Sam SS, et al. Comparative evaluation of three commercial quantitative cytomegalovirus standards by use of digital and real-time PCR. J Clin Microbiol. 2015;53:1500–5. 11. Preiksaitis JK, Pang XL, Fox JD, Fenton JM, Caliendo AM, Miller GG. Interlaboratory comparison of epstein-barr virus viral load assays. Am J Transplant. 2009;9:269–79. 12. Hayden RT, Sun Y, Tang L, Procop GW, Hillyard DR, Pinsky BA. Progress in Quantitative Viral Load Testing: Variability and Impact of the WHO Quantitative International Standards. J Clin Microbiol. 2017;55:423–30. 13. Preiksaitis JK, Hayden RT, Tong Y, et al. Are we there yet? Impact of the first international standard for cytomegalovirus DNA on the harmonization of results reported on plasma samples. Clin Infect Dis. 2016;63:583–9. 14. Hayden RT, Yan X, Wick MT, et al. Factors contributing to variability of quantitative viral PCR results in proficiency testing samples: a multivariate analysis. J Clin Microbiol. 2012;50:337–45. 15. Fryer JF, Heath AB, Minor PD. A collaborative study to establish the 1st WHO international standard for human cytomegalovirus for nucleic acid amplification technology. Biologicals. 2016;44:242–51.


387

16. Fryer JF, Heath AB, Wilkinson DE, Minor PD. A collaborative study to establish the 1st WHO international standard for Epstein-Barr virus for nucleic acid amplification techniques. Biologicals. 2016;44:423–33. 17. Greninger AL, Bateman AC, Atienza EE, et al. Copy number heterogeneity of JC virus standards. J Clin Microbiol. 2017;55:824–31. 18. Tan SK, Milligan S, Sahoo MK, Taylor N, Pinsky BA. Calibration of BK Virus Nucleic Acid Amplification Testing to the 1st WHO International Standard for BK Virus. J Clin Microbiol. 2017;55:923–30. 19. Semenova T, Lupo J, Alain S, et al. Multicenter evaluation of whole-blood Epstein-Barr viral load standardization using the WHO international standard. J Clin Microbiol. 2016;54:1746–50. 20. Bateman AC, Greninger AL, Atienza EE, Limaye AP, Jerome KR, Cook L. Quantification of BK virus standards by quantitative real-time PCR and droplet digital PCR is confounded by multiple virus populations in the WHO BKV international standard. Clin Chem. 2017;63:761–9. 21. Schibler M, Yerly S, Vieille G, et al. Critical analysis of rhinovirus RNA load quantification by real-time reverse transcription-PCR. J Clin Microbiol. 2012;50:2868–72. 22. Hoffman NG, Cook L, Atienza EE, Limaye AP, Jerome KR. Marked variability of BK virus load measurement using quantitative real-time PCR among commonly used assays. J Clin Microbiol. 2008;46:2671–80. 23. Gu Z, Belzer SW, Gibson CS, Bankowski MJ, Hayden RT. Multiplexed, real-time PCR for quantitative detection of human adenovirus. J Clin Microbiol. 2003;41:4636–41. 24. Boom R, Sol CJ, Schuurman T, et al. Human cytomegalovirus DNA in plasma and serum specimens of renal transplant recipients is highly fragmented. J Clin Microbiol. 2002;40:4105–13. 25. Tong Y, Pang XL, Mabilangan C, Preiksaitis JK. Determination of the biological form of human cytomegalovirus DNA in the plasma of solid-organ transplant recipients. J Infect Dis. 2017;215:1094–101. 26. Huggett JF, Cowen S, Foy CA. Considerations for digital PCR as an accurate molecular diagnostic tool. Clin Chem. 2015;61:79–88. 27. Vynck M, Vandesompele J, Thas O. Quality control of digital PCR assays and platforms. Anal Bioanal Chem. 2017;409:5919. 28. Hayden RT, Gu Z, Ingersoll J, et al. Comparison of droplet digital PCR to real-time PCR for quantitative detection of cytomegalovirus. J Clin Microbiol. 2013;51:540–6. 29. Sedlak RH, Cook L, Cheng A, Magaret A, Jerome KR. Clinical utility of droplet digital PCR for human cytomegalovirus. J Clin Microbiol. 2014;52:2844–8. 30. Dingle TC, Sedlak RH, Cook L, Jerome KR. Tolerance of droplet-digital PCR vs real-time quantitative PCR to inhibitory substances. Clin Chem. 2013;59:1670–2. 31. Racki N, Dreo T, Gutierrez-Aguirre I, Blejec A, Ravnikar M. Reverse transcriptase droplet digital PCR shows high resilience to PCR inhibitors from plant, soil and water samples. Plant Methods. 2014;10:42. 32. Sedlak RH, Kuypers J, Jerome KR. A multiplexed droplet digital PCR assay performs better than qPCR on inhibition prone samples. Diagn Microbiol Infect Dis. 2014;80:285–6. 33. Sedlak RH, Nguyen T, Palileo I, Jerome KR, Kuypers J. Superiority of digital reverse transcription-PCR (RT-PCR) over real-time RT-PCR for quantitation of highly divergent human rhinoviruses. J Clin Microbiol. 2017;55:442–9.

Direct Nucleotide Sequencing for Amplification Product Identification Tao Hong

Introduction The technological advances for determining the nucleotide sequence of DNA have fundamentally changed the field of biological research and medicine. For diagnostic molecular microbiology, the most precise method of identification of a PCR product (amplicon) is to determine its nucleotide sequence. Although it is not always necessary to sequence the entire amplicon for routine diagnostic procedures, DNA sequencing techniques have been used to analyze broad range PCR products for bacterial identification as well as for gene mutations related with antimicrobial resistance, for bacterial strain typing, for viral genotyping, and more. Most of the amplicons used for these applications are large (range approximately from 300 base pairs to 1500 base pairs), and the exact nucleotide sequence of these amplicons is crucial for the results. Two basic methods have been created for DNA sequencing, the ddNTP-mediated chain termination method of Sanger et al. [1] and the chemical cleavage method of Maxam and Gilbert [2]. The Sanger method has been widely performed in most research laboratories, using radioisotope-labeled nucleotide (e.g., 32P or 35 S) and a standardized manual method. This method relies on enzymatic DNA synthesis from a specific oligonucleotide primer. The primer is annealed to the complementary sequence adjacent to the DNA of interest on a genetic element [3]. In contrast, the method of DNA sequencing developed by Maxam and Gilbert is based on the specific cleavage of DNA at specific nucleotide. A homogeneous sample of DNA radiolabeled at one end is treated with four separate chemical reactions, each of which modifies a particular type of base. Conditions of the subsequent cleavage reactions are set such that cleavage occur an average of only once for each DNA molecule.

T. Hong (*) Department of Pathology, Hackensack University Medical Center, Hackensack, NJ, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_19

389

390

T. Hong

In the recent past, DNA sequence-based analyses were laborious and time-consuming. These methods were generally used only in research settings. Recent advances in the use of fluorescent dye terminator chemistry and laser scanning in a polyacrylamide gel electrophoresis (PAGE), as well as the application of a capillary electrophoresis technique combined with a fluorescent dye terminator using base-calling software, have made DNA sequencing much less labor intensive. Capillary electrophoresis allows for accurate size discrimination of fluorescently labeled nucleic acids from 50 to 1000 bases with single-base precision. Accordingly, capillary electrophoresis base nucleic acid sequencing analysis has become a routing procedure in many molecular diagnostic laboratories. Pyrosequencing is a non-gel-based DNA sequencing technique that is based on the detection of the pyrophosphate (PPi) released during DNA synthesis. In a cascade of enzymatic reactions, visible light is generated at a level that is proportional to the number of incorporated nucleotides (Fig. 1) This method Polymerase 3’----------------ATGCCGGTTTCCCAAAGTCCC -------5’ 5’ →TACGGCC dCTP Apyrase

PP ATP

dCMP + 2Pi

Light

Sulfurylase Luciferase Real-time monitoring

Nucleotide added: ACGTACGTACGTACGTACGT DNA sequence:

-------TACG------G-----C-----C

Fig. 1 Schematic diagram of pyrosequencing. The reaction mixture consists of single-stranded DNA with an annealed primer, DNA polymerase, ATP sulfurylase, luciferase, and apyrase. The four nucleotide bases are added to the reaction mixture in a particular order, e.g., A, C, G, and T. If the added nucleotide forms a base pair (in this case, two Cs base pair to the template), the DNA polymerase incorporates the nucleotide and a pyrophosphate (PPi) is released. The released pyrophosphate is converted to ATP by ATP sulfurylase, and luciferase uses this ATP to generate detectable light. This light is proportional to the number of nucleotides incorporated and is detected in real time. The pyrosequencing raw data are displayed simultaneously, and in this example, the sequence generated reads TACGGCC. Excess quantities of the added nucleotide are degraded by apyrase. If the nucleotide does not form a base pair with the DNA template, it is not incorporated by the polymerase, and no light is produced. Apyrase then rapidly degrades the nucleotide

Direct Nucleotide Sequencing for Amplification Product Identification

391

generates 30- to 40-base sequences with each primer, and the throughput is 96 samples in approximately 10 min; i.e., the throughput is much higher than that which can be achieved by conventional Sanger sequencing on a gel or by capillarybased automated sequencing instruments. The limitation of pyrosequencing is that the sequence is only accurate within the first 30–40 bases; beyond that the data is unreliable.

Methodology Four distinct steps are required to obtain a DNA sequence from a PCR product: (1) nucleic acid extraction (either RNA or DNA), (2) PCR amplification (or RT-PCR for an RNA target), (3) nucleotide sequencing, and (4) database homology search/ analysis and reporting. 1. Nucleic acid extraction. Depending on the PCR primers (i.e., broad range primers), a pure culture of a bacterial/viral agent is generally required for its identification. If the PCR primer is designed specifically for a particular microbial agent, clinical specimens may be used directly for this nucleic acid extraction step. Various DNA extraction methods can be used, including the traditional phenol chloroform method, commercial DNA extraction kits, and others. Pure culture and relatively large quantity of target DNA make contamination by background DNA from reagents and other sources negligible. In our experience, no DNA purification is necessary for most bacterial targets. Two colonies or the pellet of 1 ml of a culture-positive liquid medium is resuspended in 200 μl sterile saline; 2 μL of the suspension is used directly in the subsequent PCR reaction. Alternatively, the bacterial suspension can be boiled for 10 min and centrifuged for 5 min at 8000 g, and the supernatant (2 μL) can be used for PCR. 2. PCR. Depending on the target and primer set, the PCR condition may vary. It is important to verify the purification of PCR product by visualizing the amplified DNA on an agarose gel before starting the DNA sequencing process, especially in the assay validation stage. Once the procedure is validated and a single PCR product is routinely obtained, the agarose gel step may not be necessary. Usually PCR amplicon amplified from a pure target produces large amount of DNA sufficient for nucleotide sequencing. For a PCR reaction that generates multiple products, a gel purification procedure is necessary to purify the amplicon of interest. 3. Nucleotide sequencing. The PCR amplicon can be sequenced directly after removal of unpolymerized primers and 4-NTPs that can be achieved by enzymatic digestion with exonuclease and shrimp alkaline phosphatase. No further purification or concentration of the amplicon is generally necessary. Automated sequencing can be performed according to sequencing chemistry and sequencing instrument of one’s laboratory. Usually, one of the PCR primers is used as the primer for the sequencing reaction. If both strands of the amplicon are to be sequenced, two separate reactions are needed. For clinical microbiology

392

T. Hong

laboratories with no DNA sequencing equipment, this final step can usually be achieved by sending the purified amplicon with one of the PCR primers to an in-house core sequencing facility or to a commercial laboratory that provides DNA sequencing services. Ruano and Kidd [4] have developed a method called coupled amplification and sequencing; it is a method for sequencing both strands of the template as they are amplified. The procedure is biphasic, stage I selects and amplifies a single target from the genomic DNA, and stage II accomplishes the sequencing as well as additional amplification of the target using aliquots from the stage I reaction mixed with end-labeled primer and dideoxynucleotides. A modified procedure (CLIP) has been developed using Clipper sequencer [5]. Two characteristics of the CLIP reaction as a modification of the original coupled amplification and sequencing method by Ruano and Kidd are: (a) An engineered mutant of thermostable DNA polymerase is used which lacks 5′-3′ exonuclease activity and therefore produces uniform band intensities. (b) Different far-red fluorescent dyes are linked to the two inward-facing CLIP primers, allowing a template to be sequenced in both directions in a single run. 4. Homology search and reporting. For sequence analysis, the sequence is compared with the data in nucleotide sequences database, whether an in-house developed database or a commercial or public database (such as GenBank). The match (sometimes multiple matches) should be interpreted cautiously; specifically, consensus of the matches and/or the match with type strain should be sought. Preferably, sequences from type strains with good quality (no unresolved nucleotides or artificial gaps) and obtained from a reputable laboratory should be used. One should be aware that the nucleotide sequence data in the public data base have not been peer reviewed. Early sequencing data generated by a manual method may not be very accurate.

Application of DNA Sequencing in Molecular Diagnosis Sequencing of hsp65 for Identification of Mycobacterial Species Clinical microbiology laboratories usually use a combined molecular/conventional approach for Mycobacterium identification. Commercial probes are available for M. tuberculosis complex, M. avian/intracellular complex, M. kansasii, and M. gordonae. For other mycobacteria, conventional−/biochemical-based (and time- consuming) methods are often used for identification. Atypical biochemical reactions frequently cause problems for accurate Mycobacterium species identification. In addition, many molecular methods have been developed for identifying mycobacteria; the 16S rRNA gene sequencing is the most commonly used approach for sequence based identification of Mycobacterium species (review


393

in chapter 28 of this book). The 65-kDa heat shock protein gene (hsp65), present in all Mycobacterium species, is more variable than the 16S rRNA gene sequence and is useful only for the identification of genetically related species. Sequence variations in the hsp65 gene have been exploited to identify both slowly growing Mycobacterium species and rapidly growing mycobacteria (RGM) to the species level. Hance et al. [6] have reported amplifying a fragment of the 65-kDa heat shock protein gene (hsp65) to detect and, coupled with species-specific probes, in order to identify Mycobacterium species from clinical samples. Moreover, Plikaytis et al. [7] and Telenti et al. [8] have described the use of separate gene regions for the successful identification of Mycobacterium species by using restriction digest analysis of amplified hsp65 fragments (hsp65 PRA). Hsp65 PRA has been widely used for identification, and an algorithm based on this approach has recently been developed for differentiating 34 Mycobacterium species, including members of the RGM group. A sequence-based strategy has several potential advantages. It generates direct, unambiguous data and can distinguish medically relevant subspecific phylogenetic lineages. Recent advances in automated DNA sequencing have also made this approach much easier. To overcome the limitations of hsp65 PRA and the potential advantage of generating direct unambiguous data, Kapur et al. [9] developed a procedure for sequencing the hsp65 amplicon generated by the Telenti primers as a means for identifying Mycobacterium species. This technique has since been used by many investigators to identify species, as well as characterize and define groups within a number of mycobacteria. A study by McNabb et al. [10] assessed the viability of using hsp65 sequencing to identify all Mycobacterium species routinely isolated by a clinical mycobacteriology laboratory; in addition, the ability of an in-house database, consisting of 111 hsp65 sequences from putative and valid Mycobacterium species or described groups, to identify 689 mycobacterial clinical isolates from 35 species or groups was evaluated. The overall agreement between hsp65 sequencing and other identification methods was 85.2%. This study indicates that for hsp65 sequencing to be an effective means for identifying Mycobacterium species, a comprehensive database must be constructed. Hsp65 sequencing has the advantage of being more rapid and less expensive than conventional biochemical test panels, uses a single set of reagents to identify both rapid- and slow-growing Mycobacterium species, and can provide a more definitive identification. Due to limitations of an appropriate database, the best approach for sequence identification of Mycobacterium species is to have both the16S rRNA gene method (see chapter “Advances in the Diagnosis of Mycobacterium tuberculosis Infection” in Volume 2) and the hsp65 gene method available in the laboratory.

The recA Gene Sequence The recA gene sequence can be an alternative to the sequence analysis of the gene 16S rDNA for the differentiation of mycobacteria [11]. The RecA protein encoded by this gene exists in all bacteria. It plays an important role in homologous DNA

394

T. Hong

recombination, DNA damage repair, and induction of the SOS response. Sequencing of the fragment A (915 bp–970 bp) can distinguish M. leprae, M. aurum, and M. mucogenicum. The recA sequencing can differentiate the clinically important M. kansasii from the less clinically important M. gastri. Sequencing of the fragment B (about 1 kbp) distinguished the species M. xenopi, M. asiaticum, M. shimoidei, and MTC.

Internal Transcribed Spacer The internal transcribed spacer (ITS) region, a stretch of DNA that lies between the 16S and 23S rRNA subunit genes, has proven to show a high degree of variability in both sequence and size at the genus and species level [12, 13]. Hence, this region may allow efficient identification of species due to its enhanced variability within a genus [14]. The diversity of the intergenic spacer regions is due in part to variations in the number and type of tRNA sequences found among these spacers. Sequence of ITS region has been used for identification of Mycobacterium species, for Staphylococcus species [15], Streptococcus species, and for rapid identification of medically important yeast [16]. For molecular identification of mycobacteria, the most frequently used DNA sequence-based method is the 16S rRNA gene. However, there are instances in which the sequences of 16S rDNA genes have been found to be very similar, if not identical, between different species in a genus, making it necessary to find alternative specific sequences. The intergenic 16S-23S internal transcribed spacer (ITS) region is considered to be less prone to selective pressure and consequently can be expected to have accumulated a higher percentage of mutations than the corresponding rDNA. Sequencing of the ITS regions of diverse bacteria indicates that considerable length and primary sequence variation occurs, and this variability has been successfully used to distinguish between closely related mycobacteria, such as the M. avium-M. intracellulare complex [17, 18]; M. gastri and M. kansasii (both of them share an identical 16S rDNA sequence [19, 20]); Mycobacterium farcinogenes and Mycobacterium senegalense [21]; and Mycobacterium chelonae complex. Streptococci are a very diverse group of microorganism; many molecular techniques have been used for identification of streptococci. Chen et al. [22] have evaluated the feasibility of sequence analysis of the 16S-23S ribosomal DNA (rDNA) intergenic spacer (ITS) for the identification of clinically relevant viridans group streptococci (VS). The ITS regions of 29 reference strains (11 species) of VS were amplified by PCR and sequenced. The ITS lengths (246–391 bp) and sequences were highly conserved among strains within a species. The intraspecies similarity scores for the ITS sequences ranged from 0.98 to 1.0, except for the score for S. gordonii strains. The interspecies similarity scores for the ITS sequences varied from 0.31 to 0.93. Phylogenetic analysis of the ITS regions revealed that evolution of the regions of some species of VS is not parallel to that of the 16S rRNA genes.


395

The accuracy of using ITS sequencing for identification of VS was verified by 16S rDNA sequencing for all strains except strains of S. oralis and S. mitis, which were difficult to differentiate by their 16S rDNA sequences. It was concluded that identification of species of VS by ITS sequencing is reliable and could be used as an alternative accurate method for identification of VS. In staphylococci, there are several copies of the rrn operon. Gürtler and Barrie [23] characterized the spacer sequences of S. aureus strains, including methicillin- resistant S. aureus (MRSA) isolates, and identified nine rrn operons whose 16S23S spacer region varied from 303 to 551 bp. Three of these spacers contain the tRNAIle gene, and two contain both the tRNAIle and the tRNAAla genes, while the remaining four 16S-23S spacers have no tRNA gene. Forsman et al. [24] sequenced the 16S- 23S spacer of five staphylococcal species (S. aureus, S. epidermidis, S. hyicus, S. simulans, and S. xylosus) and found that in addition to S. aureus, S. hyicus and S. simulans also had a tRNAIle gene in some of their rrn operons. The sequence conservation of the rrn operons argues for the use of the 16S-23S spacer region as a stable and direct indicator of the evolutionary divergence of S. aureus strains. Fungi are an incredibly diverse and ubiquitous group of eukaryotes; traditional identification depends on morphological differences in their sexual or asexual reproductive structures. A PCR/sequencing-based approach may provide rapid and more accurate identification method. Coding regions of the 18S, 5.8S, and 28S nuclear rRNA genes evolve slowly, are relatively conserved among fungi, and provide a molecular basis of establishing phylogenetic relationships [25]. Between coding regions are the internal transcribed spacer 1 (ITS1) and regions 2 (ITS2), respectively. The ITS region evolves more rapidly and vary among different species within a genus. The ITS regions are located between the 18S and 28S rRNA genes and offer distinct advantages over other molecular targets including increased sensitivity due to the existence of approximately 100 copies per genome (Aspergillus species). PCR amplification may facilitate the identification of ITS region DNA sequences with sufficient polymorphism to be useful for identifying medically important fungal species; the concordance rate between phonotypical and ITS2 is greater than 98% [26, 27]. The sequence variation of ITS regions has led to their use in phylogenetic studies of many different organisms [28].

HCV Genotyping by Nucleotide Sequencing There are nearly four million persons infected with HCV in the United States, and it is estimated that 30,000 acute new infections will occur annually. Progression to chronic disease occurs in approximately 85% of individuals. Chronic HCV infection is known to progress to cirrhosis and hepatocellular carcinoma. Recently, the development of direct-acting antiviral agents (DAAs), combined with pegylated interferon (Peg-IFN) and ribavirin (RBV), has improved the sustained virologic response rates significantly compared with Peg-IFN and RBV alone. By targeting

396

T. Hong

HCV proteins or enzymes essential for viral replication, the directly acting antivirals (DAAs) have a much more potent action. They are currently used in all-oral IFN-free combinations for 3–6 months; with or without ribavirin, the cure rate is as high as 90%. The HCV genotype and subtype are recommended before anti-HCV treatment due to the variable response to antiviral therapy of the HCV types [29]. HCV is genetically diversified; they are classified into seven genotypes (1–7) with differences of 30–35% of nucleotide sequence and further classified into 67 subtypes [30]. Full genome sequencing, the reference standard for HCV genome typing, is a long and tedious process. Most of the full-length HCV sequencing assays were performed for the description or for the classification of a novel strain. For clinical testing, selected targets are used for genome typing. Commercial HCV genotyping assays are currently based on different strategies (DNA sequencing, reverse hybridization, real-time PCR), and distinct HCV genomic targets are used (5’-UTR, Core, NS5B). The Spanish Group for Viral Hepatitis Study (GEHEP) has evaluated three commercial methods – Trugene HCV genotyping kit (Siemens), VERSANT HCV Genotype 2.0 assay (Siemens), and Real-Time HCV genotype II (Abbott), compared to NS5B Sanger DNA sequencing. The overall discordance with the reference method was 34% for Trugene and 15% for VERSANT HCV2.0. The Abbott assay correctly identified all 1a and 1b subtypes, but did not subtype all the 2, 3, 4, and 5 (34%) genotypes. Major discordances were found in 16% of cases for Trugene HCV, and the majority were 1b- to 1a-related discordances; major discordances were found for VERSANT HCV 2.0 in 6% of cases [31]. With the advance of new sequencing technology, the next-generation sequencing (NGS), whole genome sequencing of HCV has become more achievable. The advantage of NGS is high throughput, read depth and sensitivity, and detection of minor strain/mutations Sanger sequence cannot achieve.

HIV-1 Genotyping The ViroSeq HIV-1 genotyping assays (Celera Diagnostics, Alameda, CA; distributed by Abbott Molecular Diagnostics, Des Plaines, IL) also use dideoxy chainterminating sequencing, but each dideoxynucleotide is labeled with a different fluorescent dye. Each reaction mixture contains one primer but all four uniquely labeled dideoxynucleotides. Separation of the terminated PCR products is done by capillary electrophoresis. The ViroSeq HIV-1 Genotyping System can be used for detecting HIV genomic mutations associated with resistance to specific types of antiretroviral drugs and facilitates monitoring and treatment of HIV infection. Specifically, the ViroSeq HIV-1 Genotyping System can be used to detect HIV-1 Subtype B viral resistance in plasma samples collected in EDTA with a viral load ranging from 2000 to 750,000 copies/mL and can genotype the entire HIV-1 protease gene from codons 1–99 and two-thirds of the reverse transcriptase (RT) gene from codons 1–335.


397

Sequence-Based Bacterial Genome Typing Spa Typing of Staphylococcus aureus Many techniques are available to differentiate S. aureus, and specifically MRSA, isolates. Conventionally, isolates were distinguished by phenotypic methods, including antibiotic susceptibility testing and bacteriophage typing. Both methods have limitations, as genetically unrelated isolates commonly have the same antibiogram, and many S. aureus isolates are nontypeable by phage typing. With the advancement of molecular biology, strain typing focused on DNA-based methods: restriction of endonuclease patterns of chromosomal or plasmid DNA, Southern blot hybridization using gene-specific probes, ribotyping, polymerase chain reaction (PCR)-based approaches, and pulsed-field gel electrophoresis (PFGE). These methods require subjective interpretation and comparison of patterns and fingerprint images. Nucleotide sequence analysis is an objective genotyping method; sequencing data can be easily stored and analyzed in a relational database. Recent advances in DNA sequencing technology have made it possible for sequencing to be considered as a viable typing method. Two different strategies have been used to provide genotyping data: multi-locus sequence typing (MLST), which compares sequence variation in numerous housekeeping gene targets, and single-locus sequence typing, which compares sequence variation of a single target. MLST has been developed for Neisseria gonorrhoeae, Streptococcus pneumoniae, and S. aureus, based on the classic multi-locus enzyme electrophoresis (MLEE) method used to study the genetic variability of a species. Sequence analysis of five to seven housekeeping genes provides a database from which to infer relationships in somewhat distantly related isolates that have had substantial time to diversify. The MLST approach is not practical to be used in a clinical laboratory because it is labor intensive, timeconsuming, and costly. A single-locus target, if discriminating, provides an inexpensive, rapid, objective, genotyping method to subspeciate bacteria. Two S. aureus genes are conserved within the species; Shopsin et al. has developed the protein A (spa) [32] and coagulase (coa) [33] procedures for sequence-based staphylococcal strain typing. DNA sequencing of the short sequence repeat (SSR) region of the protein A gene (spa) has been used as an alternative to current techniques for the typing of S. aureus. The SSR consists of a variable number of 24-bp repeats and is located immediately upstream of the region encoding the C-terminal cell wall attachment sequence. The sequencing of the spa SSR region combines many of the advantages of a sequencing-based system such as MLST but may be more rapid and convenient for outbreak investigation in the hospital setting since spa typing involves a single locus. The coagulase gene (coa) variable region has been evaluated for use in conjunction with spa sequencing for the strain typing of MRSA. The coagulase protein is an important virulence factor of S. aureus. Like spa, coa has a polymorphic repeat region that can be used for differentiating S. aureus isolates. The variable region of coa is comprised of 81-bp tandem short sequence repeats (SSRs) that are

398

T. Hong

variable in both number and sequence, as determined by restriction fragment length polymorphism analysis of PCR products. Coagulase gene (coa) short sequence repeat region sequencing was used to measure relatedness among a collection of temporally and geographically diverse methicillin-resistant Staphylococcus aureus isolates. The results show that coa typing is a useful addition to spa typing for analysis of S. aureus, including methicillin-resistant strains.

Pyrosequencing Pyrosequencing, with its ability of rapidly sequencing a short piece of DNA, has been evaluated for applications used in many areas: GC strain typing (porB gene sequencing) [34], linezolid resistance in enterococci [35], lamivudine resistance in HBV [36], monitoring HIV protease inhibitor resistance [37], rapid identification of bacteria from positive blood culture [38], and detection of HSV-1 and 2 [39]. For 16S rRNA gene-based bacterial identification, a minimum of 200 bp or more is needed for any meaningful identification. However, many investigators have tried to use short representative regions for rapid identification, notably identification for mycobacteria and rapid ID for sepsis-related bacteria. Jordan et al. [38] have studied the possibility of using pyrosequencing to identify a 15-base hypervariable region within the 16S rRNA gene for bacterial ID, as compared with 380 bp 16S rRNA fragment sequencing. The results were not very encouraging; the 380 bp sequencing can give species level identification, while the 15 bp pyrosequencing can only give semi-genus level identification, such as Staphylococcus, Streptococcus, or enteric gram-negative bacilli. The 15-bp pyrosequencing did not do much better than a simple gram stain smear reviewed by an experienced clinical microbiologist. A 30-bp pyrosequencing method was evaluated for Mycobacterium identification. When blasted against GenBank, 179 of 189 sequences (94.7%) assigned isolates to the correct molecular genus or group. Pyrosequencing of this hypervariable region afforded rapid and acceptable characterization of common, routinely isolated clinical Mycobacterium spp. However, additional sequencing primer or additional biochemical tests may be needed for more accurate identification [40]. Pyrosequencing did very well in identifying mutant genes associated with drug resistance. A pyrosequencing assay for the rapid characterization of resistance to HIV-1 protease inhibitors [37]. This sequencing approach allows parallel analysis of 96 reactions in 1 h, facilitating the monitoring of drug resistance in 8 patients simultaneously. Twelve pyrosequencing primers were designed and were evaluated on the MN strain and on viral DNA from peripheral blood mononuclear cells from eight untreated HIV-1-infected individuals. The method had a limit of detection of 20–25% for minor sequence variants. Pattern recognition (i.e., comparing actual sequence data with expected wild-type and mutant sequence patterns) simplified


399

the identification of minor sequence variants. This real-time pyrosequencing method was applied in a longitudinal study monitoring the development of PI resistance in plasma samples obtained from four patients over a 2 1/2-year period. Pyrosequencing identified eight primary protease inhibitor resistance mutations as well as several secondary mutations. A pyrosequencing method for detection and quantification of macrolide resistance mutations was developed and tested for Streptococcus pneumoniae, Streptococcus pyogenes, Mycobacterium avium, Campylobacter jejuni, and Haemophilus influenzae. The method detecting mutations at positions 2058 and 2059 (Escherichia coli numbering) of the 23S rRNA gene [41]. Pyrosequencing has also been used for fungal identification [42]. The Roche 454 platform, using pyrosequencing technology to carry out hundreds of thousands of sequencing reactions simultaneously on independent beads, is one of the technologies referred to as the “next-generation sequencing.”

Next-Generation Sequencing During the past decade, multiple new sequencing technology platforms have emerged. And they have surpassed conventional Sanger sequencing method in terms of increased total sequence production and significantly decreased cost, these new sequencing methods are referred to as next-generation sequencing, and they have considerable potential for clinical diagnostics. The five major nextgeneration sequencing platforms as of this writing are the Roche 454 GS-FLX (454, Branford, CT), the Illumina (San Diego, CA) Genome Analyzer, the ABI SOLiD (Applied Biosystems), HeliScope (Helicos), and SMRT by Pacific Bioscience. A detailed description of the mechanisms of each technology is beyond the scoop of this chapter. Briefly, these new-generation sequencing technologies have a similar general approach: breaking DNA into multiple fragments, amplifying the fragments (in some technology, no amplification needed) and then simultaneously rapidly sequencing multiple fragments, and then using a powerful bioinformatics software to align the sequences and generating final sequencing results. These new technologies are having a significant impact on human genome research, diagnosis of genetic disorders, cardiovascular disease, and cancer. In the field of clinical microbiology, the technologies will have tremendous impact on rapid whole genomic sequencing, identify new organisms, and look into strain-to-strain variations and rare mutations. Ultra-deep sequencing using the Roche 454 system can detect rare viral variants consisting of as little as 1% of the population [43] and significantly impact treatment outcomes in HIV-1 infections. This level of detection can never be achieved by traditional sequencing methods. NGS is capable of de novo sequencing directly from clinical specimens; this approach can be used for detecting unexpected and/ or unknown/novel organisms.

400

T. Hong

References 1. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A. 1977;74:5463–7. 2. Maxam AM, Gilbert W. A new method for sequencing DNA. Proc Natl Acad Sci. 1977;74:560–4. 3. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. 3rd ed. New York: Cold Spring Harber Laboratory Press; 2000. 4. Ruano G, Kidd KK. Coupled amplification and sequencing of genomic DNA. Proc Natl Acad U S A. 1991;88:2815–9. 5. Yager TD, Baron L, Batra R, Bouevitch A, Chan D, Chan K, Darasch S, Gilchrist R, Izmailov A, Lacroix J-M, Marchelleta K, Renfrew J, Rushlow D, Steinbach E, Ton C, Waterhouse P, Zaleski H, Dunn JM, Stevens J. High performance DNA sequencing, and the detection of mutations and polymorphisms, on the clipper sequencer. Electrophoresis. 1999;20:1280–300. 6. Hance AJ, Grandchamp B, Le’vy-Fre’bault V, Lecossier D, Rauzier J, Bocart D, Gicquel B. Detection and identification of mycobacteria by amplification of mycobacterial DNA. Mol Microbiol. 1989;3:843–9. 7. Plikaytis BB, Plikaytis BD, Yakrus A, Butler WR, Woodley CL, Silcox VA, Shinnick TM. Differentiation of slowly growing Mycobacterium species, including Mycobacterium tuberculosis, by gene amplification and restriction fragment length polymorphism analysis. J Clin Microbiol. 1992;30:1815–22. 8. Telenti A, Marchesi F, Balz M, Bally F, Böttger EC, Bodmer T. Rapid identification of mycobacteria to the species level by polymerase chain reaction and restriction enzyme analysis. J Clin Microbiol. 1993;31:175–8. 9. Kapur V, Li LL, Hamrick MR, Plikaytis BB, Shinnick TM, Telenti A, Jacobs WR, Banerjee A, Cole S, Yuen KY, Clarridge JE, Kreiswirth BN, Musser JM. Rapid Mycobacterium species assignment and unambiguous identification of mutations associated with antimicrobial resistance in Mycobacterium tuberculosis by automated DNA sequencing. Arch Pathol Lab Med. 1995;119:131–8. 10. McNabb A, Eisler D, Adie K, Amos M, Rodrigues M, Stephens G, Black WA, Isaac-Renton J. Assessment of partial sequencing of the 65-Kilodalton heat shock protein gene (hsp65) for routine identification of Mycobacterium species isolated from clinical sources. J Clin Microbiol. 2004;42:3000–11. 11. Blackwood KS, He C, Gunton J, Turenne CY, Wolfe J, Kabani AM. Evaluation of recA sequences for identification of Mycobacterium species. J Clin Microbiol. 2000;38:2846–52. 12. Barry T, Colleran G, Glennon M, Dunican LK, Gannon F. The 16S/23S ribosomal spacer region as a target for DNA probes to identify eubacteria. PCR Methods Appl. 1991;1:51–6. 13. Gürtler V, Stanisich VA. New approaches to typing and identification of bacteria using 16S- 23S rDNA spacer region. Microbiology. 1996;142:3–16. 14. Garcia-Martinez J, Martinez-Murcia A, Anton AI, Rodriguez-Valera F. Comparison of the small 16S to 23S intergenic spacer region (ISR) of the rRNA operons of some Escherichia coli strains of the ECOR collection and E. coli K-12. J Bacteriol. 1996;178:6374–7. 15. Couto I, Pereira S, Miragaia M, Sanches IS, de Lencastre H. Identification of clinical staphylococcal isolates from humans by internal transcribed spacer PCR. J Clin Microbiol. 2001;39:3099–103. 16. Chen Y-C, Eisner JD, Kattar MM, Rassoulian-Barrett SL, Lafe K, Bui U, Limaye AP, Cookson BT. Polymorphic internal transcribed spacer region 1 DNA sequences identify medically important yeasts. J Clin Microbiol. 2001;39:4042–51. 17. De Smet AL, Brown IN, Yates M, Ivanyi J. Ribosomal internal transcribed spacers are identical among Mycobacterium avium-intracellulare complex isolates from AIDS patients, but vary among isolates from elderly pulmonary disease patients. Microbiology. 1995;141:2739–47.


401

18. Frothingham R, Wilson KH. Sequence-based differentiation of strains in the Mycobacterium avium complex. J Bacteriol. 1993;175:2818–25. 19. Rogall T, Wolters J, Floher T, Böttger EC. Towards a phylogeny and definition of the species at the molecular level within the genus Mycobacterium. Int J Syst Bacteriol. 1990;40:323–30. 20. Roth A, Fischer M, Hamid HE, Ludwig W, Michalke S, Mauch H. Differentiation of phylogenetically related slowly growing mycobacteria based on 16S-23S rRNA gene internal transcribed spacer sequences. J Clin Microbiol. 1998;36:139–47. 21. Hamid ME, Roth A, Landt O, Kroppenstedt RM, Goodfellow M, Mauch H. Differentiation between Mycobacterium farcinogenes and Mycobacterium senegalense strains based on 16S- 23S ribosomal DNA internal transcribed spacer sequences. J Clin Microbiol. 2002;40:707–11. 22. Chen CC, Teng LJ, Chang TC. Identification of clinically relevant Viridans group streptococci by sequence analysis of the 16S-23S ribosomal DNA spacer region. J Clin Microbiol. 2004;42:2651–7. 23. Gürtler V, Barrie HD. Typing of Staphylococcus aureus strains by PCR-amplification of variable-length 16S-23S rDNA spacer regions: characterization of spacer sequences. Microbiology. 1995;141:1255–65. 24. Forsman P, Tilsala-Timisja¨rvi A, Alatossava T. Identification of staphylococcal and streptococcal causes of bovine mastitis using 16S–23S rRNA spacer regions. Microbiology. 1997;143:3491–500. 25. White TJ, Bruns T, Lee S, Taylor J. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gefland DH, Sninsky JJ, White TJ, editors. PCR protocols: a guide to methods and applications. New York: Academic Press, Inc; 1990. p. 315–22. 26. Chen YC, Eisner JD, Kattar MM, Rassoulian-Barrett SL, LaFe K, Yarfitz SL, Limaye AP, Cookson BT. Identification of medically important yeasts using PCR-based detection of DNA sequence polymorphisms in the internal transcribed spacer 2 region of the rRNA genes J. Clin Microbiol. 2000;38:2302–10. 27. Travis H, Iwen PC, Hinrichs SH. Identification of Aspergillus species using internal transcribed spacer regions 1 and 2. J Clin Microbiol. 2000;38:1510–5. 28. Guarro J, Gene J, Stchigel AM. Developments in fungal taxonomy. Clin Microbiol Rev. 1999;12:454–500. 29. Lynch SM, Wu GY. Hepatitis C virus: a review of treatment guidelines, cost-effectiveness, and access to therapy. J Clin Transl Hepatol. 2016;4(4):310–9. 30. Smith DB, Bukh J, Kuiken C, Muerhoff AS, Rice CM, Stapleton JT, et al. Expanded classification of hepatitis C virus into 7 genotypes and 67 subtypes: updated criteria and genotype assignment web resource. Hepatology. 2014;59:318–27. 31. Chueca N, Rivadulla I, Lovatti R, Reina G, Blanco A, Fernandez-Caballero JA, et al. Using NS5B sequencing for hepatitis C virus genotyping reveals discordances with commercial platforms. PLoS One. 2016;11(4):e0153754. 32. Shopsin B, Kreiswirth BN. Molecular epidemiology of methicillin-resistant Staphylococcus aureus. Emerg Infect Dis. 2001;7:323–6. 33. Shopsin B, Gomez M, Waddington M, Riehman M, Kreiswirth BN. Use of coagulase gene (coa) repeat region nucleotide sequences for typing of methicillin-resistant Staphylococcus aureus strains. J Clin Microbiol. 2000;38:3453–6. 34. Unemo M, Olce’n P, Jonasson J, Fredlund H. Molecular typing of Neisseria gonorrhoeae isolates by pyrosequencing of highly polymorphic segments of the porB gene. J Clin Microbiol. 2004;42:2926–34. 35. Sinclair A, Arnold C, Woodford N. Rapid detection and estimation by pyrosequencing of 23S rrna genes with a single nucleotide polymorphism conferring linezolid resistance in enterococci. Antimicrob Agents Chemother. 2003;47:3620–2. 36. Lindström A, Odeberg J, Albert J. Pyrosequencing for detection of lamivudine-resistant hepatitis B virus. J Clin Microbiol. 2004;42:4788–95.

