| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Clinical Virology and Molecular Microbiology Research Unit, Sir Albert Sakzewksi Virus Research Centre, Royal Children's Hospital and Clinical Medical Virology Centre, University of Queensland, Herston, Queensland, Australia 4029 2,6Department of Paediatrics and Child Health2 and Department of Medicine6, University of Queensland, Brisbane, Queensland, Australia 3Adult Cystic Fibrosis Unit, The Prince Charles Hospital, Brisbane, Queensland, Australia 4Department of Respiratory Medicine, Royal Children's Hospital, Brisbane, Queensland, Australia 5Department of Microbiology, Queensland Health Pathology Service, The Prince Charles Hospital Campus, Brisbane, Queensland, Australia
Correspondence Michael D. Nissen theniss{at}uq.edu.au
Received January 22, 2004
Accepted July 7, 2004
In this study, the suitability of two repetitive-element-based PCR (rep-PCR) assays, enterobacterial repetitive intergenic consensus (ERIC)-PCR and BOX-PCR, to rapidly characterize Pseudomonas aeruginosa strains isolated from patients with cystic fibrosis (CF) was examined. ERIC-PCR utilizes paired sequence-specific primers and BOX-PCR a single primer that target highly conserved repetitive elements in the P. aeruginosa genome. Using these rep-PCR assays, 163 P. aeruginosa isolates cultured from sputa collected from 50 patients attending an adult CF clinic and 50 children attending a paediatric CF clinic were typed. The results of the rep-PCR assays were compared to the results of PFGE. All three assays revealed the presence of six major clonal groups shared by multiple patients attending either of the CF clinics, with the dominant clonal group infecting 38 % of all patients. This dominant clonal group was not related to the dominant clonal group detected in Sydney or Melbourne (pulsotype 1), nor was it related to the dominant groups detected in the UK. In all, PFGE and rep-PCR identified 58 distinct clonal groups, with only three of these shared between the two clinics. The results of this study showed that both ERIC-PCR and BOX-PCR are rapid, highly discriminatory and reproducible assays that proved to be powerful surveillance screening tools for the typing of clinical P. aeruginosa isolates recovered from patients with CF.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
PFGE is commonly employed, and has achieved widespread recognition as the ‘gold standard’ for P. aeruginosa DNA typing (Bertrand et al., 2001; Breitenstein et al., 1997; Douglas et al., 2001; Grundmann et al., 1995; Spencker et al., 2000). However, this method is limited by technical complexity, expense and prolonged turnaround times for results (Olive, 1999). As an alternative, repetitive-element-based PCR (rep-PCR) has shown considerable potential as a DNA typing tool in the laboratory (Lau et al., 1995; Olive, 1999). Rep-PCR assays utilize primers targeting highly conserved repetitive sequence elements in the bacterial genome (Versalovic et al., 1991). Two such groups of repetitive elements are the enterobacterial repetitive intergenic consensus (ERIC) sequences common to Gram-negative enteric bacteria, and the BOX elements, originally detected in Streptococcus pneumoniae (Hulton et al., 1991; Martin et al., 1992; Tyler et al., 1997).
To our knowledge, ERIC- and BOX-PCR have not previously been used to compare P. aeruginosa strains isolated from adult and paediatric patients with CF. In this study, we characterized 163 clinical isolates collected from 50 children and 50 adults attending CF clinics in Brisbane by ERIC- and BOX-PCR, and compared the results with those obtained by PFGE. The genetic diversity of these P. aeruginosa strains was determined for our population.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Control isolates.
Two P. aeruginosa strains representing the major epidemic strains from Liverpool, UK (strain H190), and from Manchester, UK (strain C3425), were provided by Professor John Govan from the Medical Microbiology Department, Medical School, University of Edinburgh, Edinburgh, UK. One P. aeruginosa strain (pulsotype 1), representing the dominant epidemic strain identified at CF clinics at the Royal Prince Alfred Hospital, Sydney, and the Royal Children's Hospital and Monash Medical Centre, Melbourne, was provided by Dr Barbara Rose from the University of Sydney, Sydney, Australia. In addition, a control P. aeruginosa strain (ATCC 27853) was supplied in the Group 3 Control Kit from Bio-Rad.
Clinical isolates.
Sputum specimens (at least 2 g) from each patient were submitted to the pathology laboratories of the two hospitals for the routine investigation of respiratory organisms. From these, a total of 163 P. aeruginosa isolates were recovered by bacterial isolation on MacConkey agar plates (Oxoid) and tentatively identified as P. aeruginosa by colony size, pigmentation and mucoidy. Isolates were subcultured onto chocolate agar plates and nutrient agar slopes and the details of all clinical and control isolates were entered into a Microsoft Excel database allocating a unique identifier to each isolate. All isolates were then subcultured aseptically onto horse blood agar plates (Oxoid) and grown overnight at 37 °C. Once purity was confirmed, a single colony was placed in 3 ml Luria–Bertani (LB) broth and incubated overnight at 37 °C. This culture was subsequently used for extraction of bacterial DNA and molecular analysis. The remaining pure growth was stored in Protect Bacterial Preservers (Technical Service Consultants, UK) at –80 °C following the manufacturer's instructions.
