Variation of Bacterial Communities with Water Quality in an Urban Tropical Catchment

Persistence of human- and cattle-associated Bacteroidales and mitochondrial DNA markers in freshwater mesocosms

Accurate identification of the origins of non-point source pollution is essential for the effective control of fecal pollution. Host-associated Bacteroidales and mitochondrial DNA (mtDNA) markers have been developed to identify the sources of human and cattle fecal pollution. However, the differences in persistence between these two types of markers under different environmental conditions are still poorly understood. Here, we conducted mesocosm experiments to investigate the influence of indigenous microbiota and nutrients on the decay of Bacteroidales and mtDNA markers associated with humans and cattle. Raw sewage or cattle feces were inoculated into mesocosms containing natural eutrophic water, sterile eutrophic water or artificial freshwater. The Bacteroidales markers HF183 (human) and CowM3 (cattle) and mtDNA markers HcytB (human) and QMIBo (cattle) were quantified using the quantitative polymerase chain reaction (qPCR) assays. All markers but HF183 decreased the fastest in the presence of indigenous microbiota. Nutrients caused a decrease in the persistence of HF183; however, no significant nutrient effects were observed for HcytB, CowM3, and QMIBo. The time to reach one log reduction (T90) for HF183 and HcytB was similar; CowM3 reached T90 earlier than QMIBo in all the treatments but eutrophic water. E. coli persisted longer than both Bacteroidales and mtDNA markers in the mesocosms regardless of inoculum type. Additionally, 16S rRNA gene amplicon sequencing was used to determine the changes in bacterial communities accompanying the marker decay. Analysis using the SourceTracker software showed that bacterial communities in the mesocosms became more dissimilar to those in the corresponding inoculants over time. Our results indicate that environmental factors are important determinants of genetic markers' persistence, but their impact can vary depending on the genetic markers. The cattle Bacteroidales markers may be more suitable for determining recent fecal contamination than cattle mtDNA.

Human Fecal Contamination Corresponds to Changes in the Freshwater Bacterial Communities of a Large River Basin

Microbial water quality is generally monitored by culturable fecal indicator bacteria (FIB), which are intended to signal human health risk due to fecal pollution. However, FIB have limited utility in most urbanized watersheds as they do not discriminate among fecal pollution sources, tend to make up a small fraction of the total microbial community, and do not inform on pollution impacts on the native ecosystem. To move beyond these limitations, we assessed entire bacterial communities and investigated how bacterial diversity relates to traditional ecological and human health-relevant water quality indicators throughout the Milwaukee River Basin. Samples were collected from 16 sites on 5 days during the summer, including both wet and dry weather events, and were processed by 16S rRNA gene amplicon sequencing. Historical water quality at each sampling location, as opposed to upstream land use, was associated significantly with bacterial community alpha diversity. Source partitioning the sequence data was important for determining water quality relationships. Sewage-associated bacterial sequences were detected in all samples, and the relative abundance of sewage sequences was strongly associated with the human Bacteroides fecal marker. From this relationship, we developed a preliminary threshold for human sewage pollution when using bacterial community sequence data. Certain abundant freshwater bacterial sequences were also associated with human fecal pollution, suggesting their possible utility in water quality monitoring. This study sheds light on how bacterial community analysis can be used to supplement current water quality monitoring techniques to better understand interactions between ecological water quality and human health indicators. IMPORTANCE Surface waters in highly developed mixed-use watersheds are frequently impacted by a wide variety of pollutants, leading to a range of impairments that must be monitored and remediated. With advancing technologies, microbial community sequencing may soon become a feasible method for routine evaluation of the ecological quality and human health risk of a water body. In this study, we partnered with a local citizen science organization to evaluate the utility of microbial community sequencing for identifying pollution sources and ecological impairments in a large mixed-use watershed. We show that changes in microbial community diversity and composition are indicative of both long-term ecological impairments and short-term fecal pollution impacts. By source partitioning the sequence data, we also estimate a threshold target for human sewage pollution, which may be useful as a starting point for future development of sequencing-based water quality monitoring techniques.

