Development and validation of a high throughput Neisseria gonorrhoeae genotyping method

BackgroundNeisseria gonorrhoeae genotyping by whole-genome sequencing (WGS) is expensive for a large sample set, a less expensive and more efficient genotyping method is required. We developed a high-throughput genotyping method for N. gonorrhoeae to improve molecular epidemiological typing and antimicrobial-resistant identification in N. gonorrhoeae antimicrobial susceptibility surveillance. MethodsWe used multiplex-tailed PCR to amplify and sequence 15 alleles from multilocus sequence typing (MLST), N. gonorrhoeae multiantigen sequence typing (NG-MAST), and N. gonorrhoeae sequence typing for antimicrobial resistance (NG-STAR). After indexing-PCR, we sequenced the DNA library using the MiSeq platform (Illumina). Sequencing reads were de novo assembly or constructing consensus sequences of alleles, then assigned sequence type. We used 54 previously characterized strains of N. gonorrhoeae and WGS data to validate our method. ResultsThe allele identification results of MLST and NG-STAR in all strains agreed with the draft WGS. However, in NG-MAST, only 35 strains agreed. Disagreement was found in the NG-MAST of porB in 15 strains and of tbpB in seven strains. QRDR analysis perfectly predicted levofloxacin resistance. But was less successful in predicting reduced susceptibility or resistance phenotype to penicillin G, cefixime, or ceftriaxone using penA, porB, ponA, or mtrR alleles. ConclusionsThe successful performance in MLST and NG-STAR of our method was validated in this study. This method may be useful for large-scale genotyping for N. gonorrhoeae surveillance in a cost- and labor-saving manner. Phenotypic prediction of antimicrobial susceptibility by combining multiple alleles may be necessary for other than fluoroquinolones.