A rapid low-cost real-time PCR for the detection of klebsiella pneumonia carbapenemase genes

Background Klebsiella pneumonia carbapenemases (KPCs) are able to hydrolyze the carbapenems, which cause many bacteria resistance to multiple classes of antibiotics, so the rapid dissemination of KPCs is worrisome. Laboratory identification of KPCs-harboring clinical isolates would be a key to limit the spread of the bacteria. This study would evaluate a rapid low-cost real-time PCR assay to detect KPCs. Methods Real-time PCR assay based on SYBR GreenIwas designed to amplify a 106bp product of the blaKPC gene from the159 clinical Gram-negative isolates resistant to several classes of -lactam antibiotics through antimicrobial susceptibility testing. We confirmed the results of real-time PCR assay by the conventional PCR-sequencing. At the same time, KPCs of these clinical isolates were detected by the modified Hodge test (MHT). Then we compared the results of real-time PCR assay with those of MHT from the sensitivity and specificity. Moreover, we evaluated the sensitivity of the real-time PCR assay. Results The sensitivity and specificity of the results of the real-time PCR assay compared with those of MHT was 29/29(100%) and 130/130(100%), respectively. The results of the real-time PCR and the MHT were strongly consistent (Exact Sig. (2-tailed) =1. 000; McNemar test). The real-time PCR detection limit was about 0.8cfu using clinical isolates. Conclusion The real-time PCR assay could rapidly and accurately detect KPCs -harboring strains with high analytical sensitivity and specificity.


Background
Carbapenems are widely used to treat serious infections caused by multi-resistant Gram-negative bacteria. However, beginning with the initial description of a novel KPC from an isolate of K. pneumoniae in 2001 [1], carbapenem resistance in Enterobacteriaceae has been rapidly increasing. KPCs are able to hydrolyze the carbapenems, and cause resistance to multiple classes of antibiotics. Treatment of KPC-producing bacterial infection is thus a considerable challenge for clinicians. KPCs have been reported worldwide, such as North America, South America, Greece, Israel, Puerto Rico, China and so on [2][3][4][5][6]. The expanding geographic spread of KPCs underscores the importance of clinical recognition of these enzymes. In addition, KPCs have been found in bacteria other than K. pneumoniae, including K. oxytoca [7,8], P. mirabilis [9], Acinetobacter spp [6], P. aeruginosa , C. freundii [10], S. marcescens and E. coli [11]. This rapid dissemination of KPC is worrisome. Laboratory identification of KPC-harboring clinical isolates will be critical for limiting the spread.
However, detection of KPC -harboring stains in the clinical laboratory remained a difficult task. The failure of automated susceptibility testing systems to detect KPC-mediated carbapenems resistance was previously reported [12][13][14]. In 2009, the Clinical Laboratory Standards Institute (CLSI) guidelines (M100) recommended MHT to detect carbapenemase production. Wang et al. [15] noted that false positive results could occur when the MHT was used to detect carbapenemase in ESBLproducing isolates. MHT is time-consuming and not routinely tested for E. cloacae, P. aeruginosa in laboratory, so that many molecular detection of bla KPC genes were evaluated [16][17][18][19]. Rapid and sensitive bla KPC assays are critical to control the spread of bla KPC -harboring bacteria in hospitalized patients.
In this paper, we would describe the development of a low-cost real-time PCR assay to screen clinical isolates for bla KPC .
The 159 clinical isolates including K. pneumoniae, E. coli, E. cloacae, K. oxytoca, S. marcescens, P. mirabilis, MDR A.baumanii and MDR P. aeruginasa. were recovered from multiple infection sites like blood, wound, sputum, catheter, urine and pleural effusion from Beijing Tongren Hospital. These clinical isolates were resistant to several classes of β-lactam antibiotics, which were identified by antimicrobial susceptibility testing.

