ABSTRACT
In 2009, the first high-level ceftriaxone-resistant Neisseria gonorrhoeae strain (H041) was isolated in Kyoto, Japan. The present study describes an intensified surveillance (antimicrobial resistance and molecular typing) of Neisseria gonorrhoeae isolates in Kyoto and its neighboring prefecture Osaka, Japan, in 2010 to 2012, which was initiated after the identification of H041. From April 2010 to March 2012, 193 N. gonorrhoeae isolates were collected and the MICs (μg/ml) to six antimicrobials, including ceftriaxone, were determined. All isolates showed susceptibility to ceftriaxone and cefixime (MIC values, <0.5 μg/ml), and spectinomycin. The rates of resistance (intermediate susceptibility) to azithromycin, penicillin G, and ciprofloxacin were 3.6% (19.7%), 24.4% (71.0%), and 78.2% (0.5%), respectively. Multilocus sequence typing (MLST) showed that 40.9%, 19.2%, and 17.1% of isolates belonged to ST1901, ST7359, and ST7363, respectively. Furthermore, N. gonorrhoeae multiantigen sequence typing (NG-MAST) revealed that 12 (63%) of the 19 isolates with decreased susceptibility to ceftriaxone (MIC > 0.064 μg/ml) were of ST1407. NG-MAST ST1407 was also the most prevalent ST (16.1%; 31 of 193 isolates). In those NG-MAST ST1407 strains, several mosaic type penA alleles were found, including SF-A type (penicillin binding protein 2 allele XXXIV) and its derivatives. These were confirmed using transformation of the penA mosaic alleles as critical determinants for enhanced cefixime and ceftriaxone MICs. The intensified surveillance in Kyoto and Osaka, Japan, did not identify any dissemination of the high-level ceftriaxone-resistant N. gonorrhoeae strain H041, suggesting that H041 might have caused only a sporadic case and has not spread further.
INTRODUCTION
Neisseria gonorrhoeae infections are major public health concerns worldwide. In 2008, the World Health Organization (WHO) estimated 106 million gonorrhea cases among adults globally, which places the infection as the most prevalent bacterial sexually transmitted infection (STI) (with a global incidence similar to that of genital chlamydial infections) (1). Resistance in N. gonorrhoeae to previously recommended first-line antimicrobials for treatment of gonorrhea is also prevalent worldwide. During the recent decade, the susceptibility to the extended-spectrum cephalosporins (ESCs) cefixime (oral) and ceftriaxone (parenteral), which currently are the first-line antimicrobials in most countries, has decreased rapidly worldwide (2–12). Clinical treatment failures with cefixime have been verified in many countries (13–19). In regard to ceftriaxone, which is the last remaining option for empirical first-line antimicrobial monotherapy in most countries, a few cases of failure in treating pharyngeal gonorrhea despite relatively low ceftriaxone MICs of the gonococcal strains have been confirmed in Australia (20), Sweden (21), and Slovenia (22). However, these cases likely reflected the fact that pharyngeal gonorrhea commonly is harder to treat than urogenital gonorrhea and not treatment failure due to the slightly increased ceftriaxone MICs of the gonococcal strains. Nevertheless, it is most perturbing that recently the first extensively drug-resistant (XDR) gonococcal strain H041 (23, 24) in Kyoto, Japan, followed by an XDR strain in France (14) and Spain (25), which all have been confirmed to have a high-level resistance to ceftriaxone, were described. If these strains start to spread globally, gonorrhea will become untreatable in certain circumstances and especially in settings where dual antimicrobial therapy is not feasible and/or affordable. In response to this developing situation, the WHO (26, 27), the European Centre for Disease Prevention and Control (ECDC), and the Centers for Disease Control and Prevention (CDC) in the United States have published global and region-specific response plans for their respective regions, i.e., the European Union/European Economic Area (EU/EEA) countries (28) and the United States (29). One main component of these action/response plans is to enhance the surveillance of antimicrobial resistance and treatment failures. Nevertheless, for future treatment of gonorrhea it is imperative to develop new treatment options.
