Rapid detection of the fluoroquinolone resistance-associated ParC mutation in Neisseria gonorrhoeae using TaqMan probes

Authors


Akinobu Gotoh md phd, Laboratory of Cell and Gene Therapy, Institute for Advanced Medical Sciences, Hyogo College of Medicine, Nishinomiya 663-8501, Japan. Email: gotoh@hyo-med.ac.jp

Abstract

Backgrounds:  In fluoroquinolone-resistant Neisseria gonorrhoeae, the amino acid mutations in the fluoroquinolone-resistant determining region (QRDR) of the parC gene are an important factor. The aim of the present study was to develop a rapid detection method of a serine 88 to proline substitution in parC which we previously showed as having significantly higher fluoroquinolone minimal inhibitory concentrations (MIC) using the TaqMan discrimination system.

Methods:  We extracted DNA from 90 urine or urethral swab samples obtained from male patients with urethritis caused by N. gonorrhoeae. After DNA extraction, they were subjected to real-time polymerase chain reaction (PCR) using a TaqMan discrimination system and compared with the results of conventional DNA sequencing.

Results:  Of the 90 samples, the TaqMan technique result showed 13 samples that were classified as having a serine 88 to proline mutation in parC, and 77 samples that did not have a serine 88 to proline mutation in parC. The classifications of all samples completely corresponded to those determined by conventional DNA sequencing. We also found that N. gonorrhoeae with a serine 88 to proline mutation in parC have a significantly higher MIC to ciprofloxacin than that without a serine 88 to proline mutant in parC.

Conclusions:  The present genotyping method of real-time PCR using a TaqMan discrimination system could be applied to the rapid detection of a serine 88 to proline amino acid mutation in parC of N. gonorrhoeae. This point mutation is significant for the determination of fluoruquinolone resistance. This rapid detection system may lead to the prevention of use of noneffective antimicrobial agents and a decrease of resistant strains.

Introduction

Neisseria gonorrhoeae occurs worldwide and is one of the most common causes of sexually transmitted bacterial infection.1 The treatment of these infections has recently become more difficult due to their increased resistance to antimicrobial agents. The clinical isolation of strains with decreased susceptibilities to fluoroquinolones has been reported in many places in the world including Japan.2–6 Several mechanisms of fluoroquinolone resistance have been identified and characterized in N. gonorrhoeae strains.7–10 The resistance to fluoroquinolones has been shown to be primarily associated with mutations in the gyrA gene, coding for DNA gyrase, in the parC gene, coding for DNA topoisomerase9,11,12 and less often with reduced uptake and accumulation of fluoroquinolone in the cells.13

In those genes, we detected gyrA mutation in almost 90% cases of N. gonorrhoeae in Japan,14 and additionally, N. gonorrhoeae with this mutation have a variety of fluoroquinolone minimum inhibitory concentrations (MIC). For this reason, even if we detect a gyrA mutation, we can not assume the fluoroquinolone MIC. In contrast, parC mutation is associated with significantly fluoroquinolone-resistant Neisseria gonorrhoeae (QRNG).11,15 Several point amino acid mutations have been reported, and of them we have previously confirmed that the strains with serine 88 to proline mutations have significantly higher MIC of fluoroquinolones (ciprofloxacin, levofloxacin, and gatifloxacin).16

In the clinical aspects, we should diagnose the resistant strains as early as possible for the purpose of preventing antimicrobial medication from becoming useless. If the diagnosis is performed within several hours, we can start the appropriate treatment from the day when the patients come to the hospital or the infections occur, because almost all patients are outpatients. The conventional amino acid sequencing is time consuming and requires much labor, so we studied this rapid detection system with real-time polymerase chain reaction (PCR) using a TaqMan probe.

In the present study, we explored a high-throughput genotyping method for detecting a significant serine 88 to proline mutation in the parC of N. gonorrhoeae against fluoroquinolone resistance using a TaqMan allelic discrimination system.

Methods

Reagents

All reagents were purchased from Perkin-Elmer (Foster City, CA, USA) unless otherwise stated.

Bacterial strains

For this study, we used urine or urethral swab specimens that were submitted to clinical laboratories of Kobe University Hospital gathered from several urology clinics for the investigation of N. gonorrhoeae infection in 2002. Isolates were maintained in 10% skim milk at −80°C for long-term storage. We evaluated 90 gonococcal isolates, with 21 of these strains (23.3%) being acquired in Osaka and 69 (76.7%) being acquired in Hyogo. Osaka and Hyogo are adjoined prefecture each other and located in the urban area in Japan.

