Sequence-Selective Recognition of Extended-Spectrum β-Lactamase GES-2 by a Competitive, Peptide Nucleic Acid-Based Multiplex PCR Assay

Abstract

Extended-spectrum β-lactamases (ESBLs) in Pseudomonas aeruginosa, such as GES-2, which compromises the efficacy of imipenem, tend to be geographically restricted. The CC-to-AA base pair substitution at positions 493 and 494 of the bla_GES-2-coding region distinguishes this ESBL from bla_GES-1 and the bla_IBC-type genes, making it an ideal target for the development of a novel sequence-specific, peptide nucleic acid (PNA)-based multiplex PCR detection method. By using two primer pairs in conjunction with a PNA probe, this method provided an accurate means of identification of bla_GES-2 compared to standard PCR and gene sequencing techniques when it was used to test 100 P. aeruginosa clinical isolates as well as previously published, well-described control strains encompassing all presently known genes in the bla_GES-IBC ESBL family. This novel method has the potential to be used in large-scale, cost-effective screening programs for specific or geographically restricted ESBLs.

After the discovery of the extended-spectrum β-lactamase (ESBL) GES-2 from Pseudomonas aeruginosa in Pretoria, South Africa, in 2000 (24), it became clear that class 1 integron-borne ESBLs were established in the South African nosocomial setting. The same strain subsequently caused a nosocomial outbreak in three intensive care units (ICUs) in a teaching hospital, in which the patients had diverse clinical presentations and during which several patients succumbed (25). The P. aeruginosa strain described during that outbreak exhibited a tendency to colonize and infect mostly debilitated patients, including human immunodeficiency virus-positive patients, significantly increasing both their lengths of stay in the ICU and the cost of treatment (25). Detection of GES-2 and other ESBL-producing P. aeruginosa isolates in the clinical microbiology laboratory by ordinary methods is notoriously difficult for various reasons, including (i) false-negative results due to naturally occurring β-lactamases, such as the chromosome-encoded AmpC enzymes, which may be overexpressed; (ii) the simultaneous presence of metallo-enzymes with carbapenem-hydrolyzing activities (the IMP and VIM series) or with extended-spectrum oxacillinases (OXA-2 and OXA-10 derivatives); (iii) relative resistance to inhibition by clavulanate, as exemplified by GES-2; and (iv) combined mechanisms of resistance, such as impermeability and efflux (29). When an ESBL is suspected in P. aeruginosa, PCR-based molecular techniques may help to detect the gene with a series of primers designed for recognition of class A β-lactamase genes in this species (29). However, PCR experiments without further sequencing of the amplification products could not accurately differentiate between all the genes in a β-lactamase family, substantially adding to the costs of such an exercise. Other methods, such as isoelectric focusing analysis, may just indicate the presence of acquired β-lactamases rather than identify the ESBL precisely (18, 29). Descriptions of the bla_GES-1 gene (5, 26) and the bla_IBC-type genes (8, 13) from different geographical locations further complicates the laboratory identification of bla_GES-2, the more potent and, at present, the more geographically restricted gene in this family (24, 29). Since the initial description of the peptide nucleic acid (PNA) in 1991 (17), interest in this synthetic polymer as a molecular device has steadily increased. The hallmark of PNA that stimulated this interest is its intrinsic high affinity and specificity for complementary nucleic acids (17), making it the ideal nucleic acid molecule for recognition of point mutations without the need for highly specialized equipment. The fact that the GES-IBC β-lactamase family is still relatively small and the fact that it has only a few point mutations separating the genes in question (29) make it the ideal model for the development of a novel sequence-specific, PNA-based method for the identification of geographically restricted ESBLs suited to large-scale, low-cost screening programs.

MATERIALS AND METHODS

Bacterial strains.

One hundred clinical isolates of P. aeruginosa, collected and cryopreserved by the Department of Medical Microbiology, University of Pretoria, Pretoria, South Africa, were included in this study. Isolates were identified with the API 20NE system (bioMérieux, Marcy l'Etoile, France), according to the instructions of the manufacturer. The control bacterial strains used in this study are listed in Table 1.

TABLE 1.