In Japan, a sentinel surveillance system for sexually transmitted infections, including gonorrhea, was launched in 1999. This system includes approximately 1,000 sentinel sites that monthly report their gonorrhea cases. The number of reported gonorrhea cases peaked in 2002 (n = 21,921; 23.9 cases/site) but has since declined to 10,247 (10.6 cases/site) in 2011. Unfortunately, this sentinel surveillance system does not include any requirements to perform antimicrobial susceptibility testing on the identified gonococcal isolates. Nevertheless, after the identification of the first high-level ceftriaxone-resistant XDR strain H041 (23, 24) in Kyoto, Japan, surveillance of antimicrobial resistance in N. gonorrhoeae was initiated in April 2010 in Kyoto and in April 2011 in the neighboring prefecture Osaka.
The present study describes the antimicrobial resistance and molecular typing of Neisseria gonorrhoeae isolates in Kyoto and Osaka, Japan, in 2010 to 2012, a study initiated after the identification of the first high-level ceftriaxone-resistant XDR N. gonorrhoeae strain H041 in Kyoto (23, 24).
MATERIALS AND METHODS
Study settings and N. gonorrhoeae isolates.Five outpatient clinics, three in Kyoto and two in Osaka, Japan, provided mainly consecutive urethral discharge samples to the National Institute of Infectious Diseases (NIID), Tokyo, for isolation of N. gonorrhoeae. The specimens were transported using Seed swab γ no. 2 (Eiken Chemical, Tokyo, Japan) to NIID, where the samples were inoculated on selective BBL Modified Thayer Martin (MTM II) agar plates (Becton, Dickinson and Company, Sparks, MD). After 18 to 24 h of incubation, suspected colonies on the MTM II agar plates were verified as N. gonorrhoeae by identification of Gram-negative diplococci in microscopy, by the rapid oxidase reaction, and with the biochemical test ID Test HN-20 Rapid (Nissui Pharmaceutical, Tokyo, Japan). If culture was negative after 18 to 24 h of incubation, the agar plates were incubated for an additional 18 to 24 h before being classified as negative for N. gonorrhoeae. From April 2010 to March 2012, in total 154 males and 39 females were culture positive for N. gonorrhoeae, and accordingly, 193 N. gonorrhoeae isolates (one per patient) were obtained (53 from Kyoto and 140 from Osaka).
Antimicrobial susceptibility testing.The MICs (μg/ml) of ceftriaxone, cefixime, penicillin G, ciprofloxacin, azithromycin, and spectinomycin were determined using the Etest method (AB bioMérieux, Solna, Sweden) according to the manufacturer's instructions. MIC breakpoints used for determination of susceptibility, intermediate susceptibility, and resistance (Table 1) were in accordance with the Clinical and Laboratory Standards Institute (CLSI; www.clsi.org), with the exception of azithromycin, for which CLSI does not state any breakpoints. For azithromycin, the breakpoints stated by the European Committee on Antimicrobial Susceptibility Testing (EUCAST; www.eucast.org) were used. The 2008 WHO N. gonorrhoeae reference strains (30) were used for quality control in all antimicrobial susceptibility testing.
Antibiotic susceptibility of Neisseria gonorrhoeae isolates from Kyoto and Osaka, Japan, 2010 to 2012 (n = 193)
DNA extraction.The bacterial isolates were suspended in Tris-EDTA (TE) buffer (10 mM Tris, 1 mM EDTA, pH 8.0) and boiled for 10 min. After centrifugation to remove the cell debris, the supernatant was promptly used as the template DNA for the PCRs.
Molecular epidemiological characterization.Molecular epidemiological characterization by means of multilocus sequence typing (MLST) and N. gonorrhoeae multiantigen sequence typing (NG-MAST) was performed as described previously (31, 32). The diversity index for MLST and NG-MAST was calculated as described earlier (33). Neighbor-joining trees based on partial porB gene sequences (490 bp) were generated by using MEGA 4. Similarity of alleles was evaluated by individual pairwise alignment against the representative alleles in each clade (porB4, porB1059, porB206, porB1785, porB908, and porB2569) to determine the numbers of base pair differences. Alleles of porB showing ≥99% similarity (<5-bp difference) were grouped.
penA sequencing.The penA gene was PCR amplified and sequenced by using the previously described primers penA_F and penA_R (34). Briefly, the PCR mixtures were incubated for 2 min at 96°C, followed by 30 cycles of 10 s at 96°C, 10 s at 65°C, and 2 min at 72°C. The PCR products were subsequently purified with an ExoSAP IT kit (GE Healthcare Limited, Buckinghamshire, United Kingdom). Both DNA strands of the PCR products were sequenced with an ABI BigDye Terminator cycle sequencing kit (version 3.1) on an ABI 3130 xl sequencer, in accordance with the instructions from the manufacturer (Applied Biosystems, Foster City, CA).
