High prevalence of Pneumocystis jirovecii ...

Medical Mycology July 2012, 50, 556–560

High prevalence of Pneumocystis jirovecii colonization in Brazilian cystic fibrosis patients MARCO A. PEDERIVA*, GUSTAVO WISSMANN*,VICENTE FRIAZA†, RUBEM MORILLA†, CARMEN DE LA HORRA†, MARCO A. MONTES-CANO†, LUCIANO Z. GOLDANI*, ENRIQUE J. CALDERÓN† & JOÃO C. PROLLA* *Grupo de Estudos em Pneumocystis, Serviço de Infectologia, Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil, and †Instituto de Biomedicina de Sevilla, Hospital Universitário Virgen del Rocío/CSIC/ Universidad de Sevilla, and CIBER de Epidemiología y Salud Pública, Sevilla, Spain

Keywords:

Pneumocystis jirovecii, cystic fibrosis, colonization

Introduction Cystic fibrosis (CF) is the most common life-shortening autosomal recessive disorder in Caucasian populations. A recent study found that CF occurs approximately once in every 1,587 live births in southern Brazil [1]. Understanding the microbial flora of the CF respiratory tract is of considerable importance, as the morbidity and death of CF patients are primarily caused by chronic respiratory infections. However, chronically colonized airways of these patients represent a surprisingly complex and diverse ecosystem [2]. Pneumocystis jirovecii is an atypical opportunistic fungus with apparent lung tropism and worldwide distribution

that causes pneumonia in immunosuppressed persons. Pneumocystis colonization has recently been described in subjects with various lung diseases, and accumulating evidence suggests that this colonization may be an important clinical phenomenon [3]. Only a few European studies have evaluated the prevalence of Pneumocystis colonization in patients with CF, reporting ranges from 1.3–21.5% [4–6] and there is no information about Pneumocystis colonization in CF patients outside of Europe. In this investigation, we examined the prevalence of P. jirovecii colonization and the distribution of the gene(s) encoding the mitochondrial large subunit ribosomal RNA (mtLSUrDNA) genotypes of P. jirovecii among CF patients in Brazil.

Materials and methods Received 18 August 2011; Received in final revised form 1 November 2011; Accepted 29 November 2011 Correspondence: Gustavo Wissmann, Hospital de Clínicas de Porto Alegre, Rua Ramiro 2350, 90035-903, Porto Alegre, Brazil. Tel.: ⫹55 51 3359 8000; Fax: ⫹55 51 3330 9700; E-mail: [email protected]

© 2012 ISHAM

Patients This study included 34 CF patients who were admitted consecutively between March 2006 and August 2009 to the bronchoscopy unit for evaluation of their disease at DOI: 10.3109/13693786.2011.645892

Downloaded from http://mmy.oxfordjournals.org/ by guest on November 4, 2015

A high rate of Pneumocystis jirovecii colonization was observed in Brazilian cystic fibrosis (CF) patients (13 out of 34; 38.2%) who underwent bronchoscopy between March 2006 and August 2009 at the Hospital de Clinicas de Porto Alegre, Brazil. Bronchoalveolar lavage samples were collected from these patients and studied by nested PCR amplification of the mitochondrial gene coding for the large subunit ribosomal RNA (mtLSUrDNA). The observed rate of colonization was higher than that reported in European populations. Genotypic characterization of the mtLSUrDNA locus revealed a predominance of the polymorphisms 85C/248C (genotype 1) and 85T/248C (genotype 3), with all samples possessing the wild-type genotype of dihydropteroate synthase. These findings suggest that cystic fibrosis patients could be an important reservoir and source of P. jirovecii infection. Further studies are required to elucidate the role of this common fungal colonization in the evolution of CF patients.