Well-characterized bacterial strains used in this study

Strain	Relevant properties	Source or reference
P. aeruginosa GW-1	blaGES-2-producing isolate	24
P. aeruginosa Pu21	blaGES-2 transconjugant isolate	24
K. pneumoniae ORI-1	blaGES-1-producing isolate	26
E. cloacae HT-9	blaIBC-1-producing isolate	8
E. coli brcpHT-8	blaIBC-1 transconjugant isolate	8
E. coli DH5α	blaIBC-2 transconjugant isolate	13
E. coli ATCC 25922	blaGES- and blaIBC-negative strain	ATCCa

^aATCC, American Type Culture Collection, Manassas, Va.

Susceptibility testing.

Antibiotic-containing disks (Mast Diagnostics, Merseyside, United Kingdom) were used to determine routine laboratory antibiograms by the disk diffusion assay, as described and interpreted according to the guidelines of the National Committee for Clinical Laboratory Standards (NCCLS) (15). Resistance to ceftazidime was an absolute requirement for inclusion in the study.

DNA extraction.

Extraction of whole-cell DNA was performed by a precipitation-based method, as described previously (23), in combination with a repeated final washing step with ethanol to improve template purity (9). DNA pellets were dried in a DNA Speed Vac 110 instrument (Savant Instruments Inc., Farmingdale, N.Y.) and resuspended in 1 ml of TE buffer (10 mM Tris-HCl, 1 mM EDTA [pH 7.4]). Extraction of DNA from each isolate was verified by electrophoresis at 2 V/cm for 1 h in a 0.8% agarose gel containing ethidium bromide (0.5 μg/ml) in TBE buffer (45 mM Tris-borate, 1 mM EDTA [pH 7.4]), and the extracted DNA was visualized under UV light.

Standard PCR amplification. (i) Standard PCR primer construction.

A primer pair that anneals internally to the binding sites of previously published primers GES-1A and GES-1B (24) was constructed from the gene sequence of bla_GES-2 (GenBank accession number AF326355 [24]) by using Primer3 software (Whitehead Institute for Biomedical Research), which is available from http://www.inqaba.com. The following primers were generated: primer GES-C (5′-GTT TTG CAA TGT GCT CAA CG-3′; positions 176 to 195 of the coding region; melting temperature [T_m] = 54.6°C) and primer GES-D (5′-TGC CAT AGC AAT AGG CGT AG-3′; positions 527 to 546; T_m = 55.6°C) (Table 2). These primers target a 371-bp region that includes all presently known point mutations of the bla_GES and bla_IBC genes (5, 8, 13, 24, 26). The primers were synthesized and purified by Integrated DNA Technologies Inc., Coralville, Iowa.

TABLE 2.

Oligonucleotide sequences used in this study

Procedure and primer or probe	Sequencea	Function	Source or references
Standard PCR and DNA sequencing
GES-C	GTT TTG CAA TGT GCT CAA CG	Forward primer	This study
GES-D	TGC CAT AGC AAT AGG CGT AG	Reverse primer	This study
PNA-based multiplex PCR
GES-1A	ATG CGC TTC ATT CAC GCA C	Forward primer	24, 29
GES-1B	CTA TTT GTC CGT GCT CAG G	Reverse primer	24, 29
GES-E	GTG TGT TGT CGT TCA TCT C	Reverse primer complementary to blaGES-2	This study
GES-F	CCT GGC GAC CTC AGA GAT AC	Forward primer	This study
PNA probe	GTT GTC GCC CAT CTC	Complementary to blaGES-1 and blaIBC	This study

^aSequences are delineated 5′ to 3′. Relevant point mutations are depicted in boldface.

(ii) Standard PCR amplification and detection.

