Evaluation of a real-time PCR hybridization assay for rapid detection of Legionella pneumophila in hospital and environmental water samples
2003; Elsevier BV; Volume: 9; Issue: 7 Linguagem: Inglês
10.1046/j.1469-0691.2003.00666.x
ISSN1469-0691
AutoresK. Levi, J. Smedley and, K. J. Towner,
Tópico(s)Biosensors and Analytical Detection
ResumoA real-time LightCycler assay for Legionella pneumophila was evaluated with 120 water samples potentially contaminated with PCR inhibitors. Results were obtained within five hours, with a detection limit equivalent to 800 cells/L. However, 11 of 22 culture-positive samples containing 800 CFU/L. The assay seemed suitable for rapidly screening large sample numbers for heavy contamination with L. pneumophila, but conventional culture should continue to be used to detect low contamination levels. A real-time LightCycler assay for Legionella pneumophila was evaluated with 120 water samples potentially contaminated with PCR inhibitors. Results were obtained within five hours, with a detection limit equivalent to 800 cells/L. However, 11 of 22 culture-positive samples containing 800 CFU/L. The assay seemed suitable for rapidly screening large sample numbers for heavy contamination with L. pneumophila, but conventional culture should continue to be used to detect low contamination levels. Legionella pneumophila infection arises through inhalation or aspiration of contaminated water, and it has been established that water systems of hospitals and other large buildings, especially cooling towers, are frequently contaminated with these organisms. Rapid monitoring of suspected hospital and environmental water reservoirs of L. pneumophila is therefore essential for risk assessment, determination of possible sources of infection, and the prevention of further cases. As identification of L. pneumophila takes three to ten days by conventional culture, several assays based on the polymerase chain reaction (PCR) have been evaluated for use with clinical and environmental samples [1Jaulhac B Reyroille M Sodahlon YK et al.Comparison of sample preparation methods for detection of Legionella pneumophila in culture-positive bronchoalveolar lavage fluids by PCR.J Clin Microbiol. 1998; 36: 2120-2122PubMed Google Scholar, 2Koide M Saito A Kusano N Higa F Detection of Legionella spp. in cooling tower water by the polymerase chain reaction.Appl Environ Microbiol. 1993; 59: 1943-1946PubMed Google Scholar, 3Matsiota-Bernard P Pitsouni E Legakis N Nauciel C Evaluation of commercial amplification kit for detection of Legionella pneumophila in clinical specimens.J Clin Microbiol. 1994; 32: 1503-1505PubMed Google Scholar, 4Miyamoto H Yamamoto H Arima K et al.Development of a new semi-nested PCR method for detection of Legionella species and its application to surveillance of Legionellae in hospital cooling tower water.Appl Environ Microbiol. 1997; 63: 2489-2494PubMed Google Scholar, 5Villari P Motti E Farullo C Torre I Comparison of conventional culture and PCR methods for the detection of Legionella pneumophila in water.Lett Appl Microbiol. 1998; 27: 106-110Crossref PubMed Scopus (32) Google Scholar]. The advent of real-time PCR technology has made possible further reductions in analysis time, as amplification and detection can now proceed in parallel. One such approach uses LightCycler technology, in which rapid cycling in glass capillaries is combined with fluorescent monitoring and specific identification of amplification products as they are formed, thereby allowing the amplification and analysis steps to be completed within one hour. Several LightCycler assays have been described that detect Legionella DNA in respiratory specimens and potable water samples [6Ballard AL Fry NK Chan L et al.Detection of Legionella pneumophila using a real-time PCR hybridization assay.J Clin Microbiol. 2000; 38: 4215-4218PubMed Google Scholar, 7Hayden RT Uhl JR Qian X et al.Direct detection of Legionella species from bronchoalveolar lavage and open lung biopsy specimens: comparison of LightCycler PCR, in situ hybridisation, direct fluorescence antigen detection and culture.J Clin Microbiol. 2001; 39: 2618-2626Crossref PubMed Scopus (120) Google Scholar, 8Rantakokko-Jalava K Jalava J Development of conventional and real-time PCR assays for detection of Legionella DNA in respiratory specimens.J Clin Microbiol. 