Pneumocystis jirovecii in Brazilian cystic fibrosis patients

Detection of P. jirovecii Pneumocystis jirovecii detection was carried out by analyzing BAL samples with nested PCR amplification of the mtLSUrDNA gene of Pneumocystis. After samples were digested with proteinase K at 56°C, DNA was extracted from the BAL samples using a commercial kit (QIAamp DNA mini kit; Qiagen, Hilden, Germany) and the gene encoding the mtLSUrRNA was amplified. A two-step protocol was used for nested PCR which included in the first amplification round, the use of the external primers pAZ102-E (5¢-GAT GGC TGT TTC CAA GCC CA-3¢) and pAZ102-H (5¢-GTG TAC GTT GCA AAG TAC TC-3¢) to obtain a 346 bp fragment. The second round of amplification utilized the primers pAZ102-X (5¢-GTG AAA TAC AAA TCG GAC TAG G-3¢) and pAZ102-Y (5¢-TCA CTT AAT ATT AAT TGG GGA GC-3¢) and yielded a 260 bp product. Both rounds included 40 cycles of amplification. The PCR products were analyzed by electrophoresis on a 1.5% agarose gel containing ethidium bromide, and the bands were visualized by UV light. To prevent false positives due to contamination, pipette tips with filters were used at all stages. DNA extraction, preparation of the reaction mixture, PCR amplification, and detection were performed in different areas of the laboratory. In addition, a positive control was included in each reaction. To detect any cross-contamination, all PCR steps were performed with a negative control of sterile water. All experiments were repeated at least twice [8]. P. jirovecii mtLSUrDNA genotype characterization All samples that were positive according to nested PCR were sequenced; polymorphisms at the nucleotide positions 85 © 2012 ISHAM, Medical Mycology, 50, 556–560

and 248 were detected by direct sequencing [9]. The nested PCR products were purified using Sephacryl S-400 columns (Amersham Pharmacia Biotech AB, Uppsala, Sweden) and were reamplified using the ABI Prism dRhodamine Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems, Foster City, CA, USA). Each reaction included 5μl of PCR product, 4 μl of terminator ready reaction mix, and 3 pmol/l of primer. The extension products were purified by ethanol precipitation to remove residual dye terminators. Each sample pellet was resuspended in 12.5 μl of template suppression reagent and heated at 95°C for 3 min to denature the product. Electrophoresis was carried out on the ABI prism 310 sequencer (PE Applied Biosystems) in accord with the manufacturer’s recommendations. The sequenced DNA fragments were analyzed using Sequence Navigator version 1.0.1 (PE Applied Biosystems) [8]. P. jirovecii dihydropteroate synthase (DHPS) polymorphism Samples that were identified as positive by nested PCR of the mtLSUrRNA gene were also further examined by PCR restriction fragment length polymorphism (RFLP) analysis, as previously described [8]. Briefly, the single-copy DHPS gene was amplified with a touchdown-PCR protocol using the primers DHPS-3 (5¢-GCG CCT ACA CAT ATT ATG GCC ATT TTA AAT C-3¢) and DHPS-4 (5¢-GGA ACT TTC AAC TTG GCA ACC AC-3¢), yielding a 370-bp fragment. The PCR product was divided into three aliquots. One was used to confirm the presence of a 370-bp fragment from the DHPS gene. The other two aliquots were used to identify mutations in codons 55 and 57 by RFLP with the restriction enzymes AccI and HaeIII (Roche Diagnostics, Mannheim, Germany), respectively. Statistical analysis Statistical analysis was performed using SPSS software, version 15.0 (SPSS, Chicago, IL, USA). Student’s t-test, the Chi-square test with Yates’ correction, Fisher’s exact test or the Mann-Whitney U-test were used to compare the characteristics of colonized and non-colonized cases, and P ⬍ 0.05 was regarded as statistically significant. The variance test was used to confirm a normal distribution.

Results The study included 34 patients (17 males and 17 females; median age 11.0 years, range 1–35 years) from whom BAL samples were collected. Seven patients (20.5%) had a diagnosis of diabetes mellitus, and 27 patients (79.4%) suffered from pancreatic insufficiency. Moderate or severe malnutrition was found in 10 patients (29.4%). Eight patients (23.5%) had had three or more CF-related hospitalizations


Hospital de Clínicas de Porto Alegre, southern Brazil. In all cases, informed consent was obtained from the patients or their guardians. The study was approved by the hospital’s ethics committee. All patients underwent clinical, functional, and radiological evaluations. For each patient, epidemiological information was obtained by review of medical charts and included age, sex, height, weight, bacterial colonization in the last three months, pancreatic enzyme use, diabetes mellitus diagnosis, percentage from predicted forced expiratory volume in one second (FEV1-%), and the number of hospital admissions over the previous year. We noted any prior treatment with sulfa-drugs or azithromycin, which was defined as the prescription of any of these agents, regardless of duration of use, during the six months before the bronchoscopy study. The weight-for-age Z-score was used to identify cases of moderate to severe malnutrition [7]. We also performed diagnostic bronchoscopy with bronchoalveolar lavage (BAL) fluid examination.

