The investigation focused on 14 drinking water treatment plants that draw raw water directly from the Upper Mahaweli River and its tributaries. Researchers collected 45 water samples across the system and cultured the bacteria they contained, ultimately isolating 167 strains of E. coli, the classic indicator organism for fecal contamination in water. Each isolate was tested against ten clinically important antibiotics using the Kirby-Bauer disk diffusion method, the standard laboratory technique in which antibiotic-impregnated discs are placed on bacterial lawns and zones of inhibition are measured to determine susceptibility.

The results were striking. Overall, 60.48 percent of the E. coli isolates exhibited resistance to at least one antibiotic, while multidrug resistance, defined as resistance to three or more antibiotic classes, was detected in 27.54 percent of isolates. The highest levels of resistance were observed for amoxicillin, a widely used penicillin-class drug, at 47.31 percent, followed by tetracycline at 26.95 percent and co-trimoxazole, a combination of sulfamethoxazole and trimethoprim, at 24.55 percent. Notably, resistance to four of the tested antibiotics showed seasonal variation, a pattern that researchers attribute to fluctuations in rainfall, river flow, and the runoff of contaminants from surrounding land during different times of the year.

Beyond the resistance phenotypes, the team went a step further and searched for the molecular machinery behind the phenomenon. Genomic DNA was extracted from the culturable bacteria in each of the 45 water samples and screened by polymerase chain reaction for 11 antibiotic resistance genes, or ARGs. Across 210 PCR assays, 42.42 percent returned positive detections. The most abundant genes were blaTEM, which encodes a beta-lactamase enzyme that degrades penicillin-type antibiotics, found in 80.0 percent of samples, followed by tetA at 66.67 percent, and tetM and qnrS, each detected in 62.22 percent of samples. The tet genes protect bacteria from tetracycline through efflux pumping and ribosomal protection respectively, while qnrS confers reduced susceptibility to fluoroquinolones, a class of critically important human medicines.

To quantify the overall burden of resistance, the researchers calculated two indices. The multiple antibiotic resistance index, or MAR index, which reflects the average number of antibiotics to which isolates are resistant, ranged from 0.00 to 0.80, with 44.91 percent of samples exceeding the 0.2 threshold that signals a high-risk source of contamination, typically one associated with human or animal waste. The antibiotic resistance index ranged from 0.00 to 0.32, with eight sampling points above the same threshold. These values suggest that the contamination entering the river is not random background noise but reflects sustained pressure from sources where antibiotics are used and excreted, such as households lacking adequate sanitation, livestock operations, and agricultural runoff.

Hierarchical cluster analysis, a statistical technique that groups sampling sites according to the similarity of their resistance profiles, revealed a clear spatial pattern. The majority of drinking water sources fell into an intermediate category of contamination, but a few stood out at the extremes. The Kotagala and Thalawakelle-Galkanda intakes showed the lowest resistance burdens, while Haragama, Paradeka, and Nawalapitiya emerged as hotspots with the highest levels. This variability mirrors the intensity of anthropogenic interference along the river, with densely populated towns, commercial activity, and agricultural land leaving distinct fingerprints on the microbial ecology of the water at each intake.

The Upper Mahaweli River is not a marginal water body. It is the longest river in Sri Lanka and the backbone of water supply for the central hill country, feeding treatment plants that serve towns and cities across the region, including greater Kandy. The finding that antibiotic-resistant bacteria and their genes are present upstream of drinking water treatment infrastructure matters because conventional treatment is not always designed to remove them. Chlorination, the most common disinfection step, can kill susceptible bacteria but has been shown in other studies to sometimes promote the horizontal transfer of resistance genes between surviving organisms, and free DNA carrying ARGs can persist in water even after the cells that hosted them are destroyed. Resistance genes are also mobile: they ride on plasmids and transposons that can shuttle between species, meaning that harmless environmental bacteria can serve as reservoirs from which resistance traits eventually reach pathogens.

The study’s authors emphasize that the presence of AR-E. coli and ARGs in these drinking water sources represents a genuine public health risk, particularly in a tropical South Asian setting where the environmental dimensions of antimicrobial resistance remain poorly documented. Antimicrobial resistance is globally recognized as one of the leading threats to modern medicine, and aquatic environments are increasingly understood as critical corridors through which resistance genes travel between humans, animals, and ecosystems. In Sri Lanka, national surveys have documented substantial antibacterial consumption in both human and veterinary medicine, and previous work in the country has already identified resistant bacteria in hospital wastewater, recreational river water, poultry operations, and the Kelani River, another major waterway. The new findings extend this picture to the drinking water supplies of the hill country itself.

What makes the results especially consequential is the seasonal dimension. The detection of seasonal variation in resistance to four antibiotics suggests that monsoon-driven hydrology plays a role in mobilizing contamination, washing resistance determinants from soils, latrines, and farms into the river during periods of heavy rain. As climate change alters rainfall patterns across South Asia, such episodic pulses of contamination may become more frequent or intense, compounding the challenge for water utilities that must deliver safe water under increasingly variable conditions.

The researchers conclude that proactive interventions are urgently needed to safeguard the long-term safety of drinking water sources along the Upper Mahaweli. Such measures would likely include improved sanitation and wastewater management in riparian communities, better regulation of antibiotic use in agriculture and livestock, routine surveillance of resistance in source waters, and evaluation of whether existing treatment trains adequately remove both resistant bacteria and free-floating resistance genes. In a world where the pipeline from environment to clinic grows shorter every year, the study stands as a warning that the battle against antimicrobial resistance will be fought not only in hospitals and pharmacies, but in the rivers from which people drink.

Subject of Research: Antibiotic-resistant E. coli and antibiotic resistance genes in drinking water sources of the Upper Mahaweli River, Sri Lanka

Article Title: Occurrence of antibiotic-resistant E. coli and antibiotic resistance genes from culturable bacteria in drinking water sources along the Upper Mahaweli River, Sri Lanka

Article References: Occurrence of antibiotic-resistant E. coli and antibiotic resistance genes from culturable bacteria in drinking water sources along the Upper Mahaweli River, Sri Lanka. (n.d.). https://doi.org/10.1007/s10661-026-15884-z

Image Credits: AI Generated

DOI: 10.1007/s10661-026-15884-z

Keywords: antibiotic resistance, E. coli, antibiotic resistance genes, drinking water, Mahaweli River, Sri Lanka, multidrug resistance, blaTEM, water quality, public health, environmental microbiology, antimicrobial resistance