402

T. Hong

37. O’meara D, Wilbe K, Leitner T, Hejdeman B, Albert J, Lundeberg J. Monitoring resistance to human immunodeficiency virus type 1 protease inhibitors by pyrosequencing. J Clin Microbiol. 2001;39:464–73. 38. Jordan JA, Butchko AR, Durso MB. Use of pyrosequencing of 16S rRNA fragments to differentiate between bacteria responsible for neonatal sepsis. J Mol Diagn. 2005;7:105–10. 39. Adelson ME, Feola M, Trama J, Tilton RC, Mordechai E. Simultaneous detection of herpes simplex virus types 1 and 2 by real-time PCR and pyrosequencing. J Clin Virol. 2005;33(1):25–34. 40. Tuohy MJ, Hall GS, Sholtis M, Procop GW. Pyrosequencing trade mark as a tool for the identification of common isolates of Mycobacterium sp. Diagn Microbiol Infect Dis. 2005;51:245–50. 41. Haanpera M, Huovinen P, Jalava J. Detection and quantification of macrolide resistance mutations at positions 2058 and 2059 of the 23S rRNA gene by pyrosequencing. Antimicrob Agents Chemother. 2005;49:457–60. 42. Gharizadeh B, Norberg E, Loffler J, Jalal S, Tollemar J, Einsele H, Klingspor L, Nyren P. Identification of medically important fungi by the pyrosequencing technology. Mycoses. 2004;47:29–33. 43. Simen BB, Simons JF, Hullsiek KH, Novak RM, Macarther RD, Baxter JD, Hugan C, Lubeski C, Turenchalk GS, Braverman MS, Desany B, Rothberg JM, Egholm M, Kozai MJ, Beirn T, Community Programs for Clinical Research on AIDS. Low abundance drug-resistant viral variants in chronically HIV infected, antiretroviral treatment-naïve patients significantly impact treatment outcomes. J Infect Dis. 2009;199:610–2.

Solid and Suspension Microarrays for Detection and Identification of Infectious Diseases Sherry Dunbar, Janet Farhang, Shubhagata Das, Sabrina Ali, and Heng Qian

Introduction Microarray technology has been in use for over 30 years, and since its inception for use with antibodies, the concept has been further developed to utilize components such as DNA, proteins, cells, tissues, and other biomaterial, allowing microarrays to meet the specific needs of diagnostic applications in genomics, oncology, and microbial diagnostics. Microarrays can be simply described as a collection of microscopic spots (or features) of biological capture molecules arranged in a defined order on a solid substrate. For DNA microarrays, each spot contains probes consisting of short single-stranded DNA oligonucleotides or larger double-stranded DNA. While solid-state microarrays can be broadly depicted by the concept of genetic probes attached to a solid, impermeable substrate in known locations, there are a wide variety of means by which this can be accomplished [1]. Points of differentiation include the substrate material, probe immobilization techniques, signal detection methods, and various assembly methods – all resulting in a myriad of microarray types (Table 1). The sample containing target DNA molecules is usually prepared with a detectable tag, the microarray is exposed to the sample mixture under conditions favorable to hybridization, and any unbound DNA fragments are washed away. Microarrays are then read to detect those positions with DNA-DNA hybrids, indicating the presence of the target nucleic acid in the original sample. Each microarray is therefore a collection of many hybridizations done in parallel and thus functions as a highly parallel processor for biological inquiry. In this chapter, the concepts behind microarray technologies, the various platforms

S. Dunbar (*) · J. Farhang · S. Das Luminex Corporation, Austin, TX, USA e-mail: [email protected] S. Ali · H. Qian Luminex Corporation, Toronto, ON, Canada © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_20

403

Illumina

Horiba Scientific

Infinium

Description Maskless, in situ synthesis is used to create 60-mer oligonucleotide probes by an inkjet printing method on a glass substrate Low-density microarrays are generated through 3D gel droplets containing covalently bonded capture probes or antibodies (inclusion immobilization) on plastic substrate surface 4 mm x 4 mm microarray chips are placed into the bottom of an ArrayTube well. ArrayTube2 wells can be combined into the ArrayStrip platform allowing up to 96 microarrays in a typical 96-well format SPRi-Biochips™, SPRi-Slides™, and SPR Imaging Systems are available to create lab-developed tests

BeadChip, 3 μm silica beads coupled with target probes are held in fluorescent detection microwells by van der Waals forces and hydrostatic interactions. On the surface of each BeadChip array, beads with the various probes targeting specific loci of the target gene of interest are added to discrete regions. As denatured PCR-amplified or sample DNA fragments pass over the BeadChip, each probe binds to the complementary sequences in the sample DNA, if present. A single-base extension incorporates one of four fluorescently labeled nucleotides which is detected by laser scanning

Colorimetric, fluorescent and precipitation staining Surface plasmon SPRi- resonance imaging Biochips™, SPRi-Slides™ (SPRi)

ArrayTube/ ArrayTube2/ Array Strip

Alere

3D inclusion gel droplet, fluorescent detection

Technology Maskless inkjet-mediated

TruArray®

Microarray platform SurePrint

Akonni Biosystems

Company Agilent

Table 1 Comparison of commercial microarray platforms

http://www.horiba. com/scientific/ products/ surface-plasmonresonanceimaging-spri/ Michael et al. [4]; https://www. illumina.com/ science/ technology/ beadarraytechnology.html

XelPleX and OpenPlex

iScan

https://aleretechnologies.com

Cooney et al. [3]

References Hughes et al. [2]

ArrayMate

Instrument SureScan Microarray Scanner TruDx® 2000 or 3000 Imagers

404 S. Dunbar et al.

xMAP®

Luminex

GeneChip®

VereChip™

Thermo Fisher/ Affymetrix

Veredus Laboratories

NanoCHIP® Savyon Diagnostics/ Nanogen, Inc.

NanoGrid®

Luminex

ssDNA probes are spotted onto a glass substrate. Target DNA is hybridized followed by addition of poly-A-tailed target-specific oligos that bind to sections of target DNA not bound by the ssDNA probes. Poly-T-tailed gold nanoparticles bind to the poly-A tails, and addition of a silver substrate results in signal amplification by LED light scatter which is detected by camera Bead suspension, Fluorescently dyed polystyrene/magnetic beads are fluorescent detection covalently coupled with oligonucleotide probes. Following hybridization of PCR-amplified targets, fluorescent reporter is used to detect target-specific DNA by flow cytometry or charge-coupled device (CCD) imaging Electronic, Pre-synthesized probes are transported to specific locations fluorescent detection on the solid silicon surface by electric field and immobilized by streptavidin-biotin bonding. Either target DNA is fluorescently labeled and hybridized or fluorescently labeled probes are used in sandwich assays to detect target-specific DNA using CCD imaging Photolithographic synthesis of high-density DNA arrays on Mask-mediated in silica substrate by repetitive selective light exposure and situ, light-sensitive masking agents generating 25-mer chemiluminescent oligonucleotide probes. Each gene is represented by detection thousands of oligonucleotide probe pairs, consisting of a perfect match and a mismatch probe. Following hybridization of the target, the chip is stained with fluorescent reporter which is measured through confocal laser scanning “Lab-on-a-chip,” Plastic board with a mounted silicon-based microchip is fluorescent detection used for medium-density array with up to 400 sites. A combination of PCR, microarray, and microfluidics technologies is utilized to provide a “lab-on-a-chip” (LOC)

Spotted, label-free detection

Dunbar [6]

Barlaan et al. [7]; Kumar et al. [8]

LX100/200 and MAGPIX®

NanoChip XL®

VerePLEX™ Biosystem

http://vereduslabs. com

Beier and Hoheisel GeneChip® Scanner 3000 7G [9], [10]; Pease et al. [11] or GeneTitan™ HT Microarray System

Cordeiro et al. [5]

Verigene® System

Solid and Suspension Microarrays for Detection and Identification of Infectious Diseases 405

406

S. Dunbar et al.

available, methods for microarray data analysis, and descriptions of clinical applications for pathogen detection, genotyping, and antimicrobial resistance testing in the molecular microbiology laboratory will be reviewed.

Fundamentals of Microarray Platforms Substrates Substrates typically take the form of either chips (slides) or microspheres (beads). Common materials for chips include glass, silicon, nylon, plastic, and various metals, whereas bead materials consist of silica, agarose, polystyrene, and magnetic beads [12, 13]. The selected substrate may dictate not only the immobilization options available but also the fabrication and signal detection methods to be used. For example, SiOH groups (a silanol functional group in silicon chemistry) on the surface of a glass chip can be modified to react with amine or carboxylated probes for covalent immobilization. Similarly, a metal such as gold will allow bonds with thiol molecules. Polystyrene beads can be functionalized with carboxyl groups to allow for covalent attachment of nucleic acid probes. Whether a chip or a bead is used as the solid substrate can also influence the signal detection method used.

Probe Immobilization Probe immobilization is the method by which a probe or an individual nucleotide is physically bound to the selected substrate [14]. Immobilization techniques including non-covalent physical and affinity immobilizations, covalent immobilization on nanocones and metals such as gold, and 3D gel drops are described below and shown in Fig. 1. Immobilization by Physical Adsorption Adsorption utilizes non-covalent or electrostatic, polar, or hydrophobic attachment of charged molecules, as well as van der Waals interactions and hydrogen bonds, which will then allow binding at various points along the length of the probe to the substrate surface. The type of interaction is usually determined by the surface substrate selected [15, 16]. Probe adsorption was one of the first immobilization methods developed and, while simple, is also limited by the weakness of the non-covalent bonds (which can lead to desorption due to changes in temperature, pH, etc.) and the random orientation of the probes to the substrate surface. In order to have probes bind with significant strength, long molecules such as cDNA or PCR fragments are generally required. Another limitation is the incompatibility with the miniaturization of chips (substrates) utilized in today’s applications.

Solid and Suspension Microarrays for Detection and Identification of Infectious Diseases

407

Fig. 1 Schematic representation of various probe immobilization techniques. Immobilization of oligonucleotide probes (red) onto the substrate through various techniques (adsorption, covalent coupling, affinity, thiol, nanocone, and inclusion) is shown schematically

Covalent Immobilization Multiple methods of covalent immobilization have been developed. Many work with either the attachment between the sulfur atoms of a thiol (SH) molecule and a gold (Au) substrate (chemisorption), or similarly between amine (NH2) or carboxylated (COOH) probes and a modified glass substrate [17, 18]. While covalent immobilization techniques generally offer strong bond strength and good stability, disadvantages often include long hybridization times, high hybridization temperatures, or non-specific interactions. Crowding of the probes can also result in steric hindrance or non-specific binding, leading to low signal specificity coupled with high backgrounds. Another version of covalent bonding is click immobilization – named as such when it involves a cycloaddition reaction. An example would be the formation of a triazole ring in a copper-catalyzed reaction between azides and alkynes using agarose beads as the substrate [19]. Affinity Immobilization Either avidin or streptavidin can be used to bind with biotin to form a stable bond for immobilization [20]. Of the two, streptavidin is often preferred for its lower isoelectric point. To utilize this bond, the substrate may first be biotinylated using a cross-linker reagent followed by the addition of streptavidin, or the streptavidin may be immobilized to the substrate. The latter is a more complicated and a more costly process but in either case, the use of this bond tends to lower specificity and sensitivity relative to some of the other immobilization options available.

408

S. Dunbar et al.

Nanocone Immobilization To alleviate the crowding effect seen in some covalent immobilization techniques, the use of dendrons, which would occupy a relatively large space on the substrate but offer only one chemically addressable group to which a probe could be attached, was introduced by Hong et al. [21] The greater space between probes reduces nonspecific interference but the dendron structure offers little support for the probe so that it is less likely to remain relatively perpendicular to the surface of the substrate. Techniques to prevent or remove collapsed probes tend to produce very low sensitivity. Inclusion Immobilization In this method, probes are embedded in an insoluble gel. While the matrix of the gel will physically retain the probe, the smaller molecules of the targets and reagents are able to pass through the structure of the gel to access the probes [22]. The gel structure, therefore, dictates the minimum size of the probe, as probes that are too small will simply escape the matrix of the gel. An enhancement of this technique incorporates the use of covalent bonds between the probes and functional groups that are part of the gel structure. Akonni Biosystems utilizes inclusion immobilization in their TruArray® platform.

Microarray Fabrication In addition to the different combinations of components such as substrate, probe type, and immobilization method, microarrays can be assembled by a variety of methods [1]. Microarray technologies often utilized in microbial target detection include spotted (printed) arrays, electronic arrays, in situ synthesized arrays, and bead-based arrays, which come in either solid-state or suspension formats. Spotted (Printed) Arrays Spotted microarrays utilize pre-synthesized probes which are attached to the selected substrate through various delivery methods [23, 24]. The probe types include cDNA, PCR fragments, or short oligonucleotides with a combination of either covalent or non-covalent attachment chemistry (Fig. 2a). In a process known as contact printing, robots may dip small pin tools into a probe solution and “spot” the probe in the 0.5–12 nl volume range, leaving a print of a 62.5–600 μm diameter spot onto the prepared substrate. Pin tools may be solid, which require redipping prior to each application, or hollow, which allows them to hold a sufficient volume


409

Fig. 2 Schematic representation of microarray types. (a) Spotted array. This array relies on passive transport of the pre-synthesized oligonucleotide probes. (b) Electronic array. During immobilization, positive current is applied to specific test site to direct the movement of the negatively charged DNA oligonucleotide probe. (c) BeadChip array. This method utilizes self-assembly of microbeads coupled with specific oligonucleotide probes on a silicon substrate. (d–f) Three types of in situ synthesized arrays, based on (d) mask-mediated, (e) digital micromirror-mediated, or (f) inkjet-mediated fabrication methods

of probe solution through capillary action for multiple applications. Damage to the pin tools due to variations in the substrate surface can be an issue. Some spotting techniques employ the use of micro-stamps in which the probe solution is applied to the stamp and then mechanically stamped onto the substrate. Other technologies, collectively referred to as non-contact printing, deliver much smaller volumes of the probe solution to the substrate without coming into physical contact. Piezoelectric printing utilizes technology similar to that found in inkjets to deliver pico-droplets to the substrate without the risk of degrading the probes as may happen due to the high temperatures required by another non-contact application named thermic printing. Electrosprays utilize electric fields to simultaneously deliver droplets while lyophilizing the solution. After printing is complete, the slide may be modified (fixed) to make the surface more hydrophilic, thereby facilitating the hybridization process, but also to prevent non-specific binding of the DNA target to the glass slide during the hybridization.

410

S. Dunbar et al.

Electronic Arrays In the fabrication of electronic microarrays, electric currents are used to activate (positively charge) discrete sites containing streptavidin on the microarray substrate [25, 26]. By activating one site at a time while introducing negatively charged, biotinylated DNA probes, the probes are first actively attracted to the specific charged site and subsequently bound through the formation of a streptavidin-biotin bond at that location on the array (Fig. 2b). Once the probes have been bound, the electric field can be terminated from that location. By repeating this process, different probes can be attached to each of the sites available on the array. Target DNA is fluorescently labeled and introduced to the array. Hybridization is detected by fluorescence at the positive test site. This method is the basis for the NanoChip® XL by Savyon Diagnostics. BeadChip Microarray In fabrication of the BeadChip microarray, microbeads are coupled with a capture probe through the 5′-terminal base and then self-arrange on a specific region of a substrate chip that has been etched with microwells (Fig. 2c) [4]. Once in a well, the bead is held in place by van der Waals forces and hydrostatic interactions. Each discrete chip region is populated with beads representing a target-specific probe. Once prepared, the chip can be exposed to the target sample for hybridization. Following hybridization, a dNTP solution in which each of the four nucleotides has been tagged by a unique fluorophore is introduced. Specific hybridization of the target will support extension of the probe at the 3′-end by only one of the nucleotide- fluorophore combinations introduced. Laser imaging is then used to identify the target by measuring the fluorescent intensity of the desired fluorophore. Illumina’s BeadChip array utilizes this type of technology. In Situ Synthesis In contrast to spotted, electronic, and bead-chip microarrays, for in situ microarray synthesis, the probes are created by the base-by-base application of nucleotides directly onto the substrate (Fig. 2d–f) [10]. This method is able to produce very high-density microarrays. Another contrast is the direction of the probes, which are created from the 3′ end with the terminal 3′-base covalently bound to the solid surface. The bases that are added to create the oligonucleotide probes have a protective group such as DMT (dimethyloxytrityl) or other blockers on the 5′ end to prevent the addition of more than one base during each round of addition. Simply, the process of synthesizing the probes can be imagined to alternate between the conversion of the 5′ protective group (deprotection) to an OH group on the terminal base to allow the subsequent addition of a new base (protection) in a 3′ to 5′-direction.


411

The deprotection stage can be achieved in one of three ways: • Mask-mediated photolithography. • Maskless photodeprotection using digital micromirrors. • Maskless inkjet-mediated arrays. Mask-Mediated Photolithography Photolithographic masks are the basis of the Affymetrix® GeneChip® technology. Deprotection is achieved through exposure to UV light, and a physical barrier (chromium mask) is used to protect the 5′-terminal nucleotide of the probe (Fig. 2d) [9, 11]. Following the selective exposure of light to a specific array location (feature), the nucleotides that have been deprotected will add a single new nucleotide to their length when nucleotides are made available. Once the locations for the various probe sequences on the microarray are identified, a set of masks are designed for the application of the four nucleotides. By utilizing a different mask at each round of deprotection and exposing predetermined locations on the substrate to the UV light, unique oligonucleotide sequences can be achieved for each location on the substrate. For a probe of 25 nucleotides in length, a set of 100 masks are required per microarray. While the creation of the physical masks is expensive and must be done in advance, once produced this technology allows the mass production of large numbers of identical microarrays through the repeated use of the same masks. Maskless Photodeprotection Using Digital Micromirrors Digital micromirrors also use light to deprotect array features, but rather than a physical barrier preventing light from reaching certain probes, micromirrors are used to precisely direct the application of the UV light – thereby making those probes available for the addition of a new base (Fig. 2e) [27]. The absence of physical masks makes this method better suited to applications requiring smaller batch sizes for a given high-density array layout. This technology is the basis for the Roche NimbleGen Maskless Array Synthesizer (MAS). Maskless Inkjet-Mediated Array Chemicals applied via inkjet technology are the basis for Agilent Technologies’ microarrays (Fig. 2f) [2]. Agilent utilizes the precise application of trichloroacetic acid by an inkjet-style printer to deprotect DMT-protected nucleoside phosphoramidites to generate the desired probe sequences on specific array locations, one base at a time. Similar to the digital micromirror, the inkjet technology is very flexible, allowing for the creation of small batch sizes, but neither method is able to yield the same efficiency as the photolithographic mask technology for large batches.

412

S. Dunbar et al.

Suspension Bead Microarray The previously described technologies all rely on planar substrates, such as silicon or glass chips, which are fixed and therefore identify each probe based on its precise location on the chip. However, it is possible to create suspension microarrays with the use of microparticles suspended in solution. Suspension bead microarray technologies are three-dimensional arrays that use microparticles (microspheres or beads) as the solid support for the binding reaction and CCD imaging or flow cytometry for detection of the bead and associated target (Fig. 3). The particles can be distinguished into subsets by different classification features, such as internally absorbed fluorophores, size or diameter, and surface or composition characteristics. The xMAP® Technology platform by Luminex® utilizes polystyrene microspheres as the solid substrate for the covalent attachment of probes and subsequent hybridization with the target [6]. The microspheres are internally dyed with two or three spectrally distinct fluorochromes, and by using precise amounts of each of these fluorochromes in combination, an array is created consisting of different bead sets with specific spectral addresses. A specific capture oligonucleotide probe is covalently coupled to the surface of the specific bead sets, or bead sets precoupled with unique capture oligonucleotides are commercially available for microarray development. The target nucleic acid is captured by hybridization to the complementary probe sequence and labeled with a fluorescent reporter molecule to allow detection and quantitation of the target binding that has occurred at the particle surface. Multiple readings are made per bead set, providing a fluorescent output signal of the results.

Signal Detection Methods Signal detection traditionally relies upon fluorescence, chemiluminescence, or colorimetry, but more recently label-free methods have been commercially developed. Fluorescence The use of fluorophores is the most common method of target detection, and the development of new fluorophores continues to offer improved stability and sensitivity. However, the relatively weak signals generated by fluorophores necessitate a device capable of gathering very small amounts of light and resolving them in fine detail. For this reason, traditional fluorescent scanning devices tend to be relatively large machines that only scan a limited number of microarray features at a time. While microarrays have been miniaturized, the size of the fluorescent reporter molecules limits the miniaturization of microarrays and associated scanners using this signaling method and makes their use in a portable or a point-of-care solution challenging.

Solid and Suspension Microarrays for Detection and Identification of Infectious Diseases Fig. 3 Suspension bead array. Microscopic beads are dyed with different concentrations of red and infrared dyes to create up to 500 bead sets with unique spectral identities. Bead sets are coupled with specific capture molecules, such as oligonucleotide probes, which can then be hybridized to labeled targets prepared (amplified) from the original specimen. Hybridized bead sets are labeled with a fluorescent reporter dye, and the suspension array is analyzed by CCD imaging or flow cytometry to identify each target and quantify the amount bound to the bead surface

413

414

S. Dunbar et al.

Chemiluminescence Another method of signal detection can be achieved through chemiluminescence, which is based on generation of photons by a chemical reaction [28]. ThermoFisher Scientific utilizes this interaction in their GeneChip® system for a variety of DNA and RNA analysis applications [29]. However, similar to fluorescence, scanners of considerable size are required to detect the light emitted. Colorimetry Another common signal detection method used for microarrays is colorimetric analysis to determine the presence of the target. Colorimeters measure the absorbance of a light wavelength which will vary with the concentration of the target molecule. Genomica’s CLART® microarray platform and Alere’s ArrayTube technology are examples of colorimetric signal detection in commercial applications [30]. Label-Free Detection Several label-free detection methods have been developed offering the advantage of a simplified process and the potential of a portable system. Surface Plasmon Resonance Imaging (SPRi) Surface plasmon resonance (SPR) is a refractive index (RI)-based detection method. This technique works through the excitation of surface plasmon polaritons in a thin metal film caused by changes to the RI of molecules bound to the surface above the film [31]. The RI of the bound molecules changes as their mass changes during coupling. A glass substrate is coated with gold and subsequently spotted with target-specific probe molecules and immobilized onto the substrate, forming the basis of the microarray. When placed into an appropriate instrument, the target is introduced and hybridized to the probes. SPR is used during the hybridization process and measures changes to the RI of light being applied to the chip at a given location, thus distinguishing between unhybridized and hybridized probes. SPRi has been incorporated into several commercial applications, such as platforms from Horiba Scientific and GE Healthcare Biacore™, which use this label-free technology.


415

Verigene® NanoGrid Technology Recent advances in integrating biomaterial such as DNA structures, peptides, and antibodies with nanomaterial such as gold nanoparticles (AuNPs), carbon nanotubes (CNTs), magnetic nanoparticles, and graphene have provided a unique platform for the development of next-generation lab-on-chip (LOC) point-of-care diagnostic applications [30]. One commercial example of a label-free technology that has opened a door for the use of microarrays in point-of-care testing is the Luminex® Verigene® NanoGrid technology [5]. The NanoGrid system uses target-specific single-stranded DNA capture probes which are spotted onto a glass chip substrate (Fig. 4). Within the processing instrument, sample DNA is introduced to the array and put through a denaturation process, followed by hybridization where target DNA binds to the capture probe on the chip. Once hybridization of the sample is complete, two additional reagents are added to facilitate a sandwich hybridization and allow detection of the bound targets through signal amplification. The first reagent is a poly-A tagged target-specific mediator oligonucleotide that is complementary to a portion of the denatured sample DNA that is not bound to the capture probe during initial hybridization. The second reagent is a gold nanoparticle functionalized with a poly-T tail. Once both of these reagents have been bound through an additional hybridization process, a silver solution is added to the chip. The silver adheres to the gold nanoparticles creating a larger particle that easily

Fig. 4 Luminex® Verigene® NanoGrid technology. A double target-specific capture and mediator probe system is used to identify targets within a biological sample. Following hybridization, a sandwich Capture:Sample DNA:PolyA-Mediator assay is detected by poly-T-functionalized gold nanoparticles, with signal amplification process involving silver deposition

416

S. Dunbar et al.

scatters LED light applied to the array’s surface. A camera captures the light scattered off any silver particles, indicating the presence of the specific target. The NanoGrid technology offers increased sensitivity to that of fluorophores, as well as the additional benefits of non-toxicity and a long shelf-life.

Microarray Data Analysis Microarray data analysis can be divided into the core steps of image processing, transformation, and normalization of data (including how gene expression data is expressed, transformed, normalized, and compared). All available public and commercial microarray software follow these core steps. Data analysis for microarray technologies that produce nucleotide sequences (e.g., Illumina technology) will not be discussed as it lies within the scope of next-generation sequencing (NGS) data analysis; however, RNA sequence analysis would be conducted as described in sections “Transformation of Expression Ratio, Normalization for Comparison Between Samples and Composite Microarray Data Analysis”

Image Processing and Analysis The primary method by which data is extracted from the captured images forms the foundation of any subsequent analysis. An image is generated following the completion of hybridization on the microarray in a Tagged Image File Format (TIFF). In the next step, the spots are identified based on the arrangement of the sub grids or pen groups (group or array elements deposited by a single spotting pen) [32].

Determination of Local Region There are two methods for identifying the spot signal. The first method uses an area of fixed size that is centered on the spot and only pixels within this fixed area are included. In the second method, the software identifies the boundary for a given spot and thus includes pixels within the boundary. The first method is the less challenging of the two from a computational standpoint as the boundary is always a fixed radius from the center of the spot. However, the data produced can be misleading and lead to inaccurate calculation of the spot signal and the background signal. While the second method does not suffer from this challenge, it does require more computational power.


417

Background Subtraction Once the spot and background areas have been defined and measured, the mean, median, and sum values for the spot signal and the background signal are recorded. Most algorithms take the median background signal and subtract it from the median spot signal to report the true spot signal. Using the median value safeguards the reported signal from being skewed by extreme outliers at either end; however, this method requires the precise measurement of both spot and background signals. The standard method is global background subtraction and is calculated through the mean or median value of the negative controls. This method subtracts the mean/ median negative control value from all of the spots. Local background subtraction is usually used for spotted cDNA microarray where there may be variation in the array surface. This method calculates the background intensity for each spot in the array by the signal derived from the pixels in the surrounding spots.

Expression Ratios In the two-channel microarray system using Cy5 (Red) and Cy3 (Green)-labeled cDNA or oligonucleotide probes for gene expression level measurements, an intensity ratio can be generated by measuring the amount of red fluorescent light emitted and captured on the image and dividing it by the amount of green fluorescent light emitted and captured. The difference between red and green light signal can be defined as a ratio between two median signals calculated as described above and shown in the equation below: spot background Rmedian − Rmedian Qmedian = spot background Gmedian − Gmedian In this example only two colors are used, but with current microarray technology, samples can be labeled using more than two fluorophores. However, a two-color approach in which the ratio of signals on the same array is measured is much more reproducible [33].

Transformation of Expression Ratio In the two-channel system, transformation of the expression ratio is also necessary due to the misleading absolute difference between upregulated (higher ratio of red signal compared to green) and downregulated (lower ratio of red signal compared to green) ratios. On a linear scale, genes that are upregulated will be expressed with a ratio between 1 and positive infinity, whereas downregulation ratios fall between 0

418

S. Dunbar et al.

and 1. Therefore, a tenfold upregulation will appear to be significantly further away from the normal expression ratio of 1 compared to a tenfold downregulation. Traditionally, the transformation of the downregulation expression ratio is done through inverse transformation. In cases of upregulation, the fold change is equal to the expression ratio, and in cases of downregulation, the fold change is equal to the reciprocal of the expression ratio multiplied by −1. This method can be problematic as there is a discontinuity between −1 and + 1. Presently, the use of a logarithmic transformation using a logarithm base 2 value of the expression ratio [log2(expression ratio)] is more common. The advantage of a logarithmic transformation lies in its continuity so that upregulation and downregulation scales are proportionally comparable. However, the disadvantage of logarithmic transformation is that it removes absolute expression (e.g., 500/100 and 50/10 will have the same expression ratio). A comparison of expression ratios and transformations are shown in Table 2.

Normalization for Comparison Between Samples In order to compare two samples from different array preparations that may have different hybridization efficiencies and different quantities of starting material, normalization is essential. The elimination of the systemic variation that affects measurements of gene expression while leaving the indicators of biological variation intact allows the signal intensity between various microarray slides to be compared and integrated into a single analysis [34]. Normalization methods include using statistical descriptors (median, mean) and the use of control probes or housekeeping genes [35]. Statistical descriptors are often used for total/global intensity normalization. The primary assumption is that the total starting material is the same for both samples and that an equal number of molecules are hybridized to the microarray. Total/ global intensity normalization is performed by either division or subtraction of the statistical descriptors (e.g., the background median value is subtracted from all intensity values). Table 2 Comparison of expression ratios and transformations

Upregulation

Downregulation

Expression ratio (R/G) 8 4 2 0.5 0.25 0.125

Inverse transformation 8 4 2 −2 −4 −8

Logarithmic transformation (base 2) 3 2 1 −1 −2 −3


419

Normalization using control probes or housekeeping genes is the technique of including spots of known intensities or genes of a known expression level. Here the assumption is that the control probes or housekeeping genes are constant across all samples and all spot intensities can be normalized in reference to the intensity of the control spots. Other normalization methods include linear regression, nonlinear regression, and various combinations of the two methods mentioned above [36–38].

Composite Microarray Data Analysis Following the spot signal analysis steps described above, clustering is performed to display a differential signal profile between the specimen in question and the control sample, while reducing false discovery rate [39]. To do this, the data are usually represented in a matrix format and a method to quantify the similarity and dissimilarity of the data points is required with each row being the data for a single gene and each column being data for a single comparison or time point. The use of clustering for genes with similar expression profiles can be classified into two main branches: supervised and unsupervised [40]. In supervised learning, annotations are used to separate expression profiles into either “condition state” or “control state.” Then the two profiles are separated and analyzed for patterns, allowing a model to be developed based on the labels assigned. In unsupervised learning, the expression data matrix is analyzed such that the genes/ samples are grouped into clusters without annotation, thus eliminating the influence that external information might impute to the data. Cluster analysis under unsupervised learning is comprised of two aspects: a distance measurement and a cluster algorithm. The two most common ways to measure distance is to determine the Euclidean distance (ED) or the Pearson correlation coefficient (PCC). ED measures the “distance” between two variables and PCC measures the linear similarity or linear correlation between two variables. For example, ED would measure the magnitude of the difference between the variables, while PCC would measure the trend or the similarity of the direction between the variables. PCC essentially creates a line graph to represent the data (Fig. 5). The types of cluster algorithms available can be divided into two types: partitioning and hierarchical. The partitioning method divides the data into clusters that are a prespecified number “k” in size. Examples of the partitioning method include the k-means algorithm (utilizing centroids), fuzzy clustering, self-organizing map (SOM) algorithm, and the partitioning around medoids (PAM) algorithm [41–43]. The hierarchical method of clustering produces a dendrogram and does not require a prespecified number “k.” When determining which algorithm to use, it is important to understand the implications of the various alternatives. Partitioning has the

420

S. Dunbar et al.

Fig. 5 Euclidean distance and Pearson correlation coefficient graphs representing different expression profiles from different samples Table 3 Publically available microarray data analysis software Software OMICtools – a centralized database of microarray tools useful for various microarray types (e.g., ChIP-on-chip, Antibody, SNP) Cluster and Treeview – Cluster performs a variety of cluster analyses including hierarchical and partitioning clustering. Treeview allows results to be graphically displayed ExpressYourself – Integrated platform for processing microarray data. Includes features to regenerate images of the original microarray after applying various data processing steps. Facilitates identification of position- specific artifacts Expression Profiler – a set of tools for clustering, analysis and visualization of gene expression and other genomic data. Includes a feature for gene ontology categories and protein interaction studies J-Express – a package to facilitate analysis of microarray data. Includes features for hierarchical clustering and principal component analysis

Source https://omictools.com/ microarray-category

References Henry et al. [45]

http://homer.ucsd.edu/ homer/basicTutorial/ clustering.html

de Hoon et al. [46]

http://array.mbb.yale. edu/analysis/

Luscombe et al. [47]

http://ep-sf. sourceforge.net

Kapushesky et al. [48]

http://jexpress.bioinfo. Stavrum et al. no/site/downloadMain. [49] php

advantage in a controlled experiment where the initial “k” is known (i.e., working with known set number of variables), but is a disadvantage if “k” is not known. Partitioning is more computationally intensive than a hierarchical approach, and therefore, with a given data set, the hierarchical method would be computationally faster. However, the hierarchical method is rigid and the groupings made in the earlier stages cannot be corrected later and can cause unrelated genes to be grouped together at the endpoint [44]. Table 3 lists of some of the publically available microarray data analysis software.


421

Applications of Microarrays for Diagnostic Microbiology Conventional methods for detecting pathogenic microorganisms such as bacteria, viruses, and fungi mainly rely on culture-based testing that can take days or sometimes weeks to deliver results. Diagnosis by culture can also be limited by the presence of uncultivable bacteria in the sample, inappropriate culture conditions, and the presence of inhibitory substances in the patient sample [50]. Additionally, culture methods can be insensitive and difficult to perform and require extensive technologist training and may be unsuccessful due to an organism’s growth requirements. With a constant decline in the number of specialized microbiologists over the past few years, molecular methods such as microarrays, sequencing, and real-time PCR have emerged as valuable techniques for identifying pathogenic organisms that are traditionally identified by culture [51]. These tests are particularly suitable for pathogens that are not routinely or easily cultured and that are dangerous to culture and those for which serologic testing is difficult to interpret. However, sequencing by high-throughput methods can miss organisms at low concentration when coverage is insufficient, and is expensive and time-consuming, making it difficult to use for routine diagnosis. Real-time PCR, on the other hand, is inexpensive but has limited capacity for multiplexing and is not well-suited for discovering novel species or strains due to its intolerance toward primer-target mismatch [52]. In recent years microarrays have become a viable alternative, providing large-scale screening and more comprehensive target detection within a 24-h or less timeframe and at a moderate cost. Microarray technology has been used for over a decade for the simultaneous detection and identification of various microbial pathogens, pathogen discovery, antimicrobial resistance monitoring, strain typing, and monitoring host-pathogen interactions [1]. High-throughput microarrays have been successfully developed for simultaneous identification and genetic characterization of both cultivable and fastidious microorganisms in a single assay [53]. Identification of antibiotic-resistant strains using species-specific probes is also possible and can potentially alter treatment decisions [54]. With the increase in the availability of the complete genome sequences of many disease-causing organisms, a multitude of probes can be designed for detecting a broad spectrum of pathogens [55]. Low- or mid-density arrays are also used for epidemiological investigation and molecular typing of bacterial pathogens [56]. Recently, guidelines have been developed for validation, verification, quality control, and interpretation of results for microarray methods to be used for the diagnosis and monitoring of infectious diseases in clinical laboratories [57].

422

S. Dunbar et al.

Pathogen Detection Microarrays for pathogen detection are designed with a combination of high- specificity probes and probes targeting conserved regions of a species. The level of resolution offered by the microarray assay primarily depends on the degree of conservation of the marker gene and the length of the target oligonucleotide probe [58]. Selection of the ideal probe or probe sets and the ability to amplify the regions of interest using conserved primers is important for increasing the sensitivity and spectrum of detection of the assay. Ambiguous sequences or those highly similar to nontarget sequences can result in cross-hybridization and produce inaccurate results [59]. As it is not usually possible to identify a single target gene or genetic region that could allow detection of all organisms of interest, microarrays are usually designed to target one or more classes of organism. Some commonly used target genes are listed in Table 4. The 16S rRNA gene is one of the most commonly analyzed Table 4 Conserved gene targets used in microarray applications Gene 16S rDNA 23S rDNA 16S-23S rRNA internal transcribed spacer gyrB

Function Organism class Protein synthesis Bacteria, mycobacteria Protein synthesis Bacteria, mycobacteria Protein synthesis Bacteria, mycobacteria

References Liu et al. [60]; Negoro et al. [61]; Troesch et al. [62] Schnee et al. [63]; Yoo et al. [64]

Topoisomerase

Roth et al. [68]; Kakinuma et al. [69]; Fukushima et al. [70] Antwerpen et al. [71]; Yao et al. [72]; Caoili et al. [73] Yao et al. [72]; tang et al. [74]; Zhang et al. [75]

hsp65

Bacteria, mycobacteria RNA polymerase Bacteria, mycobacteria Mycobacteria Catalase- peroxidase Protein coding Chaperonin Mycobacteria

mecA cpn60

Protein coding Chaperonin

Bacteria Bacteria

parE

Topoisomerase

Bacteria

recA groEL sodA

DNA repair Chaperonin Superoxide dismutase Protein synthesis

Bacteria Bacteria Bacteria

rpoB katG, inhA

18S and 28S rRNA internal transcribed spacers

Fungi

Nubel et al. [65]; Keramas et al. [66]; Wang et al. [67]

Zimenkov et al. [76]; Ringuet et al. [77] Cleven et al. [78]; Zhu et al. [79] Masson et al. [80]; Maynard et al. [81] Roth et al. [68]; Jarvinen et al. [54] Champagne et al. [82] Hu et al. [83] Giammarinaro et al. [84]; Kooken et al. [85] Spiess et al. [86]; Huang et al. [87]; Campa et al. [88]; Landlinger et al. [89]


423

biomarkers for development of diagnostic microarrays [90]. The functional consistency of 16S rRNA enables it to serve as an evolutionary clock and taxonomic marker in bacterial systematics [91]. Several studies have described the development of broad panel microarrays using the 16S rRNA gene for detection and identification of bacterial pathogens [92, 93]. Another commonly used target is the 23S rRNA gene which offers the same advantages as the 16S gene, but demonstrates more variation between species [94]. In addition to these highly conserved nucleic acid sequences and phylogenetic markers, genes encoding a particular toxin or virulence factor and species-specific insertion elements can also be used for identifying pathogens [95]. Identification of Culture Isolates Microarrays can play a central role in diagnosis of organisms isolated by culture- based methods, particularly those which are unusual or difficult to identify by conventional tests. The performance of these assays depends on the effective selection and design of the pathogen-specific probe and the successful extraction and amplification of the organism’s DNA from the clinical isolate. Microarrays have been developed for the identification and differentiation of Mycobacterium tuberculosis complex (MTC) as well as nontuberculous mycobacterial species from positive culture isolates [70, 96, 97]. These assays are also useful for detection of mixed infections or laboratory contamination with environmental mycobacteria. For species-level detection of mycobacteria, various rapid and sensitive commercial line probe assays are available that use PCR amplification of the 16S to 23S rRNA spacer region of Mycobacterium species [98–100]. Detection of drug-resistant strains of M. tuberculosis (MTB) is important for administering appropriate antimycobacterial therapy. Numerous studies have evaluated microarrays detecting point mutations and other rearrangements in the rifampin resistance-determining region of the rpoB gene [72, 101–103]. Microarrays are also useful for the comprehensive coverage of a species and detection of virulence- and resistance-associated genes in organisms like Staphylococcus aureus and aid in investigation of the variation, evolution, and epidemiology of the pathogen [79, 104, 105]. Comparison microarrays along with sequencing for selected strains have been used for molecular typing and genomic analysis to detect genotypic changes in organisms [106]. Microarrays can also be applied for simultaneous detection of food and waterborne pathogens [81, 107– 109]. In addition, microarrays have been developed for detection and differentiation of a wide range of fungal pathogens, including different strains of Candida species and fungal pathogens involved in invasive mycoses [64, 88, 110–112]. Arrays that use conserved targets within viral families can identify atypical or highly pathogenic viruses isolated in culture [113, 114]. Sepsis is one of the most common reasons for ICU admission and plays a significant role in hospital mortality [115]. In clinical laboratories, blood culture using automated systems followed by routine microbiological biochemical tests is consid-

424

S. Dunbar et al.

ered the gold standard even though periodical false negative results might be reported due to inappropriate growth conditions or prior antimicrobial therapy [116]. Molecular techniques such as fluorescent in situ hybridization, MALDI-ToF, and microarrays have been widely implemented for rapid detection of a wide variety of microorganisms from positive blood cultures, where higher concentration of the pathogen ensures accurate identification [117]. Numerous studies have evaluated the performance of commercially available microarray assays for detection of Gram-positive and Gram-negative organisms and resistance genes directly from blood culture bottles [118–122]. The Gram-positive panels can detect the presence of resistance genes such as mecA for staphylococcal methicillin resistance and vanA for vancomycin resistance, whereas the Gram-negative panels can identify multidrug-resistant Pseudomonas aeruginosa and extended-spectrum beta- lactamase or carbapenem-resistant Enterobacteriaceae [123–129]. Various studies have evaluated the potential clinical impact of implementing these panels in terms of time to results, economic benefits and overall patient outcome. A reduction in the length of hospital stay and mean hospital cost has been observed when pathogens are identified and differentiated at earlier stages of infection [130–132]. Incorporating these highly sensitive panels in the testing algorithm increases clinical confidence, resulting in the timely administration of appropriate, narrower-spectrum antibiotic therapy [133, 134]. However, input from pharmacy practitioners is essential to realize the maximum clinical benefit that could be obtained by earlier identification and detection of bloodstream pathogens and resistant markers [135]. Application of Microarrays to Patient Samples Respiratory Infections The detection of causative pathogens responsible for respiratory infections has gained importance over the last decade. Rapid and accurate identification of the etiological agents provides a correct diagnosis and can result in better patient care and more appropriate antibiotic therapy, as well as reduces the risk of nosocomial transmission within healthcare facilities [136]. Detection of common and novel respiratory pathogens can also help in the development of new epidemiological survey systems for respiratory viral infections [137]. It has been observed that detection systems based on RT-PCR followed by low-density microarray analysis are a sensitive and specific platform suitable for rapid detection and type/subtype identification of pathogens for common and novel respiratory viruses [138, 139]. With a constant rise in the number of significant pathogens and subtypes, use of FDA-cleared commercial multiplexing panels has become routine in clinical laboratories. All of these platforms provide broad pathogen coverage for detecting multiple viruses but differ in the range of pathogens detected, ability to subtype, workflow, and ease of use [140, 141]. Numerous studies have established that commercial microarray platforms, which improve rate of diagnosis over culture and


425

laboratory-developed real-time PCR assays, can be used to diagnose pathogens in different patient populations and are valuable for patient management and public health epidemiology [142, 143]. Additionally, comparative studies evaluating assay performance have observed more than 80% concordance between different commercial platforms and greater than 90% sensitivity and specificity in most cases [144–146]. However, some differences in assay sensitivities have been observed for specific viral targets such as human rhinovirus/enterovirus and adenovirus [143, 145, 147–149]. In addition to viral targets, some platforms also detect bacterial targets, such as Streptococcus pneumoniae, Chlamydophila pneumoniae, Mycoplasma pneumoniae, Haemophilus influenzae, Bordetella pertussis, etc., and they can be adapted to diagnose both upper and lower respiratory tract infections [68, 150, 151]. Clinicians may often fail to consider the presence of additional etiological agents and in many cases order diagnostic tests for a single pathogen. It has been observed that coinfection of respiratory syncytial virus and human metapneumovirus with other respiratory viruses is common and can increase the severity of the disease [152, 153]. Various studies have demonstrated that these multiplex molecular microarray assays can also efficiently identify coinfections that are otherwise missed by traditional detection methods [143, 154]. In addition to assay performance, the turnaround time, ease of use, and cost are equally important factors to consider when choosing a microarray platform, as this is important for overall savings in technician time and hospital stay. Numerous studies have compared these parameters for commercial panels [144, 148]. The ease of use, significantly better performance, and availability of rapid results (especially for the inpatient setting) often outweigh the higher cost per test associated with these panels [155]. In many cases, implementing a molecular microarray platform is more cost-effective than traditional methods as it simplifies the workflow and in turn reduces labor cost [156, 157]. As larger microarrays are usually designed with a combination of detection and discovery probes, they tend to be robust assays and are ideal for pathogen characterization and molecular epidemiology investigations and can detect novel organisms on the basis of conserved sequence homology [52, 158, 159]. However, these panels are not yet widely implemented in clinical laboratories as they are expensive, are difficult to develop and validate, and require optimization of several factors, such as sample amplification, probe specificity, and interpretation strategy [160, 161]. Resequencing microarrays use closely overlapping probe sets to determine a target organism’s nucleotide sequence and are ideal for epidemic outbreak investigation, single nucleotide polymorphism (SNP) genotyping, and phylogenetic analysis [160]. These arrays have been used with respiratory pathogens to differentiate between similar related organisms and provide information on mutational hot spots and strain variants [162, 163]. However, these arrays have been found to have lower sensitivities due to their short oligo probes, can only detect a limited range of organisms on a single array, and cannot identify novel organisms [51].

