|Home | About | Journals | Submit | Contact Us | Français|
The quinolone resistance determining region (QRDR) of the gyrA gene in ciprofloxacin-susceptible strains (n=53) and strains of Neisseria spp. with reduced susceptibility (n=70) was determined by the pyrosequencing method. Results showed that the QRDR of the gyrA gene is an effective molecular indicator of resistance to ciprofloxacin in Neisseria gonorrhoeae, and presumably in Neisseria meningitidis, but not in all other Neisseria spp. This sequence was not unique for N. gonorrhoeae and seems unsuitable for species verification of N. gonorrhoeae. However, whether it is also possible to use this region for verification depends on the specificity of the primary screening method used.
Several commercial kits for molecular diagnosis of Neisseria gonorrhoeae in a duplex PCR with Chlamydia trachomatis are available. Many laboratories, as well as clinicians, are interested in these techniques, since they provide fast negative results, are easy to perform, and also urine samples can be analyzed (1-7).
However, most of the commercial kits suffer from low specificity (1-7), and therefore require a confirmatory PCR (1, 2, 6, 7). The AMPLICOR N. gonorrhoeae PCR (Roche Diagnostics, Indianapolis, IN, USA) targets the putative cytosine DNA methyltransferase gene. A PCR targeting the 16S rRNA gene has been recommended for confirmation (8), but we found it suboptimal for this purpose (6). Our previous results showed that the QRDR of the gyrA could be used as an indicator of ciprofloxacin resistance in N. gonorrhoeae and possibly for species verification of N. gonorrhoeae (6, 8, 9). In addition, the QRDR of the gyrA seems to be an effective indicator of ciprofloxacin resistance in N. meningitidis (10-12).
The aims of this study were to investigate if the molecular method indicating resistance to ciprofloxacin in N. gonorrhoeae, based on pyrosequencing of the QRDR of the gyrA gene, can be used as an indicator of resistance to ciprofloxacin in most Neisseria spp. and if this QRDR sequence is suitable also for species confirmation of N. gonorrhoeae.
In the present study, N. gonorrhoeae reference strains (CCUG 37534 and ATCC 43069) and clinical isolates (n=83), as well as reference strains (N. lactamica ATCC 23970 and CCUG 17628, N. meningitidis ATCC 13090 and ATCC 13077, N. subflava CCUG 30607 and CCUG 32186, N. flavescens CCUG 26476, CCUG 17913 and CCUG 806, N. cinerea CCUG 25879 and CCUG 27178, N. mucosa ATCC 9913, and N. sicca CCUG 26877) and clinical isolates (n= 27) of nine other Neisseria spp. were examined.
The species identification of the isolates was based on growth ability in different incubational conditions, colony morphology and pigmentation on different culture media, reduction of nitrate, sugar oxidation tests (glucose, maltose, fructose, and lactose) (13), RapID NH (Remel, Lenexa, USA) and/or API NH v2.0 (BioMérieux, Lyon, France), and sequencing of the 16S rRNA gene (Unemo et al. personal communication). All isolates were cultured as previously described (14). Minimum inhibitory concentration (MIC) of ciprofloxacin was determined by Etest (AB Biodisk, Solna, Sweden).
According to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) (http://www.srga.org/eucastwt/MICTAB/index.html, access date: 2006-04-21) as well as the Swedish Reference Group for Antibiotics (SRGA) http://www.srga.org/MICTAB/MICTAB2.htm, access date: 2006-04-21) N. gonorrhoeae and N. meningitidis ciprofloxacin-susceptible strains have MIC≤0.03 mg/L and resistant strains have MIC>0.06 mg/L. In the present study, susceptibility of all Neisseria spp. was interpreted according to these breakpoints. Of the Neisseria isolates that gave a sequence (n=113), 46 were susceptible to ciprofloxacin, 65 were resistant, and two isolates were intermediate (Table 1). All sequenced reference strains were susceptible to ciprofloxacin.
The pyrosequencing method gave sequences for all strains of N. gonorrhoeae (Table 1). The gyrA sequences encoding amino acids 91–95 of GyrA revealed the wild-type sequence in all ciprofloxacin-susceptible strains and six different patterns of non-synonymous substitutions in the ciprofloxacin-resistant strains.
We found the ciprofloxacin-resistant strain comprising a mutation that changed Ser91 to Thr and also a synonymous substitution in Ala92 especially interesting, since this sequence was a unique QRDR of gyrA variant in our large collections of N. gonorrhoeae, and also it was identical to the sequences of N. subflava and N. flavescens ciprofloxacin-susceptible strains. To our knowledge this sequence has not been published (Table 1).