557

558

Pederiva et al.

Discussion The results of our study reveal, for the first time, the high prevalence of P. jiroveci colonization in Brazilian patients

Prevalence of mtLSUrDNA genotypes

45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Genotype 1 (85C/248C)

Genotype 2 (85A/248C)

Genotype 3 mixed genotypes (85T/248C) 1 and 3

Fig. 1. Rates of genotypes at the mtLSUrDNA Pneumocystis jirovecii locus from 34 Brazilian cystic fibrosis patients.

with CF and show an association between P. jiroveci colonization and P. aeruginosa infection in patients with this disease, suggesting a possible interaction between both pathogens, which could be of important medical significance in CF progression. In our study, we found a prevalence of P. jirovecii colonization of 38.2% among CF-patients, whereas studies in Spain, Germany and France showed rates of 21.5%, 7.4% and 1.3%, respectively [4–6]. Prevalence of P. jirovecii colonization among CF patients is significantly higher in Brazil than in Germany (7.4% vs. 38.2%; P ⫽ 0.0001) or France (1.3% vs. 38.2%; P ⬍ 0.00001), but similar to prevalence among Spanish patients (21.6% vs. 38.2%; P ⫽ 0.1). There are several factors that could explain the high rate of P. jirovecii colonization observed in Brazilian CF patients. The incidence of acquired immunodeficiency syndrome (AIDS)-related P. jirovecii pneumonia (PcP) in a particular geographic region could contribute to the rate of colonization by this fungus in CF patients [6]. Currently there is evidence that the transmission of P. jirovecii occurs by airborne route from person to person [10] and that encounters between infected and non-infected persons could be more likely in areas with high incidence of AIDSrelated PcP [11]. Between 1980 and June 2010, 592,914 AIDS cases were reported in Brazil [Epidemiological Bulletin, Ministry of Health of Brazil, 2010]. Moreover, in Brazil, PcP was reported to be the second most common AIDS-defining condition after the introduction of combined antiretroviral therapy [12]. Several authors have shown seasonal changes in PcP incidence that seem to be associated with climate factors [13–16]. While previous studies showed PcP incidence to be maximal during the winter months [13–15], a recent study revealed the opposite pattern, with PcP incidence being maximal in the summer months [16]. Southern Brazil is a subtropical region, with an average yearly temperature of 18°C (similar to Seville, 18.6°C), and both areas had a high prevalence of P. jirovecii colonization suggesting that this situation could be more frequent in countries with warmer climates. This idea is consistent with results from a previous study that showed a higher rate of colonization in chronic obstructive pulmonary disease (COPD) patients from Sevilla (southern Spain) than from Amiens (northern France) [17]. Further studies are needed to define the relationship between climate and the prevalence of colonization by P. jirovecii. Differences in colonization rates might also be related to the different types of clinical specimens that were analyzed. While previous authors described results from analyses of sputum or oropharyngeal washes (OW), the presented study used BAL samples. BAL samples provide the highest sensitivity for the diagnosis of PcP by © 2012 ISHAM, Medical Mycology, 50, 556–560