Primers GES-C and GES-D were used under standard PCR conditions (28) to amplify a 371-bp product. The PCR mixture comprised a 0.32 μM concentration of each primer, 5 μl of Taq DNA reaction buffer, 1.5 mM MgCl₂, a 200 μM concentration of each deoxynucleoside triphosphate, 1.25 U of Taq DNA polymerase (Promega Corp., Madison, Wis.), 2 μl of DNA template, and distilled water to a final reaction volume of 50 μl. Amplification was conducted on a GeneAmp PCR system 9600 (Perkin-Elmer Cetus, Emeryville, Calif.) and consisted of an initial denaturation step at 95°C for 2 min, followed by 35 amplification cycles each comprising a denaturation step at 95°C for 30 s, followed by an annealing step at 50°C for 1 min and an extension step at 72°C for 1 min. After completion of the 35 amplification cycles, an extension step was performed at 72°C for 5 min. The PCR products were visualized by electrophoresis at 4 V/cm for 45 min in a 0.8% agarose gel containing ethidium bromide (0.5 μg/ml). The amplicon size was verified with a 100-bp DNA ladder (Promega Corp.). DNA templates obtained from P. aeruginosa GW-1, bla_GES-2 transconjugant P. aeruginosa Pu21, Klebsiella pneumoniae ORI-1, Enterobacter cloacae HT-9, bla_IBC-1 transconjugant Escherichia coli brcpHT-8, and bla_IBC-2 transconjugant E. coli DH5α (Table 1) served as positive PCR controls. E. coli ATCC 25922 was used as a bla_GES-IBC negative PCR control. All PCR experiments were performed in duplicate. Reaction mixture preparation was done in separate rooms equipped with laminar-flow cabinets to avoid cross-contamination with amplification products.

Competitive PNA-based multiplex PCR.

The novel competitive PNA-based multiplex PCR method for the detection of point mutations in ESBL genes by PCR relies on two features of PNA: (i) its intrinsically high affinity for DNA (17) and (ii) its inability to serve as a primer for DNA polymerases (21). This method uses a forward primer in conjunction with a reverse primer that is complementary to the desired mutant sequence, as well as a PNA probe similar to the reverse primer, but complementary to the wild type (wt) sequence. When the reaction contains a wt DNA template, the PNA probe outcompetes the reverse primer for binding and no amplification occurs; conversely, the opposite happens when a mutant sequence is present during the reaction (20). A second set of primers targets an area downstream from the mutation site in both wt and mutant genes; this then serves as an internal amplification control measure.

(i) Multiplex PCR primer construction.

In addition to primers GES-1A and GES-1B (24), primers for multiplex PCR analysis were constructed in a fashion similar to that described for the standard PCR method. The following primers were generated: primer GES-E (5′-GTGT GTT GTC GTT CAT CTC-3′; positions 487 to 505 of the bla_GES-2-coding region; T_m = 51.8°C) and primer GES-F (5′-CCT GGC GAC CTC AGA GAT AC-3′; positions 505 to 524; T_m = 57.4°C) (Table 2).

(ii) Peptide nucleic acid probe construction.

A mixed-sequence PNA probe was designed to be complementary to positions 487 to 501 of the coding regions of bla_GES-1, bla_IBC-1, and bla_IBC-2 (GenBank accession numbers AF156486, AF208529, and AF329699, respectively [8, 13, 26]) by using Custom PNA Probe Designer software (Applied Biosystems, Rotkreutz, Switzerland), which is available from http://www.appliedbiosystems.com. The 3′ l-lysine-labeled probe sequence (20) 5′-GTT GTC GCC CAT CTC-3′ (T_m = 70°C) was synthesized in antiparallel binding mode and was purified by Boston Probes, Bedford, Mass.

(iii) Reaction mixture preparation and conditions.

In a multiplex PCR assay, primer GES-1A with primer GES-E and primer GES-F with primer GES-1B were used to amplify 505- and 360-bp products, respectively. The PCR mixture comprised a 0.32 μM concentration of each primer, 5 μl of Taq DNA reaction buffer, 1.5 mM MgCl₂, a 200 μM concentration of each deoxynucleoside triphosphate, 1.25 U of Taq DNA polymerase (Promega Corp.), 2 μl of DNA template, and distilled water to a final reaction volume of 50 μl. Amplification and detection were performed as described above for the standard PCR procedure. Once the reaction conditions were standardized with well-described bacterial isolates (Table 1), the PNA probe was added to subsequent multiplex PCR mixtures to a final concentration of 0.32 μM to compete with reverse primer GES-E.

DNA sequencing analysis.

Sequencing of both the forward and the reverse strands of the standard PCR products was performed on a SpectruMedix model SCE 2410 automated sequencer (Spectru Medix, State College, Pa.) by incorporation of the ABI Big Dye Terminator Cycle Sequencing kit (version 3.1; Applied Biosystems, Foster City, Calif.). Electropherograms of the sequences generated were inspected with Chromas software (version 1.45; Technelysium Pty. Ltd., Helensvale, Queensland, Australia). The DNA sequences obtained were analyzed with the BLAST program (1), which is available from http://www.ncbi.nlm.nih.gov. A PCR product obtained from P. aeruginosa GW-1 (Table 1) was used as a control.