426

S. Dunbar et al.

Fungal respiratory infections are frequent in immunocompromised individuals, such as transplant patients, and result in increased rate of mortality [164]. Early and rapid diagnosis of systemic fungal infection remains challenging as cultivation of fungi from clinical samples is difficult and time-consuming, demonstrates low sensitivity, and requires extensive experience for correct identification [165]. Microarrays can aid in reliable and rapid species-specific detection of a wide spectrum of clinically important fungal pathogens such as Aspergillus spp., Cryptococcus spp., and Candida spp. in a single assay, which facilitates the selection of the most effective antifungal treatment [89]. Furthermore, it has been shown that microarrays are suitable for identifying a wide spectrum of clinically relevant fungal pathogens causing invasive infections in immunocompromised patients, including uncommon and drug-resistant fungi [86, 87]. Nontuberculous mycobacterial pathogens are commonly present in the environment and can cause pulmonary disease via aerosol inhalation [166, 167]. Highly sensitive and specific commercial line probe microarrays have been proven effective in identifying a variety of these mycobacteria at the species level [99, 168]. Gastrointestinal Infections According to the Centers for Disease Control and Prevention (CDC), about 48 million Americans suffer annually from foodborne diseases, resulting in 128,000 hospitalizations and 3000 deaths [169]. Microarray methods that can simultaneously identify a wide range of common and uncommon gastrointestinal pathogens in a single test can significantly simplify testing and deliver accurate results in a timely manner [170, 171]. Various studies have concluded that microarray test methods for gastrointestinal pathogens exhibit better sensitivity and specificity and can detect a higher number of pathogens in less time than conventional methods [172–176]. The faster turnaround time of these assays caters to the needs of high volume molecular laboratories and potentially impacts several medical decisions, resulting in better patient management and public health response [177]. Gastrointestinal microarray panels are also cost-effective and able to detect coinfections and may provide valuable information regarding the genotype of the infecting species or subspecies [178–183]. In addition, these microarrays can be useful for epidemiological surveillance and for diagnosing gastrointestinal infections in endemic geographic regions [184, 185]. Bloodstream Infections Conventional methods for identifying microorganisms causing bloodstream infections involve cultivation of the pathogens from blood with subsequent morphological and physiological characterization. These methods can be time-consuming, labor-intensive, and often incapable of providing results in a timely manner.


427

Microarray panels have demonstrated rapid detection of pathogens from positive blood cultures, thus reducing time to result and providing diagnostic information that can lead to reduction in length of hospital stay [130]. However, to realize the full benefit of microarray-based assays in sepsis diagnosis, fast and reliable detection of pathogens directly from patient blood samples is needed. Microarrays have the advantage over conventional methods in the ability to detect static or dead cells before genome degradation, which enables them to identify the causative organisms even after empirical antimicrobial therapy has been administered [116]. Furthermore, these panels can include fastidious microorganisms that are difficult to grow in culture [186]. The two major limiting factors for detection of microorganisms directly from whole blood samples are the presence of a low number of circulating organisms (1–10 CFU/ml) and the presence of a high quantity of human DNA that might interfere with microarray primers and probes [187]. It is technically difficult to extract large amounts of blood, and the low sample input volume during PCR might affect the sensitivity of the assay [188, 189]. A number of microarray-based assays have been developed for the detection of various human pathogens from whole blood samples [116, 190, 191]. Various studies have evaluated the detection capability, as well as sensitivity and specificity, of these panels over conventional culture methods and found that microarray panels detect a relatively higher number of organisms than blood culture, especially in immunocompromised patients and patients undergoing antibiotic treatment [61, 192]. Using gene-specific PCR products as capture probes, sensitive chip-based arrays have been developed that can simultaneously identify pathogens and characterize them for antibiotic resistance and virulence [78]. In cases where cellular necrosis from severe physical injury initiates inflammatory signals resulting in severe innate immune response- related sepsis, microarrays can be used to evaluate disruptions in circulating gene expression profiles that can aid in determining the status of the patient’s immune system [193]. However, despite the numerous benefits of these novel pathogen detection methods, trial data demonstrating clinical reliability of these assays in the diagnosis of bacteremia directly from blood samples is currently unavailable [188]. Central Nervous System Infections Microarrays with 16S rRNA PCR amplification have been widely implemented in detecting bacterial meningitis. These assays are rapid and highly sensitive and detect a higher number of cases as compared to conventional culture methods [60, 194, 195]. However, presence of any low-level sample contamination could increase the chance of reposting false-positives, which could result in a different treatment strategy [196]. Microarrays can also detect the most common neurotropic viruses such as herpes simplex virus, varicella zoster virus, and enteroviruses and exhibit greater than 95% sensitivity and specificity [197–199]. High-throughput specialized microarrays with genus- and species-level detection capability have also been developed for identifying highly pathogenic viruses such as chikungunya virus,

428

S. Dunbar et al.

yellow fever virus, encephalitis virus, and hantavirus [114]. In cases of cryptococcal meningitis, detection and differentiation of Cryptococcus neoformans and Cryptococcus gattii at the species, genotype, and hybrid levels from CSF specimens are possible. However, optimization of the DNA extraction method is required before the assay can be routinely used in clinical laboratories [200]. Urogenital Tract Infections Urinary tract infections (UTIs) are one of the most common reasons for prescribing antibiotics and have led to increased antimicrobial resistance among uropathogens [201]. Uropathogens are also commonly present in asymptomatic persons, depending on the population, and thus, it is important to treat only patients who have a significant bacterial load in their urine [202]. However, most of the microarray-based assays implemented for detection of uropathogens are qualitative and/or not fully quantitative, which limits the ability to differentiate a true pathogen in urine containing mixed flora [203]. Most studies have utilized species-specific oligonucleotide microarrays for genotyping and focus on the detection of resistant pathogens such as fluoroquinolone-resistant Escherichia coli, carbapenemase- resistant Enterobacteriaceae, or nosocomial bacteria such as Pseudomonas aeruginosa and Acinetobacter baumannii [204–208]. Currently, biosensors integrated with microfluidics have been applied to the diagnosis of UTIs [209]. These platforms eliminate the steps required for sample processing and thus can aid in delivering direct results at the point of care. Application of this technology in conjunction with microarrays could be very promising for diagnosing UTIs and guiding personalized treatment decisions. According to data reported in 2008, it has been estimated that sexually transmitted infections (STIs) affect about 110 million people in the United States annually and more than 20% are young men and women aged between 15 and 24 years [210]. STIs often appear to be asymptomatic in the early stages of infection, thus making reliable, timely diagnosis difficult [211]. Multiplex PCR-based microarrays using highly specific and efficient primers for common STI pathogens such as Neisseria gonorrhoeae, Chlamydia trachomatis, Mycoplasma genitalium, and Ureaplasma spp. have been developed for use in routine surveillance of infections caused by these pathogens [211–213]. The detection and differentiation of high-risk and low-risk human papillomavirus (HPV) is extremely important because of the strong association of certain HPV types with cervical cancers [214]. HPV is the most common STI in the United States and is associated with both malignant and benign tumors of the cutaneous and mucosal epithelia. According to current evidence, HPV DNA testing is more effective than cytology for cervical cancer screening and has led to the development of highly sensitive HPV tests [215]. Detection and typing of HPV genotypes using various microarray formats have been used to assess the pathogenesis and prognosis of cervical neoplasia and to aid in the planning of vaccination trials [216–219]. Several highly sensitive commercial microarray assays have been developed that


429

detect high-risk HPVs from cervical swab specimens and most of them demonstrate equal or better sensitivity as compared to the first FDA-cleared HPV test, Digene Hybrid Capture 2 [220–224]. Currently, no microarray-based assay has been FDA- cleared for primary HPV screening. Skin, Soft Tissue, and Joint Infections Skin, soft tissue, and joint infections may occur frequently after an invasive procedure and often require hospitalization. Microarrays have been widely used to simultaneously detect and differentiate fungal pathogens from cutaneous skin infections, as well as from superficial and invasive mycoses [88, 112, 225, 226]. Diagnosis of prosthetic joint infections (PJIs) is clinically challenging, and microarrays can play an important role in the diagnostic algorithm [227]. Microarrays targeting the most common pathogens involved in PJIs have been shown to be highly accurate for osteoarticular and primary sterile samples and thus are advantageous over bacterial culture in patients undergoing antimicrobial treatment [228, 229].

Pathogen Genotyping Bacterial Genotyping Bacterial genotyping, which consists of characterizing bacterial strains based on their genetic content, has become widely used for strain typing due to the availability of molecular methods. Genotyping bacterial pathogens provides important epidemiological data for surveillance programs, which can ultimately help trace the source of infections and develop strategies to help prevent and control the emergence of infectious diseases. While whole genome sequencing (WGS) can detect differences between two strains down to the single nucleotide level, it is expensive and inaccessible on a global scale. Microarrays offer a cheaper, high-throughput alternative to WGS and have been developed to identify strains, virulence factors, and resistance loci in real-time outbreak scenarios, such as foodborne outbreaks of E. coli O157:H7 and Listeria monocytogenes [230, 231]. To this end, the Food and Drug Administration (FDA) developed an E. coli identification array that can be applied to distinguish between Shiga toxin-producing serotypes and subtypes, some of which have been associated with alternative disease outcomes [232]. Typing arrays have also been developed for the surveillance and monitoring of pathogens in healthcare settings, such as Clostridium difficile, and can help guide preventative care measures [233]. The information gleaned from bacterial genotyping arrays can also influence therapeutic decision making by detecting a multitude of known antibiotic resistance loci, as discussed in section “Bacterial Resistance Testing”.

430

S. Dunbar et al.

Mycobacterial Genotyping According to the World Health Organization (WHO), tuberculosis is one of the top 10 causes of death worldwide [234]. Tuberculosis is caused the group of Mycobacteria known as the Mycobacterium tuberculosis complex (MTC) which includes species such as Mycobacterium tuberculosis (MTB) and Mycobacterium bovis. Genotyping of MTC pathogens was first performed by PCR of spacer sequences in the direct repeat genomic region of MTC species and is known as spacer oligonucleotide typing or spoligotyping [235]. Ruettger et al. adapted spoligotyping into a DNA microarray format generating a rapid, high-throughput MTC genotyping assay [236]. Most on-market array-based assays for the detection of MTB only target known antibiotic resistance genes and do not include targets that provide genetic lineage information [73, 237, 238]. A comprehensive analysis of the incidence of MTB drug-resistant mutations has linked genetic background to the development of resistance phenotypes and suggests future diagnostic strategies should incorporate the simultaneous detection of phylogenetic markers and resistance loci [239]. Fungal Genotyping Fungal infections most commonly affect immunocompromised individuals such as people diagnosed with HIV or cancer patients on immunosuppressants. As the incidence of invasive fungal disease increases, there is a growing need for genomic surveillance of circulating fungal strains. Several microarray tests will identify fungal pathogens as a minority of targets on a broader array probing bacterial pathogens and parasites [225, 240]. While most microarrays including fungal targets will only provide genetic information at the genus and species level, there are some reports of array systems that characterize virulence loci and mycotoxigenic genes [241, 242]. Fungal arrays have been applied to discover novel antifungal resistance genotypes and are discussed in section “Fungal Resistance Testing”. Viral Genotyping Similar to bacterial and fungal genotyping, viral genotyping is primarily used for epidemiological surveillance and its clinical application mostly resides in the detection of drug resistance mutations. Due to intrinsically high mutation rates and the heterogeneous nature of viral populations, the detection range of microarray- based tests is limited by the availability of genomic information for probe design. Probe sets require continuous curating and updating according to genomic databases. For respiratory viruses, early detection and genotyping data can help prevent infection transmission and avoid the unnecessary use of antibiotics. Several microarrays have been developed for influenza A genotyping, based on the hemagglutinin (HA) and neuraminidase (NA) genes [243–245]. Numerous


431

commercially available array-based assays have been developed for the detection and genotyping of viral pathogens and include the Verigene® Respiratory Pathogens Flex test (Luminex Corp.), the CLART® PneumoVir DNA array assay (Genomica), and the FilmArray® Respiratory Panel (BioFire Diagnostics). And, as mentioned previously, array-based tests are also available for HPV genotyping (e.g., CLART® HPV2 from Genomica, Hybrid Capture® 2 from Digene Corp.).

Antimicrobial Resistance Gene Detection Bacterial Resistance Testing One of the most urgent unmet needs in the fight to control the spread of antibiotic resistance is rapid, accurate diagnostics. Diagnosing antimicrobial resistance is exceptionally challenging because of the ability of microbes to continuously evolve new resistance mechanisms and acquire multiple resistance genes via horizontal gene transfer. The capacity of microarrays to analyze thousands of genes from a single reaction mixture makes microarrays an attractive laboratory approach with future point-of-care potential. Gram-Negative Bacteria The rise of multidrug-resistant Enterobacteriaceae in healthcare settings has created a demand for diagnostics that can detect and distinguish a multitude of betalactam resistance loci in a single sample. In response to this demand, a number of DNA microarray tests were released to market, such as the AMR-ve Genotyping Kit AT−1, CarbDetect AS-1 and AS-2 from Alere Technologies, and the Check-MDR CT101, 102, and 103 from Check-Points. These commercially available microarrays enable the identification of beta-lactamases, including carbapenemases, and narrow and extended-spectrum beta-lactamases. The microarray kits offered by Alere Technologies and Check-Points are built on the ArrayTube and ArrayStrip platforms, which enable the main steps of the assay to be performed in a single tube by integrating the microarray chips into the bottom of micro-tubes. An evaluation of the CarbDetect AS-1 performed with 117 phenotypically and genotypically characterized strains reported sensitivity and specificity values of 98.2% and 97.4%, respectively [246]. In a more recent version of the CarbDetect AS-1 and CarbDetect AS-2, the array was extended to detect aminoglycoside, macrolide, quinolone, and co-trimoxazole resistance loci. Evaluations of the Check-MDR CT102 and updated Check-MDR CT103 DNA microarrays have demonstrated 95–100% sensitivity and specificity, with the exception of an ESBL mutant, SHV E240K, which was not detected in same isolates [247, 248]. In another study, the Check-MDR CT103 microarray was used as the gold standard to evaluate the detection of

432

S. Dunbar et al.

extended-spectrum beta-lactamase genes by a whole genome sequencing method, signifying the acceptance of microarray assays as a standard laboratory test [249]. While beta-lactam resistance microarrays demonstrate high concordance with reference genotyping methods, disadvantages that hinder the widespread use of microarrays in routine analysis include multistep laboratory procedures with extended turnaround times (typically >4.5 h post sample preparation). To address these drawbacks, the Verigene® Gram-negative blood culture (BC-NG) nucleic acid test incorporated microarray detection into an automated workflow with a sample-to-result format. The Verigene® BC-GN assay detects nine Gram-negative bacterial genus/species and six resistance determinants directly from positive blood cultures within 2 h, with sensitivities and specificities of 89% and 100% for species identification and 96.7% and 100% for resistance genes [129]. Gram-Positive Bacteria Among Gram-positive pathogens, methicillin-resistant S. aureus (MRSA) is a major cause of both community and nosocomial infections and is a strong focus of worldwide surveillance and infection control measures [250]. In addition to the high incidence of methicillin resistance, S. aureus isolates have been characterized with resistance to penicillin, aminoglycosides, macrolides, quinolones, gentamicin, tetracyclines, streptogramin B, and lincosamide. Zhu et al. developed a microarray that covered detection of six antibiotic-resistant genes from S. aureus and reported 90% agreement with phenotypic susceptibility tests [79]. The turnaround time for the assay was 5 h post-culture isolation. A low-density oligonucleotide array approach was also undertaken by Strommenger et al., who targeted ten clinically relevant antibiotic resistance genes (mecA, aacA-aphD, tetK, tetM, vatA, vatB, vatC, ermA, ermC, and grlA mutation) using a rapid array method that combines PCR amplification, labeling, and hybridization in a single step [251]. On-market array-based tests for the detection of MRSA include TruArray® MRSA Assay from Akonni Biosystems and the Alere S. aureus Genotyping Kit 2.0. The TruArray® MRSA Assay is a low-density oligonucleotide array and uses a three-dimensional, gel-drop technology [252]. A retrospective analysis of 246 nasal swab samples reported 80.5% sensitivity and 96.6% specificity. The S. aureus Genotyping Microarray 2.0 contains 336 probes and enables identification of resistance loci, virulence genes, and typing markers. In a comparison study of the predecessor assay to S. aureus Genotyping 2.0, the Identibac S. aureus Genotyping Microarray with whole genome sequencing reported that 96.8% of all the typing results were equally identified by WGS and microarray [253]. The authors noted that while WGS offers the advantage of being an open system, the data analysis remains convoluted and requires qualified personnel.


433

Mycobacterial Resistance Testing MTB is a notably slow-growing organism, requiring between 2 and 6 weeks to culture from blood and sputum samples [254, 255]. As a result, conventional culturebased drug susceptibility tests can delay patient access to effective treatment plans. To reduce diagnostic delays, the WHO has recommended the implementation of accurate, evidence-based commercial molecular tests for the identification of multidrug-resistant tuberculosis (MDR-TB) [256]. Several microarrays have been developed to detect first-line drug resistance to rifampicin, isoniazid, and pyrazinamide [72, 237, 238, 257, 258]. Microarrays for second-line drugs can detect resistance to fluoroquinolones and injectable aminoglycosides [259]. Some microarray assays include a combination of both first-line and second-line drug resistance markers [260–262]. Commercially available MDR-TB microarray tests that have been validated include the TB-Biochip® developed at the Engelhardt Institute of Molecular Biology, the CombiChip Mycobacteria™ by GeneIn Co., and the TruArray® MDR-TB assay developed by Akonni Biosystems [73, 237, 238]. Despite the capability of microarrays to detect all known drug resistance loci in a single test, the most widely used molecular assay for MDR-TB is the real-time PCR-based Xpert® MTB/RIF assay by Cepheid, which strictly detects M. tuberculosis species and rifampicin resistance. In a comparative analysis of the TB-Biochip® with the Xpert® MTB/RIF test, the TB-Biochip detected MTB in 92% of culture positive sputum samples and identified rifampicin and isoniazid resistance with 100% concordance to the susceptibility tests [263]. In comparison, the Xpert® MTB/RIF test only detected 78% of the culture positive sputum samples and was 97% concordant with the susceptibility tests for rifampicin resistance. The uptake of microarray-based diagnostics as a standard testing practice for MDR-TB will require automated platforms, reduced sample-to-result times, and user-friendly result reports. Toward this aim, the VerePLEX™ Biosystem “lab-on-a-chip” platform for MDR-TB testing combines microfluidics and array-based detection methods into a miniaturized, portable, automated device [260]. Systems such as the VerePLEX™ Biosystem (with a reported detection accuracy of 97.8%) suggest a promising future for improved access to rapid MDR-TB diagnostics in the regions of the world where it is needed the most. Fungal Resistance Testing As with virtually any known sequences of interest, allelic variants associated with fungal resistance can be probed by microarray detection. Candida spp. and Aspergillus spp. are responsible for the majority of invasive fungal infections, and their resistance to the two most widely prescribed classes of antifungal drugs, triazoles and echinocandins, is well established [264, 265]. Several specific mutations conferring resistance to triazoles have been characterized and include alterations to the drug-binding domains of targeted proteins and mutations to promoter regions, resulting in the upregulation of efflux pumps [266, 267]. In

434

S. Dunbar et al.

addition to detecting established resistance loci, one of the most pertinent applications of microarrays is to discover novel resistance pathways. Yan et al. described an in vitro approach whereby a fluconazole-susceptible culture of Candida albicans was exposed to increasing concentrations of fluconazole until a drug- resistant phenotype was observed [268]. The gene expression patterns between of the susceptible parental C. albicans strain and the laboratory-derived resistant isolate was characterized by high-density oligonucleotide microarray analysis. Differentially expressed genes included membrane transporters, ergosterol biosynthesis genes, transcription factors, and metabolic genes, and novel mutations targeting these gene classes were reported. The future application of microarrays can both advance the diagnosis of antifungal resistant alleles and elucidate novel genetic mechanisms underlying resistance phenotypes. Viral Resistance Testing Like bacteria, viruses have the ability to rapidly mutate and become resistant to antiviral therapeutics. RNA viruses are well-known to have error-prone polymerases resulting in exceptionally high mutation rates and a large variety of genotypes and drug-resistant alleles [269]. Sequencing is widely used in developed nations as the molecular method of choice to detect antiviral drug-resistant point mutations but is prohibitively expensive for resource-limited regions often faced with the highest viral disease burdens. Several evaluations have explored microarray platforms as a rapid, cost-effective alternative to sequencing. HIV-1 mutations have been successfully identified with microarray-based tests, with a reported concordance of >90% with conventional sequencing methods [270, 271]. In a recent evaluation, a novel HIV-1 DNA genotyping microarray proved to be highly sensitive and detected low-frequency subpopulation mutants representing as little as 5–10% of the host sample [272]. Hepatitis B virus mutants resistant to the antiviral therapies lamivudine and adefovir have been positively identified by microarray-based tests [273–275]. Microarray detection has also proven to be effective at identifying influenza resistance to first-line antiviral therapies, inclusive to the neuraminidase inhibitors oseltamivir and zanamivir, and second-line antiviral therapies, the matrix protein 2 ion channel blockers amantadine and rimantadine [276].

Conclusion Microarray platforms offer a convenient method for probing specific pathogen sequences at scales of tens to thousands of genetic features. Various microarray platforms, including solid-state, electronic, and suspension bead microarrays, are available with many alternatives in probe chemistries, fabrication methods, detection methods, and data collection and analysis formats. Microarrays have been adopted in the molecular microbiology laboratory for microbial detection and genotyping


435

and antimicrobial resistance gene detection, both from isolated organisms and directly from clinical samples. Sensitivities are similar to other molecular techniques, but microarrays offer multiplexing to enable queries of multiple targets simultaneously from the same sample. In particular, microarrays offer a panel approach to diagnosis of specific clinical syndromes and infections. They can also discriminate between different genotypes of organisms, which can be useful for epidemiology and investigation of outbreaks. As we learn more about the genes that are involved in antimicrobial resistance, microarrays may help to predict susceptibility based on sequence detection in specific organisms. Microarray assays are currently limited by the availability of sequences for probe design and thus may not adequately detect sequence variants or novel organisms. Assay design and validation studies must carefully consider the ability to differentiate closely related species, especially those with differences in clinical significance, and to ensure that relevant strains can be detected and identified. Microarray probe sets should be continuously updated as new sequence data become available to address changing genotypes and strains. With further advances in automated methods for microarray processing, software for data analysis, and algorithms to predict genotype-phenotype association, microarrays will become even more useful in the clinical microbiology laboratory. The use of microarray tests as an addition to or replacement for conventional microbiology and antimicrobial susceptibility testing ultimately promises to allow more rapid and targeted therapy for infectious diseases.

References 1. Miller MB, Tang YW. Basic concepts of microarrays and potential applications in clinical microbiology. Clin Microbiol Rev. 2009;22:611–33. 2. Hughes TR, Mao M, Jones AR, et al. Expression profiling using microarrays fabricated by an ink-jet oligonucleotide synthesizer. Nat Biotechnol. 2001;19:342–7. 3. Cooney CG, Sipes D, Thakore N, Holmberg R, Belgrader P. A plastic, disposable microfluidic flow cell for coupled on-chip PCR and microarray detection of infectious agents. Biomed Microdevices. 2012;14:45–53. 4. Michael KL, Taylor LC, Schultz SL, Walt DR. Randomly ordered addressable high-density optical sensor arrays. Anal Chem. 1998;70:1242–8. 5. Cordeiro M, Ferreira Carlos F, Pedrosa P, Lopez A, Baptista PV. Gold nanoparticles for diagnostics: advances towards points of care. Diagnostics (Basel). 2016;6:43. 6. Dunbar SA. Applications of Luminex xMAP technology for rapid, high-throughput multiplexed nucleic acid detection. Clin Chim Acta Int J Clin Chem. 2006;363:71–82. 7. Barlaan EA, Sugimori M, Furukawa S, Takeuchi K. Electronic microarray analysis of 16S rDNA amplicons for bacterial detection. J Biotechnol. 2005;115:11–21. 8. Kumar S, Wang L, Fan J, et al. Detection of 11 common viral and bacterial pathogens causing community-acquired pneumonia or sepsis in asymptomatic patients by using a multiplex reverse transcription-PCR assay with manual (enzyme hybridization) or automated (electronic microarray) detection. J Clin Microbiol. 2008;46:3063–72. 9. Beier M, Hoheisel JD. Production by quantitative photolithographic synthesis of individually quality checked DNA microarrays. Nucleic Acids Res. 2000;28:E11.

436

S. Dunbar et al.

10. Beier M, Hoheisel JD. Synthesis of 5'-O-phosphoramidites with a photolabile 3'-O-protecting group. Current protocols in nucleic acid chemistry / edited by Serge L Beaucage [et al] 2004;Chapter 12:Unit 12.3. 11. Pease AC, Solas D, Sullivan EJ, Cronin MT, Holmes CP, Fodor SP. Light-generated oligonucleotide arrays for rapid DNA sequence analysis. Proc Natl Acad Sci U S A. 1994;91:5022–6. 12. Foglieni B, Brisci A, San Biagio F, et al. Integrated PCR amplification and detection processes on a lab-on-Chip platform: a new advanced solution for molecular diagnostics. Clin Chem Lab Med CCLM/FESCC. 2010;48:329–36. 13. Marzancola MG, Sedighi A, Li PC. DNA microarray-based diagnostics. Methods Mol Biol (Clifton, NJ). 2016;1368:161–78. 14. Desmet C, Marquette CA. Surface functionalization for immobilization of probes on microarrays. Methods Mol Biol (Clifton, NJ). 2016;1368:7–23. 15. Cretich M, Breda D, Damin F, et al. Allergen microarrays on high-sensitivity silicon slides. Anal Bioanal Chem. 2010;398:1723–33. 16. Piletsky S, Piletska E, Bossi A, Turner N, Turner A. Surface functionalization of porous polypropylene membranes with polyaniline for protein immobilization. Biotechnol Bioeng. 2003;82:86–92. 17. Elsholz B, Worl R, Blohm L, et al. Automated detection and quantitation of bacterial RNA by using electrical microarrays. Anal Chem. 2006;78:4794–802. 18. Tjong V, Tang L, Zauscher S, Chilkoti A. “Smart” DNA interfaces. Chem Soc Rev. 2014;43:1612–26. 19. Duckworth BP, Xu J, Taton TA, Guo A, Distefano MD. Site-specific, covalent attachment of proteins to a solid surface. Bioconjug Chem. 2006;17:967–74. 20. Andersson H, Jonsson C, Moberg C, Stemme G. Self-assembled and self-sorted array of chemically active beads for analytical and biochemical screening. Talanta. 2002;56:301–8. 21. Hong BJ, Oh SJ, Youn TO, Kwon SH, Park JW. Nanoscale-controlled spacing provides DNA microarrays with the SNP discrimination efficiency in solution phase. Langmuir ACS J Surf Colloids. 2005;21:4257–61. 22. Charles PT, Goldman ER, Rangasammy JG, Schauer CL, Chen MS, Taitt CR. Fabrication and characterization of 3D hydrogel microarrays to measure antigenicity and antibody functionality for biosensor applications. Biosens Bioelectron. 2004;20:753–64. 23. Goldmann T, Gonzalez JS. DNA-printing: utilization of a standard inkjet printer for the transfer of nucleic acids to solid supports. J Biochem Biophys Methods. 2000;42:105–10. 24. Lausted C, Dahl T, Warren C, et al. POSaM: a fast, flexible, open-source, inkjet oligonucleotide synthesizer and microarrayer. Genome Biol. 2004;5:R58. 25. Edman CF, Raymond DE, Wu DJ, et al. Electric field directed nucleic acid hybridization on microchips. Nucleic Acids Res. 1997;25:4907–14. 26. Sosnowski RG, Tu E, Butler WF, O'Connell JP, Heller MJ. Rapid determination of single base mismatch mutations in DNA hybrids by direct electric field control. Proc Natl Acad Sci U S A. 1997;94:1119–23. 27. Singh-Gasson S, Green RD, Yue Y, et al. Maskless fabrication of light-directed oligonucleotide microarrays using a digital micromirror array. Nat Biotechnol. 1999;17:974–8. 28. Seidel M, Niessner R. Chemiluminescence microarrays in analytical chemistry: a critical review. Anal Bioanal Chem. 2014;406:5589–612. 29. Dalma-Weiszhausz DD, Warrington J, Tanimoto EY, Miyada CG. The Affymetrix GeneChip® platform: an overview. Methods Enzymol. 2006;410:3–28. 30. Samanta A, Medintz IL. Nanoparticles and DNA - a powerful and growing functional combination in bionanotechnology. Nanoscale. 2016;8:9037–95. 31. Sun YS. Label-free sensing on microarrays. Methods Mol Biol (Clifton, NJ). 2017;1518:81–108. 32. Quackenbush J. Microarray data normalization and transformation. Nat Genet. 2002;32(Suppl):496–501.


437

33. Bumgarner R. Overview of DNA microarrays: types, applications, and their future. Curr Protoc Mol Biol. 2013;Chapter 22:Unit 22 1. 34. Simon RM, Korn EL, McShane LM, Radmacher MD, Wright GW, Zhao Y. Design and analysis of DNA microarray investigations. New York: Springer-Verlag; 2003. 35. Butte A. The use and analysis of microarray data. Nat Rev Drug Discov. 2002;1:951–60. 36. Edwards D. Non-linear normalization and background correction in one-channel cDNA microarray studies. Bioinformatics. 2003;19:825–33. 37. Kepler TB, Crosby L, Morgan KT. Normalization and analysis of DNA microarray data by self-consistency and local regression. Genome Biol. 2002;3:Research0037. 38. Wang Y, Lu J, Lee R, Gu Z, Clarke R. Iterative normalization of cDNA microarray data. IEEE Trans Inf Technol Biomed Publ IEEE Eng Med Biol Soc. 2002;6:29–37. 39. Reiner A, Yekutieli D, Benjamini Y. Identifying differentially expressed genes using false discovery rate controlling procedures. Bioinformatics. 2003;19:368–75. 40. Gollub J, Sherlock G. Clustering microarray data. Methods Enzymol. 2006;411:194–213. 41. Chung FL, Lee T. Fuzzy competitive learning. Neural Netw. 1994;7:539–51. 42. Kohonen. The self-organizing map. Proc IEEE. 1990;78:1464–80. 43. Kaufman L, Rousseeuw PJ. Finding groups in data: an introduction to cluster analysis. Hoboken: Wiley; 1990. 44. Vogt W, Nagel D. Cluster analysis in diagnosis. Clin Chem. 1992;38:182–98. 45. Henry VJ, Bandrowski AE, Pepin AS, Gonzalez BJ, Desfeux A. OMICtools: an informative directory for multi-omic data analysis. Database (Oxford). 2014;2014:1–5. 46. de Hoon MJ, Imoto S, Nolan J, Miyano S. Open source clustering software. Bioinformatics. 2004;20:1453–4. 47. Luscombe NM, Royce TE, Bertone P, et al. ExpressYourself: a modular platform for processing and visualizing microarray data. Nucleic Acids Res. 2003;31:3477–82. 48. Kapushesky M, Kemmeren P, Culhane AC, et al. Expression profiler: next generation--an online platform for analysis of microarray data. Nucleic Acids Res. 2004;32:W465–70. 49. Stavrum AK, Petersen K, Jonassen I, Dysvik B. Analysis of gene-expression data using J-Express. Curr Protoc Bioinformatics. 2008.; Chapter 7:Unit 7.3 50. Donatin E, Drancourt M. DNA microarrays for the diagnosis of infectious diseases. Med Mal Infect. 2012;42:453–9. 51. McLoughlin KS. Microarrays for pathogen detection and analysis. Brief Funct Genomics. 2011;10:342–53. 52. Gardner SN, Jaing CJ, McLoughlin KS, Slezak TR. A microbial detection array (MDA) for viral and bacterial detection. BMC Genomics. 2010;11:668. 53. Yoo SM, Lee SY, Chang KH, et al. High-throughput identification of clinically important bacterial pathogens using DNA microarray. Mol Cell Probes. 2009;23:171–7. 54. Jarvinen AK, Laakso S, Piiparinen P, et al. Rapid identification of bacterial pathogens using a PCR- and microarray-based assay. BMC Microbiol. 2009;9:161. 55. Yoo SM, Choi JH, Lee SY, Yoo NC. Applications of DNA microarray in disease diagnostics. J Microbiol Biotechnol. 2009;19:635–46. 56. Garaizar J, Brena S, Bikandi J, Rementeria A, Ponton J. Use of DNA microarray technology and gene expression profiles to investigate the pathogenesis, cell biology, antifungal susceptibility and diagnosis of Candida albicans. FEMS Yeast Res. 2006;6:987–98. 57. CLSI. Microarrays for diagnosis and monitoring of infectious diseases; approved guideline. CLSI document MM22-a. Wayne: Clinical and Laboratory Standards Institute; 2014. 58. Loy A, Bodrossy L. Highly parallel microbial diagnostics using oligonucleotide microarrays. Clin Chim Acta Int J Clin Chem. 2006;363:106–19. 59. Liu H, Bebu I, Li X. Microarray probes and probe sets. Front Biosci (Elite Ed). 2010;2:325–38. 60. Liu Y, Han JX, Huang HY, Zhu B. Development and evaluation of 16S rDNA microarray for detecting bacterial pathogens in cerebrospinal fluid. Exp Biol Med (Maywood). 2005;230:587–91.

438

S. Dunbar et al.

61. Negoro E, Iwasaki H, Tai K, et al. Utility of PCR amplification and DNA microarray hybridization of 16S rDNA for rapid diagnosis of bacteremia associated with hematological diseases. Int J Infect Dis IJID Off Publ Int Soc Infect Dis. 2013;17:e271–6. 62. Troesch A, Nguyen H, Miyada CG, et al. Mycobacterium species identification and rifampin resistance testing with high-density DNA probe arrays. J Clin Microbiol. 1999;37:49–55. 63. Schnee C, Schulsse S, Hotzel H, et al. A novel rapid DNA microarray assay enables identification of 37 mycoplasma species and highlights multiple mycoplasma infections. PLoS One. 2012;7:e33237. 64. Yoo SM, Choi JY, Yun JK, et al. DNA microarray-based identification of bacterial and fungal pathogens in bloodstream infections. Mol Cell Probes. 2010;24:44–52. 65. Nubel U, Schmidt PM, Reiss E, Bier F, Beyer W, Naumann D. Oligonucleotide microarray for identification of Bacillus anthracis based on intergenic transcribed spacers in ribosomal DNA. FEMS Microbiol Lett. 2004;240:215–23. 66. Keramas G, Bang DD, Lund M, et al. Development of a sensitive DNA microarray suitable for rapid detection of campylobacter spp. Mol Cell Probes. 2003;17:187–96. 67. Wang M, Cao B, Gao Q, et al. Detection of Enterobacter sakazakii and other pathogens associated with infant formula powder by use of a DNA microarray. J Clin Microbiol. 2009;47:3178–84. 68. Roth SB, Jalava J, Ruuskanen O, Ruohola A, Nikkari S. Use of an oligonucleotide array for laboratory diagnosis of bacteria responsible for acute upper respiratory infections. J Clin Microbiol. 2004;42:4268–74. 69. Kakinuma K, Fukushima M, Kawaguchi R. Detection and identification of Escherichia coli, Shigella, and Salmonella by microarrays using the gyrB gene. Biotechnol Bioeng. 2003;83:721–8. 70. Fukushima M, Kakinuma K, Hayashi H, Nagai H, Ito K, Kawaguchi R. Detection and identification of mycobacterium species isolates by DNA microarray. J Clin Microbiol. 2003;41:2605–15. 71. Antwerpen MH, Schellhase M, Ehrentreich-Forster E, Bier F, Witte W, Nubel U. DNA microarray for detection of antibiotic resistance determinants in bacillus anthracis and closely related Bacillus cereus. Mol Cell Probes. 2007;21:152–60. 72. Yao C, Zhu T, Li Y, et al. Detection of rpoB, katG and inhA gene mutations in Mycobacterium tuberculosis clinical isolates from Chongqing as determined by microarray. Clin Microbiol Infect. 2010;16:1639–43. 73. Caoili JC, Mayorova A, Sikes D, Hickman L, Plikaytis BB, Shinnick TM. Evaluation of the TB-biochip oligonucleotide microarray system for rapid detection of rifampin resistance in Mycobacterium tuberculosis. J Clin Microbiol. 2006;44:2378–81. 74. Tang X, Morris SL, Langone JJ, Bockstahler LE. Microarray and allele specific PCR detection of point mutations in Mycobacterium tuberculosis genes associated with drug resistance. J Microbiol Methods. 2005;63:318–30. 75. Zhang Z, Li L, Luo F, et al. Rapid and accurate detection of RMP- and INH- resistant Mycobacterium tuberculosis in spinal tuberculosis specimens by CapitalBio DNA microarray: a prospective validation study. BMC Infect Dis. 2012;12:303. 76. Zimenkov DV, Kulagina EV, Antonova OV, et al. Evaluation of a low-density hydrogel microarray technique for mycobacterial species identification. J Clin Microbiol. 2015;53:1103–14. 77. Ringuet H, Akoua-Koffi C, Honore S, et al. hsp65 sequencing for identification of rapidly growing mycobacteria. J Clin Microbiol. 1999;37:852–7. 78. Cleven BE, Palka-Santini M, Gielen J, Meembor S, Kronke M, Krut O. Identification and characterization of bacterial pathogens causing bloodstream infections by DNA microarray. J Clin Microbiol. 2006;44:2389–97. 79. Zhu LX, Zhang ZW, Wang C, et al. Use of a DNA microarray for simultaneous detection of antibiotic resistance genes among staphylococcal clinical isolates. J Clin Microbiol. 2007;45:3514–21.