2001; 39: 2904-2910Crossref PubMed Scopus (78) Google Scholar, 9Wellinghausen N Frost C Marre R Detection of legionellae in hospital water samples by quantitative real-time LightCycler PCR.Appl Environ Microbiol. 2001; 67: 3985-3993Crossref PubMed Scopus (192) Google Scholar]. However, the problem of amplification inhibition, particularly that caused by iron compounds (e.g. rust) sometimes present in water supplies [10Maiwald M Kissel K Srimuang S von Knebel DM Sonntag HG Comparison of polymerase chain reaction and conventional culture for the detection of legionellas in hospital water samples.J Appl Bacteriol. 1994; 76: 216-225Crossref PubMed Scopus (40) Google Scholar, 11Toze S PCR and the detection of microbial pathogens in water and wastewater.Water Res. 1999; 33: 3545-3556Crossref Scopus (162) Google Scholar], may limit the usefulness of PCR-based methods unless improved front-end DNA preparation methods can be incorporated into these assays. This study evaluated the use of an L. pneumophila LightCycler assay, in combination with a column-based DNA extraction and purification method, for the surveillance of hospital and environmental water samples prone to contamination with a range of PCR inhibitors, including humic substances and iron compounds. One hundred random environmental water samples were investigated, of which 68 came from various sites in different hospitals, 32 were from private domestic water supplies, 16 were from cooling towers, and four were from private external sites. These samples were concentrated from 1 L to about 1 mL by membrane filtration and centrifugation [12Donovan TJ Manger PA Isolation of legionellae from 1 litre of water: a modified filtration method.PHLS Microbiol Dig. 1989; 6: 134-135Google Scholar]. Three separate 100-μL aliquots of these concentrates (either untreated or treated with heat or acid [13Anonymous British Standard 6068-4.12. Water quality part 4: Microbiological methods. Section 4.12: Detection and enumeration of Legionella. British Standards Institution, London1998Google Scholar]) were cultured on GVPC (glycine, vancomycin, polymyxin, cyclohexamide) selective medium for L. pneumophila [13Anonymous British Standard 6068-4.12. Water quality part 4: Microbiological methods. Section 4.12: Detection and enumeration of Legionella. British Standards Institution, London1998Google Scholar]. Concurrently, nucleic acid was extracted from the concentrates (see below). In addition to these random samples, 20 environmental water samples found previously to be culture–positive for L. pneumophila were included in the study. Concentrates of these known culture-positive samples had been stored for varying periods at −20 °C before nucleic acid extraction. For each sample, 80 μL of concentrate was added to 100 μL of lysis (ATL) buffer in a DNA Mini Kit (Qiagen Ltd, Crawley, UK) and extracted according to the manufacturer's instructions for tissue samples. To minimize potential inhibition, 5 μL of 2% (w/v) bovine serum albumin (Sigma, Poole, UK) was added to the final 50-μL eluate [6Ballard AL Fry NK Chan L et al.Detection of Legionella pneumophila using a real-time PCR hybridization assay.J Clin Microbiol. 2000; 38: 4215-4218PubMed Google Scholar, 14Kreader CA Relief of amplification inhibition in PCR with bovine serum albumin or T4 gene 32 protein.Appl Environ Microbiol. 1996; 62: 1102-1106PubMed Google Scholar]. For positive controls and assessment of assay sensitivity, genomic DNA was extracted from L. pneumophila serogroup 1 cells grown on buffered charcoal–yeast extract (BCYE) agar for 72 h at 37 °C and resuspended in sterile distilled water. DNA concentrations were determined spectrophotometrically. A 131-bp region of the L. pneumophila macrophage infectivity potentiator (mip) gene was amplified in a model LC32 LightCycler (Bio/Gene, Cambridge, UK) with primers mip-Lpn0901F (5′-AACCGATGCCACATCATTA) and mip-Lpn1011R (5′-CTTGCATGCCTTTAGCCA). The identity of amplification products was confirmed with the hybridization probe mip-Lpn0941P (5′-Cy5-TCGGCACCAATGCTATAAGA-biotin). The combined specificity of the primers and hybridization probe has been established previously [6Ballard AL Fry NK Chan L et al.Detection of Legionella pneumophila using a real-time PCR hybridization assay.J Clin Microbiol. 