RESULTS

Detection of bla_GES-IBC by standard PCR.

Clinical isolates of P. aeruginosa collected and cryopreserved from 1998 to 2001 by the Department of Medical Microbiology, University of Pretoria, originated from a variety of specimens, including endotracheal aspirates, urine samples, pus swabs, blood cultures, tissue biopsy specimens, and intravenous catheters. Specimens originated mainly from ICUs serving internal medicine, surgery, pediatrics, and neurosurgery departments, while the remainder of the specimens originated from general hospital wards, including orthopedic and gynecologic wards. Although various zone size diameters were observed, all isolates (n = 100) were resistant to ceftazidime by the disk diffusion assay, according to the guidelines of NCCLS (15) (data not shown). Repeated standard PCR amplification resulted in 51 of the isolates testing positive for a bla_GES-IBC gene product of 371 bp. A 371-bp PCR product was obtained from each control strain (Table 1) except the bla_GES-IBC-negative control, E. coli ATCC 25922.

DNA sequencing of standard PCR products.

By sequence analysis of both the forward and the reverse strands of the standard PCR products (n = 51), all products corresponded to bla_GES-2. Each sequence was thoroughly inspected for signature point mutations, such as the 5′-GCT-3′ motif (positions 358 to 360 of the coding region), which differentiates bla_GES- from bla_IBC-type genes (26), and the 5′-GAA-3′ motif (positions 492 to 494), which differentiates bla_GES-2 from bla_GES-1 and bla_IBC-type genes (24). In addition to the specific motifs mentioned above, a stem-loop in the secondary DNA structure (positions 466 to 470 and 482 to 486 of the coding region) was noted.

PNA-based, sequence-specific detection of bla_GES-2.

Fifty of the 51 isolates positive by the standard PCR method were positive for bla_GES-2 when the 100 clinical isolates were tested by sequence-specific multiplex PCR amplification. A repeat multiplex PCR with the DNA template volume increased from 2 to 5 μl yielded a positive result for the isolate that did not initially test positive by the sequence-specific multiplex PCR. Two distinct patterns were detected when the amplification products were subjected to gel electrophoresis: the GES-2-producing clinical isolates and control strains produced two distinct bands of 505 and 360 bp, respectively, while GES-1- and IBC-producing control strains produced only the 360-bp internal amplification control band, clearly distinguishing them from GES-2-producing isolates (Fig. 1). Control strains P. aeruginosa GW-1 and P. aeruginosa Pu21 (24) both tested positive for bla_GES-2 in repeated reactions, while repetitive results for bla_GES-1- and bla_IBC-producing strains (Table 1) were obtained. The E. coli ATCC 25922 isolate tested negative in all reactions. These results corresponded with the results obtained by the standard PCR amplification and sequencing reactions.

FIG. 1.

Gel electrophoresis depicting the PNA-based, sequence-specific PCR products obtained from the well-characterized bacterial isolates listed in Table 1. Lanes MW, 100-bp marker (Promega Corp.) (the 500-bp segment is indicated); lane 1, E. cloacae HT-9; lane 2, P. aeruginosa GW-1; lane 3, E. coli brcpHT-8; lane 4, E. coli ATCC 25922; lane 5, K. pneumoniae ORI-1; lane 6, P. aeruginosa Pu21; lane 7, IBC-2 transconjugant E. coli DH5α.

DISCUSSION

Despite considerable efforts by various investigators, detection of ESBLs in P. aeruginosa still remains a problem due to the notoriously low sensitivities of easy-to-perform susceptibility tests and the laborious and often expensive molecular detection methods (29). Apart from standard PCR and gene sequencing, molecular biology-based methods for the detection of ESBLs in general have been developed for a wide array of applications most suited for the TEM- and SHV-type ESBL families (4). These methods may include oligonucleotide typing (12), PCR-restriction fragment length polymorphism analysis (2, 19), single-strand conformation polymorphism analysis (14), real-time PCR with melting-curve analysis (27), and ligase chain reaction-based tests (10, 16). In addition to the aforementioned methods, this study describes the application of a PNA-based method as a novel means for the detection of point mutations in ESBL genes of the GES-IBC family.

