High rate of fluoroquinolone resistant Neisseria gonorrhoeae detected by molecular surveillance of antimicrobial resistance determinants in Bulgaria

Abstract Antimicrobial-resistant Neisseria gonorrhoeae is a major public health concern. The surveillance of antimicrobial resistance benefits from rapid and accurate molecular techniques in molecular diagnostics to facilitate individualised medicine and antimicrobial stewardship. To support the recommendations for empirical treatment of gonococcal infections in Bulgaria, we investigated N. gonorrhoeae-positive clinical specimens from 2018 to 2021 for the presence of genetic determinants associated with antimicrobial resistance. N. gonorrhoeae-positive samples stored at the National Center of Infectious and Parasitic Diseases during the four-year study period were retrospectively analysed by polymerase chain reaction and DNA sequencing assays for resistance determinants to fluoroquinolones, third-generation cephalosporins and macrolides. The detected determinants indicated a high rate of fluoroquinolone resistance (59%), very low level of decreased susceptibility to third-generation cephalosporins (3%) but no macrolide resistance (0%). These findings validate the utilisation of the international guidelines’ recommendations for empirical dual therapy with ceftriaxone/cefixime and azithromycin in Bulgaria. Because of the high fluoroquinolone resistance rate, ciprofloxacin should only be considered as treatment if phenotypic or molecular antimicrobial susceptibility data indicate susceptibility to ciprofloxacin. For the purposes of surveillance and individualised medicine, molecular assays for resistance determinants could complement culture-based phenotypic gonococcal antimicrobial resistance testing.


Introduction
Sexually transmitted infections (STIs) caused by Neisseria gonorrhoeae remain public health concern on a global scale. According to the World Health Organization (WHO), the global prevalence of urogenital chlamydial infection, gonorrhoea, trichomoniasis and syphilis in adults of reproductive age (15-49 years) is high, with nearly one million newly diagnosed cases each day [1]. In 2020, WHO estimated 82.4 million [95% CI 47.7 million-130.4 million] new global gonococcal cases among adults yearly [2]. The rapid development of antimicrobial resistance in N. gonorrhoeae is significant, as it considerably compromises the effectiveness of treatment and, consequently, the prevention and control of gonococcal infections [3]. N. gonorrhoeae is among the top 12 bacterial pathogens on the WHO antimicrobial resistance priority pathogen list and ranks in the group of the second-highest level of concern for public health [4]. N. gonorrhoeae has developed resistance to all antimicrobials for empiric monotherapy [5,6], including the third-generation cephalosporin ceftriaxone [7]. This led to the introduction of dual antimicrobial therapy with ceftriaxone (250 mg-1 g single intramuscular dose) and azithromycin (1-2 g single oral dose) in the last decade [8][9][10], until the first report of its failure to cure a case of gonorrhoea in 2016 [11]. Soon after, the first gonococcal isolates that show resistance to ceftriaxone combined with high resistance to azithromycin emerged in England and Australia [12,13].
An additional cause of concern is that gonococcal strains with resistance or decreased susceptibility to the third-generation cephalosporins have spread worldwide [14,15]. Many cases of gonorrhoea have become difficult to treat, and some N. gonorrhoeae infections can become incurable [16]. In response, WHO published a global action plan including measures to mitigate the emergence, spread and impact of antimicrobial resistance in N. gonorrhoeae [17]. This global action plan sets as a key priority to significantly improve quality-assured gonococcal antimicrobial resistance surveillance globally, which has led to the strengthening of gonococcal antimicrobial surveillance programs at multiple levels [18][19][20][21][22][23]. In many less-resourced settings, however, the management of gonorrhoea is only syndromic, and in most well-resourced settings, molecular diagnostics have replaced culture to detect N. gonorrhoeae, resulting in a lack of culture-based phenotypic testing for antimicrobial resistance [16]. Therefore, it is essential to introduce reliable, sensitive and specific molecular detection of antimicrobial resistance determinants to predict gonococcal resistance. These tests should be suitable for monitoring antimicrobial resistance, but also be designed to guide individualised appropriate treatment for each gonorrhoea patient, reducing the selection pressure for antimicrobial resistance and ineffective use of antimicrobials [16,24].
In this context, the objectives of this study were to investigate the genetic resistance determinants in N. gonorrhoeae-positive clinical specimens from 2018 to 2021. Polymerase chain reaction (PCR) and DNA sequencing were used to detect resistance determinants implicated in gonococcal antimicrobial resistance to currently recommended dual antimicrobial therapy and to an alternative treatment option, in order to accumulate scientific evidence to back up the recommendations for treatment of gonococcal infections in Bulgaria.