Antimicrobial susceptibility testing
Antimicrobial susceptibility testing was performed with the Vitek 2 susceptibility card AST-GN13 by Vitek 2 automated system (BioMérieux Inc, Durham, NC) according to the manufacturer's instructions. Minimum inhibitory concentration (MIC) results of imipenem and ertapenem were classified as susceptible, intermediate, or resistant based on the 2010 CLSI breakpoints (susceptible, ≤ 1 μg/ml and ≤ 0.25 μg/ml; intermediate, 2 μg/ml and 0.5 μg/ml; resistant, ≥ 4 μg/ml and ≥ 1 μg/ml, respectively). However, the AST-GN13 card cannot classify organisms as susceptible to ertapenem without the dilutions less than 0.5 μg/ml. All clinical isolates were subsequently tested by MHT. The indicator strains in MHT were E. coli ATCC 25922 for Enterobacteriaceae and K.pneumoniae ATCC 700603 for non-Enterobacteriaceae like P.aeruginosa [20].

DNA isolation
Bacterial strains were grown on MacConkey agar and incubated overnight at 35°C. One colony was resuspended in 100 μl of sterile distilled water and the cells were lysed by heating at 100°C for 10 min. Cellular debris was removed by centrifugation at 13000 g for 10 min, and the supernatant was used as a source of template DNA for amplification.
For analytical sensitivity based on bacterial colonyforming unit (cfu), DNA isolation was performed using the DNeasy Blood&Tissue Kit (Qiagen Sciences, Maryland, USA) according to the protocol suggested by the manufacture. In brief, a bacterial suspension equivalent to that of a 2.0 McFarland standard was prepared in saline, then 200 μl (8.0 × 10 7 cfu ) suspension were serially diluted 10-fold in saline. Bacterial total nucleic acid was extracted from 200 μl of each dilution and then eluted in 50 μl elution buffer and stored at −20°C.

bla KPC detection by PCR -sequencing
The presence of bla KPC was confirmed by conventional PCR and sequencing [1]. The primers included the forward (5'-TGTCACTGTATCGCCGTC-3') and the reverse (5'-CTCAGTGCTCTACAGAAAACC-3') , The PCR reaction system contained 0.5 μM each primer, 2 × EasyTaq PCR SuperMix (TransGen Biotech, Beijing, China) and 2 μl DNA template. The reactions were amplified in a My Cycler thermal cycler (BIO-RAD, USA). Cycling parameters were 5 min at 95°C, followed by 35 cycles of 1 min at 95°C, 30s at 58°C, and 1 min 30 s at 72°C. The PCR amplification was ended by a final extension at 72°C for 10 min. sequencing of the PCR products was commercially performed by SinoGenoMax Co. Ltd (Beijing, China). For sequence analysis, the BLAST program from the National Center for Biotechnology Information Web site was used (http: //www. ncbi. nlm. nih.gov/BLAST).
The 25 μl real-time PCR mixture contained 12. 5 μl TransStart Green qPCR super MIX (TransGen Biotech, Beijing, China), 0.5 μl PCR enhancer (TransGen Biotech, Beijing, China), 0.2 μM each primer, 9 μl sterile distilled water and 2 μl DNA template. Real-time PCR amplification was performed using the Roche Light cycler 480 Realtime system (Roche Diagnostics, Mannheim, Germany). Cycling parameters were 5 min at 95°C, followed by 40 cycles of 15 s at 95°C, 15 s at 55°C, and 30 s at 72°C. Single fluorescence detection was performed in each cycle at 55°C . Melting curve acquisitions were done immediately after the final amplification step by heating at 96°C for 5 s, cooling to 55°C for 1 min, and heating slowly at 0.11°C per second to 96°C with continuous fluorescence recording. Melting curves were recorded by plotting fluorescence signal intensity versus temperature. Amplicon melting temperatures(Tm) were determined by calculating the derivative of the curve using Roche Light cycler 480 software. The results were visualized by plotting the negative derivative against temperature.

Specificity and sensitivity
In order to determine analytical sensitivity of our assay, bla KPC real-time PCR experiments were performed on 10-fold serial dilutions of bacterial cultures (8.0 × 10 7 cfu). To evaluate the analytical specificity, a panel of reference stains and clinical strains resistant to several classes of βlactam antibiotics was tested. For statistical analysis, we used the MHT as the reference standard. The differences between sensitivities of the real-time PCR assay and MHT were evaluated with the McNemar test.