Data from studies that used nuclear magnetic resonance (11) and X-ray crystallography (3) have shown that mixed-sequence PNA forms duplexes with cDNA sequences by way of base-specific hydrogen bonds with a 1:1 stoichiometry, which obeys Watson-Crick hydrogen-bonding rules (7). The hybridization properties of mixed-sequence PNA and, consequently, the high affinity of the interaction of PNA with DNA can be attributed to the lack of a negative charge in the PNA backbone, as demonstrated by melting-point analysis at high ionic concentrations (>1 M Na⁺) (7). Previous analysis of PNA base-mismatch recognition in DNA templates has shown that of the 12 possible single-base mismatches, all but 1 (G_PNA and T_DNA) had larger destabilizing effects on the PNA-DNA duplexes than on the corresponding DNA-DNA duplexes, regardless of the ionic concentration of the medium (22). The 2-bp mismatch targeted in this study (CC_PNA and AA_DNA) provided sufficient instability of the PNA-DNA duplex in the presence of bla_GES-2 (mutant) templates to facilitate replacement of the PNA probe by the reverse primer GES-E in order for amplification to occur (Fig. 2). In the presence of bla_GES-1 (wt) or bla_IBC-type templates, the resulting PNA-DNA duplex experienced no mismatch and was therefore stable enough to prevent binding of the reverse primer, henceforth disabling PCR amplification in a sequence-specific manner. These data, coupled with the ability of PNA to be sequence specific, regardless of the ion concentration during the PCR (7, 22), show that it is a very robust molecular tool well suited for the screening of large numbers of isolates for very specific point mutations.

FIG. 2.

Schematic diagram depicting the placement of the amplification primers and the competitive PNA probe relative to the gene sequence of bla_GES-2 (nucleotides 1 to 864). Shaded boxes, relevant point mutations; open boxes, stem-loop sequences. Note the CC_PNA-AA_DNA nucleotide mismatch that causes the instability of the PNA-DNA duplex, thereby facilitating replacement of the PNA probe with reverse amplification primer GES-E. The amplification products (505 and 360 bp) are depicted by straight lines. The sequence direction is 5′ to 3′.

The sensitivity and specificity of this novel method for the accurate detection of the GG-to-AA point mutation in bla_GES-2 matched those generated by the “gold standard” methods (PCR and DNA sequencing) perfectly. As published previously, the use of a high-quality DNA template is, however, imperative to avoid false-negative results (29), as was noted with one of the clinical isolates when it was subjected to the sequence-specific multiplex PCR method. Large-scale epidemiological use of this novel PNA-based PCR method would possibly require the use of a commercially available DNA extraction process to standardize the DNA template yield and input. A second primer pair that targets a 360-bp region downstream from the mutation site has therefore been incorporated into the reaction mixture to serve as an internal amplification control. The presence of a stem-loop structure in the secondary structure of the bla_GES and bla_IBC DNA templates used in this study (CGGCT and AGCCG motifs at positions 466 to 470 and 482 to 486 of the coding region, respectively; T_m = 82°C) was noted, creating a possible steric hindrance during annealing of reverse primer GES-E to the closely situated target site. This was overcome by using a thorough initial denaturation step in the amplification protocol and permitting the PNA probe to bind to the target sequence earlier than the competing reverse primer, due to the rapid formation of antiparallel PNA-DNA duplexes in comparison with the rate of formation of duplexes formed in the parallel binding mode (20). Additionally, by constructing the PNA probe so that it binds to the cDNA target in a highly thermostable, antiparallel mode (20), the resulting PNA-DNA duplex formation impaired the amplicon interstrand reassociation (6), thereby further stabilizing the reverse primer-binding site on the DNA template.

Conclusion.

Novel methods for the detection and identification of genes encoding ESBLs will continue to develop as understanding of molecular biology-based methods increases. However, the most powerful mechanism driving this movement will probably be the need to stay within budgetary restraints. The remarkable ability of PNA to hybridize in a sequence-specific manner with a high affinity to complementary nucleic acids and to thereby act as a PCR controller makes it the ideal molecule for use in programs for the screening of well-characterized, specific, or geographically restricted ESBLs. This method, however, does not detect point mutations outside of the chosen mutation site and therefore offers only a partial alternative to gene sequencing. Proper modification of this method to suit individual needs may prove to be highly cost-effective in resource-poor settings, not only from the savings incurred by the direct detection of epidemiologically relevant point mutations but also from the savings achieved by not having to procure expensive equipment.