Susceptibility testing

Quantitative tests of susceptibility to ciprofloxacin were performed using the agar plate dilution method, with a GC agar base (Becton Dickinson, Cockeysville, MD, USA) containing 1% Iso VitaleX (Becton Dickinson) and two-fold dilutions of antimicrobial agents. The plates were inoculated with 104 CFU per spot.16 The ciprofloxacin MIC ranges for the strains were categorized according to the National Committee for Clinical Laboratory Standards (NCCLS) guideline.17 The MIC were the lowest concentration of antibiotics that inhibited growth of the bacteria. We used N. gonorrhoeae ATCC 49226 as a quality control of susceptibility testing. The plates were incubated for 24 h at 35°C in a 5% CO2 atmosphere.

DNA extraction and polymerase chain reaction

DNA extraction was performed using a Qlamp DNA Extraction Kit (QIAGEN, Tokyo, Japan). The oligonucleotide primers for the PCR amplification were described previously.14 Briefly, PCR amplification was performed with 20 µL of a reaction mixture containing 1 × PCR buffer, 200 nmol/L of each of the two primers, 150 µmol/L of dNTP, 0.025 U/mL of Taq enzyme (Takara Shuzo, Kusatsu, Japan), and 10 µL of template DNA. The reaction mixture was subjected to 35 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 45 s. The amplicon size of the PCR product was 68 bp.

DNA sequencing and computer analysis

The PCR products were purified for sequencing using a PCR purification kit (QIAGEN). The oligonucleotide primers for the PCR amplification and sequencing were as following: for the parC QRDR, the forward primer was 5′-GTTTCAGACGGCCAAAAGCCC-3′ and the reverse primer was 5′-GGACAACAGCAATTCCGCAAT-3′; these were designed by Trees et al.16 These primers yielded a 300 bp fragment of the parC gene.

Products were sequenced directly using 4 µL (200 ng) of template DNA, 0.75 µL (7.5 pmol) of primer, and 2 µL of dye terminator premix (Amersham Pharmacia Biotech, Piscataway, NJ, USA) in 13.25 µL of distilled water. We performed 25 cycles of denaturation at 96°C for 30 s, annealing at 50°C for 15 s, and extension at 60°C for 1 min. The reaction products were then precipitated and analysed using an ABI 310 DNA sequencer (Applied Biosystems, Warrington, UK). Database searches were conducted and pair-wise alignments of DNA sequences were carried out using BLAST software.

Oligonucleotide primers and hybridization probes in TaqMan

The PCR products were hybridized with two fluorescence dye-labeled hybridization probes binding specifically to either serine 88 to proline mutants in parC of N. gonorrhoeae or non-serine 88 to proline mutants in parC of N. gonorrhoeae. The different fluorescence signals (640 and 705 nm) can be detected by the TaqMan allelic discrimination equipment, which allows the differentiation of the presence of the serine 88 to proline mutation in the parC gene of N. gonorrhoeae. The PCR primers and TaqMan probes were designed with the Primer Express Program (Applied Biosystems). Two TaqMan probes were designed according to the polymorphic site of parC of N. gonorrhoeae. Table 1 shows the sequences of the TaqMan primers and probes used. These probes emit light at different wavelengths (640 and 705 nm) and can be used for detection of the presence of serine 88 to proline mutation in parC of N. gonorrhoeae by a fluorescence signal. Based on the raw fluorescence signals of the reporter dyes FAM (640 nm, labeling dye of the non-serine 88 to praline mutant probe) and TET (705 nm, labeling dye of the serine 88 to praline mutant probe) obtained from the “dye component” function in the sequence detection software, two distinct strains were identified (Fig. 1). According to the FAM to TET ratio, the presence of serine 88 to proline mutation in parC of N. gonorrhoeae could be categorized as either allele 1 (serine 88 to proline mutant; TET >> FAM) or allele 2 (non-serine 88 to proline mutant; FAM >> TET).

Table 1.  TaqMan probes and primers
  1. PCR, polymerase chain reaction.

Forward PCR primer: 5′-GTCGGCGAGATTTTGGGTA-3′
Reverse PCR primer: 5′-ATGCGCACCATCGCCT-3′
Probe (allele 1, serine 88 to proline mutation probe): TET-5′-CACGGCGACAGTCCCGCCTATG-3′-TAMRA
Probe (allele 2, non-serine 88 to proline mutation probe): FAM-5′-CGCACGGCGACAGTTCCGC-3′-TAMRA.
Figure 1.