Transformation assays.To assess the capacity of a unique penA allele to result in increased MICs of cefixime and ceftriaxone, the full-length penA allele was PCR amplified and transformed into a recipient strain as previously described (34). The recipient gonococcal strain NG9807 was of MLST ST7363 and NG-MAST ST4093 and had a ceftriaxone as well as cefixime MIC of 0.016 μg/ml (23). Briefly, the recipient was suspended in GC broth containing 1.5% (wt/vol) proteose peptone 3 (Becton, Dickinson and Company, Sparks, MD), 0.4% (wt/vol) K2HPO4, 0.1% (wt/vol) KH2PO4, 0.5% (wt/vol) NaCl, and 1% (vol/vol) isovitalex (1 × 108 cells/100 μl) and incubated with 0.2 μg of PCR product of the specific penA allele (after purification using the High Pure PCR Product Purification kit [Roche Diagnostics GmbH, Mannheim, Germany]) for 4 h. Aliquots of 10 μl and 100 μl were inoculated on GC agar containing cefixime (0.064 μg/ml). After incubation for 18 to 24 h, the obtained colonies were subcultured on an antimicrobial-free GC agar plate for single-clone isolation. For confirmation, the full-length penA allele was sequenced in all transformants.
RESULTS
Antimicrobial susceptibility testing.The results of the antimicrobial susceptibility testing of the N. gonorrhoeae isolates (n = 193) are summarized in Table 1. All 193 examined isolates were susceptible to ceftriaxone, cefixime, and spectinomycin. The rates of resistance (intermediate susceptibility) to azithromycin, penicillin G, and ciprofloxacin were 3.6% (19.7%), 24.4% (71.0%), and 78.2% (0.5%), respectively. However, although there were no isolates resistant to cefixime and ceftriaxone according to the CLSI breakpoints (MIC, >0.25 μg/ml), 48 (24.9%) isolates were resistant to cefixime according to the European breakpoints (EUCAST; MIC, >0.125 μg/ml) and the MIC90 value for ceftriaxone was 0.094 μg/ml (Fig. 1).
Cefixime (CFM) and ceftriaxone (CRO) MIC distribution of Neisseria gonorrhoeae isolated in Kyoto and Osaka, Japan, in 2010 to 2012 (n = 193).
Multilocus sequence typing investigation and relationship between MLST and cephalosporin resistance.In Japan, the initial emerged N. gonorrhoeae strains with intermediate susceptibility and resistance to cefixime possessed a penA mosaic allele X, which encodes a penicillin binding protein 2 (PBP 2) X. These strains belonged mainly to two major MLST sequence types (ST), particularly ST7363 but also ST1901 (34). These STs and their penA alleles have subsequently further evolved. The first strain with high-level resistance to ceftriaxone identified in Kyoto, Japan (H041; 23, 24), was of ST7363. The additional high-level ceftriaxone-resistant strains isolated in France (14) and Spain (25) were of ST1901. In the present study, a total of 29 MLST STs were found among the 193 isolates from Kyoto and Osaka. ST1901 was the most common MLST ST (n = 79, 40.9%), followed by ST7359 (n = 37, 19.2%), ST7363 (n = 33, 17.1%), and ST7819 (n = 17, 8.8%) (Table 2). The diversity index for MLST of the 193 isolates was 0.76.
MLST and NG-MAST sequence types of N. gonorrhoeae isolates from Kyoto and Osaka, Japan, 2010 to 2012
Among those four most prevalent MLST STs, isolates of ST1901, ST7363, and ST7819 showed relatively high MIC values for cefixime with similar MIC distribution patterns (Fig. 2). In regard to ceftriaxone, isolates of ST1901 showed slightly higher MICs than did the ST7363 and ST7819 isolates. In contrast, all ST7359 isolates were highly susceptible to both cefixime and ceftriaxone (Fig. 2).
