Next generation sequencing for the water industry

Abstract

The wastewater industry uses biotechnology to ensure that the discharge of sewage does not have deleterious effects on the environment, yet knowledge of the underlying microbiology is poor. This leads to over engineered and inefficient processes which occasionally and unexpectedly fail. Similarly the impact of sewage on the microbiology of receiving waters is unclear. Recent developments in DNA sequencing have enabled its use where cost was prohibitive. I investigated two applications of Next Generation Sequencing (NGS); activated sludge process monitoring for nitrification, foaming and bulking, and microbial source tracking of faecal contamination in bathing waters. Samples from 32 activated sludge plants (ASPs) were collected and analysed. Cell specific ammonia oxidation rates were calculated using the equation 𝐶𝑆𝐴𝑂𝑅 =(𝐴 × 𝑀 ×106) ×𝑟𝐴𝑂𝐵 ×𝑀𝐿𝑆𝑆 ×𝑉 where A = grams of ammonia oxidised, M = the number of moles of ammonia in a gram, r = correction factor of 0.9 due to some ammonia removal by adsorption and assimilation (Daims, Ramsing, et al. 2001), MLSS = mixed liquor suspended solids in mg/L and V = the volume of the aeration basin in litres. The CSAOR in nitrifying plants ranged from one to ten mmol / cell / hour, in agreement with other CSAOR studies using alternative techniques. Biological foaming in ASPs occurs when the abundance of filamentous bacteria with hydrophobic surface membranes becomes excessive, though the exact abundance threshold above which foaming occurs has not yet been established. The relative abundance of bacteria associated with foaming was measured for all ASPs which were then categorised as non-foaming, occasionally-foaming or currently-foaming based on operator assessment. There was a significant difference in the abundance of foaming bacteria between non-foaming and occasionally foaming plants (ANOVA p < 0.001), with all non-foaming plants having less than 1% relative abundance of foaming bacteria. These results demonstrate that NGS could be a useful ASP process monitoring tool. A bathing water catchment was sampled throughout a bathing season, including a storm event. Partial least squares analysis showed there was a significant correlation between faecal indicator bacteria and the cumulative apportioned fraction of sources (using Bayesian statistics) in the bathing water community (p < 0.001, r2 = 87%). Faecal host marker analysis detected human contamination upstream of any wastewater network inputs, illustrating the impact of diffuse human pollution. Whole community analysis apportioned the bathing water microbial community to point and diffuse sources, and found that whilst human sources were dominant during storm conditions, in dry weather the primary source of faecal contamination was variable and in some cases could not be attributed to known faecal sources.