Rapid and Simultaneous Detection of Genes Encoding Klebsiella pneumoniae Carbapenemase (bla_KPC) and New Delhi Metallo-β-Lactamase (bla_NDM) in Gram-Negative Bacilli

Abstract

We present a duplex, real-time PCR assay for detection of Klebsiella pneumoniae carbapenemase (bla_KPC) and New Delhi metallo-β-lactamase (bla_NDM) genes. Accuracy was assessed with 158 Gram-negative bacillary isolates, including 134 carbapenemase producers. The assay had 100% sensitivity and specificity compared with reference methods and a turnaround time of 90 min.

TEXT

Acquired carbapenem nonsusceptibility associated with carbapenemase production in Gram-negative bacilli is increasing, compromising treatment and raising concerns about nosocomial transmission (1, 2). Several genes encode carbapenemases, including serine carbapenemases bla_KPC and bla_OXA-48 and metallo-β-lactamases bla_IMP, bla_VIM, and bla_NDM (3). In the United States, Klebsiella pneumoniae carbapenemase (KPC) is most common, followed by New Delhi metallo-β-lactamase (NDM). Plasmids carrying these genes have a range of host organisms and are spread efficiently (4, 5).

Carbapenemase detection using the modified Hodge test is neither sensitive nor specific and is subjective and requires follow-up molecular methods to characterize the underlying mechanism in positive isolates (6–12). Molecular methods can detect and characterize carbapenemases, including KPC- and NDM-mediated resistance (9). Confirmation of the PCR product by sequencing or target-specific probes in real-time assays has been described (13–21), as have microarray and loop-mediated isothermal amplification (LAMP) detection (22–25).

Here we describe a LightCycler (Roche Molecular Diagnostics, Indianapolis, IN) duplex real-time PCR assay employing fluorescence resonance energy transfer (FRET) hybridization probe-based detection of bla_KPC and bla_NDM which, when coupled with simple pre-PCR colony lysis, yields a “colony-to-result” time of 90 min, faster than any previously described assay. We validated our assay using a large number of carbapenemase-producing isolates, including Enterobacteriaceae and non-Enterobacteriaceae.

(This work was presented in part at the joint 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy [ICAAC]-46th Annual Meeting of the Infectious Diseases Society of America [IDSA], Washington, DC, 2008.)

Primers targeting bla_KPC (GenBank accession no. AF297554.1) and bla_NDM (GenBank accession no. FN396876.1) with exact sequence matches to all known genotypes of KPC (1 to 14) and NDM (1 to 7) were designed using LightCycler probe design software, version 2.0 (Roche Applied Science, Indianapolis, IN) (primers and probes are shown in Table 1). K. pneumoniae BAA-1705 (ATCC, Manassas, VA), containing bla_KPC, and K. pneumoniae NCTC 13443 (Health Protection Agency [HPA], London, United Kingdom), containing bla_NDM-1, were used as controls.

Table 1.

Primers and probes used in this study^a

Primer or probe	Sequence
Primer
KPC160F	5′ ATTGGCTAAAGGGAAACACGACC 3′
KPC160R	5′ GTAGACGGCCAACACAAT 3′
NDM1F	5′ ATTAGCCGCTGCATTGAT 3′
NDM1R	5′ GGCATGTCGAGATAGGAAGT 3′
Probe
KPC160fl	5′ GAACCGCGGAGTGTATGGCACGG FITC 3′
KPC160iLC610	5′ AAATGACTATGCCGTCGTCTGGCCCACT Red610 3′
NDMfl	5′ CAACGGTTTGGCGATCTGGT FITC 3′
NDMiLC670	5′ Red670 TCCGCCAGCTCGCACCG TPO4 3′

^abla_KPC and bla_NDM-1 10× primer-probe set (number 1716; TIB MolBiol, Adelphia, NJ).

Each 15-μl master mix aliquot contained the following components: 9 μl molecular-grade water, 1.6 μl 10× DNA Master HybProbe mix (containing Taq DNA polymerase, reaction buffer, deoxyribonucleoside triphosphates with deoxyuridine triphosphate [dUTP] substituted for deoxyribosylthymine triphosphate [dTTP], and 1 mM MgCl₂) (Roche Applied Science), 2.4-μl volume of 25 μM MgCl₂ (supplemental), and 2.0 μl 10× primer-probe set 1716 (TIB MolBiol, Adelphia, NJ). Five microliters of colony lysate was added to 20 μl LightCycler reaction cuvettes containing the master mix. Cycling conditions were as follows: 95°C for 10 min; 45 cycles of 10 s at 95°C, 15 s at 55°C, and 15 s at 72°C; melting curve analysis for 0 s at 95°C, 20 s at 59°C, 20 s at 40°C (ramp rate of 0.2°C/s), and 0 s at 85°C (ramp rate of 0.2°C/s and continuous acquisition); and cooling for 30 s at 40°C. Analytical sensitivity for both targets was 10 CFU/μl. There was no cross-reactivity against a panel of 12 organisms (see Table S1 in the supplemental material).