439

80. Masson L, Maynard C, Brousseau R, et al. Identification of pathogenic helicobacter species by chaperonin-60 differentiation on plastic DNA arrays. Genomics. 2006;87:104–12. 81. Maynard C, Berthiaume F, Lemarchand K, et al. Waterborne pathogen detection by use of oligonucleotide-based microarrays. Appl Environ Microbiol. 2005;71:8548–57. 82. Champagne J, Diarra MS, Rempel H, et al. Development of a DNA microarray for enterococcal species, virulence, and antibiotic resistance gene determinations among isolates from poultry. Appl Environ Microbiol. 2011;77:2625–33. 83. Hu Y, Liu J, Xia D, Chen S. Simultaneous analysis of foodborne pathogenic bacteria by an oligonucleotide microarray assay. J Basic Microbiol. 2012;52:27–34. 84. Giammarinaro P, Leroy S, Chacornac JP, Delmas J, Talon R. Development of a new oligonucleotide array to identify staphylococcal strains at species level. J Clin Microbiol. 2005;43:3673–80. 85. Kooken J, Fox K, Fox A, et al. Reprint of “identification of staphylococcal species based on variations in protein sequences (mass spectrometry) and DNA sequence (sodA microarray)”. Mol Cell Probes. 2014;28:73–82. 86. Spiess B, Seifarth W, Hummel M, et al. DNA microarray-based detection and identification of fungal pathogens in clinical samples from neutropenic patients. J Clin Microbiol. 2007;45:3743–53. 87. Huang A, Li JW, Shen ZQ, Wang XW, Jin M. High-throughput identification of clinical pathogenic fungi by hybridization to an oligonucleotide microarray. J Clin Microbiol. 2006;44:3299–305. 88. Campa D, Tavanti A, Gemignani F, et al. DNA microarray based on arrayed-primer extension technique for identification of pathogenic fungi responsible for invasive and superficial mycoses. J Clin Microbiol. 2008;46:909–15. 89. Landlinger C, Preuner S, Willinger B, et al. Species-specific identification of a wide range of clinically relevant fungal pathogens by use of Luminex xMAP technology. J Clin Microbiol. 2009;47:1063–73. 90. Janda JM, Abbott SL. 16S rRNA gene sequencing for bacterial identification in the diagnostic laboratory: pluses, perils, and pitfalls. J Clin Microbiol. 2007;45:2761–4. 91. Kostic T, Sessitsch A. Microbial diagnostic microarrays for the detection and typing of foodand water-borne (bacterial) pathogens. Microarrays (Basel). 2011;1:3–24. 92. Rasooly A, Herold KE. Food microbial pathogen detection and analysis using DNA microarray technologies. Foodborne Pathog Dis. 2008;5:531–50. 93. Wang XW, Zhang L, Jin LQ, et al. Development and application of an oligonucleotide microarray for the detection of food-borne bacterial pathogens. Appl Microbiol Biotechnol. 2007;76:225–33. 94. Hunt DE, Klepac-Ceraj V, Acinas SG, Gautier C, Bertilsson S, Polz MF. Evaluation of 23S rRNA PCR primers for use in phylogenetic studies of bacterial diversity. Appl Environ Microbiol. 2006;72:2221–5. 95. Yoo SM, Keum KC, Yoo SY, et al. Development of DNA microarray for pathogen detection. Biotechnol Bioprocess Eng. 2004;9:93–9. 96. Sanguinetti M, Novarese L, Posteraro B, et al. Use of microelectronic array technology for rapid identification of clinically relevant mycobacteria. J Clin Microbiol. 2005;43:6189–93. 97. Tobler NE, Pfunder M, Herzog K, Frey JE, Altwegg M. Rapid detection and species identification of Mycobacterium spp. using real-time PCR and DNA-microarray. J Microbiol Methods. 2006;66:116–24. 98. Johansen IS, Lundgren BH, Thyssen JP, Thomsen VO. Rapid differentiation between clinically relevant mycobacteria in microscopy positive clinical specimens and mycobacterial isolates by line probe assay. Diagn Microbiol Infect Dis. 2002;43:297–302. 99. Tortoli E, Mariottini A, Mazzarelli G. Evaluation of INNO-LiPA MYCOBACTERIA v2: improved reverse hybridization multiple DNA probe assay for mycobacterial identification. J Clin Microbiol. 2003;41:4418–20.

440

S. Dunbar et al.

100. Mijs W, De Vreese K, Devos A, et al. Evaluation of a commercial line probe assay for identification of mycobacterium species from liquid and solid culture. Eur J Clin Microbiol Infect Dis Off Publ Eur Soc Clin Microbiol. 2002;21:794–802. 101. Yue J, Shi W, Xie J, et al. Detection of rifampin-resistant Mycobacterium tuberculosis strains by using a specialized oligonucleotide microarray. Diagn Microbiol Infect Dis. 2004;48:47–54. 102. Mikhailovich V, Lapa S, Gryadunov D, et al. Identification of rifampin-resistant Mycobacterium tuberculosis strains by hybridization, PCR, and ligase detection reaction on oligonucleotide microchips. J Clin Microbiol. 2001;39:2531–40. 103. Guo Y, Zhou Y, Wang C, et al. Rapid, accurate determination of multidrug resistance in M. Tuberculosis isolates and sputum using a biochip system. Int J Tuberc Lung Dis Off J Int Union Against Tuberc Lung Dis. 2009;13:914–20. 104. Spence RP, Wright V, Ala-Aldeen DA, Turner DP, Wooldridge KG, James R. Validation of virulence and epidemiology DNA microarray for identification and characterization of Staphylococcus aureus isolates. J Clin Microbiol. 2008;46:1620–7. 105. Witney AA, Marsden GL, Holden MT, et al. Design, validation, and application of a seven- strain Staphylococcus aureus PCR product microarray for comparative genomics. Appl Environ Microbiol. 2005;71:7504–14. 106. Howden BP, Johnson PD, Ward PB, Stinear TP, Davies JK. Isolates with low-level vancomycin resistance associated with persistent methicillin-resistant Staphylococcus aureus bacteremia. Antimicrob Agents Chemother. 2006;50:3039–47. 107. Chen H, Mammel M, Kulka M, Patel I, Jackson S, Goswami BB. Detection and identification of common food-borne viruses with a tiling microarray. Open Virol J. 2011;5:52–9. 108. Huang A, Qiu Z, Jin M, et al. High-throughput detection of food-borne pathogenic bacteria using oligonucleotide microarray with quantum dots as fluorescent labels. Int J Food Microbiol. 2014;185:27–32. 109. Zhou G, Wen S, Liu Y, et al. Development of a DNA microarray for detection and identification of legionella pneumophila and ten other pathogens in drinking water. Int J Food Microbiol. 2011;145:293–300. 110. Farina C, Russello G, Andreoni S, et al. Microarray technology for yeast identification directly from positive blood cultures. A multicenter Italian experience. Med Mycol. 2012;50:549–55. 111. Aittakorpi A, Kuusela P, Koukila-Kahkola P, et al. Accurate and rapid identification of Candida spp frequently associated with fungemia by using PCR and the microarray-based prove-it Sepsis assay. J Clin Microbiol. 2012;50:3635–40. 112. Leinberger DM, Schumacher U, Autenrieth IB, Bachmann TT. Development of a DNA microarray for detection and identification of fungal pathogens involved in invasive mycoses. J Clin Microbiol. 2005;43:4943–53. 113. Fitzgibbon JE, Sagripanti JL. Simultaneous identification of orthopoxviruses and alphaviruses by oligonucleotide macroarray with special emphasis on detection of variola and Venezuelan equine encephalitis viruses. J Virol Methods. 2006;131:160–7. 114. Xiao-Ping K, Yong-Qiang L, Qing-Ge S, Hong L, Qing-Yu Z, Yin-Hui Y. Development of a consensus microarray method for identification of some highly pathogenic viruses. J Med Virol. 2009;81:1945–50. 115. Liu V, Escobar GJ, Greene JD, et al. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA. 2014;312:90–2. 116. Wiesinger-Mayr H, Vierlinger K, Pichler R, et al. Identification of human pathogens isolated from blood using microarray hybridisation and signal pattern recognition. BMC Microbiol. 2007;7:78. 117. Buchan BW, Ginocchio CC, Manii R, et al. Multiplex identification of gram-positive bacteria and resistance determinants directly from positive blood culture broths: evaluation of an automated microarray-based nucleic acid test. PLoS Med. 2013;10:e1001478.


441

118. Beal SG, Ciurca J, Smith G, et al. Evaluation of the nanosphere verigene gram-positive blood culture assay with the VersaTREK blood culture system and assessment of possible impact on selected patients. J Clin Microbiol. 2013;51:3988–92. 119. Sullivan KV, Turner NN, Roundtree SS, et al. Rapid detection of gram-positive organisms by use of the Verigene gram-positive blood culture nucleic acid test and the BacT/Alert pediatric FAN system in a multicenter pediatric evaluation. J Clin Microbiol. 2013;51:3579–84. 120. Wojewoda CM, Sercia L, Navas M, et al. Evaluation of the Verigene gram-positive blood culture nucleic acid test for rapid detection of bacteria and resistance determinants. J Clin Microbiol. 2013;51:2072–6. 121. Tissari P, Zumla A, Tarkka E, et al. Accurate and rapid identification of bacterial species from positive blood cultures with a DNA-based microarray platform: an observational study. Lancet (London, England). 2010;375:224–30. 122. Laakso S, Kirveskari J, Tissari P, Maki M. Evaluation of high-throughput PCR and microarray-based assay in conjunction with automated DNA extraction instruments for diagnosis of sepsis. PLoS One. 2011;6:e26655. 123. Mestas J, Polanco CM, Felsenstein S, Dien BJ. Performance of the Verigene gram-positive blood culture assay for direct detection of gram-positive organisms and resistance markers in a pediatric hospital. J Clin Microbiol. 2014;52:283–7. 124. Tojo M, Fujita T, Ainoda Y, et al. Evaluation of an automated rapid diagnostic assay for detection of gram-negative bacteria and their drug-resistance genes in positive blood cultures. PLoS One. 2014;9:e94064. 125. Dodemont M, De Mendonca R, Nonhoff C, Roisin S, Denis O. Performance of the Verigene gram-negative blood culture assay for rapid detection of bacteria and resistance determinants. J Clin Microbiol. 2014;52:3085–7. 126. Samuel LP, Tibbetts RJ, Agotesku A, Fey M, Hensley R, Meier FA. Evaluation of a microarray-based assay for rapid identification of gram-positive organisms and resistance markers in positive blood cultures. J Clin Microbiol. 2013;51:1188–92. 127. Palka-Santini M, Putzfeld S, Cleven BE, Kronke M, Krut O. Rapid identification, virulence analysis and resistance profiling of Staphylococcus aureus by gene segment-based DNA microarrays: application to blood culture post-processing. J Microbiol Methods. 2007;68:468–77. 128. Pasko C, Hicke B, Dunn J, et al. Staph ID/R: a rapid method for determining staphylococcus species identity and detecting the mecA gene directly from positive blood culture. J Clin Microbiol. 2012;50:810–7. 129. Sullivan KV, Deburger B, Roundtree SS, Ventrola CA, Blecker-Shelly DL, Mortensen JE. Pediatric multicenter evaluation of the Verigene gram-negative blood culture test for rapid detection of inpatient bacteremia involving gram-negative organisms, extended-spectrum beta-lactamases, and carbapenemases. J Clin Microbiol. 2014;52:2416–21. 130. Felsenstein S, Bender JM, Sposto R, Gentry M, Takemoto C, Bard JD. Impact of a rapid blood culture assay for gram-positive identification and detection of resistance markers in a pediatric hospital. Arch Pathol Lab Med. 2016;140:267–75. 131. Mancini N, Infurnari L, Ghidoli N, et al. Potential impact of a microarray-based nucleic acid assay for rapid detection of gram-negative bacteria and resistance markers in positive blood cultures. J Clin Microbiol. 2014;52:1242–5. 132. Walker T, Dumadag S, Lee CJ, et al. Clinical impact of laboratory implementation of Verigene BC-GN microarray-based assay for detection of gram-negative bacteria in positive blood cultures. J Clin Microbiol. 2016;54:1789–96. 133. Box MJ, Sullivan EL, Ortwine KN, et al. Outcomes of rapid identification for gram-positive bacteremia in combination with antibiotic stewardship at a community-based hospital system. Pharmacotherapy. 2015;35:269–76. 134. Hill JT, Tran KD, Barton KL, Labreche MJ, Sharp SE. Evaluation of the nanosphere Verigene BC-GN assay for direct identification of gram-negative bacilli and antibiotic resistance mark-

442

S. Dunbar et al.

ers from positive blood cultures and potential impact for more-rapid antibiotic interventions. J Clin Microbiol. 2014;52:3805–7. 135. Sango A, McCarter YS, Johnson D, Ferreira J, Guzman N, Jankowski CA. Stewardship approach for optimizing antimicrobial therapy through use of a rapid microarray assay on blood cultures positive for enterococcus species. J Clin Microbiol. 2013;51:4008–11. 136. Caliendo AM, Gilbert DN, Ginocchio CC, et al. Better tests, better care: improved diagnostics for infectious diseases. Clin Infect Dis Off Publ Infect Dis Soc Am. 2013;57(Suppl 3):S139–70. 137. Renois F, Talmud D, Huguenin A, et al. Rapid detection of respiratory tract viral infections and coinfections in patients with influenza-like illnesses by use of reverse transcription-PCR DNA microarray systems. J Clin Microbiol. 2010;48:3836–42. 138. Cannon GA, Carr MJ, Yandle Z, et al. A low density oligonucleotide microarray for the detection of viral and atypical bacterial respiratory pathogens. J Virol Methods. 2010;163:17–24. 139. Frobert E, Escuret V, Javouhey E, et al. Respiratory viruses in children admitted to hospital intensive care units: evaluating the CLART(R) Pneumovir DNA array. J Med Virol. 2011;83:150–5. 140. Krunic N, Merante F, Yaghoubian S, Himsworth D, Janeczko R. Advances in the diagnosis of respiratory tract infections: role of the Luminex xTAG respiratory viral panel. Ann N Y Acad Sci. 2011;1222:6–13. 141. Mahony JB, Petrich A, Smieja M. Molecular diagnosis of respiratory virus infections. Crit Rev Clin Lab Sci. 2011;48:217–49. 142. Pabbaraju K, Tokaryk KL, Wong S, Fox JD. Comparison of the Luminex xTAG respiratory viral panel with in-house nucleic acid amplification tests for diagnosis of respiratory virus infections. J Clin Microbiol. 2008;46:3056–62. 143. Gharabaghi F, Hawan A, Drews SJ, Richardson SE. Evaluation of multiple commercial molecular and conventional diagnostic assays for the detection of respiratory viruses in children. Clin Microbiol Infect. 2011;17:1900–6. 144. Beckmann C, Hirsch HH. Comparing Luminex NxTAG-respiratory pathogen panel and RespiFinder-22 for multiplex detection of respiratory pathogens. J Med Virol. 2016;88:1319–24. 145. Popowitch EB, O'Neill SS, Miller MB. Comparison of the Biofire FilmArray RP, Genmark eSensor RVP, Luminex xTAG RVPv1, and Luminex xTAG RVP fast multiplex assays for detection of respiratory viruses. J Clin Microbiol. 2013;51:1528–33. 146. Pillet S, Lardeux M, Dina J, et al. Comparative evaluation of six commercialized multiplex PCR kits for the diagnosis of respiratory infections. PLoS One. 2013;8:e72174. 147. Balada-Llasat JM, LaRue H, Kelly C, Rigali L, Pancholi P. Evaluation of commercial ResPlex II v2.0, MultiCode-PLx, and xTAG respiratory viral panels for the diagnosis of respiratory viral infections in adults. J Clin Virol Off Publ Pan Am Soc Clin Virol. 2011;50:42–5. 148. Babady NE, Mead P, Stiles J, et al. Comparison of the Luminex xTAG RVP fast assay and the Idaho technology FilmArray RP assay for detection of respiratory viruses in pediatric patients at a cancer hospital. J Clin Microbiol. 2012;50:2282–8. 149. Anderson TP, Werno AM, Barratt K, Mahagamasekera P, Murdoch DR, Jennings LC. Comparison of four multiplex PCR assays for the detection of viral pathogens in respiratory specimens. J Virol Methods. 2013;191:118–21. 150. Ruggiero P, McMillen T, Tang YW, Babady NE. Evaluation of the BioFire FilmArray respiratory panel and the GenMark eSensor respiratory viral panel on lower respiratory tract specimens. J Clin Microbiol. 2014;52:288–90. 151. Lodes MJ, Suciu D, Wilmoth JL, et al. Identification of upper respiratory tract pathogens using electrochemical detection on an oligonucleotide microarray. PLoS One. 2007;2:e924. 152. Harada Y, Kinoshita F, Yoshida LM, et al. Does respiratory virus coinfection increases the clinical severity of acute respiratory infection among children infected with respiratory syncytial virus? Pediatr Infect Dis J. 2013;32:441–5.


443

153. Debiaggi M, Canducci F, Ceresola ER, Clementi M. The role of infections and coinfections with newly identified and emerging respiratory viruses in children. Virol J. 2012;9:247. 154. Brotons P, Henares D, Latorre I, Cepillo A, Launes C, Munoz-Almagro C. Comparison of NxTAG respiratory pathogen panel and Anyplex II RV16 tests for multiplex detection of respiratory pathogens in hospitalized children. J Clin Microbiol. 2016;54:2900–4. 155. Ginocchio CC. Strengths and weaknesses of FDA-approved/cleared diagnostic devices for the molecular detection of respiratory pathogens. Clin Infect Dis Off Publ Infect Dis Soc Am. 2011;52(Suppl 4):S312–25. 156. Dundas NE, Ziadie MS, Revell PA, et al. A lean laboratory: operational simplicity and cost effectiveness of the Luminex xTAG respiratory viral panel. J Mol Diagn. 2011;13:175–9. 157. Mahony JB, Blackhouse G, Babwah J, et al. Cost analysis of multiplex PCR testing for diagnosing respiratory virus infections. J Clin Microbiol. 2009;47:2812–7. 158. Simoes EA, Patel C, Sung WK, et al. Pathogen chip for respiratory tract infections. J Clin Microbiol. 2013;51:945–53. 159. Quan PL, Palacios G, Jabado OJ, et al. Detection of respiratory viruses and subtype identification of influenza a viruses by GreeneChipResp oligonucleotide microarray. J Clin Microbiol. 2007;45:2359–64. 160. Lin B, Wang Z, Vora GJ, et al. Broad-spectrum respiratory tract pathogen identification using resequencing DNA microarrays. Genome Res. 2006;16:527–35. 161. Chen EC, Miller SA, DeRisi JL, Chiu CY. Using a pan-viral microarray assay (Virochip) to screen clinical samples for viral pathogens. J Vis Exp. 2011;50:2536. 162. Lin B, Malanoski AP, Wang Z, et al. Application of broad-spectrum, sequence-based pathogen identification in an urban population. PLoS One. 2007;2:e419. 163. Lin B, Blaney KM, Malanoski AP, et al. Using a resequencing microarray as a multiple respiratory pathogen detection assay. J Clin Microbiol. 2007;45:443–52. 164. Campos S, Caramori M, Teixeira R, et al. Bacterial and fungal pneumonias after lung transplantation. Transplant Proc. 2008;40:822–4. 165. Sakai K, Trabasso P, Moretti ML, Mikami Y, Kamei K, Gonoi T. Identification of fungal pathogens by visible microarray system in combination with isothermal gene amplification. Mycopathologia. 2014;178:11–26. 166. Simons S. Nontuberculous mycobacteria in respiratory tract infections, Eastern Asia. Emerging Infectious Diseases 2011. 167. Falkinham JO 3rd. Mycobacterial aerosols and respiratory disease. Emerg Infect Dis. 2003;9:763–7. 168. Suffys PN, da Silva RA, de Oliveira M, et al. Rapid identification of mycobacteria to the species level using INNO-LiPA mycobacteria, a reverse hybridization assay. J Clin Microbiol. 2001;39:4477–82. 169. Estimates of Foodborne Illness in the United States. 2016. (Accessed July, 25, 2017, at https://www.cdc.gov/foodborneburden/2011-foodborne-estimates.html). 170. Reddington K, Tuite N, Minogue E, Barry T. A current overview of commercially available nucleic acid diagnostics approaches to detect and identify human gastroenteritis pathogens. Biomol Detect Quantif. 2014;1:3–7. 171. Jin DZ, Wen SY, Chen SH, Lin F, Wang SQ. Detection and identification of intestinal pathogens in clinical specimens using DNA microarrays. Mol Cell Probes. 2006;20:337–47. 172. Freeman K, Tsertsvadze A, Taylor-Phillips S, et al. Agreement between gastrointestinal panel testing and standard microbiology methods for detecting pathogens in suspected infectious gastroenteritis: test evaluation and meta-analysis in the absence of a reference standard. PLoS One. 2017;12:e0173196. 173. Claas EC, Burnham CA, Mazzulli T, Templeton K, Topin F. Performance of the xTAG(R) gastrointestinal pathogen panel, a multiplex molecular assay for simultaneous detection of bacterial, viral, and parasitic causes of infectious gastroenteritis. J Microbiol Biotechnol. 2013;23:1041–5.

444

S. Dunbar et al.

174. Buchan BW, Olson WJ, Pezewski M, et al. Clinical evaluation of a real-time PCR assay for identification of Salmonella, Shigella, campylobacter (campylobacter jejuni and C. Coli), and Shiga toxin-producing Escherichia coli isolates in stool specimens. J Clin Microbiol. 2013;51:4001–7. 175. Halligan E, Edgeworth J, Bisnauthsing K, et al. Multiplex molecular testing for management of infectious gastroenteritis in a hospital setting: a comparative diagnostic and clinical utility study. Clin Microbiol Infect. 2014;20:O460–7. 176. Mengelle C, Mansuy JM, Prere MF, et al. Simultaneous detection of gastrointestinal pathogens with a multiplex Luminex-based molecular assay in stool samples from diarrhoeic patients. Clin Microbiol Infect. 2013;19:E458–65. 177. Janda JM, Abbott SA. Culture-independent diagnostic testing: have we opened Pandora's box for good? Diagn Microbiol Infect Dis. 2014;80:171–6. 178. Liu J, Kabir F, Manneh J, et al. Development and assessment of molecular diagnostic tests for 15 enteropathogens causing childhood diarrhoea: a multicentre study. Lancet Infect Dis. 2014;14:716–24. 179. Binnicker MJ. Multiplex molecular panels for diagnosis of gastrointestinal infection: performance, result interpretation, and cost-effectiveness. J Clin Microbiol. 2015;53:3723–8. 180. Patel A, Navidad J, Bhattacharyya S. Site-specific clinical evaluation of the Luminex xTAG gastrointestinal pathogen panel for detection of infectious gastroenteritis in fecal specimens. J Clin Microbiol. 2014;52:3068–71. 181. Goldenberg SD, Bacelar M, Brazier P, Bisnauthsing K, Edgeworth JD. A cost benefit analysis of the Luminex xTAG gastrointestinal pathogen panel for detection of infectious gastroenteritis in hospitalised patients. J Infect. 2015;70:504–11. 182. Marotta F, Zilli K, Tonelli A, et al. Detection and genotyping of campylobacter jejuni and campylobacter coli by use of DNA oligonucleotide arrays. Mol Biotechnol. 2013;53:182–8. 183. Wang Z, Orlandi PA, Stenger DA. Simultaneous detection of four human pathogenic microsporidian species from clinical samples by oligonucleotide microarray. J Clin Microbiol. 2005;43:4121–8. 184. Leva A, Eibach D, Krumkamp R, et al. Diagnostic performance of the Luminex xTAG gastrointestinal pathogens panel to detect rotavirus in Ghanaian children with and without diarrhoea. Virol J. 2016;13:132. 185. Eibach D, Krumkamp R, Hahn A, et al. Application of a multiplex PCR assay for the detection of gastrointestinal pathogens in a rural African setting. BMC Infect Dis. 2016;16:150. 186. Fenollar F, Raoult D. Molecular diagnosis of bloodstream infections caused by non-cultivable bacteria. Int J Antimicrob Agents. 2007;30(Suppl 1):S7–15. 187. Opota O, Jaton K, Greub G. Microbial diagnosis of bloodstream infection: towards molecular diagnosis directly from blood. Clin Microbiol Infect. 2015;21:323–31. 188. Huttunen R, Syrjanen J, Vuento R, Aittoniemi J. Current concepts in the diagnosis of blood stream infections. Are novel molecular methods useful in clinical practice? Int J Infect Dis IJID Off Publ Int Soc Infect Dis. 2013;17:e934–8. 189. Schrenzel J. Clinical relevance of new diagnostic methods for bloodstream infections. Int J Antimicrob Agents. 2007;30(Suppl 1):S2–6. 190. Debaugnies F, Busson L, Ferster A, et al. Detection of Herpesviridae in whole blood by multiplex PCR DNA-based microarray analysis after hematopoietic stem cell transplantation. J Clin Microbiol. 2014;52:2552–6. 191. Shang S, Chen G, Wu Y, Du L, Zhao Z. Rapid diagnosis of bacterial sepsis with PCR amplification and microarray hybridization in 16S rRNA gene. Pediatr Res. 2005;58:143–8. 192. Burdino E, Ruggiero T, Allice T, et al. Combination of conventional blood cultures and the SeptiFast molecular test in patients with suspected sepsis for the identification of bloodstream pathogens. Diagn Microbiol Infect Dis. 2014;79:287–92. 193. Sutherland A, Thomas M, Brandon RA, et al. Development and validation of a novel molecular biomarker diagnostic test for the early detection of sepsis. Crit Care (London, England). 2011;15:R149.


445

194. Esparcia O, Montemayor M, Ginovart G, et al. Diagnostic accuracy of a 16S ribosomal DNA gene-based molecular technique (RT-PCR, microarray, and sequencing) for bacterial meningitis, early-onset neonatal sepsis, and spontaneous bacterial peritonitis. Diagn Microbiol Infect Dis. 2011;69:153–60. 195. Rafi W, Chandramuki A, Mani R, Satishchandra P, Shankar SK. Rapid diagnosis of acute bacterial meningitis: role of a broad range 16S rRNA polymerase chain reaction. J Emerg Med. 2010;38:225–30. 196. Ben RJ, Kung S, Chang FY, Lu JJ, Feng NH, Hsieh YD. Rapid diagnosis of bacterial meningitis using a microarray. J Formos Med Assoc = Taiwan yi zhi. 2008;107:448–53. 197. Liu Y, Duan C, Zhang C, et al. Evaluation of a viral microarray based on simultaneous extraction and amplification of viral nucleotide acid for detecting human herpesviruses and enteroviruses. PLoS One. 2015;10:e0117626. 198. Boriskin YS, Rice PS, Stabler RA, et al. DNA microarrays for virus detection in cases of central nervous system infection. J Clin Microbiol. 2004;42:5811–8. 199. Leveque N, Van Haecke A, Renois F, Boutolleau D, Talmud D, Andreoletti L. Rapid virological diagnosis of central nervous system infections by use of a multiplex reverse transcriptionPCR DNA microarray. J Clin Microbiol. 2011;49:3874–9. 200. Bovers M, Diaz MR, Hagen F, et al. Identification of genotypically diverse Cryptococcus neoformans and Cryptococcus gattii isolates by Luminex xMAP technology. J Clin Microbiol. 2007;45:1874–83. 201. Sugianli AK, Ginting F, Kusumawati RL, et al. Antimicrobial resistance in uropathogens and appropriateness of empirical treatment: a population-based surveillance study in Indonesia. J Antimicrob Chemother. 2017;72:1469–77. 202. van der Zee A, Roorda L, Bosman G, Ossewaarde JM. Molecular diagnosis of urinary tract infections by semi-quantitative detection of uropathogens in a routine clinical hospital setting. PLoS One. 2016;11:e0150755. 203. Miller S, Karaoz U, Brodie E, Dunbar S. Solid and suspension microarrays for microbial diagnostics. In: Sails A, Tang Y-W, eds. Methods in microbiology: Academic Press; 2015:395-431. San Diego, United States 204. Dally S, Lemuth K, Kaase M, Rupp S, Knabbe C, Weile J. DNA microarray for genotyping antibiotic resistance determinants in Acinetobacter baumannii clinical isolates. Antimicrob Agents Chemother. 2013;57:4761–8. 205. Friedrich T, Rahmann S, Weigel W, et al. High-throughput microarray technology in diagnostics of enterobacteria based on genome-wide probe selection and regression analysis. BMC Genomics. 2010;11:591. 206. Keum KC, Yoo SM, Lee SY, et al. DNA microarray-based detection of nosocomial pathogenic Pseudomonas aeruginosa and Acinetobacter baumannii. Mol Cell Probes. 2006;20:42–50. 207. Peter H, Berggrav K, Thomas P, et al. Direct detection and genotyping of Klebsiella pneumoniae carbapenemases from urine by use of a new DNA microarray test. J Clin Microbiol. 2012;50:3990–8. 208. Yu X, Susa M, Weile J, Knabbe C, Schmid RD, Bachmann TT. Rapid and sensitive detection of fluoroquinolone-resistant Escherichia coli from urine samples using a genotyping DNA microarray. Int J Med Microbiol IJMM. 2007;297:417–29. 209. Mach KE, Wong PK, Liao JC. Biosensor diagnosis of urinary tract infections: a path to better treatment? Trends Pharmacol Sci. 2011;32:330–6. 210. Satterwhite CL, Torrone E, Meites E, et al. Sexually transmitted infections among US women and men: prevalence and incidence estimates, 2008. Sex Transm Dis. 2013;40:187–93. 211. Cao B, Wang S, Tian Z, Hu P, Feng L, Wang L. DNA microarray characterization of pathogens associated with sexually transmitted diseases. PLoS One. 2015;10:e0133927. 212. Aguilera-Arreola MG, Gonzalez-Cardel AM, Tenorio AM, Curiel-Quesada E, Castro- Escarpulli G. Highly specific and efficient primers for in-house multiplex PCR detection of chlamydia trachomatis, Neisseria gonorrhoeae, mycoplasma hominis and Ureaplasma urealyticum. BMC Res Notes. 2014;7:433.

446

S. Dunbar et al.

213. Shi G, Wen SY, Chen SH, Wang SQ. Fabrication and optimization of the multiplex PCRbased oligonucleotide microarray for detection of Neisseria gonorrhoeae, chlamydia trachomatis and Ureaplasma urealyticum. J Microbiol Methods. 2005;62:245–56. 214. Albrecht V, Chevallier A, Magnone V, et al. Easy and fast detection and genotyping of high-risk human papillomavirus by dedicated DNA microarrays. J Virol Methods. 2006;137:236–44. 215. Catarino R, Petignat P, Dongui G, Vassilakos P. Cervical cancer screening in developing countries at a crossroad: emerging technologies and policy choices. World J Clin Oncol. 2015;6:281–90. 216. Hwang TS, Jeong JK, Park M, Han HS, Choi HK, Park TS. Detection and typing of HPV genotypes in various cervical lesions by HPV oligonucleotide microarray. Gynecol Oncol. 2003;90:51–6. 217. Arron ST, Skewes-Cox P, Do PH, et al. Validation of a diagnostic microarray for human papillomavirus: coverage of 102 genotypes. J Nucleic Acids. 2011;2011:756905. 218. Shen-Gunther J, Rebeles J. Genotyping human papillomaviruses: development and evaluation of a comprehensive DNA microarray. Gynecol Oncol. 2013;128:433–41. 219. Klaassen CHW, Prinsen CFM, de Valk HA, Horrevorts AM, Jeunink MAF, Thunnissen FBJM. DNA microarray format for detection and subtyping of human papillomavirus. J Clin Microbiol. 2004;42:2152–60. 220. Ko K, Yu S, Kwon MJ, Lee EH, Woo HY, Park H. Comparison of advansure HPV genoblot and hybrid capture 2 assays for detection of HPV infection. J Clin Lab Anal. 2017;32:1–3. 221. Stevens MP, Garland SM, Rudland E, Tan J, Quinn MA, Tabrizi SN. Comparison of the Digene hybrid capture 2 assay and Roche AMPLICOR and LINEAR ARRAY human papillomavirus (HPV) tests in detecting high-risk HPV genotypes in specimens from women with previous abnormal pap smear results. J Clin Microbiol. 2007;45:2130–7. 222. Ermel A, Qadadri B, Morishita A, et al. Human papillomavirus detection and typing in thin prep cervical cytologic specimens comparing the Digene hybrid capture II assay, the Roche linear Array HPV genotyping assay, and the Kurabo GeneSquare microarray assay. J Virol Methods. 2010;169:154–61. 223. Park S, Kang Y, Kim DG, Kim EC, Park SS, Seong MW. Comparison of the analytical and clinical performances of Abbott RealTime high risk HPV, hybrid capture 2, and DNA Chip assays in gynecology patients. Diagn Microbiol Infect Dis. 2013;76:432–6. 224. Cho EJ, Do JH, Kim YS, Bae S, Ahn WS. Evaluation of a liquid bead array system for high- risk human papillomavirus detection and genotyping in comparison with hybrid capture II, DNA chip and sequencing methods. J Med Microbiol. 2011;60:162–71. 225. Sato T, Takayanagi A, Nagao K, et al. Simple PCR-based DNA microarray system to identify human pathogenic fungi in skin. J Clin Microbiol. 2010;48:2357–64. 226. Carlos CA, Tang YW, Adler DJ, Kovarik CL. Mycobacterial infection identified with broad- range PCR amplification and suspension array identification. J Cutan Pathol. 2012;39:795–7. 227. Saeed K. Diagnostics in prosthetic joint infections. J Antimicrob Chemother. 2014;69(Suppl 1):i11–9. 228. Hinic V, Aittakorpi A, Suter S, et al. Evaluation of the novel microarray-based Prove-it Bone&Joint assay for direct detection of pathogens from normally sterile body sites in comparison with culture and broad-range bacterial PCR. J Microbiol Methods. 2014;107:38–40. 229. Metso L, Maki M, Tissari P, et al. Efficacy of a novel PCR- and microarray-based method in diagnosis of a prosthetic joint infection. Acta Orthop. 2014;85:165–70. 230. Jackson SA, Kotewicz ML, Patel IR, Lacher DW, Gangiredla J, Elkins CA. Rapid genomic- scale analysis of Escherichia coli O104:H4 by using high-resolution alternative methods to next-generation sequencing. Appl Environ Microbiol. 2012;78:1601–5. 231. Laksanalamai P, Joseph LA, Silk BJ, et al. Genomic characterization of Listeria monocytogenes strains involved in a multistate listeriosis outbreak associated with cantaloupe in US. PLoS One. 2012;7:e42448.


447

232. Patel IR, Gangiredla J, Lacher DW, et al. FDA Escherichia coli identification (FDA-ECID) microarray: a Pangenome molecular toolbox for serotyping, virulence profiling, molecular epidemiology, and phylogeny. Appl Environ Microbiol. 2016;82:3384–94. 233. Gawlik D, Slickers P, Engelmann I, et al. DNA-microarray-based genotyping of Clostridium difficile. BMC Microbiol. 2015;15:158. 234. Tuberculosis. 2017. (Accessed December 14, 2017, at http://www.who.int/mediacentre/ factsheets/fs104/en/). 235. Kamerbeek J, Schouls L, Kolk A, et al. Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin Microbiol. 1997;35:907–14. 236. Ruettger A, Nieter J, Skrypnyk A, et al. Rapid spoligotyping of Mycobacterium tuberculosis complex bacteria by use of a microarray system with automatic data processing and assignment. J Clin Microbiol. 2012;50:2492–5. 237. Kim SY, Park YJ, Song E, et al. Evaluation of the combichip mycobacteria drug-resistance detection DNA chip for identifying mutations associated with resistance to isoniazid and rifampin in Mycobacterium tuberculosis. Diagn Microbiol Infect Dis. 2006;54:203–10. 238. Linger Y, Kukhtin A, Golova J, et al. Simplified microarray system for simultaneously detecting rifampin, isoniazid, ethambutol, and streptomycin resistance markers in Mycobacterium tuberculosis. J Clin Microbiol. 2014;52:2100–7. 239. Miotto P, Cirillo DM, Migliori GB. Drug resistance in Mycobacterium tuberculosis: molecular mechanisms challenging fluoroquinolones and pyrazinamide effectiveness. Chest. 2015;147:1135–43. 240. Palacios G, Quan PL, Jabado OJ, et al. Panmicrobial oligonucleotide array for diagnosis of infectious diseases. Emerg Infect Dis. 2007;13:73–81. 241. Lezar S, Barros E. Oligonucleotide microarray for the identification of potential mycotoxigenic fungi. BMC Microbiol. 2010;10:87. 242. Ngamskulrungroj P, Price J, Sorrell T, Perfect JR, Meyer W. Cryptococcus gattii virulence composite: candidate genes revealed by microarray analysis of high and less virulent Vancouver island outbreak strains. PLoS One. 2011;6:e16076. 243. Kang X, Li Y, Sun H, et al. Subtype identification of the novel a H1N1 and other human influenza a viruses using an oligonucleotide microarray. Arch Virol. 2010;155:55–61. 244. Li X, Qi X, Miao L, et al. Detection and subtyping of influenza a virus based on a short oligonucleotide microarray. Diagn Microbiol Infect Dis. 2009;65:261–70. 245. Ryabinin VA, Kostina EV, Maksakova GA, Neverov AA, Chumakov KM, Sinyakov AN. Universal oligonucleotide microarray for sub-typing of influenza a virus. PLoS One. 2011;6:e17529. 246. Braun SD, Monecke S, Thurmer A, et al. Rapid identification of carbapenemase genes in gram-negative bacteria with an oligonucleotide microarray-based assay. PLoS One. 2014;9:e102232. 247. Cuzon G, Naas T, Bogaerts P, Glupczynski Y, Nordmann P. Evaluation of a DNA microarray for the rapid detection of extended-spectrum beta-lactamases (TEM, SHV and CTX-M), plasmid-mediated cephalosporinases (CMY-2-like, DHA, FOX, ACC-1, ACT/MIR and CMY- 1-like/MOX) and carbapenemases (KPC, OXA-48, VIM, IMP and NDM). J Antimicrob Chemother. 2012;67:1865–9. 248. Naas T, Cuzon G, Truong H, Bernabeu S, Nordmann P. Evaluation of a DNA microarray, the check-points ESBL/KPC array, for rapid detection of TEM, SHV, and CTX-M extended- spectrum beta-lactamases and KPC carbapenemases. Antimicrob Agents Chemother. 2010;54:3086–92. 249. Veenemans J, Overdevest IT, Snelders E, et al. Next-generation sequencing for typing and detection of resistance genes: performance of a new commercial method during an outbreak of extended-spectrum-beta-lactamase-producing Escherichia coli. J Clin Microbiol. 2014;52:2454–60.

448

S. Dunbar et al.

250. WHO. Antimicrobial resistance. Global report on surveillance. Geneva, World Health Organization; 2014. 251. Strommenger B, Schmidt C, Werner G, Roessle-Lorch B, Bachmann TT, Witte W. DNA microarray for the detection of therapeutically relevant antibiotic resistance determinants in clinical isolates of Staphylococcus aureus. Mol Cell Probes. 2007;21:161–70. 252. Zimmerman CE, Stamper PD, Bryant L, et al. Development of a simple, low-density array to detect methicillin-resistant Staphylococcus aureus and mecA dropouts in nasal swabs. J Microbiol Methods. 2012;91:366–76. 253. Strauss L, Ruffing U, Abdulla S, et al. Detecting Staphylococcus aureus virulence and resistance genes: a comparison of whole-genome sequencing and DNA microarray technology. J Clin Microbiol. 2016;54:1008–16. 254. Adler H, Straub C, Frei R. Comparison of BacT/ALERT 3D, Lowenstein-Jensen medium and Middlebrook 7H10/7H11 biplate for recovering mycobacteria from clinical specimens. Eur J Clin Microbiol Infect Dis Off Publ Eur Soc Clin Microbiol. 2005;24:499–500. 255. Samra Z, Kaufman L, Bechor J, Bahar J. Comparative study of three culture systems for optimal recovery of mycobacteria from different clinical specimens. Eur J Clin Microbiol Infect Dis Off Publ Eur Soc Clin Microbiol. 2000;19:750–4. 256. WHO. Global tuberculosis report 2017. Geneva: World Health Organization; 2017. 257. Denkin S, Volokhov D, Chizhikov V, Zhang Y. Microarray-based pncA genotyping of pyrazinamide-resistant strains of Mycobacterium tuberculosis. J Med Microbiol. 2005;54:1127–31. 258. Tang P, Wang X, Shen X, et al. Use of DNA microarray chips for the rapid detection of Mycobacterium tuberculosis resistance to rifampicin and isoniazid. Exp Ther Med. 2017;13:2332–8. 259. Zimenkov DV, Antonova OV, Kuz'min AV, et al. Detection of second-line drug resistance in Mycobacterium tuberculosis using oligonucleotide microarrays. BMC Infect Dis. 2013;13:240. 260. Cabibbe AM, Miotto P, Moure R, et al. Lab-on-chip-based platform for fast molecular diagnosis of multidrug-resistant tuberculosis. J Clin Microbiol. 2015;53:3876–80. 261. Shimizu Y, Dobashi K, Yoshikawa Y, et al. Five-antituberculosis drug-resistance genes detection using array system. J Clin Biochem Nutr. 2008;42:228–34. 262. Zimenkov DV, Kulagina EV, Antonova OV, Zhuravlev VY, Gryadunov DA. Simultaneous drug resistance detection and genotyping of Mycobacterium tuberculosis using a low-density hydrogel microarray. J Antimicrob Chemother. 2016;71:1520–31. 263. Nosova E, Krasnova MA, Galkina K, Makarova MV, Litvinov VI, Moroz AM. Comparing performance of “TB-BIOCHIP”, “Xpert MTB/RIF” and “genotype MTBDRplus” assays for fast identification of mutations in the Mycobacterium tuberculosis complex in sputum from TB patients. Mol Biol. 2013;47:267–74. 264. Badiee P, Hashemizadeh Z. Opportunistic invasive fungal infections: diagnosis & clinical management. Indian J Med Res. 2014;139:195–204. 265. Goncalves SS, Souza ACR, Chowdhary A, Meis JF, Colombo AL. Epidemiology and molecular mechanisms of antifungal resistance in Candida and Aspergillus. Mycoses. 2016;59:198–219. 266. Chowdhary A, Sharma C, Kathuria S, Hagen F, Meis JF. Azole-resistant Aspergillus fumigatus with the environmental TR46/Y121F/T289A mutation in India. J Antimicrob Chemother. 2014;69:555–7. 267. Chowdhary A, Sharma C, Kathuria S, Hagen F, Meis JF. Prevalence and mechanism of triazole resistance in Aspergillus fumigatus in a referral chest hospital in Delhi, India and an update of the situation in Asia. Front Microbiol. 2015;6:428. 268. Yan L, Zhang J, Li M, et al. DNA microarray analysis of fluconazole resistance in a laboratoryCandida albicansstrain. Acta Biochim Biophys Sin. 2008;40:1048–60. 269. Combe M, Sanjuan R. Variation in RNA virus mutation rates across host cells. PLoS Pathog. 2014;10:e1003855.