in the previous year. Five patients (14.7%) had received azithromycin or co-trimoxazole in the last six months, and bacterial colonization was identified in 29 patients (85.2%) during the previous three months. The patients enrolled in the study had a mean of FEV1-% of 69.3 ⫾ 15.9%. P. jirovecii colonization was detected in 13 out of 34 CF patients (38.2%; CI 95%: 22–56%). Standard cytological staining was negative in all cases and none of the CF patients who tested positive for P. jirovecii DNA suffered from overt Pneumocystis pneumonia. Twelve of the positive samples yielded typing results for positions 85 and 248 of the mtLSUrDNA gene. Genotype 1 (85C/248C) of the mtLSUrDNA gene was observed in five cases (41.6%; CI 95%: 15–72%), genotype 2 (85A/248C) in two patients (16.6%; CI 95%: 2–48%), genotype 3 (85T/248C) in three cases (25.0%; CI 95%: 5–57%), and mixed genotype 1 and 3 in two cases (16.6%; CI 95%: 2–48%) (Fig. 1). The wild-type DHPS genotype (threonine55/proline57) was detected in all samples. The clinical and demographic data of the patients included in the study are showed in Table 1 according to P. jirovecii colonization status. Patients with CF who were colonized by P. jirovecii had a higher frequency of P. aeruginosa infection than did non-colonized subjects, and this difference was statistically significant (P ⫽ 0.02). None of the other prognostic indicators, e.g., number of CF-related hospitalizations and nutritional characteristics, was associated with P. jirovecii colonization in this study. Patients with poor nutritional status had a higher prevalence of colonization, but this difference was not statistically significant.

Pneumocystis jirovecii in Brazilian cystic fibrosis patients

559

Table 1 Demographic and clinical data from 34 Brazilian cystic fibrosis patients according to Pneumocystis jirovecii colonization.

Characteristic

Patients without P. jirovecii colonization (n ⫽ 21)

P-value

12 (1–35) 5 (38.7) 12 (92.3) 12 (92.3) 11 (84.6) 1 (7.6) 1 (7.6)

8 (1–22) 12 (57.1) 17 (80.9) 11(52.4) 16 (76.2) 3 (14.2) 3 (14.2)

0.12a 0.29b 0.63c 0.02c* 0.68c 0.99c 0.99c

2 (15.4)

2 (9.5)

0.63c

1 (7.7) 70.9 ⫾ 16.1 2 (15.4) 10 (76.9) 5 (38,4) 4 (30.7)

4 (19.0) 67.9 ⫾ 16.3 5 (23.8) 17 (81.0) 5 (23.8) 4 (19.1)

0.35c 0.78d 0.44c 0.55c 0.45c 0.68c

aMann-Whitney

U-test; bChi-square test with Yates’ correction; cFisher’s exact test; dStudent’s t-test; eColonization in the last three months; fUse in the last six months; gPercentage from predicted forced expiratory volume in 1 s; hweight-for-age Z-score; ihospitalizations in the last year; *statistically significant.

molecular methods [18]. Therefore, BAL specimens probably have the greatest capability to detect P. jirovecii colonization using PCR. It has been reported that the patient’s age may influence colonization by P. jirovecii in CF patients. Patients aged less than 18 years had a higher rate of colonization than patients over this age, as reported by Respaldiza et al. [4]. The median age of the cases studied was lower in Brazilian patients (11.0 y, this study) than in the Spanish (15.8 y), German (18.5 y) and French (23.2 y) studies. It is known that the spectrum of pathogens in CF varies with the age of the patient. In children, the most frequent bacterial colonization is by pathogens such as Staphylococcus aureus and Haemophilus influenzae. Later, there is frequent colonization by Pseudomonas aeruginosa [2]. The findings of our study might indicate that colonization by the fungus P. jirovecii is more frequent in younger CF patients. However, further studies are needed to clarify this issue. In addition, variations in the rates of colonization could be related to differences in prior exposure to co-trimoxazole or azithromycin to prevent bacterial infections among the CF patients studied [11]. Among the Brazilian patients in this study, only 14.7% (5/34) had used these drugs in the six months prior to bronchoscopy. The high rate of P. jirovecii colonization could also be an outcome of the fact that most patients did not use anti-Pneumocystis drugs. This study also presented the results of genotyping of two DNA regions of the fungus that are commonly used in molecular epidemiology studies of P. jirovecii. The mtLSUrDNA genotype distribution among Brazilian CF patients was similar to that described in a population of Spanish © 2012 ISHAM, Medical Mycology, 50, 556–560

