The Goldilocks Zone: Searching for a Phylogenetic Approach for the Recombinogenic Neisseria gonorrhoeae

(See the Major Article by Harrison et al on pages 1816–25 and the Major Article by Cehovin et al on pages 1826–36.)

Neisseria gonorrhoeae causes gonorrhoea, a sexually transmitted infection, which was highlighted as an urgent public health threat by the World Health Organization in 2012. Of the 357 million new cases of treatable sexually transmitted diseases per year (gonorrhoea, syphilis, chlamydia and trichomoniasis), N. gonorrhoeae caused 78 million cases globally [1]. The effects of gonococcal disease are seen predominantly in women and children. High levels of disability-adjusted life-years due to persistent urethritis, cervicitis, proctitis, and disseminated infections that lead to pelvic inflammatory disease, infertility, first-trimester abortion, ectopic pregnancy, and maternal death have been predicted by the World Health Organization to increase significantly because of the spread of multidrug-resistant (MDR) N. gonorrhoeae [1]. As there is no natural immunity conferred by infection, individuals can contract the disease multiple times throughout their lifetime. Antibiotic treatment has led to the development of MDR N. gonorrhoeae, which have developed resistance to all major categories of clinically used antibiotics, including first-line extended spectrum cephalosporins and azithromycin, resulting in treatment failures [1].

A useful approach for understanding the development and spread of antimicrobial resistance (AMR) is using genomic epidemiology of the bacterial pathogen to identify the association of genetic lineage with AMR phenotypes. Developing an approach to N. gonorrhoeae genomic epidemiology has been iterative over the past 3 decades. Originally, AMR plasmids were restriction mapped and named after the first geographic site of isolation [2]. Multilocus sequence typing (MLST), which was established by Bennett and colleagues [3] and which proved very useful for the related pathogen Neisseria meningitidis, was adapted for use in N. gonorrhoeae. The MSLT scheme was instrumental in uncovering instances of genetic lineages developing AMR and spreading internationally via travel, resulting in numerous global outbreaks of MDR N. gonorrhoeae [4].

However, the introduction of routine whole-genome sequencing (WGS) of N. gonorrhoeae has revealed some interesting inconsistencies. MLST has worked well in N. meningitidis owing to the structured population of stable clonal complexes, which have been conserved by restriction modification barriers preventing unlimited horizontal recombination in these naturally transformable species [5]. This is not the case for N. gonorrhoeae. This species is panmictic as result of reduced allelic diversity but increased admixture of allelic combinations derived from homologous recombination between strains during episodes of co-colonisation [5]. Frequent recombination events have disrupted the phylogenetic relationships of MLST lineages in N. gonorrhoeae such that, in some instances, an MLST type can be associated with multiple and separate phylogenies [6]. This complication has raised questions about the interpretation of MLST phylogenies. For example, is there a single epicenter for the appearance of an AMR isolate followed by international travel, versus multiple geographically independent epicenters resulting in local clusters of AMR isolates? These differences in interpretation affect public health messaging regarding therapeutic practice, which drives the development of AMR in local jurisdictions [7].

Articles by Harrison et al [8] and Cehovin et al [9] in this issue of The Journal of Infectious Diseases address the use of MLST and a unified gonococcal plasmid nomenclature for use in the field. Harrison et al [8] have shown that extension of the MLST scheme from 7 housekeeping genes to 400 core genome loci results in a phylogeny of core genome sequence types (cg-STs) that now accurately reflects the deep phylogenetic groups of N. gonorrhoeae. This new phylogeny robustly matches with other typing schemes, such as N. gonorrhoeae multi-antigen sequence typing (NG-MAST), which maps outer membrane protein diversity, and N. gonorrhoeae sequence typing for antimicrobial resistance (NG-STAR) typing, which tracks allelic variants associated with AMR. Hierarchical bayesian analysis of the cg-STs detected 9 substantial cladal structures, highlighting that there is some weak adaptation in this species, as previously proposed using single-nucleotide polymorphism–based phylogenetic approaches [10, 11]. Cehovin et al [9] propose a uniform approach to classifying the AMR plasmids using WGS and also demonstrate that the possession of AMR plasmids is associated with cg-ST clades. This association has not been noted before and suggests that there are underlying roles for these plasmids, other than AMR transference, which may be fundamental to the biology of N. gonorrhoeae.

Together, these articles make an essential contribution to the field by establishing a uniform, robust, and informative approach to the analysis of N. gonorrhoeae genomic epidemiology. The substantial advantage to the research and diagnostic communities is the easy public accessibility of the PubMLST platform on which the analysis algorithms are run daily, and which will automatically assign cg-STs, NG-STAR, NG-MAST and MLST STs. In this respect, the PubMLST platform provides a free curated service for the analysis of WGS data, in addition to access to a large longitudinal international comparison data set, which enables simple and straightforward comparisons between international macroepidemiological studies. However, for microepidemiological studies, such as sexual network analysis, single-nucleotide polymorphism–based phylogenies are likely to remain the preferred approach, because these account for recombination events [12].

The cg-ST approach for N. gonorrhoeae genomic epidemiology and the unified approach to naming and describing AMR plasmids in the articles by Harrison et al [8] and Cehovin et al [9] enable large and small studies to be collated and compared in a uniform way to examine the global movement of MDR N. gonorrhoeae. The collation of WGS within a holistic platform integrating past typing schemes and current phylogenetic approaches will enable the rapid development of diagnostic tests to improve identification of AMR isolates, especially new combinations as they arise, to optimize patient care, and to provide a resource for the detection of conserved antigens as potential vaccine candidates [1].

Note

Potential conflicts of interest. Author certifies no potential conflicts of interest. The author has submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References