Rapid and Simultaneous Detection of bla_KPC and bla_NDM by Use of Multiplex Real-Time PCR

Abstract

The increasing incidence of carbapenem nonsusceptibility among clinically important species is of global concern. Identification of the molecular mechanisms underlying carbapenem nonsusceptibility is critical for epidemiological investigations. In this report, we describe a real-time PCR-based assay capable of simultaneously detecting bla_KPC and bla_NDM, two of the most important carbapenemases, directly from culture in less than 90 min. The assay was validated with bla_KPC- and bla_NDM-carrying clinical isolates and demonstrated 100% concordance with the Carba NP test.

TEXT

The potent antimicrobial activity of the carbapenems (ertapenem, doripenem, imipenem, and meropenem) is being increasingly compromised by the emergence and international dissemination of carbapenem-hydrolyzing β-lactamases (carbapenemases) (1–3). Culture-based methods, such as the modified Hodge test, combined with antibiotic susceptibilities remain the principal methods used to detect carbapenemase activity. These methods have limitations, including the inability to differentiate the molecular mechanism involved, extended turnaround times, and inconclusive results, particularly with nonmembers of the family Enterobacteriaceae, such as Acinetobacter baumannii and Pseudomonas aeruginosa (4–6). Molecular methods for the detection of carbapenemase genes provide several advantages over culture-based techniques, including high sensitivity and rapid turnaround time. Furthermore, they provide a critical resource for epidemiological investigations. Despite their advantages, the implementation of molecular methods has been slow, primarily because of concerns about allelic variation within target genes, primer cross-reactions, and high costs (7).

bla_KPC and bla_NDM have received global attention because of their widespread distribution (2, 8, 9), broad range of activity against β-lactams (3), and association with serious clinical infections (10, 11). As of this writing, the sequences of 13 bla_KPC and 7 bla_NDM alleles have been deposited at GenBank. However, unlike other important carbapenemase genes, such as bla_VIM and bla_IMP, allelic variation in bla_KPC and bla_NDM is very low.

The Multidrug-Resistant Organism Repository and Surveillance Network (MRSN) collects and characterizes multidrug-resistant organisms across the United States Military Health System to enhance infection prevention, inform empirical therapy, and influence policy (12). Over 20 military health care facilities worldwide, including those in war zones, submit an average of 300 clinical isolates monthly, comprising methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and multidrug-resistant, extremely drug resistant, and pan-drug-resistant Gram-negative bacteria. Thirty-five to 50 are submitted as carbapenem resistant, the majority of which are A. baumannii and P. aeruginosa. One hundred thirty-seven carbapenem-resistant members of the family Enterobacteriaceae have been submitted since 2010, with 97 displaying ertapenem resistance but imipenem/meropenem sensitivity phenotypes. All carbapenem-resistant isolates are tested for bla_KPC, bla_IMP, bla_OXA-48, bla_VIM, and bla_NDM with separate, individual real-time PCR assays. In this report, we describe a novel triplex assay that simultaneously detects all variants of bla_KPC and bla_NDM.

Primers and probes (Table 1) for bla_KPC and bla_NDM were designed with Beacon Designer 7.0 (Premier Biosoft International) by using an alignment of all of the sequence variants available in GenBank. An internal control based on the 16S rRNA gene was designed by using an alignment of the 16S rRNA gene sequences from 45 members of the family Enterobacteriaceae (see Table S1 in the supplemental material), A. baumannii, A. nosocomialis, A. pitii, and P. aeruginosa. The 16S rRNA gene primer and probe amplified a product from every organism tested, including clinical isolates of A. baumannii, Enterobacter aerogenes, Enterobacter cloacae, Escherichia coli, Klebsiella pneumoniae, and P. aeruginosa. Amplification was also achieved from 32 American Type Culture Collection strains (American Type Culture Collection, Manassas, VA) and 18 clinical isolates representing 50 different bacterial species (14) of the genera Achromobacter, Aeromonas, Alcaligenes, Burkholderia, Citrobacter, Hafnia, Kluyvera, Moraxella, Morganella, Proteus, Serratia, Salmonella, Shigella, and Stenotrophomonas (see Fig. S1 in the supplemental material). Primers and probes were validated and optimized individually and in a triplex reaction by using minimum information for the publication of quantitative real-time PCR experiments guidelines (13). The optimized triplex assay was tested against 26 bla_KPC2 (E. coli n = 4, E. cloacae n = 2, K. pneumoniae n = 20), 1 bla_KPC-3 (K. pneumoniae n = 1), and 3 bla_NDM-1 (A. baumannii n = 1, A. schindleri n = 1, Providencia stuartii n = 1) clinical isolates cultured from urine (30%), respiratory (20%), wound (17%), blood (10%), sterile tissue (7%), sterile fluid (3.3%), and surveillance (groin, 10%; rectal, 3.3%) samples collected between 2010 and 2012 (see Table S2 in the supplemental material). All genes were previously detected by the uniplex assays and confirmed by sequencing.

Table 1.