2000; 38: 4215-4218PubMed Google Scholar]. LightCycler reactions were modified from those described previously [6Ballard AL Fry NK Chan L et al.Detection of Legionella pneumophila using a real-time PCR hybridization assay.J Clin Microbiol. 2000; 38: 4215-4218PubMed Google Scholar], and comprised FastStart DNA Master Hybridization reaction mix containing 0.1 U of uracil-N-glycosylase (Roche Diagnostics, Lewes, UK), 2 μL of DNA extract, 4 pmol of each primer and the hybridization probe, 5 mm MgCl2, SyBr Green I (Bio/Gene) at a final concentration of 1 : 10 000, and molecular-grade water to a final volume of 8 μL. The LightCycler amplification program was also modified to accommodate the FastStart enzyme, and consisted of a 10-min activation and denaturation period at 95 °C, followed by 50 cycles of denaturation (5 s at 95 °C), annealing (5 s at 60 °C), and extension (5 s at 72 °C), with maximum temperature transition rates of 20 °C/s. The melt program and monitoring of fluorescence was as described previously [6Ballard AL Fry NK Chan L et al.Detection of Legionella pneumophila using a real-time PCR hybridization assay.J Clin Microbiol. 2000; 38: 4215-4218PubMed Google Scholar]. L. pneumophila-specific products were characterized by SyBr Green I and Cy5 hybridization probe disassociations at mean peaks of 80 °C and 61 °C, respectively. The characteristic mean melt temperature of 61 °C was observed to be slightly lower than that reported previously (63 °C) [6Ballard AL Fry NK Chan L et al.Detection of Legionella pneumophila using a real-time PCR hybridization assay.J Clin Microbiol. 2000; 38: 4215-4218PubMed Google Scholar], and it was postulated that this could be related to use of the FastStart enzyme and the slightly longer annealing period in each amplification cycle. The presence of amplification inhibitors was assessed by spiking negative water samples with 200 fg of extracted L. pneumophila DNA, and repeating the assay in comparison with a positive control (sterile distilled water also spiked with 200 fg of extracted L. pneumophila DNA). Equal signals indicated an absence of inhibition. Using ten-fold serial dilutions of purified L. pneumophila serogroup 1 DNA, the sensitivity limit of the LightCycler assay was found to be 10 fg of DNA, equivalent to 2.3 L pneumophila cells/assay [9Wellinghausen N Frost C Marre R Detection of legionellae in hospital water samples by quantitative real-time LightCycler PCR.Appl Environ Microbiol. 2001; 67: 3985-3993Crossref PubMed Scopus (192) Google Scholar]. However, as only 2 μL of DNA extract could be included in each LightCycler reaction, this level of contamination equated to 800 CFU/L in the original unconcentrated environmental water sample. Amplification of DNA extracts prepared from serial dilutions of a known culture of L. pneumophila serogroup 1 yielded positive results with <10 CFU/assay, thereby corroborating the results obtained with purified L. pneumophila DNA. Of the 20 water samples found previously to be positive for L. pneumophila by culture, 14 were positive in the LightCycler assay (Table 1). The six LightCycler-negative samples included five samples (9, 10, 15, 17 and 19) that contained fewer viable L. pneumophila cells than the theoretical detection limit (800 CFU/L), as calculated above. In contrast, 12 of the LightCycler-positive samples also contained 800 CFU/L (Table 1).Table 1Results obtained by the LightCycler assay with hospital and environmental water samples shown previously to be positive for L. pneumophila by conventional cultureSample numberCFU/L by culture (serogroup)LightCyclerInhibition of resultSample source amplification1120 (2-14)+NoHospital wash basin hot tap230 (2-14)+NoHospital bath hot tap360 (2-14)+NoHospital wash basin hot tap4220 (2-14)+NoHospital wash basin cold tap514 160 (1)+NoPrivate shower640 (2-14)+NoHospital shower710 (1)+NoPrivate shower810 (1)+NoPrivate bath hot tap920 (1); 20 (2-14)–NoCooling tower1010 (1); 10 (2-14)–YesCooling tower1120 (2-14)+NoHospital wash basin hot tap124120 (2-14)+NoHospital sluice hot tap13260 (2-14)+NoHospital wash basin hot tap1420 (1)+NoPrivate shower15140 (2-14)–NoHospital wash basin cold tap16630 (1)+NoHospital bath hot tap1760 (1)–NoHospital sluice hot tap1890 (1)+NoHospital shower1960 (1)–YesHospital shower205520 (1)–YesCooling tower Open table in a new tab Of the 100 random environmental water samples, 