ABSTRACT
River water is an essential human resource that may be contaminated with hazardous microorganisms. However, the risk of yeast infection through river water exposure is unclear because it is highly dependant on individual susceptibility and has therefore not been well-studied, to date. To evaluate this undefined risk, we analysed the fungal communities in less polluted (LP) and highly polluted (HP) river water, as determined using principal coordinate analysis of pollution indicators. We enumerated culturable yeasts using a thermally selective isolation procedure (37 °C) and thus promoted the growth of potentially opportunistic species. Yeast species identified as clinically relevant were then tested for antifungal resistance. In addition, we propose a quantitative microbial risk assessment (QMRA) framework to quantitatively assess the potential risk of yeast infection. Our results indicated that pollution levels significantly altered fungal communities (p = 0.007) and that genera representing opportunistic and pathogenic members were significantly more abundant in HP waters (p = 0.038). Additionally, the yeast species Candida glabrata and Clavispora lusitaniae positively correlated with other pollution indicators, demonstrating the species' indicator potential. Our QMRA results further indicate that higher risk of infection is associated with increased water pollution levels (considering both physicochemical and bacterial indicators). Furthermore, yeast species with higher pathogenic potential present an increased risk of infection despite lower observed concentrations in the river water. Interestingly, the bloom of Meyerozyma guilliermondii during the wet season suggests that other environmental factors, such as dissolved oxygen levels and water turbulence, might affect growth characteristics of yeasts in river water, which consequently affects the distribution of annual infection risks. The presence of antifungal resistant yeasts, observed in this study, could further contribute to variation in risk distribution. Research on the ecophysiology of yeasts in these environments is therefore necessary to ameliorate the uncertainty and sensitivity of the proposed QMRA model. In addition to the vital knowledge on opportunistic and pathogenic yeast occurrence in river water and their observed association with pollution, this study provides valuable methods and insights to initiate future QMRAs of yeast infections.
