Respiratory infection Development and evaluation of a real-time PCR assay for detection of Pneumocystis jirovecii DNA in bronchoalveolar lavage fluid of HIV-infected patients J F Huggett,1 M S Taylor,2 G Kocjan,3 H E Evans,4 S Morris-Jones,5 V Gant,5 T Novak,1 A M Costello,6 A Zumla,1 R F Miller4,7 1 Centre for Infectious Diseases ABSTRACT lower incidence of PCP since the introduction of and International Health, Background: Pneumocystis pneumonia (PCP) is con- highly active antiretroviral therapy. Windeyer Institute for Medical Molecular diagnostic techniques using the poly- Sciences, University College ventionally diagnosed by identifying Pneumocystis jirovecii London, London, UK; 2 Welcome in lower respiratory tract samples using cytochemical merase chain reaction (PCR) are more clinically Trust Centre for Human stains. Molecular diagnosis of PCP is potentially more sensitive than staining for detection of P jirovecii in Genetics, University of Oxford, BAL fluid and induced sputum.3–6 Molecular 3 sensitive. Oxford, UK; Department of techniques, however, may identify P jirovecii DNA Histopathology, University Methods: A study was undertaken to use an extensively College London, London, UK; optimised real-time polymerase chain reaction (PCR) using in respiratory samples from patients without 7–10 4 Centre for Sexual Health and primers designed to hybridise with the P jirovecii heat clinically apparent PCP, suggesting asympto- HIV Research, Department of shock protein 70 (HSP70) gene to quantify P jirovecii DNA matic carriage or ‘‘colonisation’’. Several studies Primary Care and Population in bronchoalveolar lavage (BAL) fluid from HIV-infected have used real-time PCR for detection of P carinii,11 Sciences, University College patients with and without PCP, and to compare this assay P murina12 and P jirovecii in respiratory specimens.13– London, London, UK; 21 5 Department of Microbiology, with conventional PCR targeting the P jirovecii mito- University College London chondrial large subunit rRNA gene sequence (mt LSU Real-time PCR allows accurate quantification of Hospitals NHS Foundation Trust, rRNA). DNA and the potential to discriminate between Windeyer Institute for Medical asymptomatic carriage of P jirovecii and clinical Sciences, London, UK; Results: Sixty-one patients had 62 episodes of PCP 6 International Perinatal Care (defined by detection of P jirovecii in BAL fluid by disease based on pathogen load. The objective of Unit, Institute of Child Health, cytochemical stains and typical clinical presentation). this study was to compare real-time PCR using University College London, primers designed to hybridise to the heat shock 7 Quantifiable HSP70 DNA was detected in 61/62 (range London, UK; Department of 22 23 ,13–18 608 copies/reaction; median ,332) and was protein 70 (HSP70) gene of P jirovecii with Infectious and Tropical Diseases, conventional PCR using primers designed to the London School of Hygiene and detectable but below the limit of quantification (,5 Tropical Medicine, London, UK large subunit of mitochondrial rRNA (mt LSU copies/reaction) in 1/62. Seventy-one other patients had 3410 74 episodes with alternative diagnoses. Quantifiable rRNA) for detection of P jirovecii DNA in BAL Correspondence to: HSP70 DNA was detectable in 6/74 (8%) episodes (range fluid from patients undergoing diagnostic broncho- Professor R F Miller, Centre for scopy. Patients had either proven PCP or confirmed Sexual Health and HIV Research, ,6–590 copies/reaction; median ,14) and detectable alternative diagnoses. Department of Primary Care and but below the limit of quantification in 34/74 (46%). Population Sciences, University College London, London WC1E Receiver-operator curve analysis (cut-off .10 copies/ 6JB, UK; [email protected] reaction) showed a clinical sensitivity of 98% (95% 91% METHODS to 100%) and specificity of 96% (95% CI 87% to 99%) for Selection of a new molecular target GeneBank: Accession number diagnosis of PCP. By contrast, clinical sensitivity of mt LSU The target region of HSP70 was obtained for P DQ987621. rRNA PCR was 97% (95% CI 89% to 99%) and specificity jirovecii by designing generic primers to the flanking Received 22 March 2007 was 68% (95% CI 56% to 78%). regions of P carinii. Nucleotide sequences for all Accepted 27 July 2007 Conclusion: The HSP70 real-time PCR assay detects P Pneumocystis types available from public data Published Online First repositories were searched using BlastX24 against 10 August 2007 jirovecii DNA in BAL fluid and may have a diagnostic application. Quantification of P jirovecii DNA by real-time the non-redundant protein database. Expressed PCR may also discriminate between colonisation with P sequence tag data were assembled into contigs jirovecii and