Scatter plot of the serine 88 to proline and non-serine 88 to proline mutants in parC of Neisseria gonorrhoeae using standard TaqMan allelic discrimination assays with automatic allele calling. The parC DNA genotypes were assigned automatically according to the ratio of FAM to TET signals. Allele 1 (serine 88 to proline; FAM << TET); allele 2 (non-serine 88 to proline; FAM >> TET); no amplification.

TaqMan genotyping assay

All genomic DNA samples were analysed using the TaqMan PCR protocol. We performed the assay on 10 healthy, non-urethritis patients’ urine samples as a control. The PCR was set up in 96-well PCR plates with 40 µL in each well and a final concentration of 20 ng DNA, 250 nmol/L TET probe, 50 nmol/L FAM probe, 100 nmol/L forward primer, and 100 nmol/L reverse primer in 1× universal master mix (Applied Biosystems). The above ratio of the forward/reverse primer and TET/FAM probe concentration was optimized by using a series of primer and probe combinations to reach a maximal signal, and the balance of the two probes by reading in an ABI 7700 sequence detector (Applied Biosystems). We used optical tubes and caps (Applied Biosystems) for preventing a contamination. The TaqMan universal PCR master mix is a premix of all the components, except the primers and probes, necessary to perform a 5′ nuclease assay. The TaqMan PCR was also optimized by increasing the annealing temperature from 60°C to 65°C. The final conditions were: 50°C for 2 min, 95°C for 15 s, and 62°C for 1 min for 40 cycles. After the PCR reactions were completed, the plates were cooled to room temperature and read in ABI 7700 sequence detector, and the results were analysed by an allelic discrimination assay of the sequence detection software (Applied Biosystems). The TET and FAM signals were stable for 3 days after PCR if stored in the dark.

Statistical analysis

Median MIC values were compared by the Mann–Whitney U-test using a software package StatView (Abacus Concepts, Berkeley, CA, USA). Statistical significance was set at 0.05.

Results

Direct detection and discrimination of serine 88 to proline and non-serine 88 to proline mutations in parC of Neisseria gonorrhoeae using TaqMan probes

After sequence verification of the genotyping results, we performed standard TaqMan allelic discrimination assays using a control allele in 90 genomic DNA samples according to the manufacturer's protocol. We did not detect bacterial DNA from healthy, non-urinary tract infected patients’ urine samples (no amplification). As a whole, the 90 samples were divided into 13 samples classified as allele 1 (serine 88 to proline mutation in parC of N. gonorrhoeae), and 77 samples classified as allele 2 (non-serine 88 to proline mutatation in parC of N. gonorrhoeae) (Fig. 1). These results revealed that our TaqMan FAM and/or TET results completely corresponded to the conventional DNA sequencing results (Table 2).

Table 2.  Ciprifloxacin activities against 90 strains of Neisseria gonorrhoeae according to National Committee for Clinical Laboratory Standard
 MIC (µg/mL)
Interpretive standard
SensitiveIntermediateResistant
  1. MIC, minimal inhibitory concentration.

Ciprofloxacin≤0.06 (12/90)0.12–0.5 (18/90)≥1 (60/90)

Susceptibilities of ciprofloxacin to Neisseria gonorrhoeae

A total of 90 N. gonorrhoeae strains were isolated from the clinical samples. The MIC for ciprofloxacin ranged from <0.002 to >8 µg/mL. Twelve strains showed an MIC range of <0.002–0.06 µg/mL for ciprofloxacin and were categorized as sensitive. On the basis of higher MIC ranges, 18 strains were described as intermediate (MIC, 0.12–0.5 µg/mL for ciprofloxacin) and 60 strains were described as resistant (MIC, 1.0 to >8 µg/mL for ciprofloxacin) according to the NCCLS guideline (Table 2).16

Associations of the serine 88 to proline mutation of parC of Neisseria gonorrhoeae with susceptibilities of ciprofloxacin to Neisseria gonorrhoeae

Figure 2 shows the correlation of serine 88 to proline mutations and non-serine 88 to proline mutations in parC with their ciprofloxacin MIC, according to the data obtained from the TaqMan allelic discrimination analysis and the susceptibility data obtained from the MIC testing. Mutants had statistically significantly higher ciprofloxacin MIC than non-mutants (P = 0.0008; Fig. 2).