MIC of cefixime (CFM) and ceftriaxone (CRO) in Neisseria gonorrhoeae isolates from Kyoto and Osaka, Japan, in 2010 to 2012 belonging to the four major multilocus sequence typing (MLST) sequence types (STs), that is, ST1901 (n = 79), ST7363 (n = 33), ST7819 (n = 17), and ST7359 (n = 37).
N. gonorrhoeae multiantigen sequence typing analysis and relationship between NG-MAST and MLST.NG-MAST has a substantially higher discriminatory power than MLST, and NG-MAST has been applied for molecular epidemiologic investigations for gonococci isolated worldwide (10, 35). In recent years, NG-MAST ST1407 has been a prevalent ST in many countries, and this ST has also accounted for a substantial proportion of the decreased susceptibility and resistance to ESCs in those countries (10). Furthermore, the high-level ceftriaxone-resistant XDR isolates cultured in France (14) and in Spain (25), which might represent one and the same strain, were of NG-MAST ST1407. In the present study, a total of 95 NG-MAST STs were identified among the 193 isolates from Kyoto and Osaka (Table 2). ST1407 (n = 31, 16.1% of all isolates) was the most prevalent NG-MAST ST, followed by ST4186 (n = 10, 5.2%), ST3505 (n = 8, 4.1%), and ST6780 (n = 7, 3.6%) (Table 2). The diversity index for NG-MAST of the 193 isolates was 0.96.
N. gonorrhoeae isolates assigned to the four most common MLST STs, ST1901, ST7359, ST7363, and ST7819, were subdivided into 30 (intra-MLST ST diversity index, 0.83), 18 (diversity index, 0.89), 23 (diversity index, 0.94), and 7 (diversity index, 0.71) NG-MAST STs, respectively. These results clearly illustrate the substantially higher discriminatory power of NG-MAST compared to MLST. However, the number of different tbpB alleles within these MLST STs was relatively small, that is, MLST ST1901, ST7363, ST7359, and ST7819 possessed only six, two (tbpB10 and tbpB3), one (tbpB241), and one (tbpB27) different tbpB alleles, respectively. Accordingly, the high discriminatory power of NG-MAST was highly dependent on the diversified porB gene (see below), and the different level of genetic heterogeneity within the main MLST STs might reflect how long these STs have been spreading and thus evolving. As aforementioned, ST1407 (n = 31) was the most prevalent NG-MAST ST, and all except one (belonging to MLST ST10241) of these isolates belonged to MLST ST1901 (30/79 [38.0%] of all ST1901 isolates). The MIC50s (ranges) of cefixime and ceftriaxone for the NG-MAST ST1407 isolates were 0.19 μg/ml (0.064 to 0.25 μg/ml) and 0.064 μg/ml (0.016 to 0.125 μg/ml), respectively. Furthermore, 12 (63%) of the 19 isolates with decreased susceptibility to ceftriaxone (MIC, >0.064 μg/ml) were of ST1407.
porB and penA sequencing.To further examine the genetic relationships of the 193 N. gonorrhoeae isolates from Kyoto and Osaka, a phylogenetic analysis of the porB sequences (490 bp) used in the NG-MAST was performed.
The phylogenetic analysis identified six clades, in which common representative alleles for each clade were porB4, porB1059, porB206, porB1785, porB908 (the porB allele in NG-MAST ST1407), and porB2569, respectively (Fig. 3A). The MIC distributions for cefixime and ceftriaxone varied substantially for those different porB clades. Interestingly, the isolates in the porB908 clade (including all 31 NG-MAST ST1407 isolates but also 31 additional isolates of other STs [n = 15]) had clearly the highest MICs of cefixime and particularly ceftriaxone; in fact, this clade contained all isolates showing an MIC of 0.125 μg/ml for ceftriaxone (Fig. 3D). NG-MAST ST1407 isolates have been previously described to possess a penA mosaic allele XXXIV encoding a PBP 2 mosaic XXXIV, which results in elevated MICs of ESCs (10, 14). Accordingly, in the present study the penA gene in all isolates from the porB908 clade (n = 62) and, for comparison, isolates from the porB1059 clade (n = 17) and the porB1785 clade (n = 25), which also had substantially elevated MICs of cefixime particularly (Fig. 3B and C), were sequenced. In total, 11 penA alleles, including nine mosaic alleles, were revealed among those 104 isolates (Fig. 4). Of the nine penA mosaic alleles, five encoded PBP 2 mosaic X or closely related sequence variants (found in 50 of the isolates), which were distributed in all three examined porB clades. In contrast, all isolates possessing a penA mosaic allele XXXIV or closely related sequence variants (in 51 isolates) were in the porB908 clade, with the exception of one isolate from the porB1785 clade (Fig. 4), indicating that this penA mosaic allele might have emerged in the porB908 clade (including the 31 NG-MAST ST1407 isolates). Nevertheless, only a few (n = 5) of the isolates in the porB908 clade possessed the authentic penA mosaic allele XXXIV, and most contained PBP 2 mosaic XXXIV with an additional P551S substitution (n = 44) (Fig. 4). Furthermore, all isolates showing an MIC of 0.125 μg/ml of ceftriaxone possessed this PBP 2 mosaic XXXIV with an additional P551S alteration, with the exception of one isolate with PBP 2 mosaic XXXIV containing an additional A501V substitution. As a matter of great concern, these results indicate that the authentic penA mosaic XXXIV allele is evolving, resulting in further enhanced MICs of ESCs.