Bacterial DNA isolation and confirmation.
All isolates were confirmed as P. aeruginosa by a previously described P. aeruginosa PCR assay targeting the species-specific oprL gene (De Vos et al., 1997). For this, one colony was selected from the horse blood agar plate and placed in 0.2 ml sterile water. DNA was extracted using the High Pure Viral Nucleic Acid kit (Roche Diagnostics), following the manufacturer's instructions. Purified nucleic acid was eluted from the extraction column in 50 µl elution buffer and stored at –70 °C until analysis.
For PFGE and the rep-PCR assays, genomic DNA was extracted from bacterial cells grown overnight in LB broth using the AquaPure Genomic DNA kit (Bio-Rad) with some exceptions. Briefly, a total of 2 ml LB broth was extracted, and in the protein precipitation step, the solution was centrifuged at 13 200 r.p.m. for 30 min. Genomic DNA was measured at A260 using the Genequant Pro Calculator (Amersham Biosciences) and diluted to a final concentration of 30 ng µl–1. Extracts were used immediately or stored at –20 °C until further analysis.
Rep-PCR.
The rep-PCR assays performed were previously described (Dawson et al., 2002; Versalovic et al., 1991) with modifications. To ensure the most efficient reaction conditions, parameters such as the concentration of magnesium ions, DNA template, Taq polymerase and primers were optimized using four unrelated isolates of P. aeruginosa (as determined by PFGE). ERIC-PCR and BOX-PCR were performed separately using 0.2 ml thin-walled PCR tubes in a Perkin Elmer 9600 thermal cycler (Applied Biosystems). In each assay, the reaction mix contained the following reagents: 10x Reaction Mix (Invitrogen), 0.2 mM each deoxynucleoside triphosphate (Promega) and a total of 60 ng genomic DNA extract. The final reaction volume was adjusted to 25 µl with PCR-grade water and PCR amplification was commenced using an initial denaturation step at 94 °C for 7 min. This was followed by 30 cycles of denaturation at 94 °C for 1 min, primer annealing at 53 °C for 1 min and extension at 72 °C for 2 min and one cycle of further extension at 72 °C for 15 min. A 7 µl volume of amplicon was loaded with 2 µl 2x loading buffer (10 % glycerol, 2 mM EDTA, 0.1 % xylene cyanol, 0.1 % bromophenol blue) (Geneworks) into one well of a 15-well 1.2 % agarose gel in 1x Tris-Borate-EDTA (TBE) buffer with 0.5 µg ethidium bromide ml–1. A 6 kb DNA ladder (Geneworks) was placed at both ends and in the middle of each gel, which was run at 80 V for 3 h at room temperature.
PFGE.
All P. aeruginosa isolates were typed by PFGE as per the protocol of Spencker et al. (2000), using the CHEF-DR III apparatus (Bio-Rad), Genepath Group 3 Reagent Kit (Bio-Rad) and the Genepath Gel Kit (Bio-Rad), with the following modifications. The 0.5x TBE buffer contained 50 µM thiourea (Sigma), the run time was 19.5 h and the final switch time was 43 s.
Pattern analysis.
Gels for PFGE were stained in an aqueous solution containing 0.5 µg ethidium bromide ml–1 and photographed under UV transillumination. Banding patterns of PFGE were also visually analysed according to criteria developed by Tenover et al. (1995). Briefly, band differences of 0, < 4, 4–6 and >6 were classified as (i) identical, (ii) related, (iii) possible related and (iv) unrelated isolates, respectively. For visual analysis of rep-PCR profiles, we determined that profiles with two or more band differences were considered to be unrelated. Variation in band intensity was not considered as genetic difference. Bands that were too faint to be interpreted were not included in the analysis. Each gel photograph was also scanned using a HP Scanjet 2300c (Hewlett Packard), digitized, and saved as inverted TIFF images. Images were then stored in a database in Gelcompar software (version 4.1; Applied Maths). Each image was converted (track resolution, 250) and then normalized (resolution, 200; smoothing, 5 point; background subtraction; rolling disk intensity, 12) by using the Lambda ladder (Bio-Rad) as the reference marker. Similarity analysis of results for each assay was calculated using the Dice coefficient and the cluster analyses of the similarity matrices was generated using the unweighted pair group method using arithmetic averages (UPGMA). The criterion for related clones for all assays was taken as profiles with 85 % or more similar bands.