Figure 2.

Correlation of serine 88 to proline and non-serine 88 to proline mutations in parC of Neisseria gonorrhoeae with their ciprofloxacin minimal inhibitory concentrations (MIC), according to the data obtained from the TaqMan allelic discrimination analysis and the susceptibility data obtained from the MIC testing. Mutant, serine 88 to proline mutants of parC; Non-mutant, non-serine 88 to proline mutants of parC. Ciprofloxacin MIC of the serine 88 to proline mutant is significantly higher than that of non-serine 88 to proline mutant (P = 0.008).

Discussion

The worldwide emergence of fluoroquinolone resistance has been a disturbing feature of infection with N. gonorrhoeae. Amino acid mutations of parC are closely related to fluoroqunilone resistance in N. gonorrhoeae.9,11,18 In particular, the serine 88 to proline mutation of parC is significantly related to fluoroquinolone resistance, and is essential for assuming the MIC for fluoroquinolones. To determine the appropriate antimicrobial therapy, we need information about the antimicrobial effects on causative pathogens of urethritis within several hours, because most urethritis patients are treated as outpatients.

The conventional culture and susceptibility testing methods take several days to produce results, and are very laborious. In addition, conventional DNA sequencing is time consuming and requires higher technique, and takes more than 6 h to get the precise results. Therefore we studied this development of a rapid, sensitive, and standardized method using the TaqMan allelic discrimination system for the detection of point mutations.

Recently, a TaqMan allelic discrimination method using TaqMan probes has provided fast and easy analysis of genomic DNA. This study confirms that it is possible to detect rapidly bacterial DNA from a small volume of urine or urethral swab samples of male urethritis cases caused by N. gonorrhoeae, and to find the significant serine 88 to proline mutation in parC of N. gonorrhoeae which confers fluoroquinolone resistance in a one-run experiment. This assay uses the 5′ nuclease activity of Taq polymerase to detect a fluorescent reporter generated during or after PCR reactions.19

Automated, rapid-throughput, large-scale genotyping for DNA polymorphism is highly desirable for the accomplishment of the current study. Until now, polymerase chain reaction–restriction fragment length polymorphism (PCR–RFLP) was one of the most commonly used method for genotyping single nucleotide polymorphisms (SNP). However, this approach is time consuming and needs intensive labor. The procedures involved are hard to automate and require highly skilled staff to conduct and evaluate the tests. In addition, if the polymorphism does not involve a restriction site change, PCR–RFLP cannot be used.20

In contrast, TaqMan-based allelic discrimination offers high sample throughput, with accurate detection of SNP. This fluorescence-based genotyping procedure greatly simplifies the assay protocol by eliminating the need for restriction enzyme digestion, gel electrophoresis, resolution of PCR products, and visual assessment of bands. In addition, the 96-well plate formats and closed-tube PCR eliminate subsequent processing, thus greatly reducing the potential for contamination. This simple method only requires the preparation of a master mix by adding the appropriate primers and TaqMan probes. The time required for sample extraction is less than 1 h. Amplification and detection requires 2 h. The total time required to process 96 samples is less than 3 h. This assay is also cost-effective and could be done within $US5 per sample. While this method gives us the results rapidly and accurately, at the same time, we have to assemble and assay many samples (within 96) in one assay. We need to wait for 2 h for the next assay once the assay begins. That is the reason why this could be applied to clinical samples but has so far rarely been applied. However, in epidemiological surveys, it is necessary to assay many samples, and so this method is appropriate.

This method will be also be particularly useful for screening mutant or non-mutant of serine 88 to proline alleles of parC in N. gonorrhoeae. By the present rapid detection system, the detection of this significant amino acid mutation may make us assume the exact fluoroquinolone MIC and select the appropriate antimicrobial agents within several hours. In the near future, the application to clinical diagnosis, after some problems such as the number of samples are solved, may give us a decrease in resistant strains and reduction of patients’ costs. Now we have an important task to apply this method to other antimicrobial agents, other microbes, and other gene-related diseases.

Conclusion

The genotyping method of real-time PCR using a TaqMan discrimination system could be applied to the rapid detection of the serine 88 to proline amino acid mutation in parC of N. gonorrnoeae. This point mutation is significantly associated with fluoruquinolone resistance. This rapid detection system may prevent the use of noneffective antimicrobial agents and decrease the development of resistant strains.

Ancillary