Results
The specificities of the real-time PCR primers for the detection of bla KPC genes were evaluated by the BLAST search program, available at www.ncbi.nlm.nih.gov.
All bla KPC genes of KPC-producing isolates in this study were verified as bla KPC-2 by sequencing assay.
The bla KPC amplicon was distinguished by its specific Tm value. Under our experimental conditions, analysis of the melting curve profile of the PCR products indicated that the products peaked at about 89°C ( Figure 1).
The analytical sensitivity of the bla KPC real-time PCR assay was determined after serially diluting known concentrations (8.0 × 10 7 cfu) of clinical isolated carbapenemsresisant K. pneumoniae. The dynamic range of the assay covered nine orders of magnitude from 8.0 × 10 7 to 0.8 cfu. bla KPC specific fluorescent peaks were detected in the isolates dilutions to about 0.8 cfu (Figure 2).

Discussion
Along with the wide use of carbapenem antibiotics, KPCs appeared a major public health concern. Bacterial isolates producing KPCs are able to hydrolyze a broad spectrum of β-lactams including the penicillins, cephalosporins, carbapenems and monobactam. They have the potential to spread rapidly in hospital environments to cause nosocomial infections with high mortality rates [21]. KPC-producing Enterobacteriaceae stains are increasingly spreading throughout China [2,9,11,22]. The dominant clone of KPC-producing K. pneumoniae in China is ST11, which is closely related to ST258 reported worldwide [23]. A rapid method confirming KPCs is significant to control this spread.
In 2009, the CLSI recommended MHT to screen for the production of carbapenemase in Enterobacteriaceae  coli with MIC to imipenem as low as 1 μg/ml was confirmed as MHT ( + ) / bla KPC ( + ) in our study. Decreased ertapenem susceptibility has been considered as one of the most sensitive phenotypic indicators of KPC production, but it has been found to be nonspecific [24,25]. In our laboratory, two clinical isolates MICs to ertapenem as high as 2 μg/ml to 8 μg/ml were MHT ( − ) / bla KPC ( − ). Despite CLSI new recommendations, our laboratory continued to confirm KPC using MHT or PCR. The sensitivity and specificity of the MHT have been shown to exceed 90 %; however, several reports have noted false positive results occurred when the MHT was used to detect carbapenemase in ESBL-producing isolates [15,26]. In addition, it may not be the ideal phenotypic confirmatory test for KPCs since interpretation can be difficult for some isolates such as A. baumanii, P. aeruginasa. In our study, we adjusted the  indicator stain to K. pneumoniaeATCC 700603 for non-Enterobacteriaceae in order to eliminate the incidence of indeterminate results of MHT [20]. Thus, an alternative method may prove to be more useful. During the recent few years, molecular methods have been used to rapidly detect bla KPC genes. In particular, real-time PCR assays offered the advantage of shorter turnaround time, which were even developed to detect KPC-containing strains with high analytical specificity and sensitivity in surveillance specimens [27,28].
In this study, we validated a rapid, sensitive, and specific real-time-PCR assay for the detection of bla KPC genes. This assay can be performed in less than 4 hours, which will reduce the chance of spreading the organism in the hospital. The real-time PCR assay specifity and sensitivity were 100 % compared to phenotypic KPC activity assessed by MHT and sequencing. Thirteen KPC gene variants have been described, classified in sequential numeric order from bla KPC-1/2 to bla KPC-13 . The bla KPC genes are characterized by nonsynonymous single nucleotide substitutions [17]. Our sequencing results showed all 29 KPC-producing isolates harbored bla KPC-2 gene. KPC-2 clinical isolates were widely isolated in most parts of China [23,29]. Last year, Li et al [30] in China firstly described KPC-3-harboring E. coli and C. freundii. Although KPC-2 and KPC-3 were well described throughout China, we designed the primers in conservative areas to ensure that our assay could almost detect the variants currently described. We identified bla KPC genes by melting curve analysis of the amplification product using SYBR GreenIwith many advantages like low-cost and easy to use. The Tm value of the bla KPC gene was detected at about 89°C. Our assay sensitivity is about one cfu sufficient to detect bla KPC -containing isolates.

Conclusions
The real-time PCR assay described here provides a useful screening test to detect bla KPC genes rapidly and accurately. Although the real-time PCR assay was unable to identify the specific gene in the bla KPC family in clinical isolates, accurate and rapid identification of this kind of resistance genes is the first step to control their spread.