Presence of Bacteroidales as a predictor of pathogens in surface waters of the central California coast

Abstract

The value of Bacteroidales genetic markers and fecal indicator bacteria (FIB) to predict the occurrence of waterborne pathogens was evaluated in ambient waters along the central California coast. Bacteroidales host-specific quantitative PCR (qPCR) was used to quantify fecal bacteria in water and provide insights into contributing host fecal sources. Over 140 surface water samples from 10 major rivers and estuaries within the Monterey Bay region were tested over 14 months with four Bacteroidales-specific assays (universal, human, dog, and cow), three FIB (total coliforms, fecal coliforms, and enterococci), two protozoal pathogens (Cryptosporidium and Giardia spp.), and four bacterial pathogens (Campylobacter spp., Escherichia coli O157:H7, Salmonella spp., and Vibrio spp.). Indicator and pathogen distribution was widespread, and detection was not highly seasonal. Vibrio cholerae was detected most frequently, followed by Giardia, Cryptosporidium, Salmonella, and Campylobacter spp. Bayesian conditional probability analysis was used to characterize the Bacteroidales performance assays, and the ratios of concentrations determined using host-specific and universal assays were used to show that fecal contamination from human sources was more common than livestock or dog sources in coastal study sites. Correlations were seen between some, but not all, indicator-pathogen combinations. The ability to predict pathogen occurrence in relation to indicator threshold cutoff levels was evaluated using a weighted measure that showed the universal Bacteroidales genetic marker to have a comparable or higher mean predictive potential than standard FIB. This predictive ability, in addition to the Bacteroidales assays providing information on contributing host fecal sources, supports using Bacteroidales assays in water quality monitoring programs.

FIG. 1.

Surface water sampling sites in the Monterey Bay region of California.

FIG. 2.

Box plots of site-specific variation of universal Bacteroidales concentrations from north to south (top) and seasonal variation for universal Bacteroidales (light gray) and fecal coliforms (dark gray) (bottom). Upper and lower bounds of boxes denote the 75th and 25th percentiles. Upper and lower bars show the 90th and 10th percentiles, with outliers represented by filled circles. Note the logarithmic vertical axis.

FIG. 3.

Threshold concentration ranges for universal Bacteroidales (top) and total coliforms (bottom), with resulting PQ values greater than 66% for data from all sites combined and separated by average site salinity (freshwater salinity of <0.5 ppt). Maximum PQ values are shown in parentheses. Vibrio parahaemolyticus found at marine-influenced sites is not displayed because only one positive sample was detected. The vertical dotted line indicates the median value of all measured samples.

FIG. 4.

Threshold concentration ranges for fecal coliforms (top) and enterococci (bottom), with resulting PQ values greater than 66% for data from all sites combined and separated by average salinity (freshwater salinity of <0.5 ppt). Maximum PQ values are shown in parentheses. Vibrio parahaemolyticus found at marine-influenced sites is not displayed because only one positive sample was detected. The vertical dotted line indicates the median value of all measured samples.