Variably improved microbial source tracking with digital droplet PCR

This study addressed whether digital droplet PCR (ddPCR) could improve sensitivity and specificity of human-associated Bacteroidales genetic markers, BacHum and B. theta, and their quantification in environmental and fecal composite samples. Human markers were quantified by qPCR and ddPCR platforms obtained from the same manufacturer. A total of 180 samples were evaluated by each platform including human and animal feces, sewage, and environmental water. The sensitivity of ddPCR and qPCR marker assays in sewage and human stool was 0.85-1.00 with marginal reduction in human stool by ddPCR relative to qPCR (<10%). The prevalence and distribution of markers across complex sample types was similar (74-100% agreement) by both platforms with qPCR showing higher sensitivity for markers in environmental and composite samples and ddPCR showing greater reproducibility for marker detection in fecal composites. Determination of BacHum prevalence in fecal samples by ddPCR increased specificity relative to qPCR (from 0.58 to 0.88) and accuracy (from 0.77 to 0.94), while the B. theta assay performed similarly on both platforms (specificity = 0.98). In silico analysis indicated higher specificity of ddPCR for BacHum was not solely attributed to reduced sensitivity relative to qPCR. Marker concentrations measured by ddPCR for all sample types were consistently lower than those measured by qPCR, by a factor of 2.6 ± 2.8 for B. theta and 18.7 ± 10.0 for BacHum. We suggest that differences in assay performance on ddPCR and qPCR platforms may be linked to the characteristics of the assay targets (that is, genes with multiple versus single copies and encoding proteins versus ribosomal RNA) however further work is needed to validate these ideas. We conclude that ddPCR is a suitable tool for microbial source tracking, however, other factors such as cost-effectiveness and assay-specific performance should be considered.

Microbial Community Composition and Putative Biogeochemical Functions in the Sediment and Water of Tropical Granite Quarry Lakes

Re-naturalized quarry lakes are important ecosystems, which support complex communities of flora and fauna. Microorganisms associated with sediment and water form the lowest trophic level in these ecosystems and drive biogeochemical cycles. A direct comparison of microbial taxa in water and sediment microbial communities is lacking, which limits our understanding of the dominant functions that are carried out by the water and sediment microbial communities in quarry lakes. In this study, using the 16S rDNA amplicon sequencing approach, we compared microbial communities in the water and sediment in two re-naturalized quarry lakes in Singapore and elucidated putative functions of the sediment and water microbial communities in driving major biogeochemical processes. The richness and diversity of microbial communities in sediments of the quarry lakes were higher than those in the water. The composition of the microbial communities in the sediments from the two quarries was highly similar to one another, while those in the water differed greatly. Although the microbial communities of the sediment and water samples shared some common members, a large number of microbial taxa (at the phylum and genus levels) were prevalent either in sediment or water alone. Our results provide valuable insights into the prevalent biogeochemical processes carried out by water and sediment microbial communities in tropical granite quarry lakes, highlighting distinct microbial processes in water and sediment that contribute to the natural purification of the resident water.

Impact of nutrient addition on diversity and fate of fecal bacteria

Understanding the variations in the microorganisms associated with human fecal pollution in different types of water is necessary to manage water quality and predict human health risks. Using an Illumina sequencing method, we investigated variations in the fecal bacteria originating from fresh human feces and their decay trends in nutrient-supplemented water and natural river water. Nutrient addition contributed to the growth of heterotrophic bacteria like Comamonadaceae, Cytophagaceae, and Sphingobacteriaceae, but led to lower concentrations for Bacteroidaceae, Lachnospiraceae, and Ruminococcaceae. This result suggests that the utilization of nutrients by high-activity bacteria may suppress other bacteria via depletion of the available nutrient resources. As we did not observe proliferation of Bacteroidales, Lactobacillales, Clostridiales, or Ruminococcaceae in either supplemented or river water, we consider these groups suitable for use as indicators to determine the level of fecal pollution. Moreover, we tested the persistence of Bacteroidales markers, including general-Bacteroidales marker GenBac and human-specific Bacteroidales marker qHS601, by quantitative PCR. We observed similar trends in the decay of the Bacteroidales markers GenBac and qHS601 in the nutrient-supplemented water and natural river water, and the high R² values of the GenBac (R²_{nutrient-supplemented} = 0.93, R²_{natural river} = 0.81) and qHS601 (R²_{nutrient-supplemented} = 0.93, R²_{natural river} = 0.91) suggests they are a good fit for the first-order decay model. We also found stronger correlations between the markers and potential pathogenic anaerobes in the different types of water, demonstrating the validity of the use of GenBac and qHS601 from Bacteroidales for the identification of human-associated pollution sources.

