Genomic characterisation of Salmonella enterica from environmental, animal, food and human sources in Zimbabwe

Abstract

Salmonella enterica is a globally ubiquitous, foodborne, zoonotic pathogen. However, there is a paucity of knowledge about the sources of Salmonella responsible for clinical infections in Zimbabwe. Although classification of Salmonella isolates into serovars is an important component of epidemiological investigations, determination of Salmonella serovars by conventional serotyping has many drawbacks. To overcome these limitations, multiple methods have been investigated including development of molecular serotyping schemes, for example, pulsed-field gel electrophoresis (PFGE) and multi-locus sequence typing (MLST). While these methods resolved some of the challenges of traditional methods, resolution continued to be a limitation.

The study of infectious disease has been transformed by next-generation sequencing (NGS) substantially increasing the knowledge of Salmonella epidemiology. The purpose of this study was to determine the epidemiology and genomic characteristics of S. enterica from environmental, animal, food and human sources in Zimbabwe using whole-genome sequencing (WGS). Ninety-seven clinical S. Typhi isolates from human infections (2012 to 2019) and 245 non typhoidal Salmonella (NTS) (2016 to 2020) recovered from chicken farms and clinical infections were analysed using NGS. This study identified S. Typhi 4.3.1.1 (H58) carrying an IncN (subtype PST3) plasmid as being prevalent in Zimbabwe. The strain exhibits reduced susceptibility to ciprofloxacin but has evolved to full resistance over 6 years from 2014 to 2019. The non-H58 strains (genotype 3.3.1) lacked any known AMR determinants. Whole-genome sequencing analysis identified Harare as the source of typhoid outbreaks in Zimbabwe, hence the introduction of typhoid vaccine in hotspot areas. The topology of the global phylogeny of H58 strains suggests that Kenya is the most likely origin of strains now prevalent in Zimbabwe, with onward spread through Tanzania and Malawi. This is reiterated in multiple reports from other researchers postulating that H58 strains spread through East African countries.

The population structure of the 245 NTS isolates showed a total of 45 different serovars with S. Enteritidis (42/245, 17%) and S. Kentucky (37/245, 15%) as the most prevalent serovars. In-depth analysis of the S. Kentucky serovars showed that the majority of isolates from both poultry and clinical samples harboured the following antimicrobial resistance genes: blaCTX-M-14.1 (89.2%) and fosA3 (89.2%) as well as genetic mutation in the gyrA and parC genes (97.3%). These isolates belonged to sequence type (ST) 198, an epidemic clone. The close genetic relationship and population structure of the human clinical and poultry isolates of ST198 in Zimbabwe confirmed that poultry is an important source of clinical infection in Zimbabwe. The high prevalence of antibiotic-resistant Salmonella serovars amongst both clinical and poultry samples, underscores the necessity for evidence-based control strategies and treatment guidelines in Zimbabwe. This study is the first of its kind and aimed to characterise both typhoid and NTS in Zimbabwe.

WGS analysis provided detailed evidence on the genetic mobile elements causing AMR, prevalent strain types, possible transmission routes and genetic relatedness of Zimbabwean isolates to other global strains.