Sequence comparison of porB alleles (490 bp) from N. gonorrhoeae isolates (n = 193) from Kyoto and Osaka, Japan, and cefixime and ceftriaxone MIC profile of each porB group. (A) An internal region of the porB genes including loops 3, 4, and 5 (490 bp), which is used in NG-MAST, was utilized to create a phylogenetic tree with the MEGA 4 program. The numbers indicate porB allele numbers assigned in the NG-MAST database. The number of isolates with the same porB allele is indicated in parentheses. (B to E) Similar porB alleles are grouped as porB1059, porB1785, porB908, and porB2569 clades, and the MICs to cefixime (gray bars) and ceftriaxone (black bars) are shown in panels B to E, respectively.
Phylogenetic tree of the amino acid sequences of penicillin binding protein 2 encoded by different penA mosaic alleles. The penA genes from a total of 104 N. gonorrhoeae isolates possessing alleles belonging to the porB908, porB1785 and porB1059 clades (C) were sequenced, and a phylogenetic tree based on the PBP 2 amino acid sequences was constructed by the MEGA 4 program. Two major penA mosaic allele groups of PBP 2 were revealed, the PBP 2 X family and the PBP 2 XXXIV family. Strains HI004 and KM017 had 3 and 11, respectively, amino acid substitutions in their PBP 2 X group alleles.
Transformation of specific penA mosaic alleles.To assess the capacity of specific penA mosaic alleles to result in increased MICs of ceftriaxone and cefixime, the full-length penA mosaic alleles X, XXXIV, XXXIV-P551S, and XXXIV-A501V were separately transformed into the ESC-susceptible strain NG9807 (ceftriaxone MIC, 0.016 μg/ml; cefixime MIC, 0.016 μg/ml). As shown in Fig. 5, the transformants with penA XXXIV variants (P551S and A501V) showed ceftriaxone MICs of 0.125 μg/ml (and cefixime MICs of 0.25 μg/ml and 0.5 μg/ml, respectively), which were clearly higher than those of transformants with penA XXXIV (ceftriaxone MIC, 0.064 μg/ml; cefixime MIC, 0.125 μg/ml). The authentic penA mosaic allele X transformant showed a ceftriaxone MIC of 0.125 μg/ml and a cefixime MIC of 0.25 μg/ml. Accordingly, in regard to ceftriaxone, penA alleles XXXIV-P551S and XXXIV-A501V showed impact on MIC identical to that of the penA X allele, and the MIC was raised twice as much as that seen with the penA XXXIV allele. Interestingly, the penA allele XXXIV-A501V could increase the MIC of cefixime up to 0.5 μg/ml, which is above the resistance breakpoint (Fig. 5).
Capacity for increased MIC conferred by the mosaic PBP 2 alleles harbored by penA X and penA XXXIV, including amino acid substituted mutants identified in the present study. Each allele was introduced into the cefixime- and ceftriaxone-susceptible strain NG9807, and MIC values of each transformant are shown.
