Microbial risk assessment in recreational freshwaters from southern Brazil

In this study, total coliforms (TC), Escherichia coli, enterovirus (EV), rotavirus (RV), and human mastadenovirus species C and F (HAdV-C and HAdV-F)were evaluated in water samples from Belo Stream. For HAdV-C and F, the infectivity was assessed by integrated cell culture quantitative real-time polymerase chain reaction (ICC-qPCR). Samples were collected monthly (May/2015 to April/2016)at four sites. Viral analyses were performed for both ultracentrifuge-concentrated and unconcentrated samples. For site P4 (used for recreational purposes), QMRA was applied to estimate health risks associated with exposure to E. coli and HAdV-C and F. TC and E. coli were present throughout the collection period. EV and RV were not detected. HAdV-C were present in 8.51 percent (1.89E + 06 to 2.28E + 07 GC (Genomic Copies)/L)and 21.27 percent (2.36E + 05 to 1.29E + 07 GC/L)for unconcentrated and concentrated samples, respectively. For HAdV-F were 12.76 percent (2.77E + 07 to 3.31E + 08 GC/L)and 48.93 percent (1.10E + 05 to 4.50E + 08 GC/L)for unconcentrated and concentrated samples, respectively. For unconcentrated samples, infectivity for HAdV-C was detected in 37.20 percent (1st ICC-qPCR)and 25.58 percent (2nd ICC-qPCR). For HAdV-F, infectivity was detected in 6.97 percent (1st ICC-qPCR)and 6.97 percent (2nd ICC-qPCR). For concentrated samples, HAdV-C infectious was observed in 17.02 percent (1st ICC-qPCR)and in 8.51 percent (2nd ICC-qPCR). For HAdV-F, were present in 8.51 percent for both 1st and 2nd ICC-qPCR. Statistical analyzes showed significant difference between the collection sites when analyzed the molecular data of HAdV-F, data of TC and E. coli. Correlation tests showed direct correlation between HAdV-F with E. coli and TC. E. coli concentrations translated to the lowest estimates of infection risks (8.58E-05 to 2.17E-03). HAdV-F concentrations were associated with the highest infection risks at 9.99E-01 and for group C, 1.29E-01 to 9.99E-01. These results show that commonly used bacterial indicators for water quality may not infer health risks associated with viruses in recreational freshwaters