Repository logo
 

Molecular epidemiology of antibiotic-resistant Escherichia coli and Enterococcus spp. from agricultural soil fertilized with chicken litter in uMgungundlovu district, KwaZulu-Natal Province, South Africa.

Thumbnail Image

Date

2021

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

The application of animal manure contaminated with antibiotic-resistant bacteria (ARB) and antibiotic-resistance genes (ARGs) represents a major route by which antibiotic resistance is transmitted into the soil environment. The introduction and persistence of ARB in agricultural soil may pose a risk to public health via the consumption or handling of contaminated farm produce. Understanding the impact of animal manure application on the agricultural soil resistome and the risk it poses on public health is critical. However, such information is limited in South Africa as most antibiotic resistance research focuses on humans and food animals. This study, therefore, describes the prevalence and the genomic profiles of antibiotic-resistant Escherichia coli and Enterococcus spp. isolated from agricultural soil fertilized with chicken litter and the chicken litter. A total of 237 samples were examined and included soil before litter application, the litter-amended soil, and the chicken litter. Isolation and quantification of Escherichia coli and Enterococci were carried out using the Colilert® -18 / Quantiti-Tray® 2000 system and the Enterolert® -18® Quanti-Tray®/2000 system, respectively. The antibiotic susceptibility profiles of the isolates was determined using the Kirby-Bauer disk diffusion method. Whole-genome sequencing (WGS) and bioinformatics tools were used to determine the resistome, virulome, mobilome, clonal lineages, and phylogenies of the isolates circulating between the soil and the chicken litter. The application of chicken litter to the soil statistically significantly increased Enterococci count and the number of antibiotic-resistant enterococci in the litter-amended soil. A total of 835 enterococci (680 from soil and 155 from litter) isolates recovered from the samples was dominated by E. casseliflavus (56%), followed by E. faecalis (22%), E. faecium (8%), E. gallinarum (2%) and other Enterococcus spp 102 (12%). Overall, 55.8% (466/835) of the enterococci isolates were resistant to one or more antibiotics with the highest rate in the litter-amended soil (68.9%, 321/466), followed by chicken litter (19.9%, 93/466) and the least in the soil samples collected before the litter amendment (11.2%; 52/466). The enterococci isolates were mostly resistant to tetracycline (33%), erythromycin (25%), and trimethoprim-sulfamethoxazole (23%), among others, intimating the high usage of these antibiotics in poultry farms in South Africa. Additionally, multidrug resistance (MDR) was recorded in 27.8% (130/466) of the enterococci isolates with MAR indices ranging from 0.13 (resistance to two antibiotics) to 0.44 (resistance to seven antibiotics). A total of 63 different resistance patterns were recorded in the MDR enterococci isolates. Notably, enterococci count and the number of antibiotic-resistant enterococci in the litter-amended soil were reduced to levels comparable to the unamended soil at 50 and 28 days after soil amendment respectively. The whole-genome analysis of the few selected enterococci isolates revealed eight novel sequence types (STs) (ST1700, ST1752, ST1753, ST1754, ST1755, ST1756, ST1004, and ST1006). Several resistance genes that confer resistance to aminoglycosides (aac(6’)-Ii, aac(6’)-Iih, ant(6)-Ia, aph(3’)-III, ant(9)-Ia), macrolide-lincosamide-streptogramin AB (MLSAB) [erm(B), lnu(B), lnu(G), lsaA, lsaE, eat(A), msr(C)], trimethoprim-sulfamethoxazole (dfrE, and dfrG), tetracycline (tet(M), tet(L), and tet(S)), fluoroquinolones (efmA, and emeA), vancomycin (VanC {VanC-2, VanXY, VanXYC-3, VanXYC-4, VanRC}), and chloramphenicol (cat) were detected in the isolates. The bioinformatics analysis further revealed that the chicken litter amendment increased the number and diversity of ARGs in the soil, resulting in increased detection of tetracycline resistance genes (tet(M), tet(L)), and the macrolide resistance gene erm(B) and appearance of some ARGs (ant(6)-Ia, aph(3’)-III, lnu(G), dfrG)) that were not detected in the unamended soil. ARGs were mostly associated with diverse insertion sequences (ISs) (IS982, ISL3, IS6, IS5, IS3, IS256, IS30) and/or transposons (Tn3, Tn916, Tn6009) on plasmids or chromosome. The tet(M) and erm(B) were also co-located on Tn916-like transposons (Tn644, Tn645, and Tn659) in the three sample groups. Some of the isolates also harboured virulence genes that encoded adherence/biofilm formation (ebpA, ebpB, ebpC), anti-phagocytosis (elrA), and bacterial sex pheromones (Ccf10, cOB1, cad, and camE). Phylogenomic analysis showed that few isolates from litter-amended soil clustered with the chicken litter isolates. The isolates from this study also clustered with clinical and animal isolates from South Africa (Pretoria, Pietermaritzburg), Angola, and Tunisia. There was also an increase (albeit statistically insignificant) in E. coli count and the number of antibiotic-resistant E. coli in the soil following chicken litter amendment. A total of 126 E. coli was recovered from the soil and chicken litter samples. In total, 76% (96/126) of the E. coli isolates displayed resistance to at least one antibiotic, with the highest prevalence in the litter-amended soil (71.9%, 69/96) and the least (1%, 1/96) in soil samples collected before the litter amendment. The E. coli isolates displayed a high percentage resistance to tetracycline (78.1%), chloramphenicol (63.5%), ampicillin (58.3%), trimethoprim-sulfamethoxazole (39.6%), cefotaxime (30.2%), ceftriaxone (26.0%), and cephalexin (20.8%). Lower percentages of XVI resistance to cefepime (11.5%), amoxicillin-clavulanic acid (11.5%), cefoxitin (10.4%), nalidixic acid (9.4%), amikacin (6.3%), ciprofloxacin (4.2%), imipenem (3.1%), tigecycline (3.1%), and gentamicin (3.1%) were also recorded in the isolates. All the isolates were completely susceptible to meropenem and ceftazidime. Approximately 54% (52/96) of the resistant isolates were MDR, and the MAR indices of the isolates ranged between 0.11 (resistance to two antibiotics) and 0.56 (resistance to ten antibiotics). Overall, 38.5% (37/96) of all the resistant isolates had a MARI > 0.2, with the highest rate (51.4%) in the litter-amended soil and the least in the soil before litter amendment (2.7%). Twenty-one multidrug resistance patterns were observed among the isolates. These results show that the soil resistome was augmented by chicken litter application. Agricultural soil and chicken litter are rich reservoirs of multidrug-resistant E. coli and Enterococcus spp. that could threaten public health through contamination of food products and the surrounding water bodies. There is therefore a need for urgent and stringent measures to mitigate the spread of antibiotic resistance in the environment via prudent use of antibiotics in food animal production and treatment of animal manure before its application onto agricultural soil.

Description

Doctoral Degree. University of KwaZulu-Natal, Durban.

Keywords

Citation

DOI