DISCUSSION
In 2009, the first high-level ceftriaxone-resistant N. gonorrhoeae strain (H041) was isolated in Kyoto, Japan (23, 24). H041 was also the first extensively drug-resistant (6) N. gonorrhoeae strain, displaying resistance to most available antimicrobials (23). The present study describes an intensified surveillance (antimicrobial resistance and molecular typing) of N. gonorrhoeae isolates in Kyoto and its neighboring prefecture Osaka, Japan, in 2010 to 2012, which was initiated after the identification of H041 (23, 24). All N. gonorrhoeae isolates (n = 193), collected at five clinics, were susceptible to ceftriaxone and cefixime (MIC values, <0.5 μg/ml) and to spectinomycin. The rates of resistance (intermediate susceptibility) to azithromycin, penicillin G, and ciprofloxacin were 3.6% (19.7%), 24.4% (71.0%), and 78.2% (0.5%), respectively (Table 1). These data are basically consistent with a previous recent report from Japan, except for the azithromycin resistance (12). In a previous publication (12), only 0.4% (1 of 242 isolates) was reported to be resistant to azithromycin. Accordingly, the high-level ceftriaxone-resistant N. gonorrhoeae strain H041 might have caused only a sporadic event, and no further dissemination in the Kyoto or Osaka area could be found. The reasons for this are unknown; however, the high-level ceftriaxone resistance (i.e., the remodeled PBP 2) might affect the biological fitness of H041. Studies investigating the biological fitness of H041 in competitive experiments in vitro and in vivo (in a mouse model) are in progress. In the present study, no other ceftriaxone-resistant strain was identified either. However, only five clinics were included in the present surveillance and, due to this lack of complete surveillance, it cannot be entirely excluded that some ceftriaxone-resistant gonococcal strain was spreading in the region. The antimicrobial resistance surveillance in this region has continued, further strengthened in the area, and should now be expanded nationally.
All clinics included in this surveillance adhered to the treatment guideline from the Japanese Society for STI, which recommends using 1 g ceftriaxone intravenously or 2 g spectinomycin intramuscularly. However, in some cases also azithromycin (2 g orally) can be used. During the study period, no cases of suspected treatment failure using 1 g ceftriaxone were identified, which further supports the lack of ceftriaxone resistance in the region.
Recently, the antimicrobial resistance surveillance for N. gonorrhoeae has been enhanced worldwide (26–28), and some of those surveillance systems have also performed molecular typing of the N. gonorrhoeae isolates (36). As to molecular typing for N. gonorrhoeae, MLST (31) and particularly NG-MAST (32) have been most widely applied worldwide (35). MLST, which identifies allelic variations in seven housekeeping genes, describes the phylogenetic relationship between isolates. However, since the housekeeping genes are relatively conserved, the discrimination of MLST is rather low and the method is more suitable for long-term epidemiology, macroepidemiology (e.g., on a global scale), and evolutionary studies (35). In contrast, NG-MAST is more applicable to short-term epidemiology and microepidemiology because it is based on sequence variations of genes encoding two bacterial cell surface proteins, that is, PorB and TbpB (35). These antigens are exposed to the human immune response, and this selective pressure results in antigenic changes giving the N. gonorrhoeae strain an opportunity to escape from the human immune system. Although NG-MAST has a very high ability to discriminate N. gonorrhoeae isolates, several publications have reported a high prevalence of the NG-MAST ST1407, which appears to be disseminated basically worldwide and account for a high proportion of the decreased susceptibility and increased resistance to ESCs and multidrug resistance (10, 14, 18, 36). As found in the present study, the NG-MAST ST1407 (30 of those 31 isolates belonged to MLST ST1901) was also the most prevalent ST in Osaka and Kyoto, Japan, and as in other countries the ST1407 isolates had decreased susceptibility to ESCs and were resistant to fluoroquinolones (data not shown).