children with CF [19]; polymorphisms 85C⁄248C (genotype 1) and 85T/248C (genotype 3) were the most frequent. A recent report suggested that underlying disease may determine the presence of a specific mtLSUrDNA genotype of P. jirovecii, supporting the hypothesis that the polymorphism 85T/248C might be the best adapted to patients with CF [20]. The absence of mutations associated with sulfa resistance in the P. jirovecii DHPS gene from Brazilian CF patients is consistent with the same prior observation in Brazilian AIDS patients and is probably related to the low use of sulfa drugs in this population [21]. Recent studies have proposed that P. jirovecii colonization may play a role in the pathophysiology of COPD by inducing inflammatory changes [22]. Moreover, there is an association between P. jirovecii colonization and the severity of airflow obstruction in smokers [23]. In this series of 34 CF patients, P. jirovecii colonization was statistically associated with P. aeruginosa colonization, a respiratory infection that is a major predictor of morbidity and mortality in CF [24]. None of the other prognostic indicators, e.g., number of CF-related hospitalizations and nutritional characteristics, was associated with P. jirovecii colonization in this study. Further studies should be conducted with larger numbers of patients to verify and further explore this finding. In an animal model, molecular and histological analyses have demonstrated that immunocompetent mice that are colonized by Pneumocystis act as a reservoir and source of infection [25]. Additionally, there is molecular evidence that the transmission of P. jirovecii from colonized immunocompetent carrier hosts to susceptible subjects may occur in humans [26]. It has been speculated that CF


Age, median (range) Male sex, no. (%) Bacterial colonization, no. (%)e Pseudomonas aeruginosa colonization, no. (%)e Staphylococcus aureus colonization, no. (%)e Haemophilus influenzae colonization, no. (%)e Burkholderia cepacia Complex colonization, no. (%)e Stenotrophomonas maltophilia colonization, no. (%)e Sulfa or azithromicin use, no. (%)f FEV-1%g (mean ⫾ SD) Diabetes mellitus, no. (%) Pancreatic enzyme use, no. (%) Moderate to severe malnutrition, no. (%)h Three or more CF-related number of hospitalizations/year, no. (%)i

Patients with P. jirovecii colonization (n ⫽ 13)

560

Pederiva et al.

patients colonized by P. jirovecii can be significant reservoirs of this atypical fungus, serving as infective sources of this microorganism to susceptible persons. In summary, this study revealed a high prevalence (38.2%) of P. jirovecii colonization in Brazilian CF patients. Together with previous data from European populations, our data suggest that P. jirovecii colonization in CF is a frequent, possibly worldwide, occurrence. Because these patients could potentially be sources of infection for other susceptible persons, further research is warranted to assess the true scope of the problem and to design rational preventive strategies. Moreover, it is necessary to elucidate the role of P. jirovecii infection in the natural history of CF to improve the clinical management of this disease.

Acknowledgements

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

References 1 Raskin S, Pereira-Ferrari L, Reis FC, et al. Incidence of cystic fibrosis in five different states of Brazil as determined by screening of p. F508del, mutation at the CFTR gene in newborns and patients. J Cyst Fibros 2008; 7: 15–22. 2 Harrison F. Microbial ecology of the cystic fibrosis lung. Microbiology 2007; 153: 917–923. 3 Morris A, Wei K, Afshar K, Huang L. Epidemiology and clinical significance of Pneumocystis colonization. J Infect Dis 2008; 197: 10–17. 4 Sing A, Geiger AM, Hogardt M, Heeseman J. Pneumocystis carinii carriage among cystic fibrosis patients, as detected by nested PCR. J Clin Microbiol 2001; 39: 2717–2718. 5 Respaldiza N, Montes-Cano MA, Dapena FJ, et al. Prevalence of colonization and genotypic characterisation of Pneumocystis jirovecii among cystic fibrosis patients in Spain. Clin Microbiol Infect 2005; 11: 1012–1015. 6 Gal SL, Héry-Arnaud G, Ramel S, et al. Pneumocystis jirovecii and cystic fibrosis in France. Scand J Infect Dis 2010; 42: 225–227. 7 Lai HC, Kosorok MR, Sondel SA, et al. Growth status in children with cystic fibrosis based on the National Cystic Fibrosis Patient Registry data: evaluation of various criteria used to identify malnutrition. J Pediatr 1998; 132: 478–485.

This paper was first published online on Early Online on 3 January 2012.