Fifty-seven Enterobacteriaceae or nonfermenting Gram-negative bacillus isolates (46 of which were ertapenem nonsusceptible) were studied at Mayo Clinic (see Table S2 in the supplemental material). Isolation plate primary inoculation areas were gently swept using an inoculation loop and transferred to methicillin-resistant Staphylococcus aureus (MRSA) lysis tubes (Roche Molecular Diagnostics). Suspensions were heated and physically disrupted on a Thermomixer R (Eppendorf AG, Germany) for 6 min at 1,400 rpm, followed by centrifugation at 20,800 × g for 2 min. Thirty isolates were positive for bla_KPC, and three isolates were positive for bla_NDM. Concordant results were obtained with the Centers for Disease Control and Prevention's duplex real-time PCR assay (26).

To enrich for NDM-positive isolates, 101 isolates of Enterobacteriaceae or nonfermenting Gram-negative bacilli, which had been shown previously to produce an NDM-type carbapenemase by metallo-β-lactamase gene PCR, were additionally studied at the HPA (see Table S2 in the supplemental material). Colonies from overnight culture were suspended in 100 μl water, heated for 5 min at 95°C, and centrifuged at 8,000 rpm for 5 min; supernatant was tested by PCR. All were positive for bla_NDM (as expected) by the novel real-time PCR assay.

Recent publications have described molecular assays for carbapenemase genes. Spanu et al. evaluated 300 clinical isolates using the commercial NucliSENS EasyQ KPC (bioMérieux, Marcy l'Etoile, France) assay and found 100% sensitivity and specificity and an analytical sensitivity of 4 CFU/reaction (16). Manchanda et al. evaluated a laboratory-developed real-time PCR assay targeting bla_NDM-1 by assaying 34 clinical isolates and compared results to a conventional PCR assay; concordant results and a sensitivity of 10 copies/reaction were reported (18). Diene et al. evaluated a collection of 44 clinical isolates using a real-time PCR assay targeting bla_NDM-1; results were concordant with conventional PCR, although only a single bla_NDM-1-positive isolate was studied (19). Ong et al. described a real-time PCR assay for bla_NDM-1 using hydrolysis probes and assayed 47 isolates (12 of which were positive), reporting 100% sensitivity and specificity compared with conventional PCR and a limit of detection of 35 CFU/reaction (21). Qi et al. described a LAMP assay for detection of bla_NDM-1 and tested it on 345 veterinary isolates also characterized by conventional PCR; their study included only a single bla_NDM-1-positive isolate (22). Finally, Monteiro et al. evaluated a multiplex real-time PCR assay, which detected six resistance genes, including bla_KPC and bla_NDM-1, using high-resolution melting-curve analysis. Fifty-eight isolates, which had been previously characterized by PCR and sequencing, were evaluated with 100% concordance (20). A commercial assay, hyplex SuperBug ID (Amplex, BioSystems GMBH, Geiβen, Germany), which detects bla_KPC and bla_NDM-1 and other carbapenemase genes in 2.5 to 4 h, has been recently described (15). Although these assays performed equivalently to our assay, all except two used preparatory DNA extraction/purification and all had longer turnaround times than our 1.5-h estimate. The only reports that used a comparable lysis method either targeted only bla_KPC (16) or did not use real-time detection (15). Additionally, none of the reports detail the specificity of assay design for detection of NDM genotypes other than NDM-1; based on in silico analysis, NDM-1 through -7 would be detected by our assay.

Two evaluations utilizing microarrays that target multiple resistance genes reported similar results compared with conventional and real-time PCR methods (23, 24, 27, 28). While more information can be gained from this approach, the turnaround time is long and the instrumentation and expertise required are beyond the scope of most clinical laboratories.

In summary, this is the first description of a FRET hybridization probe-based real-time PCR assay that targets bla_KPC and bla_NDM in a single assay. It is simple to perform, as evidenced by its implementation in our laboratories in the United States and the United Kingdom. Including the rapid preparatory lysis procedure, results are obtained within 90 min, and the assay performs as well as a reference assay for use in the public health arena. It offers rapid detection and differentiation of bla_KPC and bla_NDM in multidrug-resistant Gram-negative isolates, resistance mechanisms that are important causes of carbapenem resistance worldwide, and provides information for patient care and for limiting the spread of resistant bacteria.