Thus, the discriminatory ability of NG-MAST is substantially higher than MLST (35; this study); however, the majority of this discrimination is due to the high variation of the porB allele and far fewer tbpB alleles have been described. The porB908 allele (present in the NG-MAST ST1407) and its closely related alleles (porB908 group) were widely distributed in the isolates examined in the present study; these isolates also in general had higher MICs of ESCs. The other major porB groups were the porB1059 group, the porB1785 group, and the porB2569 group (Fig. 3). The porB2569 group isolates (mostly belonging to MLST ST7359) were mainly fully susceptible to ceftriaxone and cefixime, although some isolates showed higher MICs, that is, MICs of 0.016 to 0.064 μg/ml of ceftriaxone and MICs of 0.064 to 0.25 μg/ml of cefixime. Because porB alleles might be horizontally transferred among N. gonorrhoeae strains, it is difficult to strictly link porB alleles and accordingly NG-MAST STs to antimicrobial susceptibility phenotypes. Interestingly, seven of the eight isolates in the porB2569 group that showed higher MICs of cefixime belonged to MLST ST7363 and MLST ST1901, which are the two major STs in which cefixime resistance has emerged. This might suggest that these MLST ST7363 and MLST ST1901 isolates have horizontally acquired their porB2569 group allele from an authentic MLST ST7359 strain, which is the predominant MLST type among isolates within the porB2569 group.
The first described isolate of NG-MAST ST1407 from San Francisco, CA, displayed a decreased susceptibility to cefixime, and this isolate had a mosaic type PBP 2 encoded by the penA mosaic allele SF-A, subsequently named penA allele XXXIV (23). The high-level ceftriaxone-resistant NG-MAST ST1407 isolates identified in France (14) and Spain (25) both had an identical mutated penA XXXIV allele (penA allele CI [14, 23, 25]) that encodes a PBP 2 XXXIV sequence with leucine replaced by proline at amino acid position 501 (L501P) (14). Accordingly, these high-level resistant strains have presumably evolved by a single nucleotide polymorphism in the penA XXXIV allele resulting in an A501P alteration in PBP 2 in an NG-MAST ST1407 (MLST 1901) strain. In the present study, although many NG-MAST ST1407 N. gonorrhoeae isolates with penA XXXIV allele or derivates of this allele were identified, no penA allele CI (14, 23, 25) was found. However, some other single-amino-acid-substituted alleles of penA XXXIV, i.e., penA XXXIV with P551S or A501V, were identified. The P551S and A501V amino acid substitutions have previously been associated with decreased susceptibility to ceftriaxone (10, 37, 38). In the present study, it was shown by transformation experiments that these substitutions in a penA XXXIV allele result in a 2- to 4-fold increase in the MICs of cefixime and ceftriaxone.
In conclusion, our intensified surveillance in 2010 to 2012 in the Kyoto and Osaka prefectures, Japan, did not identify any dissemination of the high-level ceftriaxone-resistant N. gonorrhoeae strain H041, suggesting that H041 caused only a sporadic case and has not further spread. Furthermore, no other ceftriaxone-resistant strain was identified. The antimicrobial resistance surveillance in this region has continued and should now be expanded nationally.
However, it is cause for great concern that N. gonorrhoeae strains with the penA XXXIV allele and its derivatives are also shown to be spreading in Japan, as in many other countries (10, 14). The penA mosaic XXXIV allele appears also to be evolving, resulting in further-enhanced MICs of ESCs, and it might be only a matter of time before an additional high-level ceftriaxone-resistant strain emerges. If this strain is not detected in a timely fashion and has a retained biological fitness, it will spread internationally and result in a large public health problem. Consequently, it is crucial not only to enhance the phenotypic antimicrobial susceptibility surveillance of N. gonorrhoeae but also to perform molecular typing and detection to monitor the spread and evolution of successful N. gonorrhoeae clones with decreased susceptibility and resistance to ceftriaxone worldwide.
ACKNOWLEDGMENTS
We thank Makiko Itoh, Haruka Tsuruzono, Tomoko Shimamura, and Kazuma Suzuki for technical assistance.
The laboratory work was performed at the National Institute of Infectious Diseases.
The present work was supported by grants-in-aid from the Ministry of Health, Labor and Welfare of Japan (H24-Shinko-Ippan-001).
Members of the Antibiotic-Resistant Gonorrhea Study Group include Mitsufumi Fujiwara, Kei-ichi Furubayashi, Shuichi Hida, Yasuhiro Ishikawa, Mikio Itoh, Hiroshi Kameoka, Tomohiro Ueda, and Ryouji Yasumoto.
FOOTNOTES
- Received 18 June 2013.
- Returned for modification 13 July 2013.
- Accepted 3 August 2013.
- Accepted manuscript posted online 12 August 2013.
- Copyright © 2013, American Society for Microbiology. All Rights Reserved.