449

270. Masimba P, Gare J, Klimkait T, Tanner M, Felger I. Development of a simple microarray for genotyping HIV-1 drug resistance mutations in the reverse transcriptase gene in rural Tanzania. Tropical Med Int Health. 2014;19:664–71. 271. Zhang G, Cai F, Zhou Z, et al. Simultaneous detection of major drug resistance mutations in the protease and reverse transcriptase genes for HIV-1 subtype C by use of a multiplex allele- specific assay. J Clin Microbiol. 2013;51:3666–74. 272. Martin V, Perales C, Fernandez-Algar M, et al. An efficient microarray-based genotyping platform for the identification of drug-resistance mutations in majority and minority subpopulations of HIV-1 Quasispecies. PLoS One. 2016;11:e0166902. 273. Gauthier M, Bonnaud B, Arsac M, et al. Microarray for hepatitis B virus genotyping and detection of 994 mutations along the genome. J Clin Microbiol. 2010;48:4207–15. 274. Hua W, Zhang G, Guo S, Li W, Sun L, Xiang G. Microarray-based genotyping and detection of drug-resistant HBV mutations from 620 Chinese patients with chronic HBV infection. Braz J Infect Dis. 2015;19:291–5. 275. Jang H, Kim J, Choi JJ, Son Y, Park H. Peptide nucleic acid array for detection of point mutations in hepatitis B virus associated with antiviral resistance. J Clin Microbiol. 2010;48:3127–31. 276. Zhang Y, Liu Q, Wang D, Chen S, Wang S. Simultaneous detection of oseltamivir- and amantadine-resistant influenza by oligonucleotide microarray visualization. PLoS One. 2013;8:e57154.

Real-Time Detection of Amplification Products Through Fluorescence Quenching or Energy Transfer Caitlin Otto and Shihai Huang

Introduction Molecular diagnostics of microbiological agents involve detection of microbial nucleic acids (i.e., DNA and/or RNA) and often require the amplification of targeted sequences due to low abundances of the microbes or relevant molecular targets. Target amplification technology became available with the advent of polymerase chain reaction (PCR) in the 1980s, which eventually revolutionized the field of clinical diagnostics. While PCR remains the mainstay ever since, several alternative isothermal target amplification technologies have also become available and been used in practice [1, 2], including nucleic acid sequence-based amplification (NASBA) [3, 4], self-sustained sequence replication (3SR) [5, 6], transcription- mediated amplification (TMA) [7], strand displacement amplification (SDA) [8, 9], loop-mediated isothermal amplification (LAMP) [10, 11], recombinase polymerase amplification [12], helicase-dependent amplification (HDA) [13, 14], single primer isothermal amplification (SPIA) [15], isothermal and chimeric primer-initiated amplification of nucleic acids (ICAN) [16], isothermal chain amplification (ICA) [17], and Ionian Technologies amplification [18]. When amplification products are detected, the signals generated are then used to derive the diagnostic results for clinical specimens. Target detection can be achieved using methods in either a heterogeneous or homogeneous format. Heterogeneous detection is performed after target amplification (i.e., “end-point” reaction) and typically involves multiple steps such as target binding to a solid phase (i.e., beads or membranes), hybridization with excess amounts of labeled probes, washing of C. Otto (*) SUNY Downstate Medical Center, Brooklyn, NY, USA e-mail: [email protected] S. Huang Abbott Molecular Inc., Des Plaines, IL, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_21

451

452

C. Otto and S. Huang

unbound probes, and signal generation. The types of signals that have been routinely detected include radioactive decay of radioisotopes, chemiluminescence, fluorescence emission, fluorescence polarization, light scattering, and others. The target specificity in signal generation is achieved by interrogating physical and chemical properties unique to the target sequence, such as size, electrostatic charge, and various biomolecular interactions (e.g., biotin/avidin, antibody/antigen, and nucleic acid binding). Heterogeneous target amplification and detection methods have been developed. and established in both research and clinical settings. There are, however, aspects intrinsic to the heterogeneous assay workflow that prevent its wider adoption in clinical microbiology in the era of expanded application of molecular testing. Association of targets to solid surfaces slows down reaction kinetics, which requires longer time for hybridization reactions. Separate reactions for target amplification, target hybridization, probe hybridization, separation of unbound molecules, and signal detection make it conceivably difficult to develop an automated assay procedure. Further, direct manipulation of amplification products increases the likelihood of contamination due to unintended carryover of amplicons. These limitations of heterogeneous assay format may lead to slower assay turnaround and elevated levels of unforced human errors or incorrect assay results. In contrast, target amplification and detection in homogeneous reactions take place in a closed reaction vessels. Targets can then be detected as they are being amplified, thus a homogeneous reaction is often referred to as a “real-time” or “kinetic” reaction. The term “real-time” will be used to represent homogeneous methods throughout this chapter. In a real-time assay, target amplification and detection simultaneously occur but are mechanistically interdependent processes. Some close-tube assays detect amplified products as a separate step after the amplification reaction is completed, but such assay technologies are not included in the discussion here. Contrary to heterogeneous methods, real-time assays bypass the requirement of multi-step post-amplification sample processing, and therefore they can provide shorter result turnaround time and are amenable to automation. In addition, because there is no need to open the reaction vessel containing amplification products throughout the assay procedure, amplicon contamination is minimized. Real-time amplification and detection methods also provide robust quantitative measurements over wide dynamic ranges. For real-time PCR reactions, the signal being generated during the course of the amplification is proportional to the amount of accumulated products at a given point during the reaction. Real-time instruments continuously monitor signal amplitude and record the number of cycles required before there is sufficient amount of product to cross a predefined threshold (threshold cycle or CT value), usually in the early exponential phase. The threshold cycle is inversely proportional to the amount of the input target (i.e., higher target level leads to fewer cycles needed to achieve detection and vice versa). The measurement of threshold cycles has been demonstrated to be highly reliable, and its relationship with input target level is well defined over a much wider dynamic range compared to the measurement of end-point PCR signals (as in the case for heterogeneous reactions).

Real-Time Detection of Amplification Products Through Fluorescence Quenching…

453

Therefore, real-time PCR has become the method of choice for the quantitative detection of microbial agents, and the recent years have witnessed the increasing adoption of homogeneous methodologies in general. It is important to note that even though there are numerous advantages of real- time assays in quantitative detection over the traditional end-point heterogeneous assays, real-time assays can be equally effective in providing accurate qualitative results. Further, an important advantage for real-time assays is that signals throughout the course of an amplification reaction can be recorded, thereby allowing the kinetics of the reaction to be analyzed. This information can then be used to detect abnormalities in the assay that could indicate potentially incorrect or unreliable results. The ability to provide assay validity criteria to ascertain reliable results with high confidence is a critical requirement in the highly regulated in vitro diagnostic field. As a result, most commercial real-time assays have been developed with sophisticated systems of validity checks around many of the kinetic characteristics of the reaction so that signals from abnormal reactions are not mistakenly used to determine patient results. This enables objective, automated data analysis and result reporting. A real-time reaction contains components that support amplification, target detection, and signal generation. The amplification reaction components typically include primers, enzymes, nucleotides, and a buffer. These components are the same in heterogeneous amplification reactions. Target detection is usually achieved with components that bind to the amplified products. Signals for the amplified products are generated via target-binding components directly or are induced indirectly by target-binding components. These signals are then converted by data reduction processes to an output that can be accurately measured and recorded. In real-time reactions, detection of the amplification products needs to take place without first removing post-amplification unbound target components. To achieve this goal, a prominent technology used is the measurement of fluorescence signal intensity through the use of fluorophore-labeled oligomers (probes or primers). These oligomers consist of nucleotides or analogs that are designed to efficiently hybridize to amplification targets in a highly sequence-specific manner. Upon hybridization, the fluorophore is modified in a manner such that the intensity of the fluorescence signal significantly increases or decreases. Several possible mechanisms can play a role in the fluorescence modification, among which the most prominent ones include radiation-free fluorescence quenching through direct contact by a quenching moiety [19, 20] and fluorescence resonance energy transfer (or Förster resonance energy transfer, FRET) [19]. Fluorescence quenching or FRET-based real-time methods are among the most widely utilized molecular diagnostic methods in clinical microbiology [21] and will be the focus of this chapter. In addition to fluorophore-labeled oligomers, fluorescent dyes that can insert themselves in the amplified targets upon which fluorescence signal is modified have also been widely used in real-time assays [21] and will be discussed in this chapter as well. Other fluorescence-related technologies besides signal intensity measurement, such as fluorescence polarization and fluorescent correlation analysis, will not be discussed in this chapter.

454


The scope of the discussions in this chapter covers the following topics related to fluorescence-based target detection and signal generation in real-time assays: (1) signal modification principles for fluorescence quenching and energy transfer, (2) target detection/signal generation technologies capable of real-time detection of amplification products, (3) instrument systems, and (4) data analysis and result reporting for real-time assays. Several other topics related to amplification and real- time assays are covered elsewhere in this book. PCR or non-PCR target amplification technologies are discussed in chapters “ Molecular Typing Techniques: State of the Art and PCR and Its Variations”. Real-time quantification and amplification product detection are discussed in chapters “Direct Nucleotide Sequencing for Amplification Product Identification and Solid and Suspension Microarrays for Detection and Identification of Infectious Diseases”. It is to note that, besides target amplification, probe amplification (e.g., Qβ replicase [22], rolling cycle amplification (RCA), and ramification amplification method (RAM) [23–26]) and signal amplification (e.g., Invader chemistry [27]) have also been used to detect very small amount of analytes of interest which often involves enzymatic cleavage or replication. Signal amplification methods involve target detection and signal modification in which signals continue to accumulate and amplify. Signal amplification can be designed with or without concurrent target amplification (either homogeneous or heterogeneous). Some real-time methods involving both target amplification and signal amplification in one assay will be discussed in this chapter. The discussions on probe and signal amplification methods without target amplification or with heterogeneous target amplification may be found in chapter “Non-PCR Amplification Techniques”.

Fluorescence Signal Modification In order to detect targets in a real-time assay, signals generated with the amplified targets need to be differentiated from signals when targets are absent or not amplified. Because fluorescence signals can be easily modified by the environment of the fluorescence moiety, detection of fluorescence signal modification has been the main technology to generate signals in real-time assays. The principle of fluorescence signal and three major mechanisms of fluorescence modifications will be discussed here.

Principle of Fluorescence Signal Fluorescence is a natural process by which certain substances can absorb light or other electromagnetic radiations at specific wavelengths and then emit light at another (usually higher) wavelength. Fluorescence emission is a dynamic process where energy conversion from the excited state to the ground state generates a fluorescence signal that competes with multiple other conversion processes, and the relevant spectroscopic rate constants of


455

these processes are influenced by the environment for the fluorescent moiety. The types of conversions through which the excitation energy is lost spontaneously include the following: intrinsic fluorescence emission radiation decay, internal conversion from excited to ground state, inter-system crossing, collisional quenching, and resonance dipole energy transfer. In real-time assays, fluorescence technologies are used together with detection technologies in ways that these energy conversions will be directly or indirectly impacted by the presence of amplified products. As a result, fluorescence is modified, and the modification can be detected and recorded as the testing data from which diagnostic results are derived. Refer to Clegg [28] and Morrison [19] for more detailed discussion on the principles of fluorescence signals. Refer to later sections of this chapter for further discussion on detection technologies.

Fluorescence Resonance Energy Transfer (FRET) FRET is a process whereby nonradiative energy transfers between two moieties, a “donor” and an “acceptor” over 10–100 Å distances through dipole–dipole coupling. The acceptor may be a fluorescent or nonfluorescent moiety. When FRET occurs, the fluorescence of the donor is reduced compared to when no acceptor is present and the fluorescence of the acceptor is increased. In order for FRET to occur, the emission spectrum of the donor must overlap with the excitation spectrum of the acceptor and that the quantum yield, or the number of photons emitted relative to the number of absorbed photons, of the donor and absorption coefficient of the acceptor are sufficiently high. The efficiency of energy transfer strongly depends on the distance between the donor and acceptor moieties [29]. As a result, measurement of the fluorescence change due to FRET has been used effectively to study the static and dynamic molecular structures within the scale of 10–100 Å as well as the thermodynamics and kinetics of biomolecular interactions. For this reason, FRET has often been referred to as a “spectroscopic ruler” [29, 30]. To be used in diagnostic methods, the change in fluorescence signal due to FRET for a given donor–acceptor system needs to be measurable (i.e., significantly above the background noise) in response to the presence of the target. Considering the theory behind FRET as described above, the design of the donor–acceptor system is often focused on the distance between the two (or more) moieties when FRET is desired to occur as well as the selection of donor–acceptor pairs. It is most common that the donor and acceptor exist as labels on an oligonucleotide and thus the distance is often expressed as the nucleotide bases between the two labels. In this context, when estimating the distance between the donor and acceptor, it is important to consider the rigidity of the structures where labels are attached, spacing between labeled bases in the oligonucleotide (e.g., the distance between adjacent base pairs in a DNA double helix is 3.4 Å), the length of linkage for the labeled moieties, the position of attachment, and the interaction between the labeled moieties and the oligonucleotide. For DNA double helix, the distance also affects the dipole–dipole orientation which may in turn affect FRET efficiency. The use of

456


nonfluorescent labels as acceptor or quencher would preclude the measurement option of monitoring the increase of signals from acceptor but may simplify the design of a FRET system. In addition to the widely used fluorescent quenchers, several unique nonfluorescent quenchers (such as nucleotides [31] and gold particles [32, 33]) have been introduced for use, and some are found to be significantly more efficient than conventional quenchers [32]. It is also important to point out that FRET occurs between moieties that do not contact each other. This is unlike the collisional quenching that requires direct contact. Refer to Tables 1 and 2 for lists of common dyes and quenchers used in oligonucleotide labeling and their fluorescence properties. To choose dyes for assay design, multiple factors have to be considered besides the fluorescence properties of the labels, such as compatibility with the optical design of existing instrument, oligonucleotide coupling chemistry where applicable, stability of the labels, non-FRET interference from other reaction components (e.g., fluorescence quenching by specific nucleotides bases or oligonucleotide sequences), and commercial availability of the labels.

Collisional Quenching Collisional quenching is another mechanism by which fluorescence can be generated. Unlike FRET, collisional quenching requires direct contact between labels, and the energy conversion efficiency depends on the effective concentration of the quencher label and the bimolecular rate constant (usually controlled by diffusion) [29]. Even though there may be common outcomes from collisional quenching and certain FRET system, i.e., the reduction of donor signals, collisional quenching has distinct spectroscopic mechanisms and conditions [20]. First, collisional quenching does not require overlapping emission spectrum and absorption spectrum between the labels and is not impacted by the extent of the overlap, whereas FRET efficiency is significantly impacted by the spectrum overlap. Second, when both labels are fluorescent, collisional quenching results in the reduction of emission of both labels, whereas FRET leads to increase of emission of one label and reduction of the other. Third, the absorption spectrums of the labels in collisional quenching are altered compared with when the labels are separate, whereas the absorption spectrums stay unchanged in the case of FRET. Fourth, when same labels are involved, the efficiency of collisional quenching is higher than FRET-mediated quenching. Collisional quenching is often achieved by the blunt-end double-stranded oligonucleotide design where both strands are terminally labeled so that the two labels are in close contact. In such designs, hybridization stability of the oligonucleotides correlates positively with quenching efficiency. It is also to note that the interaction between the two labels can further increase the stability of double-stranded oligonucleotides, leading to higher quenching efficiency [20]. Collisional quenching, when achieved by oligonucleotide hybridization, does not necessarily require a quenching label. Instead, fluorescence quenching by specific nucleotides is well documented and has been used in methods based on collisional quenching [31].


457

Table 1 Common dyes for oligonucleotide labeling Dye name Dansyl-X Alexa Fluor 350 Biosearch Blue Cascade Blue Alexa Fluor 430 Pulsar 650 Cy2 Oregon Green 488 Alexa Fluor 488 FAM Rhodol Green Orange Green 500 BODIPY FL Rhodamine Green Oregon Green 514 JOE TET CAL Fluor Gold 540 BODIPY R6G JOE Yakima Yellow Alexa Fluor 532 BODIPY 530/550 HEX VIC CAL Fluor Orange 560 BODIPY TMR-X NED Quasar 570 Cy3 Alexa Fluor 555 Alexa Fluor 546 TAMRA BODIPY 558/568 Rhodamine Red-X BODIPY 564/570 CAL Fluor Red 590 BODIPY 576/589 ROX Alexa Fluor 568 Redmond Red BODIPY 581/591

Absorption peak wavelength (nm) 335 343 352 396 434 460 489 494 495 495 495 499 502 503 506 520 521 522 528 529 531 532 534 535 538 538 544 546 547 549 555 556 557 558 560 563 569 575 575 578 579 581

Emission peak wavelength (nm) 518 441 447 410 539 650 506 517 518 520 521 519 510 528 526 548 536 544 547 555 549 554 554 556 554 559 570 575 570 566 565 573 583 569 580 569 591 588 602 603 595 591 (continued)

458


Table 1 (continued) Dye name Cy3.5 BODIPY TR-X CAL Fluor Red 610 Alexa Fluor 594 Texas Red CAL Fluor Red 635 LC Red 640 BODIPY 630/650 Alexa Fluor 633 BODIPY 650/665 Cy5 Quasar 670 Alexa Fluor 647 Alexa Fluor 660 Cy5.5 IRDye 800RS Alexa Fluor 680 LC Red 705 Quasar 705 Alexa Fluor 700 Cy7 Alexa Fluor 750

Absorption peak wavelength (nm) 581 588 590 590 597 618 625 625 632 646 646 647 650 663 675 676 679 680 690 702 743 749

Emission peak wavelength (nm) 596 616 610 617 616 637 640 640 647 660 669 670 665 690 694 786 702 710 705 723 767 775

Nucleic Acid Dyes There are dyes that can bind to nucleic acids in a sequence-independent manner (without being conjugated to an oligonucleotide), and upon binding, their fluorescence signal increases significantly. These dyes are also called nucleic acid stains. Nucleic acid dyes can be grouped into three major classes based on the mechanism of signal generation: (1) intercalating dyes such as phenanthridine compound (also known as ethidium bromide), cyanine dyes (including SYBR Green, PicoGreen, etc.), and propidium iodide; (2) minor groove binders such as DAPI and Hoechst dyes; and (3) other miscellaneous dyes that achieve external binding. Intercalating dyes are the most commonly used nucleic acid dyes. They contain planar structures that insert between stacked bases in the nucleic acids, thereby creating a large increase in fluorescence signal relative to the free dye in solution. The signal increase is proportional to the amount of nucleic acids. This property allows the dyes to be used in either direct quantification of nucleic acid samples or real-time amplification-based detection and quantification. SYBR Green I is one of the intercalating dyes that generate signals by inserting into double-stranded DNA. It has been the most widely used nucleic acid dye in real-time PCR method due to high- level sensitivity for DNA detection and relatively cheap price. In addition to


459

Table 2 Common quenchers for oligonucleotide labeling Dye name DDQ I Dabcyl QSY 35 QXL 490 BHQ-0 QXL 520 Eclipse Iowa Black FQ BHQ-1 QSY7 QSY9 QXL 570 BHQ-2 QXL 610 DDQ II Blackberry Quencher 650 Iowa Black RQ QSY 21 QXL 670 BHQ-3 QXL 680 IRDye QC-1

Absorption peak wavelength (nm) 430 453 475 490 495 520 522 531 534 560 562 570 579 610 630 650 656 661 670 672 680 737

Approximate quenching range (nm) 380–510 403–507 434–509 410–540 425–550 460–560 460–590 420–620 460–595 535–582 535–587 520–600 510–645 520–660 550–680 550–750 500–700 601–695 600–690 575–730 610–710 660–870

methods using only nucleic acid dyes to achieve non-sequence-specific detection, some methods have also been designed where nucleic acid dyes are covalently attached to an oligomer and provide increased fluorescent signal upon sequencespecific hybridization. Refer to Table 3 for a list of nucleic acid dyes that have been used in real-time assays [34–43].

Detection of Amplification Products Real-time assay design starts with selection of target region and primer sequences as well as optimization of amplification efficiency (reagent composition, concentration, and cycling conditions). To monitor a real-time assay, signals have to be generated for the target being amplified. Target detection and signal generation can be achieved either in a sequence-specific manner by using oligomers consisting of nucleotides or their analogs or in a sequence-nonspecific manner by using nucleic acid dyes that generate a fluorescent signal upon binding to double-stranded

460


Table 3 Nucleic acid dyes used in real-time PCR Dye name LC Green ResoLight BO-PRO-1 BEBO SYTO 9 SYTO 13 SYTO 16 YO-PRO-1 SYBR Green I EvaGreen TO-PRO-1 BOXTO SYTO 82 SYTOX Orange SYTO 64 TO-PRO-3 SYTO 62 SYTO 60

Absorption peak wavelength (nm) 440–470 450–500 462 467 485 488 488 491 497 500 515 515 541 546 598 649 649 652

Emission peak wavelength (nm) 470–520 487 481 492 498 509 518 509 520 530 531 552 560 570 620 655 680 678

DNA. There is a wide range of oligomer designs suitable for real-time target detection. They vary greatly in physical characteristics such as base type (DNA, RNA, or other analogs), length, and multiplicity of the components, direct or indirect signal generation (mentioned above), biochemical/biophysical mechanisms enabling signal generation by energy transfer or fluorescence quenching/enhancement (e.g., hybridization, conformational change, and enzymatic cleavage), and presence or absence of signal amplification in addition to target detection, etc. Needless to say, each design has its advantages and disadvantages, and though every design has its flexibility, certain designs may fit specific needs better than others. Multiple factors have to be considered when designing target detection/signal generation systems: (1) optimizing sensitivity for detection of low-level targets, (2) maximizing mismatch tolerance or discrimination, (3) optimizing specificity to achieve minimal background signals, and (4) minimizing pre-amplification background signals to improve signal-to-noise ratio. A potential concern with the use of nucleic acid dyes is that high level of genomic background will increase the pre-amplification background signal. In addition, any unintended nonspecific amplification such as primer dimer formation can generate signals that lead to false-positive results. Therefore, care needs to be taken during primer design to eliminate potential nonspecific signals or reduce them to an acceptable level. The sections below will include discussions on representative designs but is not intended to be a comprehensive list. Only design principles and basic performance characteristics will be described. The discussion will not be focused on the findings or conclusions related to advantage or disadvantages of certain designs based on empirical experience from specific


461

studies, with the notion in mind that each design can be optimized and each design may fit one utility better than others within specific contexts. To facilitate discussions, sequence-specific target detection technologies are roughly categorized in four groups depending upon the components enabling signal generation. The first group relies on design of oligonucleotide probes to recognize the amplified target sequences and generate signals; the second group relies on design of primers beyond their target amplification function to generate signals upon extension of the template; the third group requires both probe and primer in order to generate signals; and the fourth group requires fluorophore-labeled nucleobases. One important difference between groups one and three and groups two and four is that the former designs take advantage of both primer and probe sequences to ensure target specificity of the reaction, whereas the latter designs depend solely on target specificity provided by primer sequences. As a result, designs belonging to groups two and four are theoretically more prone to nonspecific signals due to nonspecific amplification such as primer dimers.

Probe Designs for Target Detection and Signal Generation 5′-to-3′ Exonuclease Assay Many real-time assays are designed based on the 5′-to-3′ exonuclease activity of the polymerase that amplifies target sequences [44]. Probes used in such assays are sequence-specific, linear oligonucleotides containing a fluorescent dye “reporter” on one end of the probe and a “quencher” molecule on the other end or at an internal site. When the two molecules are in close proximity as in an intact probe, light emission from the reporter is transferred to the quenching molecule. During the extension step of the PCR reaction, the exonuclease activity of the polymerase cleaves the 5′ molecule off of the probe, separating the reporter and quencher molecules. At this point, when the sample is excited with light, the reporter is able to emit a detectable signal. The fluorescence signal measured during a real-time assay is directly proportional to the amount of cleaved probes and, in turn, the accumulated amount of amplification product. Since the 5′-to-3′ exonuclease activity has also been referred to as TaqMan activity, the probes are often called TaqMan probes. Similar to many other technologies in later discussions, assays can be designed to achieve multiplex detection by using multiple TaqMan probes labeled with spectrally distinct fluorescence labels. Due to its relative simplicity, TaqMan probes have been widely adopted in research and clinical assays. A potential limitation, probe length needs to be sufficiently long to enable hybridization to the target at the relatively high polymerase extension temperature (approximately 65 °C). As a result, background signals in the absence of amplified target may be high, and the capability of relatively long TaqMan probes in detecting single-nucleotide polymorphism (SNP) mutations may be limited. One new approach in 5′-to-3′ exonuclease assay [44, 45], called Snake,

462


utilizes a primer with a 5′-flap containing an amplicon-complementary sequence that results in the formation of a stem–loop structure of the PCR amplicon. A dual- labeled probe, together with this stem–loop structure, serves as an optimal substrate for the exonuclease activity of Taq polymerase. Unlike TaqMan probe, the Snake system prefers short probes, therefore may be more efficient in detecting SNPs. However, the design of the 5′-flap can be tricky, where the overstabilization of the secondary structure inhibits PCR efficiency.

Adjacent Probes In this design, two probes bind to adjacent locations on the target sequence [46, 47]. The 3′ end of the upstream probe and the 5′ end of the downstream probe are labeled with two fluorescence moieties capable of energy transfer or quenching. When target sequences accumulate during amplification, these two probes will bind, and the fluorescence signal modification (either signal increase for the fluorescence acceptor or signal decrease for the donor) can be monitored in real time. Real-time assays using adjacent probes often can be followed with a melting curve analysis for identification of mutations. It is to note, however, that the signal is generated with a tertiary structure involving two probes and the target; thus, assay design and optimization can be challenging. Due to the requirement of two probe-binding regions and their proximity, a long contiguous sequence has to be evaluated for fitness as the probe-binding region regarding sequence context and mutations. In addition, when accumulated target level exceeds that of the probe concentration, probes will more likely bind to different target strands than to a single strand. This will lead to a decrease in real-time PCR signal complicating data analysis. A variation of the adjacent probe design has also been reported where probes bind to double-stranded DNA target (rather than single-stranded target) each forming a triplex structure [48]. A modification of adjacent probes is C-probe design [49]. C-probe is a single oligonucleotide probe that is dual-labeled with fluorophore/quencher at both ends containing two terminal regions that bind to adjacent target sequences. As a result of the binding, the two labels are brought together enabling fluorescence modification. In another related C-probe design, the two target-binding regions are in two different probe sequences that associate with each other via a double-stranded stem [49].

Molecular Beacon Probe Molecular beacon probes are single-stranded oligonucleotides that contain self- complementary ends that bind to create a hairpin loop [50, 51]. Each end of the probe is labeled with a reporter and quencher molecule. In the native probe state, in the absence of target, the two ends of the hairpin loop are in close proximity and fluorescence is inhibited. As the target accumulates during the real-time assay,


463

molecular beacon probes hybridize to the complementary sequence of the target, thus separating the reporter and quencher molecules and resulting in an increase in fluorescence signals. The thermodynamic balance between the open and closed states of the probe can be effectively modulated by adjusting the sequence and length of either the stem or the target binding/loop sequence. The opening and closing are synergistically affected by both the loop and stem sequences. Molecular beacon probes with a strong stem and relatively short loop have often been used effectively to detect SNP mutations. In fact, SNP detection is perceived to be a predominant advantage of molecular beacon due to efficient closing of the stem–loop structure. However, assay sensitivity in such cases may be negatively impacted due to the stabilized closed state. A variation in the molecular beacon design [52], i.e., tentacle probe, was developed, which contains a capture probe in addition to the stem–loop structure. The capture probe can be designed to enhance the hybridization stability and rate while maintaining the sequence specificity of the typical molecular beacon. Molecular beacon can also be designed with long loop sequence and relatively short stem to detect heterogeneous sequences in a mutation tolerant manner, though the background signal will be higher. Another variation in molecular beacon is the “wavelength shifted” design, where a “harvester” dye is introduced at a position 5–18 bases away from the fluorophore [53]. The harvester dye has an emission spectrum that overlaps well with the excitation wavelengths of multiple fluorophore, which can presumably facilitate multiplex detection on instrument where only one excitation wavelength is available. Besides these abovementioned designs, there are also other probe designs that use similar fundamental principles as the original molecular beacons, i.e., signal generation resulted from molecular conformational changes driven by target binding. These probe designs include bimolecular beacon [54], tripartite molecular beacon [55], molecular torches [56], dumbbell molecular beacon [57], and cyclicon probe [58].

Complementary Probes Complementary probes are double-stranded probes that consist of two strands of oligonucleotides with complementary sequences [59–61]. The 5′ end of one strand and the 3′ end of the other strand are each labeled with a fluorescent or quenching moiety. Similar to molecular beacons, signal generation mechanism (energy transfer or quenching) depends on the positions of the labels; and in a preferred design, there may be variations in the number of labels. In the absence of targets, due to the formation of double helix consisting of the two probe strands, the two labels are brought into the vicinity of each other. In the presence of targets, the probe strands hybridize to their corresponding complementary target sequence, thereby separating the two labels. To achieve desired assay performance (i.e., sensitivity, mismatch tolerance versus differentiation), the thermodynamic balance between probe binding and target binding can be modulated by multiple factors such as the lengths and

464


sequences of both probe strands and their respective concentrations and molar ratio. The ability to modulate the molar ratio between the two strands is an advantage of the complementary probes over the single-stranded probes (such as TaqMan probe and molecular beacon) for which sequence variation remains the only source of design flexibility. Multiple types of complementary probe designs (categorized by length symmetry) have been introduced, including partially double-stranded probes with differential lengths [59] and symmetric or near-symmetric probes with equal or near-equal lengths between the two strands [60, 61]. While symmetric or near- symmetric probes should work effectively to distinguish mutations, this is accomplished at the expense of slower binding kinetics and potentially lower sensitivity due to competition of the equal length complementary probe with the target. The partially double-stranded probe can be designed either to achieve mismatch tolerance or mismatch discrimination. One significant advantage of partially double- stranded probe is that the design features (i.e., lengths of both probe strands, their molar ratio, and read temperature) can be thermodynamically modulated with considerable independence so that multiple performance requirements such as sensitivity and mismatch tolerance or differentiation can be achieved simultaneously. Partially double-stranded probes with two labeled oligonucleotides overlapping at the complementary region and containing noncomplementary single-stranded regions on both ends have also been described [62].

Single-Stranded Dual Label Hybridization Probes Eclipse probes are singled-stranded probes labeled with a 5′ quencher, 3′ fluorophore [63], and a minor groove binder (MGB) moiety. The MGB is conjugated to the 5′ end and significantly stabilizes the double-stranded probe–target complex. When not hybridized to the target, the fluorescence of the MGB Eclipse probe is efficiently quenched by the interaction between the fluorophore and quencher moieties. Upon hybridization, the probe is stabilized in the double-stranded structure, allowing fluorescence to be emitted. The mechanism of signal generation is independent from the 5′-to-3′ exonuclease activity because the 5′ MGB blocks such an activity. Similar to the Eclipse probes, Pleiades probes are also hybridization probes labeled with a fluorophore, quencher, and a 5′ MGB [64]. The difference is that the fluorophore is labeled at the 5′ end and the quencher at the 3′ end. As a result, Pleiades probes may have lower background signal when probes are not hybridized and higher signal gain upon probe hybridization to the target.

Light-Up and FIT Probes Light-up probe and FIT (forced intercalation) probes are short peptide nucleic acid (PNA) oligomer labeled with asymmetric cyanine dye thiazole orange (TO). PNAs are synthetic oligomers that have high binding strength. TO is an indiscriminant


465

DNA intercalating dye. Light-up probes are labeled with TO at the N-terminus of the PNA [65], whereas FIT is labeled with a quinoline-linked TO as the substitution of an internal nucleobase [66]. In suspension, TO produces little fluorescence, but when bound to double-stranded DNA, the dye produces fluorescent signals that are orders of magnitude brighter [67, 68]. Furthermore, the use of PNA instead of nominal DNA bases eliminates the binding of TO to probe bases, thus reducing the amount of background signal. Because of the high binding strength of PNA oligomers, both Light-up and FIT probes can be designed to be very short (e.g., 10 base long) while still maintaining sufficient hybridization stability. As a result of the short length, Light-up and FIT probes are highly sequence specific and have the capability to differentiate between single base pair mismatches. These short probes can be used to detect/quantify highly diverse microbial strains because it is relatively easy to find short, conserved sequences within which to design probes [69]. The probes can also be designed to straddle the mismatches to detect/identify SNP mutations. It is interesting to note that the internally labeled TO in an FIT probe behaves as a universal base and can be designed at a position complementary to a mutation to achieve tolerance to such a mutation. Importantly, the similarities in both excitation and emission wavelengths between TO and fluorescein mean that these probe designs are compatible with the majority of the existing real-time PCR instruments. The utilities of Light-up and FIT probes have been demonstrated in qualitative or quantitative clinical diagnostic testing based on real-time PCR or isothermal amplification real-time assays [69, 70].

ybridization-Induced Quenching or Dequenching H of Fluorescein–Labeled Oligonucleotides Hybridization-induced dequenching probes are oligonucleotides that lack significant secondary structure, and they possess a fluorophore moiety attached to an internal nucleotide. The internal fluorescein–oligonucleotide conjugate is quenched when free but is subsequently dequenched due to photoinduced electron transfer following hybridization to the target. Due to their electron-donating properties, guanine derivatives have been found to be strong quenchers of fluorescence. As a result, labeling of fluorophores close to a G base has been discouraged in probe designs. Interestingly, it has been found that a probe labeled with G-quenched fluorescence (HyBeacon) can be dequenched upon hybridization to the complementary target strand [71]. Besides guanine, the quenching effect of intercalating dyes has also been utilized to design a probe for which the quenching of fluorescence signals is reduced upon dye’s intercalation with probe–target double-stranded complex [72]. As a result, the fluorescence is generated when targets accumulate. In addition to the quenching effect based on primary sequence adjacency, the quenching effect from G bases brought in close vicinity of the fluorophore via stem–loop formation has also been explored for the

466


design of “Smart” probe. Like molecular beacon, the stem–loop structure opens up upon binding to the target, thereby generating signals due to reduced quenching [73]. All these abovementioned phenomena of hybridization-induced fluorescence dequenching have been or can be used to design probes for real-time assay-based nucleic acid detection and genotyping. Another example of a fluorescence-based quenching probe was designed by Kurata et al. who developed BODIPY® FL-labeled probe. In the BODIPY® FL-labeled probe, fluorescence is quenched after binding by the guanine present in the target sequence [74, 75]. In this case, when the probe hybridizes with a target sequence, its fluorescence was quenched by the guanine in the target at a rate that is proportional to the amount of target DNA.

Double-Quenched Probes One of the quenching dyes often used in FRET-based probe development is TAMRA. TAMRA is an efficient quencher for reporter molecules where the maximum emission is less than 560 nm. Dyes with longer wavelength emissions will not be effectively quenched by TAMRA. In addition, TAMRA has its own fluorescence which complicates data analysis due to cross talk between the channels. Black Hole Quenchers (BHQ) are an improvement over TAMRA-based dyes because they do not emit their own fluorescence signal, thus minimizing background noise. Further, there are multiple BHQ molecules that span the spectral reporter emission range (480–580 nm (BHQ-1), 559–670 nm (BHQ-2), and 620–730 nm (BHQ-3)).

Internally Cleavable Fluorescence Probe BD Probe Tec ET is a single-stranded DNA probe consisting of a stem–loop structure where fluorescein and rhodamine labels are positioned on two opposite sides of the loop and a recognition sequence for the BsoBI enzyme is placed within the loop sequence [76]. This probe has been designed to achieve real-time detection of the nucleic acid targets generated with SDA, based on the FRET principle. Before target amplification by SDA, there is very little signal from fluorescein due to energy transfer to rhodamine. During SDA, the probes bind to the displaced single-stranded target sequences generated from the amplification reaction and are converted to double-stranded DNA as the result of extension by Bst DNA polymerase. This double-stranded product is then cleaved by the BsoBI restriction enzyme due to the presence of the restrictive site engineered in the probe sequence. This cleavage causes the physical separation of the fluorescein and rhodamine labels and loss of


467

energy transfer which in turn leads to increase in fluorescein signals. There have also been similar designs where the donor and acceptor are separated only by a restrictive enzyme site without a stem–loop structure. Besides the above examples where each copy of amplified target supports one signal generation event, signals amplification can be achieved by using a cleavable fluorescence probe where one copy of amplified target supports multiple signal generation events. CataCleave probe (catalytically cleavable fluorescence probe) is one example [77]. CataCleave probe has a chimeric DNA-RNA-DNA sequence with four contiguous internal purine ribonucleotides. Upon binding to the DNA target, the RNA portion of the DNA:RNA hybrid is cleaved by RNase H. The cleaved products dissociate from the target DNA allowing further rounds of probe hybridization and cleavage. As such, each target DNA generates many cleaved probes that can accumulate over several orders of magnitude. Because the probe is labeled with fluorophore/quencher labels adjacent to both DNA/RNA sequence boundaries, probe cleavage results in fluorescence signal modification. Application of CataCleave probe has been demonstrated in real-time assays including both isothermal amplification and PCR.

Invader Chemistry Invader chemistry detects target with probe-mediated sequence-specific hybridization and achieves signal amplification with probe cycling and structure-specific cleavage by Cleavase enzyme [27]. Invader chemistry involves two isothermal reactions. In the first reaction, an Invader oligonucleotide and a downstream probe bind to the single-stranded target and form a substrate structure for the Cleavase, which contains a one-base overlap at the 3′ end of the Invader oligonucleotide and 5′ end of the probe’s target-binding region. In addition to the target-binding region, the probe is designed with an un-complementary arm (flap) 5′ to the target-binding region. Cleavase enzyme cleaves the probe at the overlap position and generates the flap plus one nucleotide. The reaction temperature is set close to the melting temperature of the probe therefore allowing probe to cycle on and off rapidly. As a result, cleaved probes are continuously being generated as reaction proceeds. In the second reaction, the flap released by the first reaction acts as an Invader oligo and binds to a synthetic hairpin structure that is labeled with a FRET dye–quencher pair (FRET cassette). The complex can again be cleaved by the Cleavase enzyme, and as a result the FRET pair is separated allowing fluorescence signal to be detected. This second reaction involves rapid isothermal cycling of the flap on and off the FRET cassette, similar to the first reaction. As a result, Invader chemistry can generate significant signal amplification in a sequence-specific manner. Invader chemistry has been used in a wide range of application including genotyping, target

468


quantification, and copy number variation. This signal amplification technology has also recently been combined with real-time PCR amplification to achieve multiplex quantification of viral genomes [78–80].

