Background

Neisseria gonorrhoeae is the etiological agent of the sexually transmitted infection gonorrhea, which remains a major public health issue. It represents 88 million of the estimated 448 million new cases of curable sexually transmitted infections that occur globally every year [1]. Over time, N. gonorrhoeae has progressively developed resistance to a wide range of antibiotics, including penicillin, ciprofloxacin, tetracyclines, macrolides, and more recently to third-generation cephalosporins [2, 3]. Antimicrobial resistance in N. gonorrhoeae became apparent shortly after the introduction of antibiotics into clinical practice. Moreover, its ability to acquire and/or maintain antibiotic resistance genes has become a considerable problem and an obstacle to successful therapeutic treatment [4, 5]. In 1943, penicillin was introduced as the first-line treatment of gonorrhea, but it was abandoned due to the emergence of chromosomally mediated penicillin resistance and PPNG isolates [6, 7]. PPNG strains were reported for the first time in 1976 in the United Kingdom and the United States [8, 9]. These strains produce a TEM type β-lactamase, which is carried on several related plasmids. To date, eight plasmid types have been described and named according to their epidemiological origin. Strains harbouring Asian, African and Toronto/Rio plasmids have been reported throughout the world [10, 11]. Other plasmid variants have been described, including Nimes, New Zealand, Johannesburg, and Australian plasmids [1215]. The PPNG isolate in Argentina was reported in the 1980’s and has spread since then [16]. PPNG isolates are highly prevalent in our country and three plasmids types (Asian, African and Toronto/Rio plasmids) have been detected, resulting African and Toronto/Rio the most prevalent.

The first PPNG strains produced a TEM-1 β-lactamase, a class A enzyme encoded by bla TEM-1 allele [17]. TEM-1 β-lactamase efficiently hydrolyzes penicillins and many cephalosporins, but it is not an effective catalyst for extended spectrum cephalosporin turnover. For around thirty years, other β-lactamases were not described in gonococci. However, a PPNG isolate carrying a bla TEM-135 gene was reported from Thailand in the year 2009 [18]. TEM-135 was first described in 2005 in a Salmonella enterica subsp. enterica serovar Typhimurium isolate and differs from TEM-1 by one single nucleotide substitution at position 539 (T → C), resulting in a single amino acid substitution, M182T [19]. Prevalence studies from different countries revealed that PPNG isolates possessing Asian, Toronto/Rio, African and Australian plasmids carried bla TEM-135 gene [2023]. Recently PPNG isolates carrying Toronto/Rio plasmid from Argentina that possessed the novel bla TEM-220 allele were identified [24]. TEM-220 differs from TEM-135 by one nucleotide substitution at position 547 (G → A), resulting in the amino acid substitution A185T. The prevalence and characteristics of TEM-220 strains worldwide are unknown and it seems imperative to study them.

Prevalence of PPNG isolates in Argentina has been reported previously, but national data of PPNG isolates possessing bla TEM-220 is lacking. The objective of our study was to detect bla TEM-220 gene in PPNG isolates possessing Toronto/Rio plasmid recovered between 2002 and 2011 in Argentina, and to evaluate the proportion of isolates producing non-TEM-220 containing the T539C substitution in the bla TEM allele.

Methods

Collection of isolates

The N. gonorrhoeae clinical isolates investigated in this study were collected as part of Gonococcal Antimicrobial Susceptibility Surveillance Programme-Argentina (GASSP-AR). The isolates were obtained from 71 medical centers in 24 of the 24 Argentinian provinces from 2002 to 2011. We studied 151 PPNG isolates carrying Toronto/Rio plasmid previously identified as N. gonorrhoeae based on the characteristic colony morphology, Gram staining, oxidase test, superoxol test (30 % hydrogen peroxide), carbohydrate utilization test, and the Phadebact GC Monoclonal Test (MKL Diagnostic AB, Sollentuna, Sweden) [25]. All isolates were assigned a code number, preserved in tryptic soy broth (TSB) + 20 % glycerol at – 70 °C and added to the N. gonorrhoeae isolate collection of the reference laboratory. In this study only bacterial strains from clinical specimens collected from GASSP-AR were used and no patient information was accessed. Consequently, ethical approval was not required for this study.

