Genetic Relationships among Antibiotic-Resistant Escherichia Coli from Households in Kibera, Nairobi County, Kenya

Abstract

Drug-resistant bacteria kill hundreds of thousands of people every year. In 2019, an estimated 1.3 million deaths globally were attributable to bacterial antimicrobial resistance (AMR). The resistance genes in bacteria can be carried on horizontally transmissible elements such as plasmids and the dissemination of these genes, even if infrequent, to commensal or pathogenic bacteria can lead to amplification and further dissemination of AMR. This study aimed to identify antibiotic-resistance genes and assess the genetic diversity of antibiotic-resistant Escherichia coli in randomly selected households in Kibera (Nairobi), Kenya. The work formed part of a sanitation study on antimicrobial-resistant E. coli conducted under the auspices of Washington State University (WSU) in collaboration with KEMRI/CDC. A subset of isolates (n=16) with resistance phenotypes was randomly selected, genotyped, and characterized for the resistance genes. Genomic DNA was extracted from overnight cultures followed by library preparation before paired-end sequencing using an Illumina Miseq instrument. Raw data were quality assessed before de novo assembly, genome annotation, and resistome prediction using bioinformatics tools. Ten (77%) of the E. coli isolates were resistant to three or more antibiotic classes, confirming a high prevalence of antibiotic resistance. Identified E. coli harboured antibiotic resistance genes with a predominance to beta-lactam, aminoglycoside, sulphonamide, tetracycline, and trimethoprim classes of antibiotics. There was between 62-80% consensus between predicted resistance based on sequence data and the original phenotype-based resistance to all drugs tested. blaTEM-1B was the only resistance gene identified among beta-lactamase producers. Among aminoglycoside-resistant E. coli, aph(3”)-lb, aph(6)-ld, and aadA1 were identified whereas tet(A) and tet(B) were attributed to tetracycline resistance. IncF plasmids, associated with carriage of broad spectrum antibiotic resistance genes, were the predominant plasmid marker. Multilocus sequence typing (MLST) identified 12 different sequence types (STs) among 13 isolates, reflecting a high degree of genetic diversity. Diversity indices (Shannon, Simpson and Richness) further highlighted a substantial within household variation, while chi-square analysis showed no significant association between genotype distribution and household clustering (χ², p = 0.385). The detection of globally relevant STs such as ST10, ST335 and ST398 underscores the public health importance of resistant E. coli circulating in community settings. The study demonstrates widespread carriage of diverse, multidrug resistant E. coli in households in Kibera and that non-classical mechanisms contribute to phenotypic resistance beyond recognized genes. This study recommends strengthened genomic surveillance of AMR at community level to better elucidate transmission dynamics and inform targeted public health interventions and for further research to understand non classical mechanisms of antibiotic resistance in bacteria.