PFGE profiles were referred to as pulsotypes and were labelled numerically with any strains related to pulsotype 1 classified as such, and additional profiles labelled sequentially. Rep-PCR results were labelled numerically in the same manner as PFGE results and designated ERIC- or BOX-PCR profiles.
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
PFGE analysis of P. aeruginosa isolates
Overall, 58 unrelated pulsotypes were identified by PFGE analysis amongst the 163 isolates from 100 patients examined in this study. PFGE analysis also identified that a clonal group of P. aeruginosa isolates represented by pulsotype 2 was predominant in both CF clinics in Brisbane, containing 63 isolates from 39 patients (14 adults and 25 children) (Table 1). Six unrelated pulsotypes (P1, P2, P3, P5, P42 and P58) were identified containing P. aeruginosa strains isolated from more than one patient presenting to either of the two CF clinics in Brisbane. An example of typical banding patterns is shown in Fig. 1. Each of these six pulsotypes included closely related isolates with differences in band numbers ranging from none to three. Of these, pulsotype 2 showed the greatest divergence with differences of one, two and three bands (Fig. 2). All other patients (n = 41) carried a unique P. aeruginosa isolate (n = 65) representative of a single pulsotype only that could not be classified as P1, P2, P3, P5, P42 or P58. Of these 54 unique pulsotypes, 31 (57 %) were detected in adult patients and 23 (43 %) in children.
|
|
|
ERIC- and BOX-PCR analysis of P. aeruginosa isolates
Results from this study demonstrated that ERIC- and BOX-PCR are equally suitable as rapid epidemiological surveillance tools to determine the genetic diversity of clinical P. aeruginosa strains isolated from the sputa of patients with CF. Both ERIC- and BOX-PCR analysis of the 163 isolates identified the same number of unrelated and related profiles as achieved by PFGE analysis (Fig. 3). However, neither ERIC- nor BOX-PCR was able to discriminate highly related P. aeruginosa isolates that showed differences of one, two or three bands by PFGE. For example, strains representing PFGE pulsotypes 2i, 2ii and 2iii were identical to the pulsotype 2 profile in both ERIC- and BOX-PCR (Fig. 2).
|
Single versus multiple strain types
Multiple P. aeruginosa isolates were identified from sputum in 59 patients (28 adults and 31 children) (Table 2). In a majority of cases, multiple isolates of P. aeruginosa recovered from a patient at a single sampling were more likely to be related than unrelated. P. aeruginosa isolates from five adults and 11 children showed the presence of multiple clonal types in the same patient. In 12 of these (four adults and eight children), the isolates present displayed a major difference in the banding patterns, and were easily distinguished with each of the three assays.
|
Comparison with known clonal strains
None of the profiles obtained from our isolates by ERIC-PCR, BOX-PCR or PFGE were found to be related to the profiles shown by the epidemic strains from Liverpool or Manchester. However, one clonal group present in both CF clinics in Brisbane was related to the epidemic strain, pulsotype 1 (P1), identified in CF clinics in Sydney and Melbourne. This group consisted of nine P. aeruginosa strains isolated from eight patients (five adults and three children). Eight of these nine isolates showed identical PFGE pulsotype profiles to the clonal strain pulsotype 1, while the ninth differed by the absence of one band and was classified as a related isolate according to the criteria of Tenover et al. (1995). All of these strains gave identical profiles by ERIC- and BOX-PCR. Interestingly, a large majority of Brisbane patients carrying P1 isolates had previous links to Sydney and Melbourne CF clinics or had family living in those cities. The epidemiological link between patients from the adult and paediatric CF clinics in Brisbane and the adult CF clinic in Sydney is discussed elsewhere (O'Carroll et al., 2004).
Benefits of rep-PCR typing assays
Apart from discriminatory power, a suitable DNA typing assay must also have a high level of reproducibility, typeability and stability, low complexity and cost, as well as fast result turnaround times (Pfaller, 1999; Tenover et al., 1997). All three assays showed a high level of reproducibility. The reproducibility of the PFGE, ERIC-PCR and BOX-PCR assays was examined by comparing DNA profiles of 10 unrelated isolates in four independent analyses carried out on different days (days 1, 4, 159 and 484). For the rep-PCR assays, results showed that the banding profiles produced on each of the separate occasions were identical, except for some minor differences in band intensities. This is consistent with results reported previously by others (Cho & Tiedje, 2000; Kang & Dunne, 2003; Struelens, 1998). For PFGE, the banding profiles for these isolates were identical for each independent experiment.
As previously reported, the typeability of PFGE, ERIC-PCR and BOX-PCR assays proved to be excellent (Dawson et al., 2002; Romling & Tummler, 2000), with all of the 163 isolates able to be typed. The primers from both rep-PCR assays were based on highly conserved repetitive elements, which ensures the stability of the typing assay (Olive, 1999). Also, analysis of rep-PCR results was simplified by less-complex PCR banding patterns compared to PFGE patterns. The ERIC-PCR primers and the single BOX-PCR primer generated banding patterns containing 5–16 bands ranging in size from 150 to 6000 bp, and 200–6000 bp, respectively. However, PFGE analysis using SpeI enzyme generated banding patterns of 12–27 bands ranging from 15 to 780 kb.