infection. using CAP3 prior to searching in order to remove redundancy and improve sequence accuracy. Sequences with high quality matches (bit score .60) to proteins from a broad spectrum of The fungal pathogen Pneumocystis jirovecii is the eukaryotic species were considered further (mini- cause of Pneumocystis pneumonia (PCP) in mally: Caenorhabditis elegans, Drosophila melanoga- 1 humans. Diagnosis of P jirovecii infection is ster, Saccharomyces cerevisiae, Schizosaccharomyces hampered by the lack of a sustainable in vitro pombe and Neurospora crassa). culture method. Diagnosis of PCP is typically made The corresponding region was amplified using P by direct examination of respiratory samples jirovecii as template and sequenced. Real-time PCR (bronchoalveolar lavage (BAL) fluid or induced for P jirovecii HSP70 (PjHSP70a) was designed to sputum) after staining in order to detect the cyst amplify a 106 base pair region of the P jirovecii form of Pneumocystis.2 Some laboratories also use sequence and not the corresponding sequence of immunofluorescence stains to enhance sensitivity. other Pneumocystis species, or a wide range of The use of cytochemical stains for diagnosis is time potential fungal, bacterial or mycobacterial patho- consuming, and it may be difficult to maintain gens (table 1). DNA was extracted from these laboratory diagnostic expertise because of the isolates using the DNeasy tissue kit (QIAGEN, 154 Thorax 2008;63:154–159. doi:10.1136/thx.2007.081687 Respiratory infection Table 1 Fungal, bacterial and mycobacterial organisms used for cross-reactivity assessment of the specificity of the PjHSP70a PCR assay Organism Fungi Aspergillus flavus Aspergillus fumigatus Aspergillus niger Aspergillus terreus Candida glabrata Candida kruseii Candida parapsilosis Candida tropicalis Cryptococcus sp Pneumocystis carinii and Pneumocystis wakefieldiae Pneumocystis jirovecii Figure 1 Pneumocystis jirovecii DNA detected in bronchoalveolar Pneumocystis murina lavage fluid from patients with Pneumocystis pneumonia extracted with Saccharomyces cerevisiae DNAeasy (left) and QIAamp UltraSens (right). Data are shown as scatter Trichosporon sp and bar and whisker plots (median, 25th and 75th percentiles and range). Bacteria Burkholderia cepacia Haemophilus influenzae tension (PaO2) breathing room air) and number of days of Propionibacterium sp specific anti-Pneumocystis treatment before BAL were also Staphylococcus aureus recorded. All bronchoscopies were performed by one of the Streptococcus agalactiae authors (RFM); BAL was performed in a standardised manner as Streptococcus mitis previously described.25 26 All BAL samples were coded and Streptococcus oralis Streptococcus pneumoniae analyses by PCR were performed ‘‘blind’’ to the patient’s Mycobacteria Mycobacterium avium clinical and laboratory details. All patients undergoing broncho- Mycobacterium tuberculosis scopy gave informed written consent and the study was Mammalia Human DNA performed with approval from the local research ethics committee. Crawley, UK) following the manufacturer’s instructions, and DNA extraction Pj was used as template for the HSP70a. DNA extraction from BAL fluid was first done using the DNeasy tissue kit and subsequently using the QIAamp Patients and samples UltraSens Virus Kit (both QIAGEN, Crawley, UK) following One hundred and thirty-six BAL samples were obtained from the manufacturer’s instructions. Total DNA was extracted from 132 adult HIV-infected patients (112 men) undergoing diag- 200 ml (DNeasy) or 750 ml (QIAamp) BAL fluid. RNA quench nostic bronchoscopy. Sixty-one consecutive patients with PCP was also included in the latter extract, as suggested by the had 62 episodes of PCP; one patient had two episodes (interval manufacturer, and the extracted sample was eluted in 60 mlof 9 weeks). All had typical clinical and radiographic presentations, elution buffer by contrast with the DNeasy extraction (100 ml identification of P jirovecii cysts in BAL fluid by Grocott-Gomori of elution buffer). QIAamp UltraSens Virus Kit was chosen as it methenamine silver staining2 and response to specific anti-PCP was specifically designed to extract DNA from liquid samples, therapy.25 26 A further 71 consecutive patients (74 episodes) enabled a greater volume to be extracted from BAL fluid and investigated contemporaneously did not have PCP clinically or facilitated generation of a higher concentration of extracted radiographically, had negative results from Grocott-Gomori nucleic acids. Consequently, the final QIAamp eluate achieved a staining and did not receive specific anti-Pneumocystis treatment. 12.5-fold concentration compared with a twofold increase with All had alternative diagnoses, as described previously,10 26–28 the DNeasy kit. When the two