Antibiotic resistance (AR) and multidrug resistance are significant global health concerns, predicted to cause millions of deaths annually. The increase in AR bacteria (ARB) and AR genes (ARG) in water bodies is linked to increased antibiotics, nutrients, metals, and microplastics. Sources of ARG and ARB transmission include animal and human waste, wastewater, and runoff. While groundwater offers some protection, surface waters, particularly rivers, show dynamic ARG and ARB transport. Rivers in emerging economies often serve as disposal sites, leading to contamination. The Kelani River in Sri Lanka, crucial for the country's economy and society, is a case in point, receiving significant antibiotic and ARB inputs from wastewater treatment plants (WWTPs). Studies have shown AR in several rivers worldwide, with religious festivals also contributing to AR increases. Anthropogenic activities such as urban runoff, aquaculture, agriculture, animal husbandry, and hospital effluents introduce ARGs into water bodies. The catchment area, land use, and location of point sources heavily influence ARG levels in closed systems like lakes. Previous research has explored the relationship between various pollutants in tropical rivers and sewage treatment plants. This study aimed to quantify seasonal variations in *E. coli* ARB and ARG in the Kelani River and understand the factors influencing the relationship between in situ parameters, metals, *E. coli* prevalence, and antibiotic resistance.

Existing literature highlights the global threat of antibiotic resistance and the role of water bodies in its spread. Studies have linked ARGs and ARBs to various pollutants in water, including antibiotics, nutrients, metals, and microplastics. The contribution of wastewater treatment plants and other sources like animal and human waste has been documented. Research has also investigated seasonal variations in water quality and antibiotic resistance in various rivers across the globe, but comprehensive studies focusing on the seasonal effects on AR in monsoon-dominated tropical countries are limited. Existing studies on AR in rivers primarily focus on land-use patterns as a governing factor. However, the importance of seasonal variations and factors like rainfall, discharge, and WWTP effluent are not completely understood.

Four locations along the Kelani River in Sri Lanka were selected based on in situ analyses of pH, temperature, electrical conductivity (EC), dissolved oxygen (DO), oxidation-reduction potential (ORP), and bicarbonate. Water samples were collected during the wet season (October 2017) and dry season (March 2018) for analysis of ARB, ARG, and metals (As, Cd, Co, Cr, Cu, Mn, Ni, Pb, Zn). Metals were analyzed using inductively coupled plasma-mass spectrometry (ICP-MS). *E. coli* was isolated using the membrane filtration method with Chromocult Coliform Agar. Antibiotic susceptibility testing was performed using the Kirby-Bauer disc diffusion method for six antibiotics (three fluoroquinolones and three non-fluoroquinolones). ARG screening was conducted using PCR to amplify genes conferring resistance to fluoroquinolones, tetracyclines, sulfonamides, and β-lactams. Real-time quantitative PCR (qPCR) quantified the selected ARGs. Principal component analysis (PCA) and hierarchical cluster analysis were used to explore relationships between in situ parameters, metals, *E. coli*, ARB, and ARG.

*E. coli* prevalence ranged from 10 to 27 CFU ml⁻¹ in the Kelani River, with most isolates resistant to multiple antibiotics. Resistance to tetracycline and sulfamethoxazole was significantly higher than to other antibiotics. Older antibiotics showed higher resistance percentages than newer ones. Genes conferring resistance to tetracyclines, sulfonamides, β-lactams, and fluoroquinolones were detected. Fluoroquinolone resistance showed higher seasonal variation than non-fluoroquinolones, but the effect was antagonistic. Multivariate statistical analyses (PCA and cluster analysis) revealed four principal components explaining 84.74% of the total variance, indicating associations between in situ parameters, metals, ARB, and ARG. The number of components decreased in the dry season, suggesting the importance of seasonal variability. The study observed a higher concentration of ARGs in more urbanized sites. Two out of three genes (*blaCTX1* and *qnrs*) were considerably higher in the dry season than in the wet season. Genes conferring resistance to old antibiotics (tetracyclines and sulfonamides) were detected in all samples. Higher EC and increased metals potentially enhanced the occurrence of ARGs. A conceptual model depicted the influence of seasonality on antibiotic resistance, encompassing various factors like contaminant enrichment, seasonal diseases, and antibiotic consumption patterns.

The findings indicate that seasonality significantly influences antibiotic resistance in the Kelani River. The higher resistance to older antibiotics suggests the persistence of these ARGs despite reduced use in human medicine. The antagonistic effect observed between fluoroquinolone resistance and non-fluoroquinolone resistance suggests a complex interaction of factors influencing AR development. The association between metals and ARGs underscores the impact of environmental factors on AR development and spread. The seasonal variation observed is attributable to factors like rainfall, discharge, runoff, combined sewer overflows, and changes in water quality parameters. The reduction in the number of governing components in the dry season implies greater consistency in water quality in terms of fecal and antibiotic contamination. However, *E. coli* prevalence alone is not a sufficient indicator of ARG and ARB levels, highlighting the need for identifying appropriate markers. This study provides evidence for the complex interplay of factors driving AR in tropical rivers, highlighting the limitations of using *E. coli* prevalence as a sole indicator of AR levels.

This study demonstrates the significant influence of seasonality on the prevalence of antibiotic-resistant bacteria and genes in the Kelani River. The findings underscore the need for improved wastewater management practices, increased public awareness, and effective treatment solutions to mitigate the spread of AR. Further research is needed to identify effective markers for ARB and ARG prevalence and to explore the effectiveness of various treatment strategies in reducing AR in tropical river systems. Expanding data on seasonal variations in ARB and ARG globally, particularly in emerging economies, is critical for effective management and control of antibiotic resistance.

The study's scope was limited to four sampling locations and two seasons. A more extensive spatial and temporal sampling strategy would provide a more comprehensive understanding of AR dynamics. The study focused on a limited set of antibiotics and ARGs; a broader range could provide a more complete picture of AR. While the study identified associations between various parameters, establishing direct causal relationships requires further investigation.