Ethical approval and informed consent
The study was reviewed and approved by the institutional review board (IRB) 00006384 and written informed consent was obtained from all patients for personal data collection and microbiological sample testing.

Clinical samples
A total of 1205 urogenital samples (first void urine, urethral and endocervical swabs) were referred to the National Center of Infectious and Parasitic Diseases (NCIPD, Sofia, Bulgaria) from 2018 to 2021. The samples were collected from individuals attending the Sexual Health Center in Sofia for voluntary and confidential HIV testing. All samples were initially tested by molecular methods (AmpliSens® for detection of Neisseria gonorrhoeae-screen-FRT and confirmatory porA/opa assay [25]), and were classified as N. gonorrhoeae PCR-positive and PCR-negative. Following the diagnostic testing, DNA extracts from each sample were stored at NCIPD frozen at −80 °C for further analysis.

Detection of genetic determinants associated with antimicrobial resistance
Two real-time PCR assays (one of them carried out as a multiplex assay) and 23S rRNA sequencing were used in this study for the detection of genetic determinants associated with antimicrobial resistance to fluoroquinolones, third-generation cephalosporins and macrolides.
First, stored DNA extracts from N. gonorrhoeae-positive samples were analysed by real-time PCR for determinants associated with resistance to fluoroquinolones (gyrA/parC) and third-generation cephalosporins (PBP-2) using the primers and probes described in Table 1. PCRs were performed using standard conditions or conditions previously described [26,27]. Positive and negative control samples were included in each run. All samples were tested in duplicate.
Previously validated assays targeting the gyrA and parC genes were used in a multiplex format to detect mutations in the quinolone Resistance-Determining Region (qRDR) of N. gonorrhoeae, which are associated with resistance to fluoroquinolones [27]. The assay exclusively detected the wild-type qRDR of the gyrA and parC genes. When mutations were present in the Table 1. primer and probe sequences for the molecular detection of N. gonorrhoeae antimicrobial resistance determinants. assay/ target primers and probes primers and probe sequence 5′-3′ ng23s1905-F acggtcctaaggtagcga ng23s2769-R tctcatcttcaggcgagtt a assay adapted from giles et al. [27]. b assay adapted from ochiai et al. [26]. c assay adapted from allen et al. [28]. * the gyrA and parC assays were carried out as a qRDR multiplex assay.
qRDR, the gyrA and ParC probes did not recognise the target sequence, resulting in negative PCR results [27]. Subsequently, the Penicillin-binding protein 2 (PBP-2) assay was conducted. The PBP-2 assay targeted the mosaic structure of the N. gonorrhoeae penA gene, which encodes PBP-2 and has been shown to be associated with decreased susceptibility to the third-generation cephalosporins [26].
Lastly, mutations in the four copies of the 23S rRNA genes associated with macrolide resistance in N. gonorrhoeae were analysed using a strategy previously described by Allen et al. [28]. Briefly, a common primer inside the four copies of the 23S rRNA genes and specific external primers close to each of them were used for the first round of PCRs. using the amplicons obtained in these first PCRs as DNA templates, nested PCRs were performed using common primers internal to the 23S rRNA gene (including the peptidyltransferase loop in domain V) to identify point mutations associated with azithromycin resistance (Table 1). All the amplicons were sequenced with the same specific primers as those used for the nested PCR under the DTCS quickStart Cycle sequencing methodology in the genomeLab geXP™ genetic Analysis System (Sciex, Framingham, MA, uS). Sequence editing and multiple sequence alignments were performed using the software CLC Main Workbench, version 22.0.2 (https:// digitalinsights.qiagen.com).

Neisseria gonorrhoeae-positive samples
Of the 1205 urogenital samples referred for N. gonorrhoeae testing, 2.4% (29/1205) were found positive and further analysed. The median age of the patients with gonococcal infection was 27 (age range 20-42); 25 (86.2%) were men and 4 (13.8%) were women. The prevalence of the genetic determinants associated with fluoroquinolone resistance, decreased susceptibility to third-generation cephalosporins and macrolide resistance among the detected N. gonorrhoeae-positive samples are shown in Figure 1.

Decreased susceptibility to the thirdgeneration cephalosporins
The prevalence of mosaic penA gene associated with decreased susceptibility to cephalosporins was very low in this study. Only one sample (3%) tested positive in the PBP-2 assay (Table 2).