Each frozen culture suspension was freshly subcultured on Difco GC Medium Base agar (BD, Franklin Lakes, NJ, USA) supplemented with 1 % Britalex enrichment supplement (Britania Lab., Argentina) and incubated at 35 °C in a humidified environment and enriched with 5 % CO2 during 24 to 48 hours. Cultures were examined once a day and colonies were subcultured using the same medium and incubation parameters as above. The identity of resurrected isolates was confirmed by Gram staining and the production of beta-lactamase by chromogenic Nitrocefin disc method (BD, Franklin Lakes, NJ, USA). Plasmid profile was studied using a boiling plasmid extraction method followed by agarose gel electrophoresis detection [26, 27].

Antimicrobial susceptibility testing

The minimal inhibitory concentration (MIC) values of penicillin, ampicillin, cefuroxime, tetracycline, ciprofloxacin, ceftriaxone, cefixime and azithromycin were determined using the reference agar dilution method (CLSI) [28]. The N. gonorrhoeae ATCC 49226 and WHO reference strains were used as quality control for the MIC determinations [29]. The results were interpreted in accordance with CLSI breakpoints, except for azithromycin, for which the European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoints were applied [30]. For ampicillin, as CLSI and EUCAST do not describe any breakpoints, the susceptibility categories were inferred from the penicillin breakpoints.

DNA isolation

DNA was extracted from a fresh subculture by the boiling method. Colonies were suspended in 3 ml of ultrapure water and adjusted to a turbidity equivalent to a 1 McFarland standard using a calibrated turbidimeter. An aliquot of 500 μl of this suspension was transferred to a microcentrifuge tube with a capacity of 1.5 ml, then heated for 20 min at 98 °C, and cooled to 4 °C for 3 min. The mixture was centrifuged for 5 min at 10000 rpm and the supernatant was transferred to a new tube. The extracted DNA was used directly as template in the PCR and then stored at −20 °C.

Mismatch amplification mutation assay - PCR

In order to detect N. gonorrhoeae strains possessing bla TEM-220 allele, a MAMA-PCR protocol was developed to identify a single nucleotide substitution at position 547 (G → A) in the bla TEM gene. The MAMA TEM-220 primers used in this study were designed using free online Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/primer-blast/). Assay design consisted of a forward primer (T220-F, 5′-TGACACCACGACGCCTGCTA-3′), which annealed a bla TEM-220 allele-specific polymorphism, and a conserved reverse primer (T220-R, 5′-ATGATACCGCGAGACCCACG-3′). A ‘destabilizing’ mismatch at the penultimate base position from the 3′end of the forward primer was introduced to enhance the 3′ mismatch effect (Fig. 1) [31, 32]. Amplification was carried out in a 25 μl reaction mixture containing 1 μl DNA template, 1X reaction buffer, 1.0 mM MgCl2, 0.2 mM (each) deoxynucleoside triphosphate, 0.5 μM of each primer, and 0.75 unit of Taq polymerase (Invitrogen/Life Technologies, California, USA). The parameters of the amplification were as follows: initial denaturation at 96 °C for 2 min, followed by 30 cycles of 96 °C 10 s, 62 °C for 15 s, and 72 °C for 30 s, with final extension at 72 °C for 2 min in a MyCycler thermal cycler (Bio-Rad, California, USA). The amplification products were analyzed by means of electrophoresis on a 2 % (w/v) agarose gel. The specificity of the primer was evaluated by testing the previously described N. gonorrhoeae strains: IM4519 and IM4520 containing bla TEM-220, IM5923 and IM4540 containing bla TEM-135, and IM4629 and IM5257 containing bla TEM-1. These strains were used as a control in all PCRs.

Fig. 1
figure 1

Alignment and difference of DNA sequence between the bla TEM-1, bla TEM-135, and bla TEM-220 gene. The nucleotide substitutions found in the different types of bla TEM alleles are in bold type. The forward primer MAMA TEM-220 used in this study is shown schematically with arrows. Underlined letter indicates the nucleotide alteration near the 3′ end of the primer to enhance the 3′ mismatch effect

Detection of the T539C substitution in the bla TEM gene

An allele-specific PCR method previously described was used to investigate N. gonorrhoeae isolates containing bla TEM gene with the substitution at position 539 (T → C) [33]. N. gonorrhoeae strains IM4540 and IM5923 containing bla TEM-135 and IM4629 and IM5257 containing bla TEM-1 were used as controls in the PCR.