Primer Designs for Target Detection and Signal Generation Stem-Loop Primers Three types of primers containing a stem–loop structure have been described in real-time PCR assays: AmpliFluor primer (also called Sunrise primer) [81], LUX primer (Light Upon eXtension) [82], and Scorpion primer [83]. Besides the stem– loop structure, the common features between these primer types are that (1) they are labeled with at least one fluorophore and generate signals upon opening of the stem–loop structure (probes are not needed to generate signals) and (2) the 3′ end of the primer sequence can bind to the complementary target sequence and be extended by the DNA polymerase. AmpliFluor primers are labeled with fluorophore/quencher pair at the two ends of the stem–loop structure, so that no fluorescence is generated due to efficient quenching. When PCR amplification on the complementary strand extends through the AmpliFluor primer region, the primer adopts an extended conformation due to formation of double-stranded DNA separating the two labels. As a result, energy transfer or quenching effect reduces, resulting in increase in fluorescence signal. LUX primer is singly labeled with a fluorophore at the C-5 position of the thymidine close to the 3′ end. In addition, the primer needs to have G or C base around the 3′ end so that the fluorescence is efficiently quenched due to the nucleobase- dependent quenching. Similar to AmpliFluor, LUX primer will fluoresce when its complementary sequence is extended during PCR causing the stem–loop structure to open. It is to note that only those fluorophores that can be efficiently quenched by nearby dG–dC and dC–dG base pairs such as fluorescein, JOE, HEX, TET, ROX, and TAMRA can be used in the LUX design, whereas certain other dyes such as Texas Red, Cy3, and Cy5 cannot be used in LUX primers. The loop in both AmpliFluor and LUX primers simply serves as a connector between the two stem-forming sequences so that a stable stem–loop structure can be formed. In contrast, Scorpion primer contains a loop sequence that is complementary to the extended target sequence downstream of the primer. When the new region is synthesized in the PCR, the loop will hybridize to the newly synthesized region on the same DNA strand and form a large stem–loop structure. As a result, the original stem–loop structure in the primer opens up leading to signal generation. It should be noted that a blocker moiety is inserted in the primer between the stem–loop sequence and target complementary sequence so that the extension of the reverse strand will not open up the primer. These above primer systems can be designed to tolerate or discriminate mutation dependent upon the 3′ end of the primer sequences.


469

Duplex Primer Several duplex primer designs have been introduced. The common features between these duplex primer designs include the following: (1) two complementary strands are used including a primer strand that is extendable from the 3′ end, and (2) both strands are labeled, the primer strand with a fluorescence donor and the other with a fluorescence acceptor or quencher. In one such design, PCR extension of the region complementary to the primer strand displaces the other strand labeled with a quencher and thus allowing signal generation [84]. Another design is called duplex Scorpion primer [85], which is a derivative to the original Scorpion primer design described above. Instead of having a stem as part of the stem–loop structure, the stem for the duplex Scorpion primer is formed by two separate oligonucleotides. The stem sequence in the primer strand is complementary to the PCR-extended target sequence downstream of the primer. Therefore, PCR amplification causes the stem on the primer strand to bind to the extended target region, displacing the other stem bearing the quencher and thus resulting in signal generation.

DzyNA-PCR DzyNA-PCR involves in vitro amplification of target sequences using a DzyNA primer that contains the complementary sequence of a DNAzyme [86, 87]. Upon amplification of the target, the reverse strand is generated that contains the active DNAzyme sequence. DNAzyme sequence consists of a catalytic domain of 15 nucleotides flanked by two substrate recognition domains, and it can cleave nucleic acid substrate at specific RNA phosphodiester bonds (between an unpaired purine and a paired pyrimidine). A chimeric oligonucleotide that contains RNA substrate sequence flanked by DNA-binding sequences and is labeled with FRET donor/ acceptor pair is included as the reporter substrate. During PCR, the reporter substrate will transiently associate with DNAzyme through Watson–Crick pairing yielding multiple cleavage product and signal amplification. This method obviously requires that the reaction conditions support both the target amplification and enzymatic reaction. In addition, the design of substrate recognition domains in the DNAzyme sequence needs to ensure sufficient hybridization at assay conditions and thus efficient cleavage. Similar to DzyNA-PCR, methods have been described that utilize the hammerhead ribozyme sequence either existing in a target sequence or introduced as a primer tag to achieve real-time target detection and signal amplification in an isothermal amplification reaction such as NASBA or 3SR [88]. It is to note that, similar to any assay solely relying on primer for target recognition, nonspecific primer amplification such as false priming or primer dimer may generate nonspecific signals.

470


Primer Labeled with Unnatural DNA Bases A real-time PCR method has been developed that uses a primer containing a 5-methylisocytosine nucleotide base (isoC) labeled with a fluorophore and an unlabeled primer in the reverse direction [89]. PCR amplification is performed in the presence of diGTP (isoG) labeled with a dabcyl quencher. Incorporation of the quencher through isoc–isoG pairing leads to quenching of the fluorescence. Fluorescence signal decreases along the course of reaction, which is similar to a typical real-time PCR method, albeit with signal change in the opposite direction.

Energy Transfer Between Primers or Between Primer and Probe Several methods have been described that involve energy transfer between primer and probe each labeled with a fluorescence donor or acceptor [90, 91]. The primer is labeled either at the 5′ end or at an internal position, while the probe is labeled at the 3′ end. After primer extension generates the probe-binding sequence, the probe will hybridize to the extended region bringing two labels within a short distance. As a result, energy transfer can occur and fluorescence signal change can be measured. One publication also described a method involving energy transfer between forward and reverse primers each labeled with donor and acceptor/quencher [92]. The labels are located at such positions in the primers that once the double-stranded amplicon is formed as the result of primer extension, energy can be efficiently transferred from the donor to acceptor or quencher.

RET Mediated Through Acceptor-Labeled Nucleobases F Incorporated in the Amplified Product Several methods are designed based on acceptor fluorophore-labeled nucleobases. These nucleobases are incorporated in the amplified products via primer extension. Upon target binding of certain nucleic acid dyes or cationic conjugated polymers that can act as fluorescence donors, acceptor fluorescence will increase due to FRET [93, 94]. In an alternative design called template-directed dye-terminator incorporation, the primer is labeled with fluorescence donor [95]. As a result, incorporation of the next acceptor-labeled nucleobase leads to change in fluorescence signal.


471

Instruments for Real-Time Amplification and Detection Instrument systems supporting automated real-time amplification and detection are becoming increasingly available in both research and clinical diagnostic spaces. Real-time PCR is the main target amplification and detection technology that underlies the core capability of the majority of systems, while real-time isothermal amplification and detection are also represented in a small number of systems. This discussion will not include instrumentation required for sample management or nucleic acid extraction/purification. Each instrument supporting real-time assay and fluorescence detection has three basic technical functionalities: (1) supply excitation energy, (2) detect emission energy, and (3) control and/or cycle temperature. There are currently three main types of excitation energy supply, namely, lamp, light-emitting diode (LED), and laser. Lamp provides relatively broad spectrum of lights, while LED and laser provide narrow wavelengths. Detection of the emission light is achieved with three main types of devices, namely, charge-coupled device (CCD) camera, photomultiplier tube, and other types of photodetectors. It is to note that multiplex detection is achieved by using multiple discrete filters or channels at the emission detection as well as sometimes the excitation end of the instrument. It is more difficult to achieve multiplex performance when there is only one excitation spectrum with a narrow wavelength range. For multiplex detection, filter- or channel-specific signals often have to be determined through mathematical decomposition to determine the analyte-specific signals due to the often-overlapping spectrums of fluorescence dyes. Requirements of thermal control are different for real-time PCR and isothermal amplification. While any thermal device capable of maintaining specific temperatures suffices for the isothermal amplification, real-time PCR requires a device that can cycle temperature with speed, accuracy, and uniformity. Common types of thermal cycling devices include heating block (e.g., Peltier based) and heated air (for either tube or capillary-based reaction vessels). It is obvious that the performance of real-time assays depends on factors in assay chemistry (amplification and signal generation), instrument performance (optical and thermal), and the intricate interactions between assay and instrument factors. Critical assay performance characteristics include detection sensitivity, specificity, and reproducibility (e.g., sample-to-sample, run-to-run, reagent lot-to-lot, and instrument-to-instrument). For example, the sample-to-sample reproducibility is highly dependent upon the instrument’s robustness against variables including sample positions and plate setup, which would be determined by the optical and thermal uniformity. Design features, functionalities, and technical capabilities of an instrument system are complex and multifaceted. Institutional adoption of a system typically involves comparison and consideration in various areas including instrument cost, physical dimension and other facility requirements, test menu/performance and regulatory status, open application capability, assay throughput and labor burden, service and support capability of the manufacturer, user interface, and laboratory information systems. The central consideration among all these factors has to be

472


performance of the test and instrument. Interested readers are encouraged to review a comprehensive review by Espy et al. [21] and a recent summary of automated molecular platforms for additional information [96]. The assessment of instrument systems for the purpose of assay or instrument development is typically focused on the technical design features and capabilities, such as excitation and detection systems, multiplex capability, reaction volume, throughput, thermal control, etc. It is understood that when designing an instrument, logistical, infrastructural, and other nontechnical aspects of the total system offering as stated previously have to be comprehended and optimized for the end-use laboratories and personnel. The technical features of the main instrument systems currently available in the commercial clinical diagnostic field are summarized in Table 4. The future in molecular diagnostic instrumentation lies in areas such as higher degree of process integration, menu access flexibility (random access), sample processing flexibility (continuous access and STAT testing), testing throughput, quicker turnaround time, and less hands-on time. In addition, an area of increasing importance and growing needs is robust point-of-care (POC) systems, especially for the resource-limited areas. Such POC systems need to possess these following capabilities: full integration of assay process (sample in and result out), error-proof operation, environmental robustness, and minimal logistic requirements (e.g., power, water, reagent ambient storage, and waste containment).

Data Analysis and Result Reporting Data analysis and result reporting are a critical and final process of the diagnostic testing. This process typically involves the following steps: (1) recording of optical measurements, (2) determining dye responses (analyte signals), (3) data normalization and amplification curve analysis, (4) baseline setting, (5) determining cycle number (for PCR) or time to positivity (for isothermal amplification), (6) determining qualitative and/or quantitative results based on outputs from step 5 as well as signal intensity, and (7) reporting results. The automation in data analysis and the ease of use for reported results (i.e., objectivity without human intervention) are among the hallmarks of modern commercial clinical diagnostic systems. The objectivity of reported results is the outcome of both robust data analysis algorithms and sophisticated data validity criteria. Validity criteria include checks on multiple aspects of the assay data, including amplification curve, dye intensity, signal/noise abnormality, cycle number (or time to positivity) abnormality, as well as, depending on control and calibration strategies, various performance characteristics of controls and calibrators. The output of data analysis is the generation of some sorts of actionable values from which to determine assay results. These actionable values may include cycle number (or time to positivity) or signal intensity. The algorithms in formulating assay results from these actionable values vary depending upon the diagnostic utilities that assay results are expected to fulfill. For a quantitative or semiquantitative

Model M2000rt

7500 Fast Dx

QuantStudio Dx

Max

iQ5

EasyQ

SmartCycler

3 M Integrated Cycler

Aries

Company Abbott

Applied Biosystems Applied Biosystems

BD

Bio-Rad

BioMérieux

Cepheid

Focus

Luminex

Real-time PCR

Real-time PCR

Real-time PCR Real-time PCR Real-time NASBA Real-time PCR

Real-time PCR Real-time PCR

Technology Real-time PCR

Read system excitation Halogen lamp; optical filters; full-plate illumination Halogen lamp

Individual custom cartridge

Disc

Individual custom chamber

Optical filters, CCD camera Optical filters, CCD camera

Read system detection Optical filters; CCD camera

–

LED; optical filters

4 colors

8 colors

5 colors

2 colors

2–6 colors

5 colors

Multiplex capability 5 colors

Optical filters; PMT; 4 colors individual well detection (one reader) Photodiode 6 channels

Discrete detector for each well Optical filters, CCD camera Optical filter; PMT; individual tube LED; individual well Optical filters; illumination silicon detector for each reaction

Halogen lamp 96-well plate, 384-well, and TaqMan Array Card Microfluidic card Individual well illumination 96-well plate Tungsten–halogen lamp Reaction tube Filters

96-well plate

Reaction format 96-well plate

Table 4 Real-time instruments used in clinical microbiology

(continued)

Peltier device

Individual reaction heating circuit; solid-state heater and forced-air cooling Infrared heating; air cooling

Individual reaction heating circuit 96-well block Peltier device Isothermal device

96-well block Peltier device 96-well block Peltier device

Thermal control 96-well block Peltier device

Real-Time Detection of Amplification Products Through Fluorescence Quenching… 473

Cobas TaqMan

LightCycler 4800

Versant kPCR Real-time PCR

Roche

Roche

Siemens

Real-time PCR

Real-time PCR

Model Technology Rotor-Gene Q Real-time PCR

Company Qiagen

Table 4 (continued)

96-well plate

96-well plate

Micro tube

Reaction format Reaction tube

Halogen lamp; optical filters; individual well illumination Xenon lamp; optical filters; full-plate illumination Halogen lamp; optical filters; individual well illumination

Read system excitation 6 LED; individual well illumination

4 colors

Multiplex capability 6 colors

5 color excitation; 6 color emission Optical filters; PMT; 5 colors individual well detection

Optical filters; CCD camera

Read system detection Optical filters; PMT; individual well detection (one reader) Fiber optic channels; optical filters; photo ASIC

96-well block Peltier device



Thermal control 36, 72, 100 tube rotor with air heating/cooling

474 C. Otto and S. Huang


475

assay, analyte quantities may be calculated by comparing the cycle number (or time to positivity) of the analyte against external calibrator(s) or internal quantitative standard(s). Similarly, relative quantification may be calculated by comparing the cycle time (or ∆ time to positivity) between the analyte and the endogenous control against calibrators/standards. For a qualitative assay, the positive or negative assay result may be determined by comparing the cycle number (or time to positivity) and/or signal intensity with respective cutoff values. For a genotyping assay, qualitative results from one or multiple analytes may be combined to determine the genotype profile of the sample.

Conclusions Several main topics regarding fluorescence-based detection of real-time amplification and detection assays have been discussed in this chapter, including fluorescence principles, target detection/signal generation technologies, real-time instrument systems, and data analysis and result reporting. While existing real-time assays are playing an important role in clinical microbiology, new technology platforms and instrument systems for amplification and detection as well as sample management and preparation will continue to emerge. These new technologies hold great promises in further improving established clinical utilities as well as addressing emerging clinical needs or new microbiological agents. It is therefore imperative for the diagnostic community, including researchers, laboratories, clinicians, and device manufacturers, to make concerted and continued effort to develop, commercialize, and utilize more sophisticated and accurate diagnostic tools to fight against increasing burden of diseases and infections.

References 1. Ginocchio CC. Life beyond PCR: alternative target amplification technologies for the diagnosis of infectious diseases, part I. Clin Microbiol Newsl. 2004;26:121–8. 2. Ginocchio CC. Life beyond PCR: alternative target amplification technologies for the diagnosis of infectious diseases, part II. Clin Microbiol Newsl. 2004;26:129–36. 3. Deiman B, van Aarle P, Sillekens P. Characteristics and applications of nucleic acid sequence- based amplification (NASBA). Mol Biotechnol. 2002;20:163–79. 4. Romano JW, van Gemen B, Kievits T. NASBA: a novel, isothermal detection technology for qualitative and quantitative HIV-1 RNA measurements. Clin Lab Med. 1996;16:89–103. 5. Fahy E, Kwoh DY, Gingeras TR. Self-sustained sequence replication (3SR): an isothermal transcription- based amplification system alternative to PCR. PCR Methods Appl. 1991;1:25–33. 6. Mueller JD, Putz B, Hofler H. Self-sustained sequence replication (3SR): an alternative to PCR. Histochem Cell Biol. 1997;108:431–7. 7. Hill CS. Molecular diagnostic testing for infectious diseases using TMA technology. Expert Rev Mol Diagn. 2001;1:445–55.

476


8. Walker GT. Empirical aspects of strand displacement amplification. PCR Methods Appl. 1993;3:1–6. 9. Nycz CM, Dean CH, Haaland PD, Spargo CA, Walker GT. Quantitative reverse transcription strand displacement amplification: quantitation of nucleic acids using an isothermal amplification technique. Anal Biochem. 1998;259:226–34. 10. Notomi T, Okayama H, Masubuchi H, Yonekawa T, Watanabe K, Amino N, Hase T. Loop- mediated isothermal amplification of DNA. Nucleic Acids Res. 2000;28:E63. 11. Parida M, Sannarangaiah S, Dash PK, Rao PV, Morita K. Loop mediated isothermal amplification (LAMP): a new generation of innovative gene amplification technique; perspectives in clinical diagnosis of infectious diseases. Rev Med Virol. 2008;18:407–21. 12. Piepenburg O, Williams CH, Stemple DL, Armes NA. DNA detection using recombination proteins. PLoS Biol. 2006;4:e204. 13. Jeong YJ, Park K, Kim DE. Isothermal DNA amplification in vitro: the helicase-dependent amplification system. Cell Mol Life Sci. 2009;66:3325–36. 14. Vincent M, Xu Y, Kong H. Helicase-dependent isothermal DNA amplification. EMBO Rep. 2004;5:795–800. 15. Kurn N, Chen P, Heath JD, Kopf-Sill A, Stephens KM, Wang S. Novel isothermal, linear nucleic acid amplification systems for highly multiplexed applications. Clin Chem. 2005;51:1973–81. 16. Mukai H, Uemori T, Takeda O, Kobayashi E, Yamamoto J, Nishiwaki K, Enoki T, Sagawa H, Asada K, Kato I. Highly efficient isothermal DNA amplification system using three elements of 5'-DNA-RNA-3' chimeric primers, RNaseH and strand-displacing DNA polymerase. J Biochem. 2007;142:273–81. 17. Jung C, Chung JW, Kim UO, Kim MH, Park HG. Isothermal target and signaling probe amplification method, based on a combination of an isothermal chain amplification technique and a fluorescence resonance energy transfer cycling probe technology. Anal Chem. 2010;82:5937–43. 18. Van Ness J, Van Ness LK, Galas DJ. Isothermal reactions for the amplification of oligonucleotides. Proc Natl Acad Sci. 2003;100:4504–9. 19. Morrison L. Fluorescence in nucleic acid hybridization assays. In: Lakowicz J, editor. Topics in fluorescence spectroscopy: DNA technology, vol. 7. New York: Springer; 2003. p. 69–103. 20. Marras SAE, Kramer FR, Tyagi S. Efficiencies of fluorescence resonance energy transfer and contact-mediated quenching in oligonucleotide probes. Nucleic Acids Res. 2002;30:e122. 21. Espy MJ, Uhl JR, Sloan LM, Buckwalter SP, Jones MF, Vetter EA, Yao JDC, Wengenack NL, Rosenblatt JE, Cockerill FR, et al. Real-time PCR in clinical microbiology: applications for routine laboratory testing. Clin Microbiol Rev. 2006;19:165–256. 22. Tyagi S, Landegren U, Tazi M, Lizardi PM, Kramer FR. Extremely sensitive, background- free gene detection using binary probes and beta replicase. Proc Natl Acad Sci U S A. 1996;93:5395–400. 23. Banér J, Nilsson M, Mendel-Hartvig M, Landegren U. Signal amplification of padlock probes by rolling circle replication. Nucleic Acids Res. 1998;26:5073–8. 24. Demidov VV. Rolling-circle amplification in DNA diagnostics: the power of simplicity. Expert Rev Mol Diagn. 2002;2:542–8. 25. Zhang DY, Brandwein M, Hsuih T, Li HB. Ramification amplification: a novel isothermal DNA amplification method. Mol Diagn a journal devoted to the understanding of human disease through the clinical application of molecular biology. 2001;6:141–50. 26. Zhang DY, Brandwein M, Hsuih TC, Li H. Amplification of target-specific, ligation-dependent circular probe. Gene. 1998;211:277–85. 27. de Arruda M, Lyamichev VI, Eis PS, Iszczyszyn W, Kwiatkowski RW, Law SM, Olson MC, Rasmussen EB. Invader technology for DNA and RNA analysis: principles and applications. Expert Rev Mol Diagn. 2002;2:487–96. 28. Clegg RM. Fluorescence resonance energy transfer and nucleic acids. Methods Enzymol. 1992;211:353–88. 29. Clegg RM. Fluorescence resonance energy transfer. Curr Opin Biotechnol. 1995;6:103–10.


477

30. Stryer L. Fluorescence energy transfer as a spectroscopic ruler. Annu Rev Biochem. 1978;47:819–46. 31. Nazarenko I, Pires R, Lowe B, Obaidy M, Rashtchian A. Effect of primary and secondary structure of oligodeoxyribonucleotides on the fluorescent properties of conjugated dyes. Nucleic Acids Res. 2002;30:2089–195. 32. Dubertret B, Calame M, Libchaber AJ. Single-mismatch detection using gold-quenched fluorescent oligonucleotides. Nat Biotechnol. 2001;19:365–70. 33. Dulkeith E, Morteani AC, Niedereichholz T, Klar TA, Feldmann J, Levi SA, van Veggel FC, Reinhoudt DN, Moller M, Gittins DI. Fluorescence quenching of dye molecules near gold nanoparticles: radiative and nonradiative effects. Phys Rev Lett. 2002;89:203002. 34. Mao F, Leung WY, Xin X. Characterization of EvaGreen and the implication of its physicochemical properties for qPCR applications. BMC Biotechnol. 2007;7:76. 35. Sutherland JWH, Patterson DR, inventors; Johnson & Johnson Clinical Diagnostics, Inc, assignee. Homogeneous method for assay of double-stranded nucleic acids using fluorescent dyes and kit useful therein. EP 0 684 316 B1; 1996. 36. Gudnason H, Dufva M, Bang DD, Wolff A. Comparison of multiple DNA dyes for real-time PCR: effects of dye concentration and sequence composition on DNA amplification and melting temperature. Nucleic Acids Res. 2007;35:e127. 37. Horakova H, Polakovicova I, Shaik GM, Eitler J, Bugajev V, Draberova L, Draber P. 1,2-propanediol-trehalose mixture as a potent quantitative real-time PCR enhancer. BMC Biotechnol. 2011;11:41. 38. Wittwer CT, Reed GH, Gundry CN, Vandersteen JG, Pryor RJ. High-resolution genotyping by amplicon melting analysis using LCGreen. Clin Chem. 2003;49:853–60. 39. Bengtsson M, Karlsson HJ, Westman G, Kubista M. A new minor groove binding asymmetric cyanine reporter dye for real-time PCR. Nucleic Acids Res. 2003;31:e45. 40. Lind K, Stahlberg A, Zoric N, Kubista M. Combining sequence-specific probes and DNA binding dyes in real-time PCR for specific nucleic acid quantification and melting curve analysis. Biotechniques. 2006;40:315–9. 41. Takano EA, Mitchell G, Fox SB, Dobrovic A. Rapid detection of carriers with BRCA1 and BRCA2 mutations using high resolution melting analysis. BMC Cancer. 2008;8:59. 42. Monis PT, Giglio S, Saint CP. Comparison of SYTO9 and SYBR Green I for real-time polymerase chain reaction and investigation of the effect of dye concentration on amplification and DNA melting curve analysis. Anal Biochem. 2005;340:24–34. 43. Ishiguro T, Saitoh J, Yawata H, Yamagishi H, Iwasaki S, Mitoma Y. Homogeneous quantitative assay of hepatitis C virus RNA by polymerase chain reaction in the presence of a fluorescent intercalater. Anal Biochem. 1995;229:207–13. 44. Kutyavin IV. New approach to real-time nucleic acids detection: folding polymerase chain reaction amplicons into a secondary structure to improve cleavage of Forster resonance energy transfer probes in 5'-nuclease assays. Nucleic Acids Res. 2010;38:e29. 45. Kutyavin IV. Use of base modifications in primers and amplicons to improve nucleic acids detection in the real-time snake polymerase chain reaction. Assay Drug Dev Technol. 2011;9:58–68. 46. Espy MJ, Uhl JR, Mitchell PS, Thorvilson JN, Svien KA, Wold AD, Smith TF. Diagnosis of herpes simplex virus infections in the clinical laboratory by LightCycler PCR. J Clin Microbiol. 2000;38:795–9. 47. Wittwer CT, Rasmussen RP, Lay M, inventors; assignee. Monitoring amplification with fret probes. US 7,160,998 B2; 2007. 48. Mergny JL, Garestier T, Rougee M, Lebedev AV, Chassignol M, Thuong NT, Helene C. Fluorescence energy transfer between two triple helix-forming oligonucleotides bound to duplex DNA. Biochemistry. 1994;33:15321–8. 49. Rosmarin D, Pei Z, Blaser M, Tyagi S, inventors; assignee. C-shaped probe. US 2006/0040275 A1; 2006.

478


50. Tyagi S, Kramer FR. Molecular beacons: probes that fluoresce upon hybridization. Nat Biotechnol. 1996;14:303–8. 51. Bonnet G, Tyagi S, Libchaber A, Kramer FR. Thermodynamic basis of the enhanced specificity of structured DNA probes. Proc Natl Acad Sci U S A. 1999;96:6171–6. 52. Satterfield BC, Kulesh DA, Norwood DA, Wasieloski LP Jr, Caplan MR, West JA. Tentacle Probes: differentiation of difficult single-nucleotide polymorphisms and deletions by presence or absence of a signal in real-time PCR. Clin Chem. 2007;53:2042–50. 53. Tyagi S, Marras SA, Kramer FR. Wavelength-shifting molecular beacons. Nat Biotechnol. 2000;18:1191–6. 54. Chen AK, Davydenko O, Behlke MA, Tsourkas A. Ratiometric bimolecular beacons for the sensitive detection of RNA in single living cells. Nucleic Acids Res. 2010;38:e148. 55. Nutiu R, Li Y. Tripartite molecular beacons. Nucleic Acids Res. 2002;30:e94. 56. Becker MM, Schroth GP, inventors; assignee. Molecular torches. US 6,534,274 B2; 2003. 57. Lv C, Yu L, Wang J, Tang X. A dumbbell molecular beacon for the specific recognition of nucleic acids. Bioorg Med Chem Lett. 2010;20:6547–50. 58. Kandimalla ER, Agrawal S. 'Cyclicons' as hybridization-based fluorescent primer-probes: synthesis, properties and application in real-time PCR. Bioorg Med Chem. 2000;8:1911–6. 59. Huang S, Salituro J, Tang N, Luk KC, Hackett J Jr, Swanson P, Cloherty G, Mak WB, Robinson J, Abravaya K. Thermodynamically modulated partially double-stranded linear DNA probe design for homogeneous real-time PCR. Nucleic Acids Res. 2007;35:e101. 60. Li Q, Luan G, Guo Q, Liang J. A new class of homogeneous nucleic acid probes based on specific displacement hybridization. Nucleic Acids Res. 2002;30:E5. 61. Yu G, Niu J, Shen M, Shao H, Chen L. Detection of Escherichia coli O157 using equal-length double-stranded fluorescence probe in a real-time polymerase chain reaction assay. Clinica Chim Acta Int J Clin Chem. 2006;366:281–6. 62. Keys DA, Farooqui F, Reddy MP, inventors; Beckman Coulter, Inc., assignee. Oligonucleotide probes for detecting nucleic acids through changes in flourescence resonance energy transfer. US 6,593,091 B2; 2003. 63. Afonina IA, Reed MW, Lusby E, Shishkina IG, Belousov YS. Minor groove binder-conjugated DNA probes for quantitative DNA detection by hybridization-triggered fluorescence. Biotechniques. 2002;32:940–4. 946-949 64. Lukhtanov EA, Lokhov SG, Gorn VV, Podyminogin MA, Mahoney W. Novel DNA probes with low background and high hybridization-triggered fluorescence. Nucleic Acids Res. 2007;35:e30. 65. Svanvik N, Westman G, Wang D, Kubista M. Light-up probes: thiazole orange-conjugated peptide nucleic acid for detection of target nucleic acid in homogeneous solution. Anal Biochem. 2000;281:26–35. 66. Bethge L, Jarikote DV, Seitz O. New cyanine dyes as base surrogates in PNA: forced intercalation probes (FIT-probes) for homogeneous SNP detection. Bioorg Med Chem. 2008;16:114–25. 67. Silva GL, Ediz V, Yaron D, Armitage BA. Experimental and computational investigation of unsymmetrical cyanine dyes: understanding torsionally responsive fluorogenic dyes. J Am Chem Soc. 2007;129:5710–8. 68. Rye HS, Yue S, Wemmer DE, Quesada MA, Haugland RP, Mathies RA, Glazer AN. Stable fluorescent complexes of double-stranded DNA with bis-intercalating asymmetric cyanine dyes: properties and applications. Nucleic Acids Res. 1992;20:2803–12. 69. Tonelli A, Tedeschi T, Germini A, Sforza S, Corradini R, Medici MC, Chezzi C, Marchelli R. Real time RNA transcription monitoring by Thiazole Orange (TO)-conjugated Peptide Nucleic Acid (PNA) probes: norovirus detection. Mol Biosyst. 2011;7:1684–92. 70. Wirgart BZ, Andersson P, Grillner L. Evaluation of the ReSSQ assay in relation to the COBAS AMPLICOR CMV MONITOR test and an in-house nested PCR method for detection of cytomegalovirus DNA. J Clin Microbiol. 2005;43:4057–63. 71. French DJ, Archard CL, Brown T, McDowell DG. HyBeacon probes: a new tool for DNA sequence detection and allele discrimination. Mol Cell Probes. 2001;15:363–74.


479

72. Yamane A. Smart probe: a novel fluorescence quenching-based oligonucleotide probe carrying a fluorophore and an intercalator. Nucleic acids symposium series 297–298; 2000. 73. Knemeyer JP, Marme N, Sauer M. Probes for detection of specific DNA sequences at the single-molecule level. Anal Chem. 2000;72:3717–24. 74. Crockett AO, Wittwer CT. Fluorescein-labeled oligonucleotides for real-time pcr: using the inherent quenching of deoxyguanosine nucleotides. Anal Biochem. 2001;290:89–97. 75. Kurata S, Kanagawa T, Yamada K, Torimura M, Yokomaku T, Kamagata Y, Kurane R. Fluorescent quenching-based quantitative detection of specific DNA/RNA using a BODIPY((R)) FL-labeled probe or primer. Nucleic Acids Res. 2001;29:E34. 76. Little MC, Andrews J, Moore R, Bustos S, Jones L, Embres C, Durmowicz G, Harris J, Berger D, Yanson K, et al. Strand displacement amplification and homogeneous real-time detection incorporated in a second-generation DNA probe system, BDProbeTecET. Clin Chem. 1999;45:777–84. 77. Harvey JJ, Lee SP, Chan EK, Kim JH, Hwang ES, Cha CY, Knutson JR, Han MK. Characterization and applications of CataCleave probe in real-time detection assays. Anal Biochem. 2004;333:246–55. 78. Hosono N, Kubo M, Tsuchiya Y, Sato H, Kitamoto T, Saito S, Ohnishi Y, Nakamura Y. Multiplex PCR-based real-time invader assay (mPCR-RETINA): a novel SNP-based method for detecting allelic asymmetries within copy number variation regions. Hum Mutat. 2008;29:182–9. 79. Tadokoro K, Akutsu Y, Tanaka K, Saito T, Yamaguchi T, Egashira T, Ishiwata I, Hara T. Comparative quantitative analysis of 14 types of human papillomavirus by real-time polymerase chain reaction monitoring Invader reaction (Q-Invader assay). Diagn Microbiol Infect Dis. 2010;66:58–64. 80. Tadokoro K, Yamaguchi T, Egashira T, Hara T. Quantitation of viral load by real-time PCR- monitoring Invader reaction. J Virol Methods. 2009;155:182–6. 81. Nazarenko IA, Bhatnagar SK, Hohman RJ. A closed tube format for amplification and detection of DNA based on energy transfer. Nucleic Acids Res. 1997;25:2516–21. 82. Nazarenko I, Lowe B, Darfler M, Ikonomi P, Schuster D, Rashtchian A. Multiplex quantitative PCR using self-quenched primers labeled with a single fluorophore. Nucleic Acids Res. 2002;30:e37. 83. Whitcombe D, Theaker J, Guy SP, Brown T, Little S. Detection of PCR products using self- probing amplicons and fluorescence. Nat Biotechnol. 1999;17:804–7. 84. Li J, Wang F, Mamon H, Kulke MH, Harris L, Maher E, Wang L, Makrigiorgos GM. Antiprimer quenching-based real-time PCR and its application to the analysis of clinical cancer samples. Clin Chem. 2006;52:624–33. 85. Solinas A, Brown LJ, McKeen C, Mellor JM, Nicol J, Thelwell N, Brown T. Duplex Scorpion primers in SNP analysis and FRET applications. Nucleic Acids Res. 2001;29:E96. 86. Santoro SW, Joyce GF. A general purpose RNA-cleaving DNA enzyme. Proc Natl Acad Sci U S A. 1997;94:4262–6. 87. Todd AV, Fuery CJ, Impey HL, Applegate TL, Haughton MA. DzyNA-PCR: use of DNAzymes to detect and quantify nucleic acid sequences in a real-time fluorescent format. Clin Chem. 2000;46:625–30. 88. Singh KK, Hanne A, Krupp G. Real-time monitoring of RNA and DNA reactions by fluorescence detection. Methods Mol Biol (Clifton, NJ). 2004;252:33–48. 89. Sherrill CB, Marshall DJ, Moser MJ, Larsen CA, Daude-Snow L, Jurczyk S, Shapiro G, Prudent JR. Nucleic acid analysis using an expanded genetic alphabet to quench fluorescence. J Am Chem Soc. 2004;126:4550–6. 90. Lay MJ, Wittwer CT. Real-time fluorescence genotyping of factor V Leiden during rapid-cycle PCR. Clin Chem. 1997;43:2262–7. 91. Rasmussen TB, Uttenthal A, de Stricker K, Belak S, Storgaard T. Development of a novel quantitative real-time RT-PCR assay for the simultaneous detection of all serotypes of foot- and-mouth disease virus. Arch Virol. 2003;148:2005–21.

480


92. Ahmad AI, Ghasemi JB. New FRET primers for quantitative real-time PCR. Anal Bioanal Chem. 2007;387:2737–43. 93. Takatsu K, Yokomaku T, Kurata S, Kanagawa T. A FRET-based analysis of SNPs without fluorescent probes. Nucleic Acids Res. 2004;32:e156. 94. Duan X, Yue W, Liu L, Li Z, Li Y, He F, Zhu D, Zhou G, Wang S. Single-nucleotide polymorphism (SNP) genotyping using cationic conjugated polymers in homogeneous solution. Nat Protoc. 2009;4:984–91. 95. Chen X, Zehnbauer B, Gnirke A, Kwok PY. Fluorescence energy transfer detection as a homogeneous DNA diagnostic method. Proc Natl Acad Sci U S A. 1997;94:10756–61. 96. Automated molecular platforms, October 2016 [http://www.captodayonline.com/product guides/instruments/automated-molecular-platforms-october-2016.html].

PCR/Electrospray Ionization-Mass Spectrometry as an Infectious Disease Diagnostic Tool Volkan Özenci and Kristoffer Strålin

Microbial Detection and Identification Severe infectious diseases are an increasing concern for global healthcare and have a major impact on the morbidity and mortality of the general population. Even in today’s modern healthcare environments, the mortality rates due to these infections are unacceptably high and may reach as high as 50% in some populations. Studies have repeatedly demonstrated that the rapid administration of correct antimicrobial treatment is crucial for patient survival [1, 2]. Thus, receiving timely and accurate information as to the nature of causative agent and its antimicrobial susceptibilities is of outmost importance for proper clinical management as well as enabling optimal targeted antimicrobial therapy. Outcome-based studies assessing the effect of rapid reporting of identification and antimicrobial susceptibility results have shown a faster modification of antimicrobial therapy with associated decrease in antimicrobial therapy duration, mortality, hospital stay, and healthcare costs [3, 4]. The microbiological diagnosis of severe infections is challenging and changing. Since the nineteenth century, cultures have been the method of choice for detecting microbial pathogens by the clinical microbiology laboratory. Culturebased methods continue to provide cost-effective detection, isolation, and susceptibility testing of viable microorganisms. The major advantages of isolation of V. Özenci (*) Department of Clinical Microbiology, Karolinska University Hospital, Stockholm, Sweden Division of Clinical Microbiology, Department of Laboratory Medicine, Karolinska Institute, Stockholm, Sweden e-mail: [email protected] K. Strålin Department of Infectious Diseases, Karolinska University Hospital, Stockholm, Sweden Unit of Infectious Diseases, Department of Medicine Huddinge, Karolinska Institutet, Stockholm, Sweden © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_22

481

482

V. Özenci and K. Strålin

viable microorganisms by culture-based methods include high specificity and enabling further microbiologic analysis such as epidemiological typing and antibiotic susceptibility testing. However, these methods are limited to detection of viable microorganisms that can grow on defined culture media within a given period of time. Clinical samples that include minimal amounts of viable microorganisms, fastidious or slow-growing microorganisms, and/or nonviable microorganisms because of preemptive antibiotic treatment and/other processes may not be detected by culture and will thus provide a false-negative result. Another major problem with conventional culture-based methods is that they often take too long time to efficiently and realistically impact clinical treatment decisions for infected patients. Altogether, these shortcomings with culture-based methods have stimulated clinical microbiology laboratories to provide alternative methods that are reliable and rapid for the laboratory diagnosis of severe infections. The alternative methods that best accomplish these goals are molecular methods. The development of such molecular diagnostic methods over the past two decades has played an increasingly important role in clinical microbiology and has made it possible to implement many of these rapid methods in routine diagnostics [5]. Molecular methods targeting a single pathogen, such as MRSA, were the first tests to be successfully implemented in routine clinical practice but had the obvious limitation of only focusing on one microorganism [6]. The next step was to develop disease-specific panel-based molecular methods with a coverage of a limited number of the most common pathogens causing particular infections [7]. This syndromic approach has been implemented using multiplex PCR methods. However, as in the case of pathogen-specific PCR methods, the panel-based molecular tests cover, in the best-case scenario, only the most common pathogens. Furthermore, the dramatic improvements in medical care such as transplantation in combination with globalization have resulted in increased rates of invasive infections with esoteric microorganisms [8]. As a consequence, there has been a desperate need for rapid diagnostic methods that provide an extensive broad range for the identification of microbial pathogens for use in modern clinical microbiology laboratories.

CR/Electrospray Ionization-Mass Spectrometry P (PCR/ESI-MS) The PCR/ESI-MS system is built on the principle of universal detection and specific identification. The technology detects a wide variety of organisms using broad-spectrum PCR primers designed to amplify regions that vary in sequence (and therefore in base pairs) composition among the targeted organisms. Mass measurements achieved by electrospray ionization-mass spectrometry (ESI-MS) analysis of the resulting amplicons provide data that is computationally analyzed to generate base pair composition signatures [9]. Identification is achieved by automated digital matching of species- (or strain-) specific base composition signatures to signatures contained in an integrated database (Fig. 1). These comparison signatures are derived from either existing sequence databases or from previous

PCR/Electrospray Ionization-Mass Spectrometry as an Infectious Disease Diagnostic Tool 483

Fig. 1 Identification of microorganisms by the PCR/electrospray ionization-mass spectrometry

empirical testing of well-characterized type strains using PCR/ESI-MS [9]. Each specific PCR/ESI-MS assay is designed to detect and identify all organisms within a particular interrelated phylogenetic group. Multiple primer pairs are used to amplify multiple genetic loci from most targets, increasing the resolution of identifications to the desired level [10] and precluding loss of sensitivity due to sequence variations in the primer target sequences [11]. Any group of related organisms that share conserved regions of sequence with sufficient homology to support broadspectrum priming can be targeted by a universal PCR/ESI-MS detection and identification assay (Fig. 1). This includes groups as large as the entire kingdom of bacteria, for which 16 primer pairs can provide enough organism-specific data to provide genus-specific, and often species-specific, identification [12]. This is also the case for certain families of viruses such as influenza, for which nine primer pairs are able to provide type, subtype, and lineage-specific identification of all recognized strains [13].

Brief History of PCR/ESI-MS The PCR/ESI-MS technology was initially designed for biodefense and surveillance by the Defense Advanced Research Projects Agency (DARPA). The first instrument was called TIGER (triangulation identification for the genetic evaluation of risk). Shortly after, Ibis Biosciences in the second half of the 2000s commercialized this technology as the Ibis T5000 system [10]. Ibis Biosciences was then incorporated

484


into the Abbott group, and an upgraded system, the PLEX-ID, was commercialized. The PLEX-ID system was subsequently redesigned into the IRIDICA system [14]. The IRIDICA system was CE-marked and became commercially available in Europe in December 2014. After that, the manufacturer of the system, Abbott, performed a multi-site clinical trial of the IRIDICA BAC assay on whole blood, with 17 patient specimen collection sites and 4 clinical testing sites. This study, with 1501 included patients enrolled between December 2014 and March 2016, became the basis for an application to FDA in 2016 for FDA approval. However, in April 2017, Abbott withdrew their application to FDA and ceased producing IRIDICA instruments and all IRIDICA assays, essentially finalizing the clinical use of this technology [15].