In our hands, the cost of reagents for PFGE was estimated as $AUD 35.00 per sample, whereas for each rep-PCR assay this was $AUD 10.00 per sample. Currently PFGE equipment costs approximately $AUD 28 000.00 whilst PCR equipment (thermal cycler, gel electrophoresis) costs less than half this amount. Also, the shorter hands-on time for the rep-PCR assays means that labour costs were significantly less, and training of personnel in this technology was simpler and more generic, compared to PFGE. The result turnaround time for the rep-PCR assays was less than 10 h, which was considerably faster than PFGE (4–5 days).
Also, as with all PCR-based techniques, the chance of generating artefact rather than detecting true genetic variation is greater if low-stringency PCR conditions are used such as those used in arbitrarily primed-PCR (Tyler et al., 1997). Rep-PCR assays use highly stringent conditions and therefore are more easily standardized (Olive, 1999). However, optimization of all parameters of any DNA typing assay is essential to ensure optimal inter- and intra-laboratory standardization (Pfaller, 1999). Overall, therefore, rep-PCR assays combine maximum discriminatory power, reproducibility, typeability and stability with cost-effective use of reagents and operator time.
Applications of rep-PCR assays
The benefits of rep-PCR methods have now been widely recognized in the research of bacterial diversity of clinical isolates as well as strains of industrial, agricultural and environmental organisms (Abd-El-Haleem et al., 2002; McSpadden Gardener et al., 2000; Whiteley et al., 2001). Both ERIC-PCR and BOX-PCR have shown high discriminatory power with reproducibility, stability and fast turnaround times, and are cost-effective alternatives for typing bacteria such as Xanthomonas campestris, Xanthomonas oryzae and Pseudomonas syringae (Louws et al., 1994; Struelens, 1998). BOX elements have since been identified in P. aeruginosa as well as Enterococcus faecalis (Malathum et al., 1998; McSpadden Gardener et al., 2000). ERIC-PCR has been successfully applied to P. aeruginosa as well as Listeria monocytogenes, Serratia marcescens, Escherichia coli, Staphylococcus aureus, Enterococcus faecalis, Staphylococcus epidermidis and Acinetobacter baumannii (Jersek et al., 1999; Kang & Dunne, 2003; Lau et al., 1995; Patton et al., 2001).
This study demonstrated that ERIC-PCR and BOX-PCR were equally effective in characterizing clinical isolates of P. aeruginosa. Also, it was the first to apply these methods to show cross-colonization of a dominant clone within and between Australian adult and paediatric patients attending two separate CF clinics. Our results suggest that DNA typing tools such as rep-PCR may play an important role in routine epidemiological surveillance, outbreak surveillance and in the identification of the source of transmission of P. aeruginosa in CF patients (Olive, 1999; Pfaller, 1999). Rep-PCR methods have previously been applied in infection control surveillance of pathogens in other disciplines (Jersek et al., 1999; Lau et al., 1995). However, before significant decisions are made on infection control issues, these typing methods should ideally be combined with additional genetic, phenotypic or other epidemiological data (Pfaller, 1999).
In conclusion, the results of this study suggest that BOX-PCR and ERIC-PCR are suitable, inexpensive, fast, reproducible and discriminatory DNA typing tools for effective epidemiological surveillance of potential transmissible P. aeruginosa isolates between patients with CF.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This is CMVC/SASVRC publication number 194.
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. I. Lutter, M. M. P. Faria, H. R. Rabin, and D. G. Storey Pseudomonas aeruginosa Cystic Fibrosis Isolates from Individual Patients Demonstrate a Range of Levels of Lethality in Two Drosophila melanogaster Infection Models Infect. Immun., May 1, 2008; 76(5): 1877 - 1888. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Vu-Thien, G. Corbineau, K. Hormigos, B. Fauroux, H. Corvol, A. Clement, G. Vergnaud, and C. Pourcel Multiple-Locus Variable-Number Tandem-Repeat Analysis for Longitudinal Survey of Sources of Pseudomonas aeruginosa Infection in Cystic Fibrosis Patients J. Clin. Microbiol., October 1, 2007; 45(10): 3175 - 3183. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Atkinson, J. J. LiPuma, D. B. Rosenbluth, and W. M. Dunne Jr. Chronic Colonization with Pandoraea apista in Cystic Fibrosis Patients Determined by Repetitive-Element-Sequence PCR. J. Clin. Microbiol., March 1, 2006; 44(3): 833 - 836. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