Sequencing of the bla TEM-220 isolates

In all isolates that were positive to MAMA TEM-220 PCR, the whole bla TEM was amplified and sequenced to confirm the bla TEM-220 allele. The bla TEM allele was amplified as previously described [34]. The PCR products were purified using AccuPrep® PCR Purification Kit (Bioneer, Daejeon, Republic of Korea). DNA sequencing was performed using the BigDye® terminator v3.1 cycle sequencing kits (Applied Biosystems, Foster City, CA, USA) on an ABI 3500 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). The sequences were compared and aligned to previously described bla TEM-220 allele [GenBank: accession number KM998962] using the BioEdit (version 7.2.5) software.

Genotyping

The NG-MAST analysis was performed exclusively on TEM-220-producing isolates. The porB and tbpB alleles were sequenced, and sequence types (STs) were assigned at the NG-MAST website (http://www.ng-mast.net) following the interpretative procedures previously described [35].

Results

Over the period from 2002 to 2011, 3895 N. gonorrhoeae isolates were submitted from GASSP-AR for antimicrobial susceptibility testing. Overall, 476 (12.2 %) isolates were β-lactamase positive. The proportion of PPNG isolates was over 5 % in all years and ranged between 8.9 and 17.9 %. During the period of study two plasmid types were found. The African plasmid type was the most common in all years, comprising 325/476 isolates (68.3 %), followed by Toronto/Rio plasmid 151/476 (31.7 %). The serotyping of N. gonorrhoeae strains carrying Toronto/Rio plasmid revealed that the serogroup PorB1b (WII/III) predominated among these strains 134/151 (88.7 %), followed by serogroup PorB1a (WI) 17/151 (11.3 %).

Use of the MAMA-PCR protocol for detection of the G547A substitution

A MAMA-PCR protocol was successfully designed for the identification of the substitution at position 547 (G → A) in the bla TEM-220 gene. Implementation of the corresponding PCR protocol showed that the primers were able to differentiate between PPNG strains with the G547A mutation in the bla TEM allele and those without it (Fig. 2).

Fig. 2
figure 2

Discrimination of PPNG isolates with the G547A substitution in bla TEM allele by using T220-f/T220-r primers. Lanes 1 and 2, indicate the 204 bp PCR products from strains IM4519 and IM4520 with the G547A mutation in bla TEM allele; lanes 3, 4, 5 and 6 indicate PPNG control strains containing bla TEM-1 and bla TEM-135, respectively, without amplification; lane M, molecular size marker

Four of the 151 PPNG isolates (2.6 %) showed a PCR product of the expected size (204 bp), suggesting that these isolates carried the bla TEM-220 gene. PCR amplification and sequencing analysis of the whole bla TEM gene revealed a substitution at position 539 T → C and 547 G → A, confirming that these four isolates possessed the bla TEM-220 allele.

Identification of the T539C substitution in the bla TEM gene

The 147/151 MAMA TEM-220 PCR-negative isolates were investigated for the T539C substitution in the bla TEM gene. From those, 137/147 (93.2 %) isolates showed a MAMA TEM-135 PCR product of the expected size (231 bp), which indicate the presence of the T539C substitution in the bla TEM gene. The 10/147 (6.8 %) MAMA-TEM135 PCR-negative isolates, were TEM-1 PCR positive, suggesting the presence of bla TEM-1 allele in these isolates.

MICs of Isolates

The MICs of 137 PPNG isolates carrying the T539C substitution in the bla TEM gene and all (n = 4) PPNG isolates carrying bla TEM-220 allele are summarized in Table 1. All isolates were resistant to penicillin and ampicillin and susceptible to cefuroxime, cefixime and ceftriaxone. No differences in MICs of five beta-lactam antibiotics tested were observed in PPNG isolates producing TEM-220 and PPNG isolates producing non-TEM-220 carrying the T539C substitution in the bla TEM gene. The 97.1 % of PPNG isolates containing the T539C substitution in the bla TEM gene were susceptible to ciprofloxacin (MIC: ≤0.016 μg/ml), whereas 100 % of PPNG strains producing TEM-220 were susceptible. All isolates were susceptible to azithromycin (MIC: ≤ 0.25 μg/ml).