Macrolide resistance
The targeted region of the four copies of the 23S rRNA gene was successfully amplified and sequenced in 27 (93.1%) samples. The samples that did not produce PCR amplification products in the macrolide resistance testing were late positive in the initial diagnostic PCR. All four copies of the 23S rRNA gene were successfully  sequenced in 25 (86.2%) samples. The 23S rRNA A2059g and 23S rRNA C2611T mutations (in one or several alleles) associated with macrolide resistance were not detected in any of the samples analysed in this study (Table 2). Multiple sequence alignment of partial sequences of 23S rRNA genes of N. gonorrhoeae-positive samples were mapped to pan-susceptible reference N. gonorrhoeae strain (NCBI Reference Sequence: NC_011035) and are shown in comparison to previously characterised as macrolide resistant WHO reference strains (Figure 2) [29].

Discussion
This molecular surveillance study of the antimicrobial resistance determinants in N. gonorrhoeae-positive clinical specimens from Bulgaria indicated a high rate of fluoroquinolone resistance (59%), very low level of decreased susceptibility to the third-generation cephalosporins (3%) and no macrolide resistance (0%) among gonococcal strains in this country from 2018 to 2021 (Figure 1.). unfortunately, most of Bulgaria's neighbouring countries do not report gonococcal antimicrobial resistance data, with the exception of greece with concerningly higher than the average European rates (71% fluoroquinolone resistance, 6% decreased susceptibility to third-generation cephalosporins and 29% macrolide resistance [30]).
Molecular assays for detection of N. gonorrhoeae antimicrobial resistance determinants are more and more common for molecular surveillance [31], as knowledge of the genetic determinants of resistance has greatly improved with the implementation of high-throughput DNA sequencing methodology such as whole genome sequencing (WgS) [16,31]. ultimately, detection assays for N. gonorrhoeae antimicrobial resistance determinants could be used to guide the individualised antimicrobial therapy, which should increase the treatment efficacy and alleviate the further spread of antimicrobial resistance in N. gonorrhoeae [16]. Compared to cultural diagnostics and phenotypic susceptibility testing, molecular methods have many advantages, including superior sensitivity and high specificity, shorter turnaround time, automation, high throughput and potential for use as rapid tests [6,16,31]. However, a disadvantage of the molecular assays for the detection of resistance determinants is that they cannot detect new determinants of antimicrobial resistance. They also cannot provide a complete profile of antimicrobial resistance with minimum inhibitory concentrations (MICs) of the antimicrobials. In addition, it can be very labour intensive to detect more than one resistance determinant [16]. Therefore, molecular assays for antimicrobial resistance testing could not completely replace culture, but only supplement it. Nevertheless, the increasing use of molecular assays for the identification of N. gonorrhoeae from clinical samples has entailed a reduction in the cultured isolates available for phenotypic susceptibility testing. Thus, the molecular approaches for the detection of resistance determinants in N. gonorrhoeae-positive clinical specimens become essential [32]. In the present study, PCR and DNA sequencing approaches were used to detect resistance determinants in four different genes implicated in gonococcal antimicrobial resistance. The investigated genetic determinants were those that could detect resistance to the currently recommended dual antimicrobial therapy with ceftriaxone (250 mg-1 g single intramuscular dose) and azithromycin (1-2 g single oral dose) [9] and to a potentially useful alternative of the older recommendations for gonorrhoea such as ciprofloxacin [33]. The current use of ciprofloxacin for the treatment of gonococcal infections should be considered after confirmed susceptibility to ciprofloxacin and has the advantages of providing effective oral treatment of both urogenital and extragenital infections, limited side effects, and reduced selective pressure for emergence and spread of resistance to dual therapy with ceftriaxone and azithromycin [16,24]. Ciprofloxacin is a synthetic broad-spectrum fluoroquinolone that inhibits bacterial DNA synthesis by targeting two major type II topoisomerases, DNA gyrase and topoisomerase IV [34]. In N. gonorrhoeae, the resistance to ciprofloxacin mainly results from mutations that reduce the target affinity of fluoroquinolones [16]. The most effective targets for the detection of ciprofloxacin resistance in N. gonorrhoeae are mutations in the gyrA gene [16], specifically the one encoding gyrA S91F, and concomitant mutations in the parC gene (e.g. ParC D86N), resulting in high-level ciprofloxacin resistance [34]. The results of this study showed 59% fluoroquinolone resistance among N. gonorrhoeae in Bulgaria, which is higher than the mean level of fluoroquinolone resistance in the European region (49%) [30]. Therefore, in Bulgaria ciprofloxacin should only be recommended for treatment of gonorrhoea after susceptibility verification with either phenotypic or molecular testing.
