Genomics and metagenomics analysis of non-typhoidal salmonella from children below five years and their immediate environment in disease-endemic Mukuru settlement in Nairobi, Kenya

Abstract

Non-typhoidal Salmonella (NTS) remains a leading cause of morbidity and mortality in Sub-Saharan Africa, particularly among vulnerable children within informal settlements where safe water, sanitation, and hygiene (WASH) infrastructure are inadequate. The rising prevalence of antimicrobial-resistant NTS strains and the role of household environments as potential reservoirs sustaining endemic transmission remain poorly characterized in these high-risk informal settlements. This study aimed to investigate the occurrence, genetic diversity, antimicrobial resistance, pathogenic potential, and spatial distribution of NTS strains from environmental and clinical sources within Mukuru informal settlement, Nairobi, Kenya. By linking clinical and environmental NTS strains, the study provides insights into the persistence and transmission of antimicrobial-resistant strains. A case-control study conducted between June 2021 and June 2023 analyzed 2,675 clinical samples from children and 268 environmental samples, including drinking water, effluent, soil, and raw sewage, collected from case and control households. NTS isolation from clinical samples was performed using standard microbiological and biochemical techniques, followed by confirmation and serotyping through serological assays and whole-genome sequencing. Environmental samples underwent metagenomic sequencing and binning of NTS genomes. Whole-genome and metagenomic sequences were analysed to characterize genetic diversity, antimicrobial resistance genes (ARGs), and pathogenicity determinants. Core-genome SNP-based phylogenetic reconstruction assessed genomic relatedness, while spatial analyses mapped transmission patterns over time. NTS was detected in 1.57% of clinical samples, predominantly Salmonella Enteritidis (52.4%) and Salmonella Typhimurium (47.6%). Environmental samples from case households exhibited higher detection rates of NTS (33.3%) than those from homes of controls (7.1%), with effluent (45.2%) and drinking water (33.3%) showing the highest contamination levels. Core-SNP phylogenetic analysis revealed strong genomic relatedness between clinical and environmental NTS strains within households (F = 0.8171, R² = 0.0496, p = 0.459), affirming intra-household environmental-human transmission cycles. Effluent-derived strains often occupied basal phylogenetic positions, implicating household wastewater as a potential ancestral source. Antimicrobial resistance profiling revealed aminoglycoside resistance genes as the most prevalent across clinical and environmental NTS strains, with aac(6’)-Iaa_1 detected in 100% of clinical and 97.6% of environmental strains. Additional resistance genes included aph(6), aph(9)-Ia, oqxA, sul_1, cat2, and blaTEM-104_1. Notably, aph and oqxA genes were co-located within conserved resistance cassettes embedded in IncFIB(S)_1 and IncFII(S)_1 plasmids. These cassettes were flanked by mobile genetic elements, IS3, IS15, IS256, ISPm12, and transposons Tn3 and TnAs3, forming modular, horizontally transferable resistance islands. Pairwise comparisons revealed overlapping ARG profiles within household-case clusters, suggesting intra-household environmental-to-human ARG spillover (F = 1.816, R² = 0.131, p = 0.062). IncFIB(S)_1 (97.2% clinical, 29% environmental) and IncFII(S)_1 (97.2% clinical, 31.6% environmental) plasmids, harboring compact Type IV secretion systems, were the dominant replicons across clinical and environmental strains, suggesting high conjugation potential and ecological persistence. PERMANOVA revealed significant differences in plasmid profiles between clinical and environmental compartments (p_adj=0.01), except raw sewage (p_adj=0.154). Virulence gene profiling showed substantial overlap between clinical and effluent, drinking water, and soil NTS strains, with significant differences only between clinical and raw sewage strains (F=10.625, R²=0.202, p_adj=0.05). Spatial analysis revealed that high-intensity hotspots were consistently mapped to areas of infrastructural deficiency. These spatial patterns suggest that environmental matrices serve as reservoirs for antimicrobial-resistant NTS, underscoring the need for targeted WASH improvements and routine One Health surveillance to interrupt transmission in informal settlements.