Figure 2 illustrates an ethidium bromide stained agarose gel of the 60 bp products generated from the template calibration series (range: 1 molecule to.
TCA-020
Nucleic Acid Quantification by Chemiluminescence Assay of Polymerase Chain Reaction Products Jeremy Garson and Kevin Whitby, Division of Virology, University College London Medical School, London, U.K.
Abstract In addition to its use in assays based on isotopic labels, the TopCountTM Microplate Scintillation and Luminescence Counter may be employed as an automated luminometer for luminescence assays in the 96-well format. In this report the TopCount in single photon counting (SPC) mode is used to quantify viral nucleic acid following in vitro amplification, hybridization and signal generation by means of a glow-type luminescent chemistry (xanthine oxidase/luminol system). Advantages of the chemiluminescence assay run on the TopCount include single molecule detection capacity, quantification over a 4-5 log10 dynamic range, enhanced throughput, and the convenience and safety of non-isotopic detection.
Introduction The polymerase chain reaction (PCR) is now widely employed for the amplification, detection and quantification of nucleic acids in many areas of biomedical science. In virology, for example, PCR is rapidly becoming the method of choice for the detection of viruses that are difficult to identify by “classical” techniques such as immunoassay and tissue culture. DNA genomes can be amplified directly following viral nucleic acid extraction, whereas RNA viruses require an in vitro reverse transcription step prior to PCR. Viral load and response to anti-viral therapy can be estimated by quantitative PCR techniques, but these are generally very labor intensive and therefore unsuitable for handling large throughputs. Presented here are the results of a quantitative PCR assay for viral nucleic acid which employs a standard 96-well microplate format, making it suitable for automation and therefore capable of processing large numbers of samples. The very wide dynamic range
PAN0056 2/94 Printed in U.S.A.
achieved by use of a chemiluminescent substrate, in conjunction with the TopCount in single photon counting (SPC), is comparable to that generated by isotopic methods, but without the associated handling hazards and disposal problems. Although the following protocol describes quantification of hepatitis C viral RNA (HCV-RNA), the principles of the method are of course applicable to the quantification of nucleic acid of any origin.
Methods RNA Extraction, cDNA Synthesis and PCR HCV-RNA was extracted from a 200 µL aliquot of serum by the guanidinium isothiocyanate method of Chomczynski and Sacchi.1 cDNA was synthesized from the extracted viral RNA using random hexamer primers and recombinant Moloney murine leukemia virus reverse transcriptase (Promega). The PCR was performed in a 50 µL reaction using 5 µL of the cDNA template, four units of exo- Pfu DNA polymerase (Stratagene®) and 10 ng each of a primer pair designed to amplify a 60 base pair (bp) segment of the highly conserved 5' non-coding region of the HCV genome. The antisense primer was biotinylated at the 5' end. Optimum results were obtained with 35 cycles (95 oC/1 minute, 53 oC/30 seconds, 75 oC/2 seconds, final seven minute extension) using a Techne® PHC3 automated thermal cycler. Capture and Detection of PCR Products The capture and detection strategy is represented diagrammatically in Figure 1. Biotinylated PCR products (5 µL) were captured on recombinant streptavidin coated (Sigma, 2.5 mg/mL), black 96-well microplates (Nunc® Maxisorb®). NaOH (0.15 M) was applied for five minutes to denature the double-stranded product. Following neutralization and washing (Tris pH7.6) to remove the second strand, the immobilized single-stranded product was
© 1994 Packard Instrument Company
Packard Instrument Company 800 Research Parkway, Meriden, CT 06450 Tel: 1-800-323-1891 203-238-2351 TX: 643251 FAX: 203-639-2172
Figure 1. Diagrammatic representation of PCR product capture and detection strategy.
hybridized with a dinitrophenol labeled (DNP at 5' terminal) 20 mer oligonucleotide probe. Hybridization was for ten minutes at 55 oC (Tm -5 oC) in 1/3 strength phosphate buffered saline (PBS), 0.1% Tween 20 and 1% bovine serum albumen (BSA). Following washing (x 5) to remove excess probe, the bound DNP was detected by sequential incubations with monoclonal anti-DNP (diluted 1 in 500) and xanthine oxidase conjugated rabbit anti-mouse IgG (diluted 1 in 500). The chemiluminescent signal was generated using the luminol/hypoxanthine/Fe3+/ EDTA substrate described in Baret and Fert.2 The count time on TopCount in SPC mode at 20 oC was six seconds per well. Luminescence is counted two wells at a time, so a 96-well plate is counted in five minutes. Using this substrate, the light generated can be measured for up to 24 hours after initiation of the luminescence reaction, and correction for signal half life is unnecessary.