Primers and probes used in this study

Namea	Sequence (5′-3′)b	Concn (nM)c	Locationd	Efficiency (%)e
16SMP-F	AAGTCGGAATCGCTAGTAATCG	150	1335–1356
16SMP-R	ATGGTGTGACGGGCGGT	150	1401–1417	102.1
16SMP-P	6FAM-TGTACAAGG-ZEN-CCCGGGAACGTATTCA-IABkFQ	100	1375–1399
KPCMP-F	CTGTGCAGCTCATTCAAG	300	199–216
KPCMP-R	CATGCCTGTTGTCAGATA	300	331–348	95.4
KPCMP-P	HEX-TTCTTGCTG-ZEN-CCGCTGTGCTG-IABkFQ	250	221–240
NDMMP-F	GCCCAATATTATGCACCC	250	562–582
NDMMP-R	GTCGCCAGTTTCCATTTG	250	610–626	96.2
NDMMP-P	TEX615-CGTTGGGATCGACGGCACC-BHQ2	200	585–603

^aKPC primers and probe target all 13 variants of bla_KPC; NDM primers and probe target all 7 variants of bla_NDM.

^b16SMP-Probe and KPCMP-Probe contain a double internal ZEN quencher. 6FAM, 6-carboxyfluorescein; BHQ2, black hole quencher 2; HEX, hexachlorofluorescein; IABkFQ, Iowa black fluorescence quencher; TEX615, Texas Red 615.

^cOptimal primer and probe concentrations in the triplex assay were determined by titration as described previously (13). For uniplex reactions, all primer and probe concentrations should be 250 and 200 nM, respectively.

^dRelative to the first base pair of the coding region of bla_KPC-2 and bla_NDM-1. The 16S rRNA location is approximate.

^eEfficiency was calculated from the triplex assay. The 16S rRNA primers and probe have an efficiency of >100%, as the positive control was a combination of DNA from K. pneumoniae ATCC 1705 (bla_KPC-2) and A. baumannii NDM2 (bla_NDM-2). All primers have an efficiency of ≥98% when used alone. The R² values for all primers and probes were ≥98%.

Rapid DNA extraction (23-min protocol) was performed using Lyse-and-go (Thermo Scientific, Waltham, MA) as described previously (14). Real-time PCR was performed on a CFX96 cycler (Bio-Rad Laboratories, Hercules, CA) using iQ Multiplex Powermix (Bio-Rad) in 20-μl volumes. The cycling protocol was 95°C for 5 min and 40 cycles of 95°C for 10 s and 56°C for 40 s. Appropriate positive (K. pneumoniae ATCC 1705, bla_KPC-2⁺; K. pneumoniae NCTC 13443, bla_NDM-1⁺; A. baumannii NDM2, bla_NDM-2⁺), negative (ATCC 1706), and no-template (water) controls were incorporated into every plate. In parallel, all isolates were tested for carbapenemase production by using the Carba NP test as described previously (15). The total time for the entire procedure was 88 min.

bla_KPC- and bla_NDM-carrying isolates were recovered from a variety of clinical sites (see Table S2 in the supplemental material). All isolates were classified as nonsusceptible to ertapenem (≥2 μg/ml) (16) (see Table S2). Two isolates, MRSN 10266 and MRSN 11906, were intermediately susceptible to imipenem (2 μg/ml) but nonsusceptible to meropenem (≥4 μg/ml). Conversely, MRSN 10189 was susceptible to meropenem (≤1 μg/ml) but nonsusceptible to imipenem (≥4 μg/ml). These data support previous studies that have shown various susceptibilities to carbapenems in producers of K. pneumoniae carbapenemase and New Delhi metallo-β-lactamase (17–19). In our surveillance network, nonsusceptibility to ertapenem in members of the family Enterobacteriaceae is used as the main criterion to select isolates for gene screening. However, this criterion is not suitable for P. aeruginosa and Acinetobacter species and in our experience has led to unnecessary screening of Enterobacteriaceae with altered outer membrane porins or AmpC enzyme producers (18, 20). By using the Carba NP test, the number of isolates selected for real-time PCR can be greatly reduced.

The triplex assay was capable of detecting <100 genome copies of bla_KPC and bla_NDM and amplified the expected product from all of our bla_KPC- and bla_NDM-carrying isolates. In contrast, only the 16S primer and probe amplified a product from 97 ertapenem-resistant but imipenem/meropenem-sensitive Enterobacteriaceae and 45 imipenem-resistant clinical isolates of A. baumannii and Pseudomonas aeruginosa that were previously determined to be bla_KPC and bla_NDM negative by uniplex real-time PCR. All bla_KPC- and bla_NDM-carrying members of the family Enterobacteriaceae in this study were positive for carbapenemase production by the Carba NP test (15). The test was unable to detect New Delhi metallo-β-lactamase 1 production in Acinetobacter species, most likely because of weak enzyme production by Acinetobacter species. In Acinetobacter, the bla_NDM gene is disproportionally carried as a single copy on the chromosome, rather than on plasmids, as in many members of the family Enterobacteriaceae, where it may be present in higher copy numbers (21).

The triplex assay provides a rapid and accurate method for detecting bla_KPC and bla_NDM. The assay provides considerable advantages over methods that employ melting curve analysis and double-stranded DNA binding dyes (22), including the presence of an internal control to reduce false negatives, the ability to detect both bla_KPC and bla_NDM in the same target, and removal of the inherent variation in melting curves associated with allelic variation in target genes. After initial capital investments, the average cost per reaction is about $0.90. The MRSN now initially selects isolates for carbapenemase gene testing based on antibiotic susceptibilities, followed by the Carba NP test where appropriate, and finally real-time PCR.