The IRIDICA Platform The IRIDICA platform consisted of separate instruments for processing and analysis. The pre-amplification instruments were the IRIDICA bead beater and IRIDICA sample prep, and the post-amplification instruments include the IRIDICA thermal cycler, the IRIDICA desalter, the IRIDICA mass spectrometer, and the IRIDICA analysis computer (AC). The total space requirement for the platform was ∼30 m2. Two separate rooms were required, one for sample preparation and DNA extraction and the other for PCR, desalting, and mass spectrometry [15].

The Clinical PCR/ESI-MS Assay The IRIDICA instrument included five different diagnostic assays for detection and identification of different microorganisms and available for diverse sample types. Table 1 describes the five IRIDICA assays, their coverage, and the sample types that are required. The assay was user friendly and included relatively few manual steps. For all assays, the samples were prepared to yield a sample volume imput of 5 mL homogTable 1 Sample types, sample volumes, and coverage of the IRIDICA PCR/electrospray ionization-mass spectrometry Assay BAC BSI BAC SFT BAC LRT Fungal Viral

Sample type EDTA whole blood Sterile fluids/ tissue BAL or ETA BAL Plasma

Pre-process volume Coverage 5 mL 780 bacteria, Candida spp., and 4 resistance markers: mecA, vanA, vanB, and BLAKPC 500 uL/25 mg 780 bacteria, Candida spp., and 4 resistance markers: mecA, vanA, vanB, and BLAKPC 100 uL 780 bacteria, Candida spp., and 4 resistance markers: mecA, vanA, vanB, and BLAKPC 5 mL >200 fungi >130 viruses in 13 reporting groups


Fig. 2 The IRIDICA BAC bacteria panel including more than 750 spp. (Republished with permission from [16])

enized sample. Tissue samples were prepared with initial use of a bead beater prior to dilusion. The assay turn-around time was approximately 8 h, and it was possible to obtain the results during the same working day as when the sample was run. The capacity was 5 patient samples at a time, permitting a maximum of 15 samples per 24 h for an IRIDICA instrument. The major advantage with the IRIDICA PCR/ESI-MS assay was the extensive broad coverage. The BAC assay could detect and identify over 780 bacterial/Candida species (Fig. 2). The assay could also detect resistance markers (mecA, vanA, vanB, and BLAKPC). The IRIDICA PCR/ESI-MS assay was also capable of identifying a broad range of other medically important fungi, including clinically relevant yeasts and molds (Fig. 3).

The Clinical Performance of the PCR/ESI-MS There are several clinical studies analyzing the performance of the IRIDICA PCR/ ESI-MS. In a prospective multicenter observational study, the BAC BSI assay has been evaluated in nine European intensive care units (ICUs) [18]. In total, 616 blood

486


Fig. 3 The IRIDICA fungal panel including more than 200 spp. Colored boxes indicate the coverage attained by each primer pair whose name is reported within square brackets. (Republished with permission from [17])

samples from 529 patients with suspected or proven infection were analyzed. The number of blood samples that were positive with PCR/ESI-MS was 228/616 (37%) as compared to blood cultures, where 68/616 (11%) were positive. The authors analyzed the same material in a separate study in order to describe the possible clinical relevance of the obtained PCR/ESI-MS results. Interestingly, the 28-day mortality was higher in patients with a positive than a negative PCR/ESI-MS result (42% vs. 26%, p = 0.001) [19]. In contrast, the mortality rates were similar between patients with positive and negative blood cultures. The performance of the PCR/ESI-MS BAC assay was also analyzed in 285 patients with suspected BSI from the emergency department. The PCR/ESI-MS BAC assay could detect microorganisms in 85/285 (30%), whereas blood cultures were positive in 45/285 (16%) samples. The detailed analysis of the data showed that 34 potential pathogenic organisms were detected only by PCR/ESI-MS but not by blood cultures [20]. In a similar study, the performance of IRIDICA BAC BSI assay was analyzed in 410 patients admitted to the emergency room (ER) and intensive care unit (ICU) with clinical suspicion of sepsis. The overall positive and negative agreement of IRIDICA BAC BSI assay with blood culture in the analysis by specimen was 74.8% and 78.6%, respectively. When compared with the clinical infection criterion, the positive and negative agreement was increased to 76.9% and


87.2%, respectively. Interestingly, IRIDICA detected 41 clinically significant microorganisms missed by culture, most of them from patients under antimicrobial treatment. When ICU patients were analyzed separately in comparison with the clinical infection criterion, the sensitivity, specificity, and positive and negative predictive values compared with blood culture were 90.5%, 87.2%, 64.4%, and 97.3%, respectively [21]. Detection and identification of microorganisms from sterile fluid and tissues by IRIDICA BAC SFT assay have also been analyzed. The multicenter European ICU study [18] included 110 sterile fluid and tissue samples. IRIDICA BAC SFT assay could detect microorganisms in 78/110 (71%) samples, whereas regular culture was positive in 53/110 (48%) samples. Our group performed a small comparative study using 45 such clinical samples and showed that microorganisms were detected by IRIDICA BAC SFT in 28/45 (62%) cases and by 16S rRNA gene PCR in 25/45 (56%) cases (manuscript in preparation). In a prospective study, the performance of IRIDICA assay was studied for detection of microorganisms from sterile surgical specimens obtained from 128 patients receiving antibiotic treatment [22]. PCR/ESI-MS detected bacterial pathogens in 89/128 (69.5%) patients, but cultures were positive in only 41/128 (32%) [22]. The clinical performance of IRIDICA BAC LRT assay applied to lower respiratory tract secretions has been analyzed in several studies. A multicenter European ICU study included 185 lower respiratory tract samples from patients with suspected pneumonia. PCR/ESI-MS LRT assay could detect microorganisms in 117/185 (63%) samples where conventional culture was positive in 81/185 (44%) samples [18]. Our group has reported that PCR/ESI-MS detected more pathogens than did culture in BAL fluid from 51 mechanically ventilated patients with suspected pneumonia [23]. In 8/10 pneumonia patients with a positive PCR/ESI-MS result and negative cultures, the detected pathogens had been detected by other microbiological methods. We also showed that, as expected, PCR/ESI-MS detected several members of the normal flora alongside relevant pathogens, in 121 consecutively received BAL samples [24]. Overall, several published studies have shown that the PCR/ESI-MS had higher detection rates than the conventional culture-based methods. Analysis of the PCR/ ESI-MS-positive, but culture-negative, findings showed in general that the positive PCR/ESI-MS data were supported by clinical findings [25] or other microbiological findings [26].

The Use of PCR/ESI-MS in Clinical Practice Between October 2015 and April 2017, IRIDICA was a routine diagnostic method at the Karolinska University Laboratory, for detection of bacterial and fungal pathogens in normally sterile fluids/tissues and lower respiratory secretions, as well as in blood samples in selected cases, at the discretion of the treating clinician. The laboratory receives routine microbiological samples from three

488


tertiary care hospitals (Karolinska University Hospital Huddinge, South General Hospital in Stockholm, and Södertälje Hospital in Södertälje), which together have >1500 hospital beds. During this time of routine analyses, the IRIDICABAC system was run on >1500 routine patient samples, including 452 whole blood samples, 839 sterile fluid/tissue samples, and 165 BAL/endotracheal aspirate samples, with positivity rates of 25, 50, and 84%, respectively (overall 46%), and with invalid results in only 3% of the cases overall. The summary of the positive results of these routine runs has been published [15]. In addition, the IRIDICA fungal assay was run on 103 BAL samples with positive results for fungal species in 76 cases (74%). Detected fungal species included Aspergillus species, Mucorales, and Pneumocystis jirovecii (data not shown). In our experience, IRIDICA has been a useful clinical tool, yielding information beyond that provided by standard microbiological methods and guiding clinical decisions regarding antimicrobial therapy and surgical interventions. In many cases, positive results had enabled targeted antimicrobial therapy, and negative results had supported discontinuation of antibiotics.

Conclusion PCR/electrospray ionization-mass spectrometry (PCR/ESI-MS) is a technology that combines nucleic acid amplification using multiple broad-range and specific PCR assays with electrospray ionization-mass spectrometric detection and precise mass measurement of amplified DNA for each of the PCR reactions. The exact mass measured enables determination of the base composition of amplified DNA and deconvolution of that information into potential sequences of the amplified DNA based on comparison against well-constructed broad databases. Ultimately, once the information from the multiple PCR assays is considered together, a final result, for example, detection of one or more microorganisms and also select resistance genes, is ascertained. The performance studies and our experience of PCR/ESI-MS in clinical practice indicate that this method provided useful clinical information, and thus, modern molecular methods deserve further development for broad implementation into clinical practice for improved care of severe infections [27]. The detection of extensive broad spectrum of microorganisms directly from clinical samples with a series of tests that are run automatically with a turn-around time of 8 h would be a much desirable diagnostic tool for the clinical microbiology laboratories. We believe that the demise of IRIDICA PCR/ESI-MS is a significant loss for clinical microbiology and infectious medicine. We hope that this will not hinder the development and establishment of such state-of-the-art diagnostic technologies in medicine in the future.


References 1. Seifert H. The clinical importance of microbiological findings in the diagnosis and management of bloodstream infections. Clin Infect Dis. 2009. published online EpubMay 15;48(Suppl 4):S238–45. https://doi.org/10.1086/598188. 2. Seymour CW, Gesten F, Prescott HC, Friedrich ME, Iwashyna TJ, Phillips GS, Lemeshow S, Osborn T, Terry KM, Levy MM. Time to treatment and mortality during mandated emergency care for Sepsis. N Engl J Med. 2017. published online EpubJun 8;376:2235–44. https://doi. org/10.1056/NEJMoa1703058. 3. Kollef MH. Broad-spectrum antimicrobials and the treatment of serious bacterial infections: getting it right up front. Clin Infect Dis. 2008. published online EpubSep 15;47(Suppl 1): S3–13. https://doi.org/10.1086/590061. 4. Banerjee R, Teng CB, Cunningham SA, Ihde SM, Steckelberg JM, Moriarty JP, Shah ND, Mandrekar JN, Patel R. Randomized trial of rapid multiplex polymerase chain reaction-based blood culture identification and susceptibility testing. Clin Infect Dis. 2015 published online EpubOct 1;61:1071–80. https://doi.org/10.1093/cid/civ447. 5. Buchan BW, Ledeboer NA. Emerging technologies for the clinical microbiology laboratory. Clin Microbiol Rev. 2014. published online EpubOct;27:783–822. https://doi.org/10.1128/ CMR.00003-14. 6. Banerjee R, Ozenci V, Patel R. Individualized approaches are needed for optimized blood cultures. Clin Infect Dis. 2016. published online EpubNov 15;63:1332–9. https://doi.org/10.1093/ cid/ciw573. 7. Opota O, Croxatto A, Prod'hom G, Greub G. Blood culture-based diagnosis of bacteraemia: state of the art. Clin Microbiol Infect. 2015. published online EpubApr;21:313–22. https://doi. org/10.1016/j.cmi.2015.01.003. 8. Patel R. MALDI-TOF MS for the diagnosis of infectious diseases. Clin Chem. 2015. published online EpubJan;61:100–11. https://doi.org/10.1373/clinchem.2014.221770. 9. Ecker DJ, Sampath R, Li H, Massire C, Matthews HE, Toleno D, Hall TA, Blyn LB, Eshoo MW, Ranken R, Hofstadler SA, Tang YW. New technology for rapid molecular diagnosis of bloodstream infections. Expert Rev Mol Diagn. 2010. published online EpubMay;10:399–415. https://doi.org/10.1586/erm.10.24. 10. Ecker DJ, Sampath R, Massire C, Blyn LB, Hall TA, Eshoo MW, Hofstadler SA. Ibis T5000: a universal biosensor approach for microbiology. Nat Rev Microbiol. 2008. published online EpubJul;6:553–8. https://doi.org/10.1038/nrmicro1918. 11. Metzgar D. Adaptive evolution of diagnostic resistance. J Clin Microbiol. 2011. published online EpubJul;49:2774–5. https://doi.org/10.1128/JCM.02334-10. 12. Baldwin CD, Howe GB, Sampath R, Blyn LB, Matthews H, Harpin V, Hall TA, Drader JJ, Hofstadler SA, Eshoo MW, Rudnick K, Studarus K, Moore D, Abbott S, Janda JM, Whitehouse CA. Usefulness of multilocus polymerase chain reaction followed by electrospray ionization mass spectrometry to identify a diverse panel of bacterial isolates. Diagn Microbiol Infect Dis. 2009. published online EpubApr;63:403–8. https://doi. org/10.1016/j.diagmicrobio.2008.12.012. 13. Sampath R, Russell KL, Massire C, Eshoo MW, Harpin V, Blyn LB, Melton R, Ivy C, Pennella T, Li F, Levene H, Hall TA, Libby B, Fan N, Walcott DJ, Ranken R, Pear M, Schink A, Gutierrez J, Drader J, Moore D, Metzgar D, Addington L, Rothman R, Gaydos CA, Yang S, St George K, Fuschino ME, Dean AB, Stallknecht DE, Goekjian G, Yingst S, Monteville M, Saad MD, Whitehouse CA, Baldwin C, Rudnick KH, Hofstadler SA, Lemon SM, Ecker DJ. Global surveillance of emerging Influenza virus genotypes by mass spectrometry. PLoS One. 2007. published online EpubMay 30;2:e489. https://doi.org/10.1371/journal.pone.0000489. 14. Jordana-Lluch E, Gimenez M, Quesada MD, Ausina V, Martro E. Improving the diag nosis of bloodstream infections: PCR coupled with mass spectrometry. Biomed Res Int. 2014;2014:501214. https://doi.org/10.1155/2014/501214.

490


15. Ozenci V, Patel R, Ullberg M, Stralin K. Demise of polymerase chain reaction/electrospray ionization-mass spectrometry as an infectious diseases diagnostic tool. Clin Infect Dis. 2018 published online EpubJan 18;66:452–5. https://doi.org/10.1093/cid/cix743. 16. Metzgar D, Frinder M, Lovari R, Toleno D, Massire C, Blyn LB, Ranken R, Carolan HE, Hall TA, Moore D, Hansen CJ, Sampath R, Ecker DJ. Broad-spectrum biosensor capable of detecting and identifying diverse bacterial and Candida species in blood. J Clin Microbiol. 2013. published online EpubAug;51:2670–8. https://doi.org/10.1128/jcm.00966-13. 17. Massire C, Buelow DR, Zhang SX, Lovari R, Matthews HE, Toleno DM, Ranken RR, Hall TA, Metzgar D, Sampath R, Blyn LB, Ecker DJ, Gu Z, Walsh TJ, Hayden RT. PCR followed by electrospray ionization mass spectrometry for broad-range identification of fungal pathogens. J Clin Microbiol. 2013 published online EpubMar;51:959–66. https://doi. org/10.1128/jcm.02621-12. 18. Vincent JL, Brealey D, Libert N, Abidi NE, O'Dwyer M, Zacharowski K, Mikaszewska- Sokolewicz M, Schrenzel J, Simon F, Wilks M, Picard-Maureau M, Chalfin DB, Ecker DJ, Sampath R, Singer M, T. Rapid Diagnosis of Infections in the Critically Ill. Rapid diagnosis of infection in the critically ill, a multicenter study of molecular detection in bloodstream infections, pneumonia, and sterile site infections. Crit Care Med. 2015. published online EpubNov;43:2283–91. https://doi.org/10.1097/CCM.0000000000001249. 19. O'Dwyer MJ, Starczewska MH, Schrenzel J, Zacharowski K, Ecker DJ, Sampath R, Brealey D, Singer M, Libert N, Wilks M, Vincent JL. The detection of microbial DNA but not cultured bacteria is associated with increased mortality in patients with suspected sepsis-a prospective multi-centre European observational study. Clin Microbiol Infect. 2017. published online EpubMar;23:208.e201–6. https://doi.org/10.1016/j.cmi.2016.11.010. 20. Metzgar D, Frinder MW, Rothman RE, Peterson S, Carroll KC, Zhang SX, Avornu GD, Rounds MA, Carolan HE, Toleno DM, Moore D, Hall TA, Massire C, Richmond GS, Gutierrez JR, Sampath R, Ecker DJ, Blyn LB. The IRIDICA BAC BSI Assay: rapid, sensitive and culture-independent identification of bacteria and candida in blood. PLoS One. 2016;11:e0158186. https://doi.org/10.1371/journal.pone.0158186. 21. Jordana-Lluch E, Gimenez M, Quesada MD, Rivaya B, Marco C, Dominguez MJ, Armestar F, Martro E, Ausina V. Evaluation of the broad-range PCR/ESI-MS technology in blood specimens for the molecular diagnosis of bloodstream infections. PLoS One. 2015;10:e0140865. https://doi.org/10.1371/journal.pone.0140865. 22. Farrell JJ, Wang H, Sampath R, Lowery KS, Bonomo RA. The effect of empiric antimicrobial treatment duration on detection of bacterial DNA in sterile surgical specimens. PLoS One. 2017;12:e0171074. https://doi.org/10.1371/journal.pone.0171074. 23. Stralin K, Ehn F, Giske CG, Ullberg M, Hedlund J, Petersson J, Spindler C, Ozenci V. The IRIDICA PCR/Electrospray ionization-mass spectrometry Assay on Bronchoalveolar Lavage for bacterial etiology in mechanically ventilated patients with suspected pneumonia. PLoS One. 2016;11:e0159694. https://doi.org/10.1371/journal.pone.0159694. 24. Ullberg M, Luthje P, Molling P, Stralin K, Ozenci V. Broad-range detection of microorganisms directly from Bronchoalveolar lavage specimens by PCR/electrospray ionization-mass spectrometry. PLoS One. 2017;12:e0170033. https://doi.org/10.1371/journal.pone.0170033. 25. Jordana-Lluch E, Rivaya B, Marco C, Gimenez M, Quesada MD, Escobedo A, Batlle M, Martro E, Ausina V. Molecular diagnosis of bloodstream infections in onco-haematology patients with PCR/ESI-MS technology. J Infect. 2017. published online EpubFeb;74:187–94. https://doi.org/10.1016/j.jinf.2016.11.011. 26. Geraci J, Karrasch M, Sachse S, et al. Use of PCR/electrospray ionization mass spectrometry for rapid identification and antibiotic treatment adaptation in patients suffering from sepsis [abstract 2442]. 27th European congress of clinical microbiology and infectious diseases. Vienna, 2017. 27. Reinhart K, Daniels R, Kissoon N, Machado FR, Schachter RD, Finfer S. Recognizing sepsis as a global health priority - a WHO resolution. N Engl J Med. 2017. published online EpubAug 3;377:414–7. https://doi.org/10.1056/NEJMp1707170.

Nucleic Acid Amplicons Detected and Identified by T2 Magnetic Resonance Jessica L. Snyder, Heather S. Lapp, Zhi-Xiang Luo, Brendan Manning, and Thomas J. Lowery

Introduction The sensitive and specific detection of nucleic acids by T2 magnetic resonance (T2MR®) is enabled by magnetic particles (MPs) that behave as magnetic relaxation switches [1–3]. First demonstrated in 2001, oligonucleotide conjugated MPs were shown to hybridize to target DNA sequences and agglomerate, causing a change in the T2 relaxation [4]. Subsequent reports extended the variety of analytes amenable to testing by this method, including proteins, viruses, and small molecules [5–7], and highlighted that the method is agnostic to sample opacity by evaluating various matrices, including blood, plasma, and urine [1, 2]. Further efforts have been dedicated to MP synthesis and functionalization, the relaxation mechanism of the MP system, and the development of low-field, miniature nuclear magnetic resonance (NMR) devices. Clinical applications of T2MR have focused on the identification of bloodstream infections, specifically through the detection of pathogen DNA carried by intact cells present within a whole blood sample [8–10]. T2MR detection requires an NMR device that consists of a strong magnet and electronics for radio frequency (RF) signal generation and reception. The NMR device is used to measure the spin-spin relaxation time, or T2, of the water protons (1H nuclei) in the sample. The small volume sample, typically 30uL, is placed within the magnet of the NMR device, and the protons align within the external magnetic field. A Car-Purcell-Meiboom-Gill (CPMG) pulse sequence is used to flip the magnetization vector into the transverse x-y plane via radio frequency pulse. The proton spins oscillate in the x-y plane and decay due to dephasing. The decay time is measured with a series of RF pulses and echo measurements as the T2 signal. J. L. Snyder · H. S. Lapp · Z.-X. Luo · B. Manning · T. J. Lowery (*) T2 Biosystems, Lexington, MA, USA e-mail: [email protected]; [email protected]; [email protected]; [email protected]; [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_23

491

492

J. L. Snyder et al.

MPs used in T2MR measurements can range from tens of nanometers to microns and are typically prepared from superparamagnetic iron oxide with functionalized biocompatible surfaces. When magnetized in the NMR device, they exhibit strong magnetic dipole field and produce microscopic local field inhomogeneities. The water protons are highly sensitive to nonuniformities within a magnetic field, and the MPs strongly influence the T2 response of the sample. In the absence of MPs, the protons are in phase, and signal decay is slow in CPMG measurements, resulting in a long T2 signal. When MPs are present in the solution, fast signal decay, and short T2 signal, occurs because of the dephasing caused by the magnetic dipole field of MPs. This effect has long been utilized as a contrast agent in medical magnetic resonance imaging (MRI) [11]. Functionalized MPs can be induced to cluster in the presence of an analyte, a process that is the foundation of analyte-specific T2MR detection. The influence of clustering is dependent on the size of the MPs. Early T2MR detection studies used MPs on the order of tens of nanometers, and the T2 signal was shown to increase or decrease depending on the clustering size, complicating the interpretation of T2 signals [12, 13]. Later developments employ MPs on the scale of hundreds of nanometers which consistently resulted in an increase in T2 signal upon clustering and a higher signal to noise ratio [3]. The mechanism of nucleic acid detection by T2MR is illustrated in Fig. 1. Suspensions consisting of a mix of two populations of MPs are prepared: one type of MP is conjugated with oligonucleotide probes complimentary to the 5′ end of the target sequence and a second type of MP is conjugated with oligonucleotide probes complimentary to the 3′ end. The MP suspension is hybridized to the sample, and the single-stranded target sequence cross-links the MPs and causes the formation of interconnected clusters. The agglomeration of the particles results in a sharp increase in T2 signals, often >10x higher than negative samples. In this format the T2MR measurement is qualitative, and a positive sample is determined as a sample in which the T2 signal is higher than a predetermined cutoff. While the T2MR system can be employed for detection of biological analytes besides nucleic acids, great strides have been made in the use of T2MR in molecular diagnostics. Because the biological fluids are transparent to the RF signal, the T2MR detection can be used on opaque samples such as blood, urine, sputum, etc. without upfront purification, unlike standard molecular methods such as the polymerase chain reaction (PCR). In Fig. 2, similar dose-response curves are shown for oligonucleotide targets in buffer and blood lysate, demonstrating the lack of interference from the background materials in the blood (Fig. 2). The implementation of T2MR in the identification of pathogens within blood is described in the following sections. The high sensitivity and specificity afforded by this method have resulted in limits of detection (LoD) ≤10 colony-forming units (CFU)/mL for fungal and bacterial species in whole blood [8, 14]. Emerging clinical reports have underscored the benefit of this rapid and sensitive diagnostic tool.

Nucleic Acid Amplicons Detected and Identified by T2 Magnetic Resonance

493

Fig. 1 Principles of nucleic acid detection by T2MR. (a) MPs are functionalized with oligonucleotide probes that cause the clustering of particles in the presence of target DNA. (Used with permission from Neely et al. 2013) [8]. (b) When the normalized magnetic signal (M(t)) is plotted against time, clustered MPs (B) display a slower signal decay, or longer T2MR signal, than non- clustered MPs (A). (Used with permission from Luo et al. 2016) [3]

ensitive Detection of Fungal and Bacterial Pathogens S in Complex Matrices The T2MR molecular diagnostic methodology detects pathogens through the identification of intact cells in whole blood samples by their DNA. It is important to note that the sample preparation steps described here prior to detection of nucleic acids by existing T2MR methods avoid the detection of free pathogen DNA within the blood, i.e., DNAemia or degraded pathogen DNA. The starting sample for the T2MR detection method is whole blood directly from the patient (Fig. 3). Host blood cells in the sample are lysed with a detergent-based lysis buffer and the intact target cells are concentrated through centrifugation. Any transient DNA is removed during this step. The target cells are then lysed using mechanical methods and the released DNA is amplified using a multiplex primer mixture. After amplification, the amplicon is split and mixed with particles directed against each species being

494

J. L. Snyder et al.

Fig. 2 Change in T2MR signal upon addition of DNA is similar for buffer and blood. As low as 1e9 DNA copies post-amplification can be detected with T2MR in both buffer and blood. (Used with permission from Neely et al. 2013) [8]

detected. Detection of the amplicon/particle mixture is performed on a T2MR reader. Limits of detection (LoDs) ≤10 CFU/mL have been reported for fungal and bacterial species using this method, therefore eliminating the necessity for a pre- amplification step like blood culture, which can increase the time to result by hours or days depending on the species. The use of minimally processed samples increases sensitivity through the reduction of target loss during sample purification procedures. A proprietary amplification system was designed to amplify target DNA directly within the blood lysate [8], which contains a high amount of cellular debris and protein. Most molecular methods require DNA isolation prior to amplification, and the blood in particular contains a high amount of PCR inhibitors, including hemoglobin, lactoferrin, and immunoglobulin G [15, 16], that must be removed from the sample prior to amplification. No standardized method exists for DNA isolation, and variability between methods of DNA isolation can cause discrepant results [17]. By combining a blood- tolerant amplification system with a detection system able to function in opaque samples and not influenced by common interferents, sensitive results can be obtained directly from whole blood samples. The total sample volume within the amplification reaction is an important constraint on the analytic sensitivity as well; larger volumes analyzed can attain lower potential LoD. A large whole blood sample (>1.5 mL) can be evaluated by T2MR with few loss-prone processing steps by using the method described above. In


495

Fig. 3 Steps in the T2MR assay using a whole blood sample. The blood sample is added to a lysis tube, where blood cell lysis and concentration of the target cells occur. The target or multiple targets (co-infections) are then lysed, and the opaque lysate is passed to an amplification tube for multiplex amplification of the target(s). After amplification, the product is divided into separate detection tubes and hybridized with particles each directed against a specific species. Many more individual detections can be performed than shown in this graphic. The entire assay takes 3–5 h (individual steps not drawn to scale) and can be performed directly from drawn blood

essence, the entire initial sample is concentrated into a small volume of blood lysate containing the target pathogen cells, allowing for the achievement of LoD in the single cells/mL range. Often, PCR-based methods utilize small sample volumes or pass only a small percentage of the total isolated DNA through amplification after the loss-prone purification steps [18–21], thus hindering the analytical sensitivity by raising the LoD achievable by the preparation. The determination of analytical sensitivity in the range of ≤10 CFU/mL necessitates the preparation of spiked samples at reproducibly low titers. All positive controls and positive research samples were prepared via seeding microorganisms into a bulk human whole blood pool or other matrix, a process which has been optimized to ensure accurate cell titers, and aliquoting the seeded blood into individual samples for T2MR analysis (Fig. 4). This process minimizes variation across a sample set and allows confirmation of cellular inputs. Each species of interest was analyzed under various growth and storage conditions to ensure healthy and robust cell growth and to minimize cell death. Microbes were prepared as liquid cultures and grown to mid-log phase. The mid-log cultures were enumerated via microscopy and seeded immediately into human whole blood or other medium of choice or were prepared as 100–300 μL aliquots frozen down in glycerol to store for future use. Frozen aliquot cell titers were determined by quality control (QC) plating onto solid growth media for all organisms except

496

J. L. Snyder et al.

Fig. 4 Flow charts of preparation of positive control and positive research human whole blood samples. (a) Sample preparation for Candida spp. and Bacteria species able to be grown on solid growth media from either frozen aliquots or fresh cultures. (b) Sample preparation for Borrelia spp. or any species unable to grow on solid growth media

Borrelia spp.; Borrelia titers were determined via microscopy as these organisms cannot grow on solid growth media. To prepare seeded blood samples, either a fresh culture was prepared or a frozen aliquot was thawed. The organisms were enumerated via microscopy or had a known cell titer determined previously. Organisms were then diluted and seeded into the bulk blood or other matrix at the desired concentration. For those able to grow on solid media, samples of the diluted organisms were QC plated onto appropriate growth media and incubated until countable colonies were visible (species- specific). This step allows for further confirmation of cell titer in the samples. Analytical sensitivity experiments were performed by preparing whole blood samples spiked with fungal or bacterial cells at several known concentrations using the sample preparation method as described above. This sample preparation method minimizes variation and allows consistent and reproducible samples at known and verified concentrations. In the analytical sensitivity analyses using the T2Dx® Instrument, the LoD was defined as the lowest concentration which maintained a 95% positivity rate. In a 2013 study with the five Candida spp. causing the majority of candidemia infectious, C. albicans, C. glabrata, C. krusei, C. parapsilosis, and C. tropicalis, LoDs between 1 and 3 CFU/mL were demonstrated with an automated T2MR assay run on the T2Dx Instrument (Table 1) [8]. A multiplex panel of six bacterial species that cause the majority of blood stream infections was shown to have a preliminary LoD between 1 and 8 CFU/mL using a manual T2MR assay [22]. In a 2017 study, three North American and European Lyme disease-causing Borrelia spp., B. afzelii, B. burgdorferi, and B. garinii, were tested, and preliminary LoDs of 5–8 cells/mL were found using a manually processed method (Table 2) [14].


497

Table 1 LoD for five Candida spp. as determined by analysis on the T2Dx Instrument (8, T2Candida Instructions for Use) Species C. albicans C. glabrata C. krusei C. parapsilosis C. tropicalis

CFU/mL 2 2 1 3 1

N 21 20 20 20 20

Average T2MR signal 737 554 676 1194 822

%CV T2 28 11 16 19 15

%Positive 100 100 100 100 100

Table 2 Preliminary LoD of six sepsis causing bacterial species (in CFU/mL) and three Lyme disease-causing Borrelia spp. (in cells/mL) performed using a manual T2MR assay Species Acinetobacter baumannii Escherichia coli Enterococcus faecium Klebsiella pneumoniae Pseudomonas aeruginosa Staphylococcus aureus Borrelia afzelii Borrelia burgdorferi Borrelia garinii

CFU/mL or cells/mL 2 8 3 6 1 3 8 5 8

N 20 20 20 20 20 20 78 60 60

Average T2MR signal 275 591 284 515 517 351 374 379 347

%CV T2 23 28 14 26 12 10 43 15 10

%Positive 95 95 100 100 100 100 97 97 100

Adapted from Snyder et al. [14] and Neely et al. [22]

pecificity Achieved Through Primer and Probe Design S Coupled with Discriminatory Detection System The specificity of the T2MR detection system can be adjusted at either the primer or probe level. The detection step requires hybridization of two probes to a given amplified target, and studies have shown that single-base mismatches interrupt hybridization-based agglomeration [5]. Therefore, non-specific products from off- target amplification have minimal influence on the detection of the specific targets, and specificity can be fully maintained by probes alone. Off-target amplification does not significantly impact probe performance, allowing the use of non-specific primers and probes that bind specifically to the DNA from the target species. Therefore many options exist for assay design, whether using a single primer set to amplify more than one target and differentiate products with specific probes or by having specific primers for individual targets and combinations thereof. All primer and probe designs are first performed in silico, and genomic databases are used to evaluate inclusivity or exclusivity of near-neighbor strains. Then primer and probe performance in singleplex and multiplex formulations are optimized with bench testing.

498

J. L. Snyder et al.

The T2MR detection system displays high signal to noise, allowing for unambiguous determination of positive samples. Particle cross-reactivity itself is low with optimally designed probes, as the presence of nontarget amplicon does not raise signal above baseline values. Multiplex amplification reactions can be used to test a panel of species, and species identification can be performed on samples containing either single or multiple species. Because T2MR requires independent, separate detection reactions, the amplified product is aliquoted with each unique detection particle formulation. The volume of hybridization can be small (≤30 uL), and the minimum amplicon concentration is low (~1e9); therefore then the amplified reaction can be diluted and aliquoted into several different detection reactions to allow for multiplexed detection on the order of 10–30 targets per amplification reaction. In addition, individual species can be detected in competition assays, in which one species is spiked at or near LoD and a second species is 100–1000x higher in concentration, within a single reaction. Contamination is a concern in any high sensitivity assay, and strict contamination controls are implemented as part of sample preparation and the T2MR assay. In a laboratory situation, physical separation of work areas is used to reduce the potential of cell or amplicon-induced false positives. To this end, strict separation is maintained between negative and spiked blood sample preparation and pre- and post-amplification work areas. Exceptionally low rates of contamination were observed using such processes during the analysis of Borrelia species in whole blood [14]. In addition, the detection of commensal species requires highly clean laboratory practices and reagents.

I ntegration NMR System with Automated Sample Processing Promotes Short Time to Result T2 relaxation measurements can be carried out on nearly all NMR/MRI instruments; however due to the simplicity of CPMG measurements, compact, affordable, low-field NMR instruments can be designed for T2MR measurements. One such T2MR reader designed by T2 Biosystems is the size of a shoebox and contains small permanent magnets that produce a magnetic field of ~0.55 T. 1H NMR is conducted using CPMG pulse sequences generated and detected by an internal spectrometer and a coiled RF probe around the sample tube. While nucleic acid detection by T2MR can be performed within a laboratory using a T2MR reader as described above, a rapid sample-to-answer molecular diagnostic test requires an integrated system that contains subsystems for sample preparation, DNA extraction, amplification, and detection while maintaining strict contamination controls. The T2Dx Instrument is a multipurpose, bench-top instrument that fully automates the entire T2MR assay from sample preparation to T2MR detection. Five major processes are performed by the system: specimen loading, specimen concentration and lysis, amplification of target DNA, hybrid-


499

ization of the amplified DNA to target-specific oligonucleotide probes bound to superparamagnetic nanoparticles, and analysis of the clustered nanoparticles by T2MR detection. Moderate throughput is possible with the instrument as seven individual samples can be tested at once with a time to result per sample of 3–5 h. Disposables and reagents are provided in an easy-to-use single-test cartridge that is designed to be able to generate up to 8 results from a single patient sample. An expanding menu of molecular tests is available for the device. The T2Dx Instrument enables direct-from-blood pathogen identification that can be performed routinely with minimal technical expertise in a hospital’s microbiology laboratory. The standard microbiology laboratory procedure for species identification starts with the use of blood culture, a technology that hasn’t changed much since its initial development in the early 1900s. The time to positivity for blood cultures depends heavily on the species being grown. Following blood culture, methods such as Gram stain, PCR, matrix-assisted laser desorption/ionization time- of- flight (MALDI-TOF), peptide nucleic acid-fluorescence in situ hybridization (PNA-FISH), or sequencing may be used to make a final identification, a process that can add hours or days following the already time-intensive blood culture. Many newer species identification tests aim to speed up the process post blood culture and for sensitivity reasons cannot be performed directly from the blood, unlike the T2MR method. While such a test can save time and labor over the traditional identification methodologies, they still inherit the long time to result intrinsic to blood culture. Two studies highlighting the improvement in time to identification of fungal and bacterial species using MALDI-TOF over traditional methods reported total mean times of 55.9 and 58.7 h, and indeed the majority of this time requirement is due to blood culture [23, 24]. By enabling results direct from the patient sample within 3–5 h, T2MR detection represents a paradigm shift in a field where time savings can mean substantial reductions in mortality.

igh Sensitivity and Specificity Demonstrated in Clinical H Samples Using the T2Candida® Panel The automated T2Candida Panel on the T2Dx Instrument was compared to the industry standard for diagnosis of Candida infections, blood culture, in 2015 [9]. During this study, 1501 blood samples with concurrent draws for blood culture were tested on the T2Dx Instrument. An additional 300 contrived samples were prepared and tested by T2MR and blood culture, 250 of which were spiked with one of the 5 Candida spp. detected by T2MR based on the FDA-accepted clinically relevant titer and 50 of which were negative. The clinical trial demonstrated an overall 91.1% sensitivity and 99.4% specificity per assay when compared against blood culture. While blood culture is generally considered to be the standard for detection of pathogens associated with sepsis, some experts consider it to

500

J. L. Snyder et al.

be only 50% positive due to its inability to detect deep-seated candidiasis [25], and an analysis of patients with candidemia or bacteremia showed only 65.1% of blood cultures were positive on the first draw [26]. In addition, time to result for T2MR tests during the trial was found to be 4.4 h as compared to the 2–5 days of blood culture. A rapid time to result is an important factor for both reducing mortality and directing use of appropriate antifungals as part of an antimicrobial stewardship program. An analysis of 142 patient blood samples from the Henry Ford Hospital system taken between 2015 and 2016 demonstrates the benefits of the T2Dx Instrument and the T2Candida Panel [27] in a hospital system where samples are being transported to a central laboratory for analysis. In this sample set analysis, the median time to infecting species identification decreased from 41.75 h with standard blood culture to 25.25 h after the implementation of T2MR. The median time to active appropriate therapy was found to decrease from 39.6 to 26.6 h after T2MR was available. The study further found that length of stay for patients with candidemia in the ICU decreased by 7 days. Not all patients were diagnosed by T2MR after it was implemented; thus the 25.5 h to identification and 26.6 h to appropriate therapy are medians of a population where 44% of patients were being diagnosed by more lengthy blood culture methods. Within the group of patients tested by T2MR, the median time to identification was 7.1 h and the median time to active therapy was 13.4 h. All times were reported as the time from order of testing to actionable result, and inter-hospital transport of samples within the Henry Ford Hospital system may have led to an increased time to result in some cases. In a separate study, 59 patient samples were tested by T2MR at Riverside Community Hospital (RCH) in Riverside, CA [28]. Here, RCH demonstrated that by using the T2Dx Instrument and T2Candida Panel, infections were both detected and identified earlier than by using blood culture, and because of this, 100% patients were put on focused and appropriate antifungal therapies within 9 h. Both physician education and a strong antimicrobial stewardship program are necessary to affect positive reaction based on rapid T2MR results, as evidenced by time to de-escalation exhibited by Huntsville Hospital [29]. While an average time to de-escalation of 42.6 h was measured for samples with negative T2MR results, three separate physician teams involved in the study had average times of de-escalation of 15, 52.4, and 54 h, respectively. This inconsistent utilization of the T2Candida Panel results shows the need for consistent medical education programs to ensure a uniform understanding of the T2Dx Instrument and the T2Candida Panel impact on patient care. Benefits of a rapid, direct-from-blood infectious disease test like the T2Candida Panel can also be realized in the economic impact of reduction of length of stay and de-escalation of antifungals. Using a 1-year decision tree model, savings of $5.8 million were estimated for a hospital with 5100 high-risk patients when using the T2Candida Panel as a diagnostic tool rather than standard blood culture [30]. When results can be achieved in 3–5 h using T2MR as opposed to 3.6–


501

6.4 days for blood culture, it is estimated that the major cost savings would be due to reducing of ICU length of stay for patients treated with the appropriate therapy within 24 h of diagnostic testing. In addition, it was expected that mortality would be reduced by 60.6% through the use of a rapid test. An unexpected outcome of recent studies was that T2MR is capable of detecting deep-seated, or invasive, candidiasis. Deep-seated infections are localized to typically sterile sites within the body, and circulation of the fungal cells within the blood stream does not always accompany such an infection. Of 12 cases determined to be deep-seated candidiasis in follow-up visits, 12/12 were detected by T2MR and 0/12 were detected by blood culture [10]. This discrepancy may be due to low LoD of circulating cells or the sloughing off of dead cells from the site of infection for infections that are at the breakthrough or metastatic stage, both of which may be detected by the highly sensitive T2MR method and not by blood culture.

Conclusion and Outlook Nucleic acid detection can be performed rapidly and in complex solutions using the T2MR technology. High sensitivity and specificity are afforded by primer and probe design and a highly discriminatory detection system. Targets can be detected directly in whole blood prior to culture without the use of complex sample purification techniques due to the use of a detection system that functions in opaque samples. Promising clinical results have been achieved for the detection of Candida spp. in whole blood. Initial studies have shown high sensitivity and specificity when compared to blood culture and a reduction in time to result and treatment due to the rapid nature of the test. Development of novel uses and additional target species is underway with the T2MR technology. Research on alternative sample matrices has shown strong performance in both dialysate [31] and urine [32] spiked with Candida spp. on the T2Dx Instrument with standard reagents. It was demonstrated that low volumes (0.5 mL) of blood spiked with Candida spp. can be tested with high sensitivity (LoD ~ 2.5 CFU/mL) if samples were diluted to a working volume prior to testing [33]. The ability to test low volumes of blood is crucial for pediatric and neonatal patients suspected of having sepsis. Additional test panels beyond the T2Candida Panel are currently being developed and automated on the T2Dx Instrument. As shown in Table 2, preliminary manual LoDs are less than 10 CFU/ mL for sepsis causing bacterial species and less than 10 cells/mL for Lyme disease causing Borrelia spp. Further research into sample matrices and target species will enable T2MR to be a wide reaching diagnostic platform for many forms of infectious disease.

