Microbe-level investigation of sequential anaerobic/aerobic bioremediation of polychlorinated biphenyls in soil

Abstract

Environmental polychlorinated biphenyl (PCB) contamination is often comprised of mixtures of congeners with a range of degree of chlorination. Complete biodegradation of PCB mixtures is rarely achieved, since highly chlorinated PCBs are not metabolized under aerobic conditions and less chlorinated congeners are not degraded by anaerobic microorganisms. Sequential anaerobic-aerobic treatment for PCB contamination is promising, yet little is known about specific parameters important for its successful outcome. A combination of engineering and microbial ecology was used in this work to understand the underlying mechanisms in the bioremediation of PCBs. Laboratory-scale, PCB-spiked, soil slurry microcosms were used to investigate the effects of engineering parameters on PCB biodegradation and the microbial community in each phase and in the overall sequential process. The overall goal of this work was to not only effect greater overall PCB degradation in the soil microcosms but to gain knowledge of the microbial populations responsible for this degradation so that this information could be utilized for other sites. In microcosms incubated anaerobically, dechlorination of several hexachlorobphenyts to penta- and tetrachlorobiphenyls by indigenous organisms was detected in soil microcosms within 12 weeks. The dechlorination pattern was strongly dependent on the congener being degraded in addition to the electron donor utilized. Phospholipid fatty acid/fatty acid methyl ester (PLFA/FAME) analysis indicated small differences in community structure between different electron donor treatments. Biphenyl-amended microcosms achieved the highest extent of dechlorination with 22.5% dechlorination achieved over the 60-wk anaerobic phase. Unamended, butyrate-, methanol-, butyrate+biphenyl-, and methanol+biphenyl-amended microcosms exhibited less dechlorination over the 60-wk anaerobic phase with unamended microcosms exhibiting the highest initial rate of dechlorination. Use of a denaturing gradient gel electrophoresis method combined with sequencing showed that most of the cloned sequence types were related to the I6S rRNA sequences of aerobic species, and several to acidiphilic bacteria. In addition, several sequence types were similar to known microorganisms involved in either iron reduction or iron oxidation. These microbial types fit the soil characteristics; i.e., slightly acidic and high in iron. Both 2,3',4'-PCB and biphenyl were completely degraded in aerobic microcosms in 7 wk. Microcosms not amended with biphenyl showed decreased degradation rates compared with those amended with biphenyl during the aerobic phase. Although total biomass remained constant in all treatments (as shown by PLFA), changes in microbial community composition occurred in biphenyl-amended microcosms: the anaerobic microbial community decreased in size and there was a subsequent increase in aerobic lipid biomarkers. Soil incubated with no amendment showed no difference in community composition from time zero samples. A key question in developing a sequential remediation treatment was how to switch from anaerobic to aerobic treatment conditions. Aerobic microorganisms in microcosms treated with both air and hydrogen peroxide were able to partially degrade 2,3',4'-PCB in 7 wk. In all treatments, the rate and extent of degradation were less than that observed in the aerobic treatment alone. It was hypothesized that this was due to the harsher method used to initiate anaerobic conditions before the oxygenation treatment and not due to the oxygenation treatment itself. No difference in degradation rates or microbial populations was observed between treatments, but a slightly higher rate of dissolved oxygen (DO) utilization in H2O2-treated microcosms prompted the use of a 0.25% H2O2 treatment in the sequential experiments. The sequential treatment process was more effective for overall PCB removal than either the anaerobic or aerobic processes alone. A. total reduction in PCB mass of 57.5% in sequential microcosms was observed versus 27% and 25% for anaerobic and aerobic microcosms alone. Because the soil was aerobic at the onset of the experimentation, highly reducing conditions were never obtained. In fact, degradation of the trichlorobiphenyl and biphenyl occurred during the first 12 wk of the anaerobic phase, essentially making the treatment process an aerobic/anaerobic/aerobic process. The less-reducing conditions of the anaerobic process allowed the aerobic microorganisms to survive the 60 wk and rebound to degrade the less chlorinated congeners produced. Few changes were observed in community structure between the anaerobic and aerobic phase samples.