Table 1 Susceptibility of four PPNG isolates producing TEM-220 β-lactamase and PPNG strains containing the T539C substitution in the bla TEM allele

Genotyping

The four PPNG isolates possessing bla TEM-220 were isolated in 3 of the 24 provinces of Argentina. The geographic distribution of these isolates was in the north and southern part of the country. All of this isolates were indistinguishable by NG-MAST and shared alleles por 6406 and tbpB 21, which make up NG-MAST type 10972.

Discussion

This study was conducted to analyze the prevalence of bla TEM-220 and to detect the T539C substitution in the bla TEM allele in PPNG isolates carrying Toronto/Rio plasmid. Prevalence data on plasmid profile and TEM-type beta-lactamase of PPNG isolates from South America is limited [36]. In Argentina, penicillin is not recommended as first-line therapy for the treatment of gonorrhea since the early 1990s [37]. However, a high percentage of isolates are beta-lactamase positive. In this study, the prevalence of PPNG strains was 12.2 % between 2002 and 2011 and the plasmid profile revealed two types of circulating plasmids, resulting the African-type plasmid the most frequent in all years.

PPNG isolates possessing bla TEM-220 have recently been described [24]. Here, we describe a MAMA PCR method for the specific detection of the G547A substitution in the bla TEM-220 allele. The results showed that the MAMA PCR successfully detected four PPNG isolates carrying bla TEM-220, which were confirmed by sequencing the whole bla TEM gene. The bla TEM-220 allele was found in 2.6 % of all isolates. These strains were resistant to penicillin and ampicillin, but susceptible to cefuroxime, cefixime and ceftriaxone. According to this, it is believed that these substitutions have no effect on the substrate spectra of the enzyme. Moreover, the β-lactamase phenotype of bla TEM-220 gene does not correspond to an ESBL phenotype. All isolates were assigned to serogroup PorB1b (WII/III) and NG-MAST was used to investigate the diversity and relatedness of the four TEM-220-producing isolates. All TEM-220 isolates were assigned the same ST 10972, indicating that these strains had originated from a common ancestor. In a previous study of bla TEM-135 possessing isolates in Argentina, ST 10972 was found to be prevalent in 19.0 % of the TEM-135 isolates carrying Toronto/Rio plasmid [24]. We did not observe different patterns of resistance when comparing susceptibility or resistance to penicillin, cefixime, ceftriaxone, ciprofloxacin, tetracycline and azithromycin between TEM-220 isolates and previously described TEM-135 isolates with ST 10972 (data not shown). Based on these observations, other factor(s) than antimicrobial selective pressure may have been the selective force that drove the emergence of TEM-220 beta-lactamase. Although bla TEM-220 allele might have been acquired through horizontal gene transfer from an unknown bacterial source, the highly similar contexts between TEM-135 and TEM-220 isolates suggest that bla TEM-220 allele could have evolved from bla TEM-135 gene by a single nucleotide polymorphism (SNP).

The study of the substitution at position 539 (T → C) in MAMA TEM-220 PCR-negative isolates revealed that 93.2 % of the isolates contained this mutation in the bla TEM gene. The T539C mutation results in a single amino acid substitution (M182T), which is believed to restore the stability of the enzyme affected by substitutions near the active site [38, 39]. M182T accompanies additional mutations that may extend the substrate specificity of the enzyme to ESBL such as TEM-20 and TEM-52 [40, 41]. Nowadays, extended spectrum cephalosporins are used as first line therapy in our country, and resistance has not been detected. However, an additional SNP added to the bla TEM-220 or bla TEM containing M182T substitution could lead to the emergence of ESBL as a response to the selective pressure induced by extended spectrum cephalosporin. This leads to the imperative need for monitoring for the possible emergence and spread of ESBL-producing N. gonorrhoeae.

A limitation of the current study was the lack of demographic and epidemiological information which limited the comparison between PPNG isolates and patient characteristics.

Conclusions

In summary, this study shows the prevalence of TEM-220 in N. gonorrhoeae isolates carrying Toronto/Rio plasmid in Argentina. The bla TEM-220 allele does not appear to be associated with an ESBL phenotype, and high percentage of PPNG isolates carrying Toronto/Rio plasmid contained the T539C substitution in the bla TEM gene. Knowledge about the diversity and prevalence of β-lactamases in PPNG strains is crucial in order to detect the possible emergence of resistant variants to extended-spectrum cephalosporin.