Effects of the toxin 3-chloroaniline at low concentrations on microbial community dynamics and membrane bioreactor performance.

The effects of toxins at ambient concentrations on microbial activity and community dynamics are poorly understood. We operated 4 membrane bioreactors (MBRs) in parallel; two reactors were continuously exposed to the toxin 3-chloroaniline (3-CA) at environmentally relevant levels, representing 25% of the total chemical oxygen demand (COD; Total COD = 400 mg l(-1) d(-1)), and two reactors received no 3-CA. During the 70 d exposure to 3-CA the microbial communities never adapted as evidenced by a 48% and 14% reduction in COD and ammonia removal, respectively, compared to over 92% reduction for both measurements in the controls. The bacterial 16S rRNA gene was monitored using terminal restriction fragment length polymorphism (T-RFLP) analysis (n = 15 temporal grab samples per reactor) over the 70 d period. T-RFLP spectra analysis compared the rapid species turnover rate (STR) approach with the more computationally intensive non-metric multi-dimensional scaling (NMS) complemented with multi-response permutation procedure (MRPP). The methods revealed comparable findings and the presence of 3-CA selected for a more convergent community with less bacterial turnover. In contrast, the control MBRs were more divergent as evidenced by greater bacterial turnover variability. The importance of studying replicate reactors is highlighted by the fact that one of the two controls was significantly different from the treatment MBRs (p-value = 0.01, α = 0.05) whereas the other one was not (p-value = 0.24, α = 0.05). The study suggests that analysis of community dynamics with the rapid STR approach and with NMS/MRPP can lead to comparable results when targeting the 16S rRNA gene. The use of replicate bioreactors is essential for meaningful interpretation of microbial community patterns.