Poisson distribution method. A ten-fold dilution series covering the 1 molecule - 105 molecule range was prepared from this product and used to produce standard calibration curves. The number of cDNA molecules in any given test sample was calculated with reference to the standard calibration curve generated in the same experiment. A conversion factor based on the known efficiency of the reverse transcription reaction was then used to derive the HCVRNA titer of the test serum.
Results Black microplates were found to give better signal to noise ratios than white plates. S:N ratios >2:1 were obtained with as little as 1.4 pg of “mock product.” Figure 2 illustrates an ethidium bromide stained agarose gel of the 60 bp products generated from the template calibration series (range: 1 molecule to 105 molecules) by the optimized PCR protocol. The specificity of the amplification is confirmed by the presence of bands of the expected size (the extreme left hand lane contains the molecular weight marker; phi X 174/HaeIII digest). Figure 3 shows the chemiluminescence signals generated from the standard template dilution series. The signal to noise ratio generated from one molecule of template DNA is at least 2:1, and the signal increases over an input range of 4-5 logs10. This dynamic range is greater than that reported for most alternative systems.
Preparation of Templates for Standard Calibration Curves For initial pilot experiments designed to permit optimization of the detection protocol, a “mock product” was employed. This consisted of a single stranded random sequence 30 mer oligonucleotide labeled at the 5' end with biotin and at the 3' end with DNP. For subsequent experiments a 340 bp PCR product encompassing almost the entire 5' noncoding region of HCV was generated. The titer of this product was estimated by the limit dilution/
2
Number of Molecules Molecular Weight Markers
105
104
103
102
101
1
Figure 2. Ethidium bromide stained agarose gel showing 60 bp products generated from ten-fold dilution series of DNA template.
Conclusions PCR. 3-7Throughput is enhanced by the use of stackers which hold over 20 microplates. The luminescent labeling system employed retains the sensitivity and dynamic range advantages of isotopic systems without the associated problems of safety, handling, and disposal.
Using the method described above, it is possible to measure viral titers with a sensitivity of approximately 500 genomes/mL serum (equivalent to one molecule of cDNA), over a dynamic range of at least four logs10. Using TopCount greatly increases sample throughput when compared with methods such as end point serial dilution or competitive
Figure 3. Quantification of template DNA by xanthine oxidase generated chemiluminescence. Typical calibration curve.
3
Acknowledgments Labeled oligonucleotides and anti-DNP reagents were kindly provided by Drs. T. Brown and J. Grzybowski (Oswel DNA Service, University of Edinburgh). We would like to thank Dr. A. Baret for his helpful advice and for making xanthine oxidase conjugates and chemiluminescence substrates available. Financial support for this study was provided by Murex Diagnostics Ltd.
References 1. Chomczynski P., Sacchi N. (1987). “Single-Step
Method of RNA Isolation by Acid Guanidinium Thiocyanate-Phenol-Chloroform Extraction.” Anal. Biochem. 162: 156-159. 2. Baret A, Fert V (1990). “Application of a Long-
Term Enhanced Xanthine Oxidase-Induced Luminescence in Solid-Phase Immunoassays.” Anal. Biochem. 187: 20-26. 3. Simmonds P et al. (1990). “Hepatitis C
Quantification and Sequencing in Blood Products Hemophiliacs and Drug Users.” Lancet 336: 1469-1472. 4. Brillanti S et al. (1991). “Effect of Alpha Interferon
Therapy on Hepatitis C Viremia in CommunityAcquired Chronic Non-A Non B Hepatitis A Quantitative Polymerase Chain Reaction Study.” J. Med. Virol. 34: 136-141. 5. Garson JA et al. (1992). “Hepatitis C Viraemia
Rebound After Successful Interferon Therapy in Patients with Chronic Non-A Non-B Hepatitis.” J. Med. Virol. 37: 210-214. 6. Kaneko S et al. (1992). “Quantification of Hepatitis
C Virus RNA by Competitive Polymerase Chain Reaction.” J. Med. Virol. 37: 278-282. 7. Hagiwara H et al. (1993) Gastroenterology 104: 877-883.
Strategene is a trademark of Strategene, Inc. Techne is a trademark of Techne, Inc. Nunc and Maxisorb are trademarks of Nunc, Inc.
4