14 were positive for L. pneumophila by culture, of which only four were positive in the LightCycler assay (Table 2). All ten culture-positive samples that were not detected by the LightCycler assay contained <200 CFU/L. However, one of the LightCycler-positive samples (118) contained only 30 L pneumophila serogroup 2-14 CFU/L by culture, while one culture-negative sample (77) was repeatedly positive for L. pneumophila DNA in the LightCycler assay (Table 2). Inhibition of amplification was demonstrated with one culture-positive sample (49), but this sample contained only 30 CFU/L.Table 2Random hospital and environmental water samples (n = 100) positive for L. pneumophila by either conventional culture or the LightCycler assaySample numberCFU/L by culture (serogroup)LightCyclerInhibition of resultSample source amplification2230 (2-14)–NoHospital shower2340 (2-14)–NoHospital shower2510 (1)–NoHospital wash basin cold tap2740 (1)–NoHospital wash basin cold tap2840 (1)–NoHospital wash basin hot tap2910 (1)–NoHospital wash basin cold tap4930 (1)–YesCooling tower6040 (2-14)–NoHospital wash basin hot tap62180 (2-14)–NoHospital shower6510 (2-14)–NoHospital wash basin hot tap77Negative+NoHospital shower1125520 (2-14)+NoHospital shower11830 (2-14)+NoHospital hot water supply1198080 (2-14)+NoHospital hot water supply120740 (1)+NoHospital hot water supply Open table in a new tab A recent report [15van der Zee A Peeters M de Jong C et al.Qiagen DNA extraction kits for sample preparation for Legionella PCR are not suitable for diagnostic purposes.J Clin Microbiol. 2002; 40: 1126Crossref PubMed Scopus (55) Google Scholar] has indicated that some batches of Qiagen columns might occasionally be contaminated with DNA from Legionella spp. other than L. pneumophila. The assay used in the present study is specific for L. pneumophila, and negative controls extracted and amplified in parallel with environmental water samples gave negative results throughout the study. Although rapid molecular amplification techniques for detecting L. pneumophila have been reported to detect very small numbers of legionellae in each individual assay, the need to concentrate large volumes of water samples into the assay format invariably means that these assays are relatively insensitive in real terms. In addition, concentrated environmental water samples are particularly prone to contamination with amplification inhibitors such as humic compounds, rust, divalent cations in high concentrations, and salts [10Maiwald M Kissel K Srimuang S von Knebel DM Sonntag HG Comparison of polymerase chain reaction and conventional culture for the detection of legionellas in hospital water samples.J Appl Bacteriol. 1994; 76: 216-225Crossref PubMed Scopus (40) Google Scholar, 11Toze S PCR and the detection of microbial pathogens in water and wastewater.Water Res. 1999; 33: 3545-3556Crossref Scopus (162) Google Scholar]. The present study combined conventional water concentration techniques with column-based nucleic acid purification and minimization of inhibition with bovine serum albumin [6Ballard AL Fry NK Chan L et al.Detection of Legionella pneumophila using a real-time PCR hybridization assay.J Clin Microbiol. 2000; 38: 4215-4218PubMed Google Scholar, 14Kreader CA Relief of amplification inhibition in PCR with bovine serum albumin or T4 gene 32 protein.Appl Environ Microbiol. 1996; 62: 1102-1106PubMed Google Scholar]. The subsequent LightCycler assay made it possible to screen 1-L environmental water samples within five hours of receipt of the sample, with a theoretical assay sensitivity corresponding to 800 L. pneumophila cells/L in an unconcentrated water sample. This compared favorably with other amplification-based techniques, which have only detected L. pneumophila-positive environmental waters containing at least 34 000 CFU/L [16Chang HR Loo LH Kuah BG Heng BH Comparison of multiplex PCR and culture for detection of legionellae in cooling tower water samples.Southeast Asian J Trop Med. 1995; 26: 258-262PubMed Google Scholar], 1800 CFU/L [17Yamamoto H Hashimoto Y Ezaki T Comparison of detection methods for Legionella species in environmental water by colony isolation, fluorescent antibody staining, and polymerase chain reaction.Microbiol Immunol. 