Occurrence of Traditional and Alternative Fecal Indicators in Tropical Urban Environments under Different Land Use Patterns

This study evaluated the geospatial distribution of fecal indicator bacteria (FIB) (i.e., Escherichia coli, Enterococcus spp.) and the alternative fecal indicator pepper mild mottle virus (PMMoV) in tropical freshwater environments under different land use patterns. Results show that the occurrence and concentration of microbial fecal indicators were higher for urban than for parkland-dominated areas, consistent with land use weightage. Significant positive correlations with traditional FIB indicate that PMMoV is a suitable indicator of fecal contamination in tropical catchments waters (0.549 ≤ rho ≤ 0.612; P < 0.01). PMMoV exhibited a strong significant correlation with land use weightage (rho = 0.728; P < 0.01) compared to traditional FIB (rho = 0.583; P < 0.01). In addition, chemical tracers were also added to evaluate the potential relationships with microbial fecal indicators. The relationships between diverse variables (e.g., environmental parameters, land use coverage, and chemical tracers) and the occurrence of FIB and PMMoV were evaluated. By using stepwise multiple linear regression (MLR), the empirical experimental models substantiate the impact of land use patterns and anthropogenic activities on microbial water quality, and the output results of the empirical models may be able to predict the sources and transportation of human fecal pollution or sewage contamination. In addition, the high correlation between PMMoV data obtained from quantitative real-time PCR (qPCR) and viral metagenomics data supports the possibility of using viral metagenomics to relatively quantify specific microbial indicators for monitoring microbial water quality (0.588 ≤ rho ≤ 0.879; P < 0.05).IMPORTANCE The results of this study may support the hypothesis of using PMMoV as an alternative indicator of human fecal contamination in tropical surface waters from the perspective of land use patterns. The predictive result of the occurrence of human fecal indicators with high accuracy may reflect the source and transportation of human fecal pollution, which are directly related to the risk to human health, and thereafter, steps can be taken to mitigate these risks.

Geospatial distribution of viromes in tropical freshwater ecosystems

This study seeks to understand the general distribution of virome abundance and diversity in tropical freshwater ecosystems in Singapore and the geospatial distribution of the virome under different landuse patterns. Correlations between diversity, environmental parameters and land use patterns were analyzed and significant correlations were highlighted. Overall, the majority (65.5%) of the annotated virome belonged to bacteriophages. The percentage of Caudovirales was higher in reservoirs whereas the percentages of Dicistroviridae, Microviridae and Circoviridae were higher in tributaries. Reservoirs showed a higher Shannon-index virome diversity compared to upstream tributaries. Land use (urbanized, agriculture and parkland areas) influenced the characteristics of the virome distribution pattern. Dicistroviridae and Microviridae were enriched in urbanized tributaries while Mimiviridae, Phycodnaviridae, Siphoviridae and Podoviridae were enriched in parkland reservoirs. Several sequences closely related to the emerging zoonotic virus, cyclovirus, and the human-related virus (human picobirnavirus), were also detected. In addition, the relative abundance of PMMoV (pepper mild mottle virus) sequences was significantly correlated with RT-qPCR measurements (0.588 < r < 0.879, p < 0.05). This study shows that spatial factors (e.g., reservoirs/tributaries, land use) are the main drivers of the viral community structure in tropical freshwater ecosystems.