Molecular and genomic analysis of clinical multidrug-resistant coagulase-negative staphylococci from the uMgungundlovu District in the KwaZulu-Natal Province, South Africa.

Abstract

Coagulase-negative staphylococci (CoNS) are among the most commonly recovered bacteria in clinical specimens. They are usually colonisers (commensals) of the skin and nasal passages and considered contaminants of microbial cultures. However, they have been recognised as emerging pathogens, frequently causing opportunistic infections. The frequent use of indwelling medical devices and long-term hospitalisation present an increased risk of exposure to CoNS, resulting in infections usually caused by multidrug-resistant pathogens. Few studies focus on CoNS, including characterisation of their mechanisms of resistance, virulence, and persistence. Therefore, this study describes the molecular and genomic profiles of clinical CoNS from public sector hospitals in the uMgungundlovu District in KwaZulu-Natal, South Africa. Eighty-nine clinical CoNS isolates collected from three hospitals within the uMgungundlovu District between October 2019 and February 2020, constituted the sample. Isolates were speciated using the Vitek 2 system. Antibiotic susceptibility testing was done against a panel of 20 antibiotics according to Clinical and Laboratory Standards Institute (CLSI) guidelines using the Kirby-Bauer disk-diffusion method and minimum inhibitory concentration (MIC) was determined using the broth microdilution method for penicillin G, cefoxitin, ceftaroline, ciprofloxacin, moxifloxacin, azithromycin, erythromycin, gentamicin, amikacin, chloramphenicol, tetracycline, doxycycline, teicoplanin, tigecycline, linezolid, clindamycin, rifampicin, sulphamethoxazole/trimethoprim, nitrofurantoin and vancomycin. PCR was used to detect the presence of the mecA gene to confirm phenotypic methicillin resistance. Based on their resistance profiles, a sub-sample of isolates were subjected to wholegenome sequencing (Illumina MiSeq) to ascertain the resistome, virulome, mobilome, clonality and phylogenomic relationships using bioinformatic tools. The SPAdes software was used for the assembly of the raw reads. ResFinder 4.1 and CARD were used to identify antibiotic resistance genes in the isolates, while the virulence factor database (VFDB), Center for Genomic Epidemiology‘s MLST 2.0 server and MobileElementFinder v1.0.3 were used to identify virulence genes, sequence types and mobile genetic elements, respectively. Mutations in fluoroquinolone and rifampicin resistance genes were identified by manual curation using BLASTn alignment which was also used to determine the genetic environment of the resistance genes.S. epidermidis was the most abundant CoNS species isolated. Phenotypic methicillinresistance was detected in 76.4% (n=68) of isolates, 92.6% (n=63) of which were genotypically confirmed by PCR. Multidrug resistance (MDR) was observed in 76.4% (n=68) of isolates, with 51 antibiograms observed. The resistance genes mecA, blaZ, erm(A), erm(B), erm(C), msr(A), aac(6')-aph(2'') and fosB, among others, were detected and corroborated the observed phenotypes. Molecular mechanisms of resistance to tigecycline, teicoplanin, linezolid and nitrofurantoin were not detected even though some isolates were resistant to them. There was no association between ARG type and hospital/department. The ica operon known to facilitate biofilm formation was detected in 7/16 isolates sequenced. Known and putatively novel mutations in the gyrA, parC, parE and rpoB genes were also detected for fluoroquinolone- and rifampicin-resistant isolates. Prediction of isolates’ pathogenicity towards human hosts yielded a high average probability score (Pscore ≈ 0.936), which, together with the several virulence genes detected (including atl, ebh, clfA, ebp, icaA, icaB,icaC), support their pathogenic potential to humans. Seven MLST types were found, while the community-acquired SCCmec type IV was the most common SCCmec type detected. Mobile genetic elements (MGEs) haboured by isolates included plasmid replicon Rep10 and insertion sequence IS256. Defense systems such as arginine catabolic mobile element (type I and III), CRISPR system (16), and the restriction-modification system (type II) were detected. Genetic analysis showed that resistance genes were frequently bracketed by MGEs such as transposons (such as Tn554) and insertion sequences (such as IS257 and IS1182) that facilitated their mobility. Phylogenetic studies showed that the distribution of genes did not coincide with the phylogenetic clades. Despite the relatedness of isolates (clades A and B), there is still considerable variation within individual strains that can facilitate adaptation to local environments. The isolates exhibited several permutations and combinations of ARGs, virulence genes and MGEs, pointing to a complex milieu of mobilized antibiotic resistance and pathogenic characteristics in clonal and multiclonal strains. The study necessitates surveillance of CoNS as emerging pathogens.