© 2012 ISHAM, Medical Mycology, 50, 556–560


This study was partially supported by the Spanish Ministry of Science and Innovation (FISPS09/00957) and Brazilian Ministry of Science and Technology (CNPq – Conselho Nacional de Desenvolvimento Científico e Tecnológico, 200739/2007-7).

8 Montes-Cano MA, de la Horra C, Martin-Juan J, et al. Pneumocystis jiroveci genotypes in the Spanish population. Clin Infect Dis 2004; 39: 123–128. 9 Beard CB, Roux P, Nevez G, et al. Strain typing methods and molecular epidemiology of Pneumocystis pneumonia. Emerg Infect Dis 2004; 10: 1729–1735. 10 Beard CB, Carter JL, Keely SP, et al. Genetic variation in Pneumocystis carinii isolates from different geographic regions: implications for transmission. Emerg Infect Dis 2000; 6: 265–272. 11 Calderon EJ, Friaza V, Dapena FJ, et al. Pneumocystis jirovecii and cystic fibrosis. Med Mycol 2010; 48: S17–21. 12 Marins JR, Jamal LF, Chen SY, et al. Dramatic improvement in survival among adult Brazilian AIDS patients. AIDS 2003; 17: 1675–1682. 13 Calderon EJ, Varela JM, Medrano FJ, et al. Epidemiology of Pneumocystis carinii pneumonia in southern Spain. Clin Microbiol Infect 2004; 10: 673–676. 14 Varela JM, Regordán C, Medrano FJ, et al. Climatic factors and Pneumocystis jirovecii infections in southern Spain. Clin Microbiol Infect; 2004; 10: 770–772. 15 Lubis N, Baylis D, Short A, et al. Prospective cohort study showing changes in the monthly incidence of Pneumocystis carinii pneumonia. Postgrad Med 2003; 79: 164–166. 16 Sing A, Schmoldt S, Laubender RP, et al. Seasonal variation of Pneumocystis jirovecii infections: analysis of underlying climatic factors. Clin Microbiol Infect 2009; 15: 957–960. 17 Nevez G, Magois E, Duwat H, et al. Apparent absence of Pneumocystis jirovecii in healthy subjects. Clin Infect Dis 2006; 42: e99–101. 18 Durand-Joly I, Chabe M, Soula F, et al. Molecular diagnosis of Pneumocystis pneumonia. FEMS Immunol Med Microbiol 2005; 45: 405–410. 19 Montes-Cano MA, de la Horra C, Respaldiza N, et al. Polymorphisms in Pneumocystis jirovecii strains in Spanish children with cystic fibrosis. J Infect Dis 2006; 193: 1332–1333. 20 Montes-Cano MA, de la Horra C, Dapena FJ, et al. Dynamic colonisation by different Pneumocystis jirovecii genotypes in cystic fibrosis patients. Clin Microbiol Infect 2007; 13: 1008–1011. 21 Wissmann G, Alvarez-Martinez MJ, Mesnhick SR, Diehl AR, Prolla JC. Absence of dihydropteroate synthase mutations in Pneumocystis jirovecii from Brazilian AIDS patients. J Eukaryot Microbiol 2006; 53: 305–307. 22 Norris KA, Morris A, Patil S, Fernandes E. Pneumocystis colonization, airways inflammation and pulmonary decline function in acquired immunodeficiency syndrome. Immunol Res 2006: 36: 175–187. 23 Morris A, Sciurba FC, Lebedeva IP, et al. Association of chronic obstructive pulmonary disease severity and Pneumocyctis colonization. Am J Respir Crit Care Med 2004; 170: 408–413. 24 Emerson J, Rosenfeld M, McNamara S, et al. Pseudomonas aeruginosa and others predictors of mortality and morbidity in young children with cystic fibrosis. PediatricPulmonology 2002; 34: 91–100. 25 Chabé M, Dei-Cas E, Creusy C, et al. Immunocompetent hosts as a reservoir of Pneumocystis organisms: histological and RT-PCR data demonstrate active replication. Eur J Clin Microbiol Infect Dis 2004; 23: 89–97. 26 Rivero L, de la Horra C, Montes-Cano MA, et al. Pneumocystis jirovecii transmission from immunocompetent carriers to infant. Emerg Infect Dis 2008; 14: 1116–1118.