a b s t r a c tIt is widely demonstrated that antibiotics in the environment affect microbial community structure. However, direct evidence regarding the impacts of antibiotics on microbial functional structures in wastewater treatment systems is limited. Herein, a highthroughput functional gene array (GeoChip 3.0) in combination with quantitative PCR and clone libraries were used to evaluate the microbial functional structures in two biological wastewater treatment systems, which treat antibiotic production wastewater mainly containing oxytetracycline. Despite the bacteriostatic effects of antibiotics, the GeoChip detected almost all key functional gene categories, including carbon cycling, nitrogen cycling, etc., suggesting that these microbial communities were functionally diverse. Totally 749 carbon-degrading genes belonging to 40 groups (24 from bacteria and 16 from fungi) were detected. The abundance of several fungal carbon-degrading genes (e.g., glyoxal oxidase ( glx), lignin peroxidase or ligninase (lip), manganese peroxidase (mnp), endochitinase, exoglucanase_genes) was significantly correlated with antibiotic concentrations (Mantel test; P < 0.05), showing that the fungal functional genes have been enhanced by the presence of antibiotics. However, from the fact that the majority of carbon-degrading genes were derived from bacteria and diverse antibiotic resistance genes were detected in bacteria, it was assumed that many bacteria could survive in the envi- Available online at www.sciencedirect.com journal homepage: www.elsevier.com/locate /wa tres w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 2 9 8 e6 3 0 8

A full-scale biosystem consisting of two anaerobic reactors (HA and BF1) and four aerobic ones (BF2-BF4 and OD) in succession and receiving antibiotic-bearing (mainly streptomycin) wastewater was used for studying the impacts of antibiotics on microbial community structures. Significant decreases of streptomycin (from 3955 ± 1910 to 23.1 ± 4.7 μg L −1 ) and COD Cr were observed along the treatment process. Cloning results show that the anaerobic reactors (HA and BF1) were dominated with Deltaproteobacteria (51%) mainly affiliated with sulfate-reducing bacteria (SRB), while the aerobic BF2 receiving streptomycin of 408.6 ± 59.7 μg L −1 was dominated with Betaproteobacteria (34%), Deltaproteobacteria (31%) and Bacteroidetes (14%). Gammaproteobacteria (15.9−22.4%), Betaproteobacteria (10.0−20.3%), and Bacteroidetes (4.5−29.7%) became the major bacterial groups in aerobic BF3-OD receiving streptomycin of ≤83 ± 13 μg L −1 . Archaea affiliated with Methanomethylovorans hollandica-like methylotroph was abundant in HA and BF1 (archaea/bacteria, 0.54−0.40; based on specific gene copy number), suggesting the coexistence of SRB and methanogens in degrading pollutants. Fungi were abundant (fungi/bacteria, 0.15; based on specific gene copy number) with the dominance of Ascomycota (clone ratio of Ascomycota/eukarya, 25.5%) in BF2, suggesting that fungi could be an important player in pollutant removal under high levels of antibiotics. This study demonstrates that under high antibiotic levels, wastewater treatment communities may maintain system stability through adjusting bacterial, archaeal, and eukaryal compositions.

No abstract