Most isolates of the species N. cinerea, N. elongata, N. mucosa, and N. sicca showed a sequence identical to one of the N. gonorrhoeae ciprofloxacin-resistant sequences (Table 1). The same gyrA sequence was found in both ciprofloxacin-susceptible as well as in intermediate strains of N. cinerea. The susceptible strain of N. sicca showed the same sequence as one of the N. gonorrhoeae ciprofloxacin-resistant gyrA sequences; in contrast, the two ciprofloxacin-resistant strains of N. sicca showed the N. gonorrhoeae ciprofloxacin-susceptible wild-type pattern (Table 1). N. subflava and N. flavescens showed the same sequence as that of one of the N. gonorrhoeae ciprofloxacin-resistant strains (Table 1).
Regarding N. gonorrhoeae subsp. kochii, both ciprofloxacin-susceptible as well as ciprofloxacin-resistant strains showed an identical gyrA sequence, and this sequence was also the same as for ciprofloxacin-susceptible strains of N. meningitidis and N. lactamica (Table 1).
The following 12 strains did not give a PCR product and were excluded from further analysis: one clinical isolate of N. mucosa and ATCC 9913, one clinical isolate of N. cinerea and CCUG 27178, CCUG 25879, one clinical isolate of N. sicca and CCUG 26877, three clinical isolates of N. subflava and two clinical isolates of N. meningitidis. Of these 12 strains, 7 were susceptible, 2 were intermediate, and 3 were resistant to ciprofloxacin.
The results presented in Table 1 implicate that, in Neisseria spp., also other mechanisms of resistance to ciprofloxacin than gyrA mutations at position Ser91 and Asp95 are involved. The sequence variation in the QRDR in the gyrA of N. gonorrhoeae may be due to genetic exchange among Neisseria spp.
Additional strains need to be examined in order to draw conclusions respecting those Neisseria spp. only represented by a few ciprofloxacin-susceptible strains in this study, but it seems that the correlation between ciprofloxacin susceptibility and mutations in the QRDR of the gyrA is unique for N. gonorrhoeae and probably also for N. meningitidis. Unfortunately, no N. meningitidis strain with reduced susceptibility to ciprofloxacin was available to us. According to previous studies, sequencing of the QRDR in the gyrA will presumably also identify the rare strains of N. meningitidis with reduced susceptibility to ciprofloxacin (10-12). Meanwhile, since two clinical isolates of N. meningitidis were not amplified with the primers that we used, our method needs to be further optimised for this purpose.
Despite the fact that a few ciprofloxacin-resistant N. gonorrhoeae strains without alterations in the QRDR of the gyrA have been reported (15-17), the present study supports previous studies (6-9) of pyrosequencing of the QRDR in the gyrA gene as an effective molecular method to indicate resistance to ciprofloxacin in N. gonorrhoeae. However, the method is useful for N. gonorrhoeae and presumably for N. meningitidis, but not for other Neisseria spp.
Neisseria spp. are highly recombinogenic and naturally transformable (18-19). Clinical methods to distinguish between Neisseria spp. are designed mainly to discriminate N. gonorrhoeae and N. meningitidis from other Neisseria spp. Neisseria multilocus sequence typing (MLST) showed that in Neisseria spp. the definition of a species will not be clear cut, especially among those species that colonize the same body site, such as N. meningitidis and N. lactamica. Thus, both these species colonize the human pharynx and there are also mosaic overlaps due to recombination (19). Also N. gonorrhoeae can colonize the human pharynx and thus recombination between species probably occurs. This might explain the atypical gyrA sequence of one N. gonorrhoeae strain, the same sequence as that of N. subflava and N. flavescens. Even if future commercial kits are further optimised for N. gonorrhoeae diagnostics the results should be evaluated bearing in mind the high recombination frequency.
The clinical application of this method as a molecular indicator of ciprofloxacin resistance adds valuable information to any commercial molecular N. gonorrhoeae diagnostic test. So far, all N. gonorrhoeae samples have been amplified using this method, but some strains of other Neisseria spp. gave no PCR products. Consequently, samples that give no PCR product can be verified as non-N. gonorrhoeae. In a previous study Airell et al. (6) showed that out of 24 N. gonorrhoeae culture-negative pharynx samples, 16 were positive in AMPLICOR N. gonorrhoeae PCR (Roche Diagnostics). N. gonorrhoeae species verification using the 16S rRNA gene was negative in 14 of these samples and in all 16 with verification with the gyrA gene. With further optimisations of the PCR protocol, it is possible that Neisseria spp. other than N. gonorrhoeae can be excluded with acceptable accuracy.
In conclusion, the examined QRDR of the gyrA gene may not seem suitable for species verification of N. gonorrhoeae, since this sequence only discriminates N. gonorrhoeae from N. meningitidis, N. gonorrhoeae subsp. kochii and N. lactamica (Table 1). However, whether it is possible to use this region for species verification still depends on the specificity of the primary screening method used. In addition, this sequence is an effective indicator of ciprofloxacin resistance in N. gonorrhoeae and probably in N. meningitidis, but not all other Neisseria spp.