1993; 37: 617-622Crossref PubMed Scopus (56) Google Scholar], or 1000 CFU/L [2Koide M Saito A Kusano N Higa F Detection of Legionella spp. in cooling tower water by the polymerase chain reaction.Appl Environ Microbiol. 1993; 59: 1943-1946PubMed Google Scholar]. However, the assay also gave positive results with environmental water samples containing as few as 10 CFU/L by culture, with 11 of 22 culture-positive samples containing 800 CFU/L by culture, although L. pneumophila DNA was also detected by the LightCycler in a further 14 samples. Similarly, a quantitative LightCycler assay for potable water samples found that cell number estimates obtained following DNA amplification were generally 25-fold higher than those obtained following conventional culture [9Wellinghausen N Frost C Marre R Detection of legionellae in hospital water samples by quantitative real-time LightCycler PCR.Appl Environ Microbiol. 2001; 67: 3985-3993Crossref PubMed Scopus (192) Google Scholar]. Algorithms for the investigation of water samples have been proposed in which PCR-based techniques can be used for rapid screening of samples in potential outbreak situations. PCR-positive samples would then be cultured to obtain quantitative data and isolates for epidemiology studies, while larger volumes of PCR-negative samples would be cultured to identify low concentrations of L. pneumophila [6Ballard AL Fry NK Chan L et al.Detection of Legionella pneumophila using a real-time PCR hybridization assay.J Clin Microbiol. 2000; 38: 4215-4218PubMed Google Scholar, 16Chang HR Loo LH Kuah BG Heng BH Comparison of multiplex PCR and culture for detection of legionellae in cooling tower water samples.Southeast Asian J Trop Med. 1995; 26: 258-262PubMed Google Scholar]. In the UK, guidance for the surveillance of cooling towers and hot and cold water systems for legionellae states that the detection protocol should have a theoretical detection limit of 100 CFU/L [18Anonymous Approved code of practice and guidance: the control of Legionella bacteria in water systems. Health and Safety Commission, London2000Google Scholar]. Although this theoretical limit cannot be achieved with the current LightCycler assay, the results of the present study suggest that this assay would also detect legionellae in certain samples containing very low numbers of viable cells, including occasional samples that might appear to be culture negative. No false-positive results were obtained with the negative controls included in every batch of reactions, but it would be prudent always to include sufficient negative controls to detect any potential contamination introduced during the assay procedure [15van der Zee A Peeters M de Jong C et al.Qiagen DNA extraction kits for sample preparation for Legionella PCR are not suitable for diagnostic purposes.J Clin Microbiol. 2002; 40: 1126Crossref PubMed Scopus (55) Google Scholar]. Despite the partial purification of DNA and incorporation of bovine serum albumin into nucleic acid eluates, four (11.4%) of 35 culture-positive samples (three of which were from cooling towers) still contained compounds inhibitory to amplification. This compared favorably with a previous study in which 30.6% (15 of 49) of cooling tower waters contained PCR inhibitors [9Wellinghausen N Frost C Marre R Detection of legionellae in hospital water samples by quantitative real-time LightCycler PCR.Appl Environ Microbiol. 2001; 67: 3985-3993Crossref PubMed Scopus (192) Google Scholar]. However, the results obtained in this and other studies suggest that amplification-based assays should continue to be used in conjunction with culture for the present time. Thus the main role of the LightCycler assay would seem to be in potential outbreak situations, when there is a need to screen large numbers of water samples for heavy contamination with L. pneumophila in as short a time as possible. A LightCycler-positive result could then be used to instigate an early targeted review of control measures or initiate immediate disinfection procedures while the results of culture are still awaited. Development of resistance in Pseudomonas aeruginosa obtained from patients with cystic fibrosis at different timesClinical Microbiology and InfectionVol. 9Issue 7PreviewTables 1 and 5 in the above article have been published incorrectly. The correct tables are below. We apologise for any inconvenience caused. Full-Text PDF Open Archive
Referência(s)