Azithromycin is used to treat a broad range of infections including soft tissue, gastrointestinal, respiratory and sexually transmissible infections, and can be used during pregnancy. It is commonly prescribed in the sexual health setting as a single dose due to the favourable pharmacokinetics of the drug (e.g. excellent side-effect profile, with low rates of 'mild' gastrointestinal adverse events [33,34]). Azithromycin has good activity against N. gonorrhoeae and its use as a component of dual therapy in combination with ceftriaxone is widespread, with the aim of slowing the development and spread of resistance [35]. Azithromycin is a semisynthetic macrolide that inhibits RNA-dependent peptide synthesis in the 50 ribosomal subunits by binding near 23S rRNA. The major azithromycin resistance determinants in N. gonorrhoeae are mutations in the 23S rRNA target [16,34]. Moderate and high-level resistance to azithromycin is conferred by 3-4 mutations in the 23S rRNA C2611T and 23S rRNA A2059g, respectively, and may be selected in isolates with low-level resistance due to mutations in just one or two of the mutated alleles [32,34]. Thus, as pointed out by golparian and unemo [16], it is important to detect strains with only one mutated allele since, under macrolide selective pressure, the 23S rRNA mutations can be transferred to the other alleles [36]. Our study showed no 23S rRNA C2611T or A2059g target mutations, which is in contrast to the overall increase of macrolide resistance in the European union (Eu) and European Economic Area to 13.3% in 2018 from 7.5% in 2016 (30). The lack of macrolide resistance among the tested samples in our study can be attributed to several factors, including less frequent N. gonorrhoeae testing compared to many European countries, syndromic management in some settings and unfavoured azithromycin usage for monotherapy of gonorrhoea. Although these results suggest azithromycin is a valid option against gonococcal infections in Bulgaria, it should only be used in conjunction with another effective agent (as in dual therapy) and the gonococcal macrolide resistance should be further monitored.
The third-generation cephalosporins cefixime and ceftriaxone have been commonly used for treatment of gonococcal infections [16]. As afore-mentioned, dual antimicrobial therapy consists of a single intramuscular injection of ceftriaxone plus a single oral dose of azithromycin [8][9][10]. Oral cefixime serves as an alternative to ceftriaxone and may provide a more acceptable route of administration in some patients. Because of lower and less persistent blood bactericidal levels, cefixime should be indicated only if ceftriaxone is not available [37,38]. Third-generation cephalosporins target penicillin-binding proteins (PBPs), which strengthen the cell wall by cross-linking peptidoglycan units. The β-lactam ring of third-generation cephalosporins binds primarily to PBP2 (encoded by the penA gene) in N. gonorrhoeae, which inhibits PBPs activity [16,39]. Decreased susceptibility to third-generation cephalosporins is mainly caused by penA mosaic alleles, which is the major resistance determinant [16]. However, many additional genes and mutations in these genes contribute to the emergence of resistance to third-generation cephalosporins. In addition, new resistance determinants and mutations are continuously being identified [16,40], which makes the detection of resistance to the third-generation cephalosporins more complicated than to many other antimicrobials. In Bulgaria, the detection of only one gonococcal case (3%) with decreased susceptibility to the third-generation cephalosporins was in accordance with the overall low prevalence of cephalosporin resistance (2.4%) in the Eu and the European Economic Area countries [30]. Importantly, although detected decreased susceptibility does not ground treatment failure, it is a molecular stepping stone for the subsequent development of resistance [7]. Because of the threat of resistance development and treatment failure with third-generation cephalosporins, the current recommended treatment in Bulgaria should remain ceftriaxone (cefixime if ceftriaxone is unavailable) in combination with azithromycin.

Conclusions
This molecular surveillance study investigated the antimicrobial resistance determinants in N. gonorrhoeae-positive clinical specimens in Bulgaria from 2018 to 2021. High rate of fluoroquinolone resistance (59%), very low level of decreased susceptibility to third-generation cephalosporins (3%) and no macrolide resistance (0%) was detected using PCR and DNA sequencing approaches. These findings validate the utilisation of the international guidelines' recommendations for empirical dual therapy with ceftriaxone/cefixime and azithromycin in Bulgaria. Because of the high fluoroquinolone resistance rate, ciprofloxacin should only be considered as treatment if phenotypic or molecular antimicrobial susceptibility data indicate susceptibility to ciprofloxacin. Accordingly, for the purposes of surveillance and individualised medicine, molecular assays for resistance determinants could complement culture-based phenotypic gonococcal antimicrobial resistance testing.