502

J. L. Snyder et al.

References 1. Lowery TJ. Nanomaterials-based magnetic relaxation switch biosensors. In: CSSR K, editor. Nanomaterials for the life sciences, Magnetic nanomaterials, vol. 4. Weinheim: Wiley-VCH Verlag GmbH & Co. KGaA; 2009. p. 3–53. 2. Skewis LR, Demas V, Lowery TJ. Nuclear magnetic resonance nanotechnology: application in clinical diagnostics and monitoring. Encycl Anal Chem. 2013:1–26. 3. Luo Z-X, Fox L, Cummings M, Lowery TJ, Daviso E. New frontiers in in vitro medical diagnostics by low field T 2 magnetic resonance relaxometry. Trends Anal Chem. 2016;83:94–102. 4. Josephson L, Perez JM, Weissleder R. Magnetic nanosensors for the detection of oligonucleotide sequences. Angew Chem. 2001;113:3304–6. 5. Perez JM, Josephson L, O'Loughlin T, Högemann D, Weissleder R. Magnetic relaxation switches capable of sensing molecular interactions. Nat Biotechnol. 2002;20:816–20. 6. Perez JM, Simeone FJ, Saeki Y, Josephson L, Weissleder R. Viral-induced self-assembly of magnetic nanoparticles allows the detection of viral particles in biological media. JACS. 2003;125:10192–3. 7. Perez JM, Josephson L, Weissleder R. Use of magnetic nanoparticles as nanosensors to probe for molecular interactions. Chembiochem. 2004;5:261–4. 8. Neely LA, Audeh M, Phung NA, et al. T2 magnetic resonance enables nanoparticle-mediated rapid detection of candidemia in whole blood. Sci Transl Med. 2013;5:182ra54. 9. Mylonakis E, Clancy CJ, Ostrosky-Zeichner L, et al. T2 magnetic resonance assay for the rapid diagnosis of candidemia in whole blood: a clinical trial. Clin Infect Dis. 2015;60:892–9. 10. Pfaller MA, Wolk DM, Lowery TJ. T2MR and T2Candida: novel technology for the rapid diagnosis of candidemia and invasive candidiasis. Future Microbiol. 2015;11:103–17. 11. Na HB, Song IC, Hyeon T. Inorganic nanoparticles for MRI contrast agents. Adv Mater. 2009;21:2133–48. 12. Brooks RA. T2-shortening by strongly magnetized spheres: a chemical exchange model. Magn Reson Med. 2002;47:388–91. 13. Demas V, Lowery TJ. Magnetic resonance for in vitro medical diagnostics: superparamagnetic nanoparticle-based magnetic relaxation switches. New J Phys. 2011;13:025005. 14. Snyder JS, Giese H, Bandoski-Gralinski C, et al. T2 magnetic resonance-based direct detection of three Lyme disease-related Borrelia species in whole blood samples. J Clin Microbiol. 2017;55:2453. 15. Al-Soud WA, Jonsson LJ, Radstrom P. Identification and characterization of immunoglobulin G in blood as a major inhibitor of diagnostic PCR. J Clin Microbiol. 2000;38:345–50. 16. Al-soud WA, Radstrom P. Purification and characterization of PCR-inhibitory components in blood cells. J Clin Microbiol. 2001;39:485–93. 17. Arvanitis M, Anagnostou T, Fuch BB, Claiendo AM, Mylonakis E. Molecular and nonmolecular diagnostic methods for invasive fungal infections. Clin Micro Rev. 2014;27:490–526. 18. Khot PD, Fredricks DN. PCR-based diagnosis of human fungal infections. Expert Rev Anti- Infect Ther. 2009;7:1201–21. 19. Shin JH, Nolte FS, Morrison CJ. Rapid identification of candida species in blood cultures by a clinically useful PCR method. J Clin Microbiol. 1997;35:1454–9. 20. Jordan JA, Durso MB. Real-time polymerase chain reaction for detecting bacterial DNA directly from blood of neonates being evaluated for sepsis. J Mol Diagn. 2005;7:575–81. 21. Liveris D, Schwartz I, McKenna D, et al. Comparison of five diagnostic modalities for direct detection of Borrelia burgdorferi in patients with early Lyme disease. Diagn Microbiol Infect Dis. 2012;73:243–5. 22. Neely L, Plourde D, Suchocki A, et al. T2Bacteria: rapid and sensitive detection and identification of sepsis pathogens in whole blood specimens by T2MR®. Poster presented at 2015 AMP annual meeting, Austin.


503

23. Huang AM, Newton D, Kunapuli A, et al. Impact of rapid organism identification via matrix- assisted laser desorption/ionization time-of-flight combined with antimicrobial stewardship team intervention in adult patients with bacteremia and candidemia. Clin Infect Dis. 2013;57:1237–45. 24. Schneiderhan W, Grundt A, Worner S, Findeisen P, Neumaier M. Work flow analysis of around-the-clock processing of blood culture samples and integrated MALDI-TOF mass spectrometry analysis for the diagnosis of bloodstream infections. Clin Chem. 2013;59:1649–56. 25. Clancy CJ, Nguyen MH. Finding the “missing 50%” of invasive candidiasis: how nonculture diagnostics will improve understanding of disease spectrum and transform patient care. Clin Infect Dis. 2013;56:1284–92. 26. Cockrill FR, Wilson JW, Vetter EA, et al. Optimal testing parameters for blood cultures. Clin Infect Dis. 2013;38:1724–30. 27. Wilson NM, Kenney RM, Tibbetts RJ, et al. T2 magnetic resonance improves timely management of candidemia. Poster presented at 2016 IDWeek, New Orleans. 28. Patel F, Young E. Antifungal prescribing during initial implementation of candidemia early detection and species identification testing with T2Candida panel. Poster presented at 2016 IDWeek, New Orleans. 29. Kateon H, Edwards J, Sawyer A, Ross J, Hassoun A. Utilization of T2Candid panel for the rapid detection of Candida species in a large community hospital. Poster presented at 2016 IDWeek, New Orleans. 30. Bilar SP, Ferrufino CP, Pfaller MA, Munakata J. The economic impact of rapid Candida species identification by T2Candida among high-risk patients. Future Microbiol. 2015;10:1133–44. 31. Kieffer T, Schmitt BH. Diagnosis of fungal peritonitis caused by Candida spp. using T2 magnetic resonance technology on dialysate matrix. Poster presented at 2017 ASM Microbe, New Orleans. 32. Kissel S, Kieffer T, Schmitt BH. Clinical urine specimens as a sample matrix for T2 magnetic resonance technology. Poster presented at 2017 ASM Microbe, New Orleans. 33. Kieffer T, Schmitt BH. T2 magnetic resonance technology for the diagnosis of candidemia in low volume whole blood specimens. Poster presented at 2017 ASM Microbe, New Orleans.

Molecular Contamination and Amplification Product Inactivation Susan Sefers and Jonathan E. Schmitz

Introduction “With great power comes great responsibility.” Although perhaps not what the comic book character had in mind, this popular quote aptly describes the role of the clinical microbiology laboratory in the age of molecular diagnostics. The power rests in the ability of the polymerase chain reaction (PCR) and other nucleic acid amplification tests (NAATs) to produce a massive quantity of amplicon from (theoretically) a single molecule of template, allowing for the direct identification of viral, bacterial, fungal, and parasitic pathogens in clinical specimens. Methods and platforms for detecting microbial nucleic acids are becoming increasingly more diverse. With this expanding functionality, NAATs are now seen as the new standard of laboratory care for many suspected infections, with real-time decision-making hanging in the balance. For instance, identification of herpes simplex virus 2 (HSV- 2) in the cerebrospinal fluid of an infant with suspected meningitis is now routinely performed via PCR [1], allowing for targeted administration of antiviral therapy to help prevent severe neurologic sequelae [2]. Before the advent of molecular diagnostics, there was minimal chance of recovering the virus from CSF with traditional, culture-based methods [3]. At the same time, the responsibility that comes with these powerful tests is the laboratory’s duty to ensure result accuracy. Decisions both to initiate and discontinue anti-infective therapy are frequently based on NAATs, and spurious results The chapter has been adapted from a previous edition: Sefers S, Stratton CW, and Tang Y-W. Chapter 26: Amplification product inactivation. In: Tang, Y-W and Stratton CW, eds. Advanced techniques in diagnostic microbiology. https://doi.org/10.1007/978-1-4614-3970-7_26, © Springer Science+Business Media New York 2013. S. Sefers · J. E. Schmitz (*) Department of Pathology, Microbiology, and Immunology, Vanderbilt University Medical Center, Nashville, TN, USA e-mail: [email protected] © Springer International Publishing AG, part of Springer Nature 2018 Y.-W. Tang, C. W. Stratton (eds.), Advanced Techniques in Diagnostic Microbiology, https://doi.org/10.1007/978-3-319-33900-9_24

505

506

S. Sefers and J. E. Schmitz

can contribute to adverse patient outcomes. Unfortunately, the high analytic sensitivity of molecular diagnostics can serve as a double-edged sword, especially when coupled with conditions that predispose to false positivity. Not only can pathogen DNA/RNA be present in clinical materials over an extremely wide range of concentrations (many orders of magnitude), but the amplicons themselves can serve as detectable targets for future reactions. Minute quantities of cross-contamination, either specimen-to-specimen or product-to-specimen, can give the false impression that a pathogen is present. In the clinical scenario described above, if trace amplicon from a previously positive individual made its way into the infant’s specimen and caused a spurious HSV-2 detection, antiviral therapy would be administered unnecessarily, and the child would likely remain hospitalized for additional time. Along with the continued stress for the parents and the concern for vertical transmission, thousands of dollars would be wasted on inpatient care. Even more concerning is when a false-positive NAAT for one pathogen influences clinicians to de-escalate therapy—or the halt the diagnostic work-up—for another agent that is truly responsible for a patient’s symptoms. Given what is at stake, it is crucial for molecular infectious disease laboratories to be aware of the circumstances that give rise to false-positive results, as well as common measures to prevent them. The current chapter is dedicated to these topics, including techniques for inactivating amplicons so that they cannot serve as detectable targets in future reactions. These methods are particularly relevant to laboratories that validate their own molecular diagnostic tests for on-site use (i.e., laboratory-developed tests or “home brews”). By contrast, for commercially prepackaged in vitro diagnostics, reagents and protocols are typically fixed per regulatory requirements. At the same time, many considerations for minimizing false positives pertain equally to home brews and commercial platforms. In fact, these measures are simply part of how clinical laboratories must adapt their quality assurance programs to the emerging demands of a new molecular era.

Causes of False-Positive Molecular Results All NAATs for infectious diseases share two basic components: amplification of nucleic acid targets(s) and detection of the amplified product. Although PCR and real-time PCR (qPCR) are the most pervasive examples of targeted amplification, other methods in common clinical use include loop-mediated isothermal amplification (LAMP) [4], transcription-mediated amplification (TMA) [5], and the ligase chain reaction [6]. Detection techniques are even more diverse, including fluorometric [7], colorimetric [8], turbidometric [4], electrochemical [9], optical [10], and magnetic resonance [11] technologies. Regardless of the specific amplification and detection method, the steps of a NAAT where false positivity might arise are conceptually well conserved. Some potential causes are intrinsic to the assay itself, such as off-target priming or the spurious detection of signal when no amplicon is actually present. A detailed discussion on preventing these phenomena is beyond

Molecular Contamination and Amplification Product Inactivation

507

the scope of the chapter, although it suffices to say that an assay’s primer/probe design and a robust validation are critical in ensuring high analytic specificity [12, 13]. Assuming that an assay’s basic molecular biology is sound, a far more common cause of false positivity is the introduction of target DNA/RNA into a negative specimen within the laboratory—in other words, contamination [14–16]. In this context, amplicon contamination from a previous positive specimen (product-to-specimen carryover) is the most widely feared culprit. NAATs create billions of copies (or more) of target nucleic acid, such that microscopic aerosols generated during routine handling represent more-than-ample sources for carryover. For instance, when a post-amplification microcentrifuge tube is snapped opened, amplicon can aerosolize into the immediate vicinity, with eventual introduction into the pre-amplification pool. False positivity can also arise if a high-titer positive specimen contaminates a negative sample in its immediate vicinity (specimen-tospecimen carryover). Returning to the case of herpes meningitis, one could envision a genital scraping with abundant HSV-2 being processed next to the infant’s CSF (in fact, cutaneous lesions can have notably high levels of herpesvirus [17]). In this scenario, a small drop of the transport media containing the genital scraping might come into contact with a technologist’s glove; if the CSF specimen is manipulated next, the virus could be transferred. Intact pathogens (pre-amplification) and nucleic acids (pre- or post-amplification) can likewise be carried on pipette tips, vortexers, centrifuges, etc. essentially any inanimate or bodily surface in close proximity to specimen processing and analysis [18]. In most cases, the contaminating fluid is not visible to the naked eye. One should also note that the source of specimen-tospecimen carryover does not have to be a high-titer patient sample, as positive controls and calibrators for quantitative assays can also fill this dubious role [19]. Various examples of laboratory-derived molecular contamination have been described in the literature. For instance, false-positive detections of HSV-1 were observed in the CSF of patients with glioblastoma, although (as often the case) it was not possible to pinpoint the exact step of contamination within the laboratory [20]. As part of a study utilizing dried blood spots to test for HIV, comprehensive cleaning of the punching instrument between use reduced the possibility of specimen-to-specimen carryover [21]. In principal, contamination can even occur before a sample arrives in the laboratory. Multi-subtype influenza co-detections were observed in a clinic due to environmental contamination of clinic surfaces with live attenuated vaccine [22]. Another rare, but theoretically possible, scenario involves the purity of the commercial reagents utilized in a diagnostic assay. For example, Coxiella burnetii DNA was once identified in the PCR master mix employed for a laboratory-developed Coxiella test [23]; Legionella DNA was likewise observed within a commercial extraction kit [24]. In a separate incident, an HIV-based vector was found to contaminate the reagents of an HIV genotyping assay manufactured by the same vendor [25]. (Of course, these cases underscore the need for routine negative controls in all molecular diagnostic testing.) One particular consideration involves NAATs that target sequences conserved broadly across microbial clades (e.g., pan-eubacterial 16S PCR); this scenario is discussed further in Special Considerations, below.

508


Molecular Quality Assurance Preventing false-positive detections and identifying them rapidly when they do occur are key components of a molecular laboratory’s quality assurance (QA) program. As defined by the Clinical and Laboratories Standards Institute (CLSI), QA “encompasses the overall and ongoing monitoring and evaluation of the quality of the total testing process, [including] defining policies and procedures; identifying and correcting problems to ensure accurate, reliable, and prompt reporting of test results; and ensuring the adequacy and competency of staff” [13]. Various other texts, guidelines, and regulatory documents are likewise devoted to this vitally important topic [26–28]. In this context, one can imagine how few events would comprise quality in molecular microbiology more than carryover contamination. As part of the MM03 document Molecular Diagnostic Methods for Infectious Disease, the CLSI outlines the components of a molecular QA program, as they pertain to both false positivity and various other concerns [13]. Over the following sections, we detail specific logistical and methodologic aspects of QA in relation to molecular contamination. Broadly speaking, these measures can be divided into (1) general laboratory workflow considerations, (2) biochemical techniques incorporated into an assay itself, and (3) post-analytic data review. These topics are addressed here in turn.

Workflow Considerations Molecular Laboratory Configuration Basic laboratory design remains one of the most important considerations for preventing false-positive results due to molecular contamination [13, 14, 29, 30]. Of course, these issues can also be among the most challenging, as the structural features of a laboratory space are not always easily altered (at least within fixed financial parameters). Nevertheless, to the best of its ability, laboratory personnel should strive to adapt their space to the particular needs of molecular testing. Figure 1 presents an idealized configuration for a molecular laboratory. A fundamental principle is the physical separation of the post- amplification area—where amplicons are generated, manipulated, and detected— from the remainder of the laboratory. The post-amplification workspace should be set apart by at least a door and, ideally, down a short hallway from the pre- amplification area where specimens are prepared.1 The latter should also reside in its own dedicated room (with a door), as should the location where reagents are 1 Although not the focus of the current discussion, biosafety considerations are also critical for the specimen preparation area of any molecular infectious disease laboratory [32]. Universal (standard) precautions are required for direct manipulation of patient-derived material, including the use of a biological safety cabinet. Additional measures may also be necessary if the laboratory performs molecular testing on cultured pathogens, depending on the recommended biosafety level for those particular agents.


509

Fig. 1 Ideal design of a molecular laboratory. The pre-amplification area is a separated room from the post-amplification area, and the reagent preparation area is separated from the remainder of the lab. There is a unidirectional workflow moving from pre-amplification to post-amplification, with no opposite flow. The reagent preparation area is ideally kept at positive pressure relative to the setup area (indicated in blue), with the post-amplification at negative relative pressure (indicated in red), further preventing transfer of any aerosolized nucleic acid

initially prepared. Each of these spaces requires its own refrigerators, freezers, centrifuges, computer workstations, lab coats, pens, pipettes, etc. And these items must not be moved from room to room, as they represent potential “molecular fomites.” Independent ductwork for the separate areas is likewise advisable. Technologists should follow a “one-way traffic” flow throughout the workday. Sample processing and extraction are performed within the pre-amplification area, after which the technologist moves to the post-amplification area for all subsequent steps. He/she should not subsequently return to the pre-amplification during the same shift. Within these workspaces, additional instrumentation and engineering practices provide further protection against contamination. A variety of automated nucleic acid extraction platforms are available for liquid specimens that reduce both hands-on time and the potential for carryover [31]. These instruments are computer programmed, with pipetting steps conducted inside an enclosed unit that prevents aerosols from entering or escaping. They are often equipped with an ultraviolet (UV) lamp that can be illuminated at the end of the workday to inactivate residual nucleic acids. For manual manipulation steps, many laboratories employ dead air boxes (depicted in Fig. 2a) within the pre-amplification and post-amplification rooms. A dead air box provides exactly what it describes: an area of uncirculated air to contain any aerosols that are inadvertently generated. This is especially significant when considering that a facility’s ventilation systems can disseminate aerosols

510


Fig. 2 (a) Molecular laboratory “dead air box.” The dead air box creates an environment that will decrease the chance of contamination between samples. Aerosols are blocked from entering the unit from the surrounding environment. Before and after usage, the unit can be wiped down and cleaned. Following daily use, the unit can be closed and further decontaminated via UV irradiation. A representative example is depicted here, although the exact design can vary from manufacturer to manufacturer. (b) PCR workstation. Variant technology incorporates laminar flow from HEPA- filtered air, providing additional protection against sample-to-sample cross-contamination from generated aerosols. An idealized schema of a PCR workstation is presented here, although the exact design can again vary depending on the brand

around a workspace. These apparatuses are also typically equipped with a lamp for UV treatment (further details of UV light inactivation are discussed below). Some cabinets, typically referred to as PCR workstations, also include laminar flow and airborne filtration as additional means of protection (Fig. 2b). Along with these larger pieces of equipment, incorporating certain disposable supplies/reagents into the workflow is equally important. Laboratories should utilize aerosol-resistant tips for handheld micropipettes, which include a filter below the point of attachment to block aerosolized material from collecting on the end


511

of the pipette. Replacing snap-cap microfuge tubes with screw-cap equivalents is likewise a straightforward way to reduce aerosols. With positive control specimens for qualitative tests, targets should not be present at exaggerated concentrations but rather closer to the assay’s limit of detection. Not only does this practice reduce a potential source of specimen-to-specimens contamination, but it provides assurance against any drift in an assay’s analytic sensitivity. Other basic recommendations, including routine changing of gloves and the order of preparing reactions (with the specimen added last), are summarized in the CLSI MM03 document [13]. Finally, it is advisable for laboratories to routinely and strategically sample environmental surfaces for the presence of targeted nucleic acids, to gauge whether the above measures are functioning effectively. Instrument Design As molecular assays become the standard of care for diagnosing many infections, the need for testing arises outside the context of reference laboratories and academic medical centers. NAATs are increasingly common in community hospitals, and, looking ahead, point-of-care molecular assays in clinics are undoubtedly a wave of the future (with the first examples now reaching the market) [33, 34]. At the same time, the workflow considerations discussed above may not be feasible in less specialized settings. In this context, the instruments themselves must be engineered to reduce the need for dedicated pre- and post- amplification workspaces. Indeed, a majority of commercial molecular platforms are now closed-tube systems, in which amplification and detection (and sometimes even initial extraction) proceed within a self-enclosed vessel, which remains unopened post-amplification. The potential for product-to-specimen carryover plummets when amplicons are never directly manipulated and analysis vessels are discarded after use. Many such platforms depend upon qPCR (for DNA targets) or qRT-PCR (for RNA targets); these inherently closed-tube technologies utilize fluorescence detection throughout the amplification cycle [35, 36]. Non-qPCR-based platforms have also been developed that maintain a closed system through the instruments’ unique design features, often coupled with the detection methodology itself [7, 11, 37, 38]. One notable advantage of qPCR is the relative ease with which individual laboratories can incorporate the technology into locally developed tests. Previously, molecular “home brews” often relied upon detection methods such as agarose gel visualization, EIA-linked oligonucleotide probes, or other hybridization methods, all of which involve open handling of amplicons [39]. Transitioning such assays to a qPCR format maintains similar analytic sensitivity (as well as the potential for quantification) while keeping the entire process within a microwell plate. For certain analyses, of course, some manipulation of amplicons remain unavoidable. Nevertheless, to the extent that it is possible analytically and logistically, molecular laboratories benefit greatly closed-tube approaches. Other Considerations Finally, regardless of a laboratory’s specific design and instrumentation, regular cleaning of the workspace is of utmost importance to prevent contamination. A 10% dilution of household bleach (sodium hypochlorite) is

512


perhaps the most common cleaning solution employed in molecular laboratories [40, 41]. Technologists should wipe down the work area (including pipettes, pens, and tube racks) with a freshly prepared solution before and after manipulating samples. Bleach oxidizes any nucleic acids present, although it can be corrosive to surfaces over time, so rinsing with 70% ethanol can help minimize such damage. Also, for safety purposes, wiping appropriate surfaces with water prior to bleach can prevent the generation of toxic gases (including HCN and HCl) from the reaction of hypochlorite with guanidinium isothiocyanate, a common component of DNA/ RNA extraction reagents [42]. Additional reagents are available for removing nucleic acids from surfaces, including DNA Away™ (Molecular BioProducts) and similar products from other manufacturers [43]. DNA Away™ is an alkaline solution that degrades nucleic acids. Surfaces must be rinsed with water after cleaning with these reagents, as residual inhibitory residues can cause false-negative reactions if not removed. Their use in a spray bottle is thus discouraged. It must also be emphasized that, for commercial instruments and testing platforms, the manufacturers’ recommended cleaning practices should be followed. In addition to these measures, surface cleaning may also involve nonchemical treatments. Specifically, UV light is commonly utilized to reduce nucleic acid carryover [44–46]. UV irradiation produces dimerization between adjacent pyrimidine bases on the same strand of double-stranded nucleic acid. These adducts inhibit processing by polymerase and render the nucleic acid unsuitable as a template for future amplification reactions. As mentioned previously, UV lights are typically incorporated into dead air boxes and automated extraction instruments, including transparent shields/windows to protect technologists’ eyes. UV treatment at 254 nm for 10 min is often suitable to eliminate contaminating amplicons, although the nucleic acid length, its percentage of pyrimidine bases, and the distance from the UV source could affect this efficiency. In one study, contaminating DNA was found to be eliminated at a distance of 5 cm, with reduced efficiency at longer distances [47]. Longer amplicons are inherently easier to inactivate than shorter amplicons, with greater success observed at lengths greater than 500 base pairs [48].

Amplification Product Inactivation For laboratories that develop and validate their own molecular diagnostics, additional biochemical “tricks” (often enzymatic) can be applied to reaction mixtures that alter amplicons such that they cannot serve as templates for future reactions. These methods of amplification product inactivation are especially valuable when an assay’s detection step requires open manipulation of amplicons. Moreover, even in closed-tube assays like qPCR, incorporating such techniques can provide an additional degree of protection against product-to-specimen carryover. Many commercially available molecular platforms likewise incorporate some form of


513

Table 1 Comparison of inactivation protocols used in nucleic acid tests Method UV light irradiation

Application Conditions Pre- Lower G + C amplification content, >500 b.p. amplicon

UNG

Pre- Lower G + C amplification content, >100 b.p. amplicon

Photochemical cross-linkers

Post- Lower G + C amplification content, >100 b.p. amplicon

Primer hydrolysis

Post- No specific amplification requirements

Hydroxylamine Post- Higher G + C treatment amplification content, >100 b.p. amplicon

Restriction endonuclease

Pre- No specific amplification requirements

Advantages Inexpensive Simple procedure No changes to protocol Simple procedure

Disadvantages Efficacy varies

Changes necessary in amplification cycling and master mix More expensive than other methods Simple Additional procedure instruments required Can change molecular mass of products which affects electrophoresis No effect on Necessary for reagent addition amplicon post-amplification analysis Efficacy varies Effective on Necessary for reagent addition some post-amplification amplicons Some amplicon changes may affect analysis Hydroxylamine is a mutagenic agent Effective on May lengthen processing time some amplicons

References [16, 18, 19]

[21, 22]

[21, 22, 40]

[21]

[41]

[38]

amplification product inactivation into their protocols. The basic theory and chemistry of these inactivation techniques are discussed here (and summarized in Table 1). Uracil-DNA N-Glycosylase One of the most commonly employed methods for amplicon inactivation involves uracil-DNA N-glycosylase (variably abbreviated as UDG or UNG—the latter here). This bacteria-derived enzyme (including from E. coli and other species) cleaves uracil bases from DNA at the glycosidic bond between the base and the sugar phosphate backbone [49]. While not a natural component of DNA, uracil can be incorporated into amplicons through the presence of dUTP during PCR. The resultant products are susceptible to cleavage by UNG, yielding them unsuitable as future PCR templates. UNG can only modify single- or double-stranded DNA, without any activity on RNA. The biochemical

514


U G

DNA Template C T G C A T A

A A

Amplification

C

U A

A C

G U

U G

G

C A

T U

A

C

G

U

U

G

A

C A U A

X G C A X

A C

G

UNG inactivation

A

C

G U

Contamination U G

X

C

A

A U

A C

G U

A

Fig. 3 Biochemistry of UNG. UNG and dUTP are included in the polymerase chain reaction (PCR) master mix. During amplification, dUTP is included in the amplicon. If contamination of a new PCR reaction occurs with this amplicon, UNG will cleave the apyrimidinic sites, and the amplicon is not suitable as a template for further amplification

steps of this process are summarized in Fig. 3. Overall, UNG inactivation requires two basic alternations to the traditional components of a PCR reaction mixture: (1) s ubstitution of dTTP with dUTP, so that the amplification products are susceptible to UNG, and (2) inclusion of UNG itself, to inactive uracil-labeled DNA from previous amplifications that are contaminating the current reaction. Within an amplicon, not all thymidine residues need to be substituted by uracil for UNG inactivation to be effective, although all amplicons must include substitutions. Accordingly, dUTP may completely or partially replace dTTP in a reaction mixture, although the optimal ratio is not necessarily the same from case to case. Some commercial UNG master mixes completely replace dTTP with dUTP. The nucleotide concentrations can also be varied empirically for a given reaction—along with the thermocycling parameters—to confirm which conditions ensure both efficient amplification and complete inactivation [50]. The ability of UNG to inactivate contaminating amplicons, while not inhibiting the current reaction, stems from the enzyme’s temperature dependence. UNG demonstrates optimal activity at 55 °C and is inactivated at 95 °C. Typical protocols thus include two incubations prior to PCR amplification cycles, first at 55 °C (to inactivate contaminating amplicons with

Molecular Contamination and Amplification Product Inactivation 55 C for 10 minutes

Insert tube in thermalcycler

Primers, Polymerase, dNTPs (including dUTP) Target nucleic acid,etc.

515

95 C for 10 minutes

Proceed with amplification

Fig. 4 Steps used in UNG amplification product inactivation protocol. The master mix has been formulated to contain UNG and dUTP (as opposed to dTTP). The master mix is allowed to incubate for 10 min to allow for UNG activity. Then, the mixture is warmed to 95 C for 10 min to inactivate UNG. The amplification and detection are allowed to proceed as normal

UNG) and then at 95 °C (to inactive UNG itself). Thermocycling can proceed after these steps, although it is still best for annealing temperatures to remain above 55 °C to limit any residual UNG activity [51]. For this same reason, amplicons should be held at either 72 °C or 4 °C after thermocycling is complete [52]. The entire UNG protocol is summarized in Fig. 4. On the order of 1 unit of UNG is included per 50–100 μl reaction volume, although optimal concentrations can vary from reaction to reaction and depending on the manufacturer. Inactivation of up to 3 × 109 copies of contaminating DNA has been observed, although some inhibition of amplification can occur also as UNG concentrations increase [53]. As the success of a UNG strategy depends upon uracil incorporation, AT-rich sequences and longer amplicons (>100 bp) are inherently more susceptible to inactivation [54]. A heat-labile form of UNG is also commercially available, derived from the Atlantic cod [55]. The enzyme undergoes more thorough heat denaturation and is particularly applicable to reverse transcriptase PCR (RT-PCR) reactions, in which RNA is the amplification target. Heat-labile UNG can be inactivated at 50 °C, the temperature commonly utilized for initial cDNA synthesis; some RNA degradation has been reported with this enzyme [50]. For amplicon detection after PCR, UNG protocols typically have no effect on gel mobility, ethidium bromide staining, sequencing, blotting, and other hybridization reactions. However, amplicon melting temperatures may be altered slightly in assays that utilize this step as a means quality control [56]. UNG protocols have been published for the detection of numerous viral, bacterial, fungal, and parasitic pathogens; these include enterovirus [53, 57, 58]; Mycobacterium tuberculosis [59]; Toxoplasma gondii [60]; human herpesvirus 6 (HHV-6) [61]; Histoplasma capsulatum [62]; Bartonella henselae, Bartonella quintana, and Coxiella burnetti [63]; HSV [1]; and cytomegalovirus [64]. In fact, many commercial assays for pathogen detection/ quantification include UNG as part of the platform’s fixed reagents [64, 65]. Other Enzymatic Methods Although UNG methods are the most widespread, several additional enzymatic techniques for amplicon inactivation have been described. One approach involves the use of double-stranded DNase (dsDNase), a

516


shrimp-derived nuclease commercially available in standard and heat-labile versions [66]. With specific activity against dsDNA—and negligible effect against ssDNA or RNA—dsDNase can remove amplicons (or other contaminating DNA) from PCR master mixes. For assays with dsDNA targets, the enzyme must be inactivated prior to adding the template itself (although this is not necessarily the case for RNA targets). In a systematic analysis of PCR contamination, Champlot et al. treated the polymerase, dNTPs, and primers with heat-labile dsDNAse prior to master mix formulation [67]. This step was combined with UV irradiation of other reagents (water, buffers, glycerol, etc.) to yield an efficient decontamination protocol; by contrast, direct UV irradiation of the polymerase/dNTPs/primers adversely affected the efficiency of amplification. Yet another theoretical inactivation strategy involves the use of restriction endonucleases (RE), which cleave DNA palindromes in a sequence-specific manner [68]. As with dsDNase, an appropriate RE can be incubated with the PCR reaction mixture prior to addition of a DNA template, followed by RE inactivation and thermocycling. Potential shortcomings of this approach include the need to troubleshoot which RE (or combination of REs) is effective for a given reaction, as well as longer incubation times [69]. Nonenzymatic Amplicon Inactivation In the earlier days of molecular diagnostics, several nonenzymatic methods for amplicon inactivation were also employed. Although they are no longer commonplace, these techniques are discussed here briefly for the sake of completeness (and if UNG is not possible in a particular scenario). Members of the coumarin family of natural products, small molecules known as psoralens—in particular, isopsoralen and methoxypsoralen—can be incorporated into amplification protocols [70]. They are added to the initial PCR reaction mixture along with the template. While these compounds are bystanders during the thermocycling process, the reaction vessel is exposed to UV light after amplification to activate the psoralens, which create adducts between pyrimidine residues and prevent extension by polymerase in any future PCR reactions [71, 72]. This photochemical method has been shown to inactivate, like UNG, up to 3 × 109 copies of contaminating DNA [52]. Drawbacks include possible alteration of electrophoretic migration [54], the need for a UV box, and the chemical safety risks associated with the compounds. After PCR is complete, the addition of hydroxylamine hydrochloride to the amplicon mixture represents an additional means of chemical inactivation. Hydroxylamine modifies cytosine residues (with effective working concentrations >250 mM), blocking interactions with guanine and rendering the amplicon incompetent as a future template [73]. The inherent flaw of this strategy is that it requires the reaction vessel to be opened when amplicon is present; in itself, this represents an opportunity to create aerosols and contamination. Yet another post-amplification strategy involves targeted primer hydrolysis [52]. Here, amplification primers are synthesized with ribose residues at the 3′ end. PCR products are treated with NaOH, resulting in bond cleavage at the ribose; the amplicons are again made unsuitable for future amplifications with the same primer set. As with hydroxylamine hydrochloride, primer hydrolysis requires the open manipulation of PCR products, which risks spreading amplicons before inactivation.


517

The Role of Data Review Despite the best efforts to prevent carryover, laboratories may still occasionally face scenarios where contamination arises. Although routine negative controls provide a modicum of assurance against massive contamination, carryover can still occur into patients’ specimens with the negative controls unaffected. In these instances, routine data review and statistical analysis by laboratory staff represent a final opportunity to prevent or minimize the clinical impact of a false-positive result [13, 74]. Specifically, the analysis of positivity trends for an assay—either within a run, between runs, or over time—can raise suspicion for contamination. For instance, the only positive detections on a large run could be clustered with respect to their order of extraction or position on a multi-well reaction plate. In the context of a quantitative assay, moreover, one specimen may demonstrate an extremely high load, while adjacent (potentially contaminated) specimens are positive just above the limit of detection. And for multiplex assays, certain targets can be encountered with greater frequency/seasonality; if an infrequently detected pathogen is abruptly positive in multiple specimens on the same day, a red flag is again raised. Unfortunately, signs of molecular contamination might not be so egregious. Increases in positivity rates can be subtle, with trends only becoming apparent over weeks or months. Likewise, attributing these changes to specimen contamination is not always straightforward, as trends could reflect drift in the assay’s performance characteristics (also problematic, but for other reasons). Alternately, a test’s performance could be completely normal, with changes in positivity resulting from stochastic fluctuations or evolving pathogen epidemiology. Another significant challenge is determining precisely when data review will occur and who should conduct it. Ideally, some component of review would take place before results are communicated to physicians and finalized in the patient’s medical record. Review by more experienced personnel (e.g., supervisors, pathology house staff, or medical directors) is also advantageous. Of course, all clinical laboratories face logistical constraints, and what is ideal is not always practicable. The desire to conduct a more extensive data review prior to releasing results must be weighed against the resultant increases in turnaround time. Ultimately, in the absence of specific guidelines, it is up to an individual laboratory to determine how and when to incorporate trend analysis into a quality assurance program. The thresholds for what constitutes an abnormal trend must be gauged in light of the particular assay and the patient population being served. Regardless of how these red flags are chosen, it is important that they are explicitly defined in the laboratory’s procedures and communicated to staff. A data trend that is blatantly concerning to a medical director or senior technologist might not automatically raise the suspicion of less experienced staff. The response to a suspicious data trend must also be well defined and proceduralized, typically involving notification of the laboratory’s medical and operational leadership. These individuals can then determine an appropriate action in light of the specific testing scenario, correlation with the patients’ clinical history, and other laboratory findings. Often, retesting the specimens in question is performed from the initial extraction step. If inconsis-

518


tent results support a contamination event, appropriate follow-up could involve a more extensive review of recent data, additional environmental sampling/cleaning of surfaces, and (if necessary) referral of specimens to an outside laboratory until the issue is resolved. For commercial platforms, the manufacturer can serve as a source of troubleshooting and assistance. Most importantly, if these actions entail additional delays in reporting—and, especially, if a false-positive result was published in a patient’s record—the managing clinicians must be notified to mitigate potential harm.

Special Considerations Molecular diagnostics in microbiology are evolving rapidly, and certain testing scenarios are becoming increasingly commonplace. Several of these situations merit specific discussion with regard to contamination and false positivity. Multiplex Molecular Diagnostics To begin, one of the most prominent innovations in molecular diagnostics has been the advent of multiplex platforms that allow for a syndromic approach to pathogen testing. A patient with diarrhea, for instance, can simultaneously undergo testing for numerous bacterial, viral, and parasitic agents of gastroenteritis. Commercially available platforms currently include panels for respiratory, gastrointestinal, and central nervous system pathogens [75–77], and the number of targets will assuredly increase in coming years. Although the same theoretical principles of carryover-prevention apply, syndromic infection testing can lead to several additional complexities. Due to multiplexing, explicit analyses for less common pathogens are included more frequently than ever before, facilitating diagnoses that previously went unmade. At the same time, a false-positive result can hold additional implications if it occurs for a target that would have otherwise gone untested. Fundamentally, the predictive value of any positive result depends on a pathogen’s prevalence and the assay’s pretest probability [78]. For instance, a physician may order a stool pathogen panel out of concern for viral gastroenteritis. Despite no clinical suspicion for a parasitic infection, several protozoal targets are automatically included on the panel. If molecular contamination with Cryptosporidium were to occur, the unexpected positive result could lead to both mismanagement of the patient and an unnecessary public health investigation (as Cryptosporidium is reportable in many locations). The downstream effects of contamination for an uncommon target can thus become magnified. An additional issue pertains to the ability of a multiplex panel to replace several monoplex assays (either molecular or traditional) within a laboratory. The analytic merits of this practice are debated [79], and it leads to the proverbial p lacement of many eggs in one basket. On a logistical level, when a greater portion of a laboratory’s diagnostic enterprise rests on a signal instrument, the impact of unanticipated downtime is magnified. Nucleic acid contamination is exactly the sort of event that can temporarily force a molecular assay off-line at an institution. For


519

multiplex platforms, moreover, the reportability of one pathogen becomes interconnected with the validity of the other targets. The net result, for instance, is that contamination of an instrument or workspace with rhinovirus amplicon could theoretically impact the laboratory’s ability to test for influenza. Perhaps not surprisingly, the closed-system nature of certain multiplex panels—where specimens do not have to be manipulated between the amplification and detection steps—has been highlighted as a notable strength [80]. Quantitative Molecular Testing Along with multiplexing, an additional trend in infectious disease diagnostics is the qPCR-based quantification of pathogen levels in the blood and, occasionally, other body fluids. Monitoring viral loads for human immunodeficiency virus (HIV) is now well established as a necessary component of care [81], and quantitative assays are in widespread use for cytomegalovirus (CMV), Epstein-Barr virus (EBV), BK virus (BKV), hepatitis B virus (HBV), and hepatitis C virus (BKV) [reviewed in 82]. For some of these tests (in particular, HIV, HBV, and HCV), a false-positive result may carry additional weight, depending on why the test was originally ordered. To explain, it is important to note that current IVD-marked platforms for quantifying viral loads only carry explicit FDA approval (in the United States) for monitoring patients in whom the diagnosis is already established. They are not indicated for screening asymptomatic individuals or diagnosing suspected infections, either by themselves or as part of multistep algorithms [83]. At first glance, this distinction may seem like a regulatory technicality, as these assays possess extremely high analytic sensitivity. Indeed, many clinicians still order viral loads (intentionally or inadvertently) for the purposes of screening/diagnosis. Moreover, given the paucity of commercial platforms that are explicitly approved for diagnosis of these agents, it is not uncommon for an institution’s only in-house molecular test to be quantitative. A more subtle concern can emerge, however, when one considers the theoretical possibility of specimen-tospecimen contamination. The majority of specimens for these tests will come from patients with known infections, where the test is being used for the intended purpose. As a result, a substantial burden of positive specimens (including high loads) can be present within the laboratory workspace. In itself, this raises the theoretical possibility of carryover. The downstream consequences of such an event can be more severe, however, if it happens to occur for someone who is not undergoing monitoring. It is illustrative to consider the following two hypothetical individuals: (1) a known HIV-positive man whose recent viral loads have been completely suppressed by anti-retroviral therapy and (2) an HIV-negative pregnant woman undergoing routine first-trimester screening, for whom the initial HIV antigen-antibody screening assay was falsely positive (not uncommon), with the confirmatory antibody differentiation assay negative. Per current fourth-generation CDC guidelines, an HIV NAAT is the appropriate follow-up diagnostic to clarify the results in the latter individual [83]. In both scenarios, a viral load is ordered, and slight carryover occurs from a highpositive specimen, such that the correct result of “no target detected” instead becomes “detected,