Potential and limitations of metagenomics as a public health risk assessment tool in a study of natural creek sediments influenced by agricultural and livestock runoff.

Little is known about the public health risks associated with natural creek sediments that are affected by runoff and fecal pollution from agricultural and livestock practices. For instance, the persistence of foodborne pathogens originating from these practices such as Shiga Toxin-producing Escherichia coli (STEC) remains poorly quantified. Towards closing these knowledge gaps, the water-sediment interface of two creeks in the Salinas River Valley of California was sampled over a nine-month period using metagenomics and traditional culture-based tests for STEC. Our results revealed that these sediment communities are extremely diverse and comparable to the functional and taxonomic diversity observed in soils. With our sequencing effort (∼4Gbp per library), we were unable to detect any pathogenic E. coli in the metagenomes of 11 samples that had tested positive using culture-based methods, apparently due to relatively low abundance. Further, there were no significant differences in the abundance of human- or cow-specific gut microbiome sequences in the downstream, impacted sites compared to upstream, more pristine (control) sites, indicating natural dilution of anthropogenic inputs. Notably, a high number of metagenomic reads carrying antibiotic resistance genes (ARGs) was found in all samples that was significantly higher compared to ARG reads in other available freshwater and soil metagenomes, suggesting that these communities may be natural reservoirs of ARGs. The work presented here should serve as guide for sampling volumes, amount of sequencing to apply, and what bioinformatics analyses to perform when using metagenomics for public health risk studies of environmental samples such as sediments. IMPORTANCE Current agricultural and livestock practices contribute to fecal contamination in the environment and the spread of food and water-borne disease and antibiotic resistance genes (ARGs). Traditionally, the level of pollution and risk to public health is assessed by culture-based tests for the intestinal bacterium, E. coli. However, the accuracy of these traditional methods (e.g., low accuracy in quantification, and false positive signal when PCR-based) and their suitability for sediments remains unclear. We collected sediments for a time series metagenomics study from one of the most highly productive agricultural regions in the U.S. in order to assess how agricultural runoff affects the native microbial communities and if the presence of STEC in sediment samples can be detected directly by sequencing. Our study provided important information on the potential for using metagenomics as a tool for assessment of public health risk in natural environments.