Abstract:Polycyclic aromatic hydrocarbons (PAHs), being nowadays the major environmental pollutants, can bring about adverse effects on soil microbial community. Classical biological indicators, such as dehydrogenase activity, substrate induced respiration rate and community level physiological profiles, are widely used to evaluate toxicity of PAHs to soil ecosystems. However, it is not feasible to use them to monitor soil pollution in a real-time continuous way. Microbial fuel cells (MFCs) are a kind of devices that convert chemical energy of organic substrates into electrical power through the catalysis of electrogenic bacteria. As the current which is closely related to metabolic activity of the electrogenic bacteria can be recorded immediately and continuously, MFCs have already been used as a biosensor to monitor wastewater treatment. Electrogenic bacteria are known to be widely distributed in soil. They generate electricity while decomposing soil organic matter and are extremely sensitive to environmental change. Pollutants creating stress on soil microbes may also inhibit the activity of electrogenic bacteria. Thus electrical signals generated by soil micro-organisms could be used as an indicator to detect soil pollution. The objectives of the study are to determine feasibility of the use of the electrical signals to evaluate pyrene toxicity in soil, and understand how soil electrogenic bacteria respond to pyrene pollution. Synthetic PAHs contaminated soils were prepared, different in amount of pyrene spiked, i.e. 0 (as control), 60, 120, 180 and 240 mg kg-1 pyrene and then packed into MFCs anode chambers, separately. To accelerate the electricity generation, 4% (w/w) glucose was thoroughly mixed into the soil before packing. Voltage of MFCs was real time monitored every 10 min for 110 h. To examine reliability of the use of electrical signals in toxicity detection, dehydrogenase activity in the pyrene contaminated soil was also measured simultaneously. After the MFCs operated for 110 h, electrochemical activity of the soil microbes was determined using cyclic voltammetry. Bacterial community diversity on the MFCs anode biofilm was determined through phylogenetic analysis of 16S rRNA genes with the PCR-DGGE and sequencing methods. All the DGGE band sequences were submitted to the GenBank and assigned with accession number, KJ128061 - KJ128073. Results show that the cells started to generate electricity after 6 h of operation. Peak voltages from MFCs of pyrene-contaminated soils were monitored varying between 240 ~ 270 mV, while that from control reached 305 mV. The coulomb production of the MFCs within the 110 h decreased significantly with the rate of pyrene added, and significantly (p < 0.01) correlated with soil dehydrogenase activity. Cyclic voltammogram shows that the soil of control treatment had higher redox peaks than the soil spiked with 240 mg kg-1 pyrene with ranging at potentials around -100 mV and 100 mV while no peaks were observed in the fumigated and sterilized soil, illustrating that pyrene addition weakened electrochemical activity of the soil micro-organisms. DGGE patterns show that after 110 h of operation, the bacterial community on the MFCs anode biofilm differed significantly from that in the soil in structure. Sequencing and phylogenetic analysis of the DGGE bands reveal that the bacteria on the anode biofilm was highly similar to the known electrogenic bacteria, including Sporolactobacillus, Clostridium, Enterobacter, Bacillus and Ethanoligenens. Pyrene addition decreased the abundance of Bacillus. This study demonstrates that the electrical signals generated by soil micro-organisms could satisfactorily be used to evaluate pyrene toxicity in the soil. The mechanisms of pyrene reducing electricity generation include inhibition of electrical activity of soil micro-organisms and alteration of the structure of the electrogenic bacterial community on the anode. In future the study in this field should be oriented toward monitoring of more pollutants varying in a wider range of concentrations, and optimization of the configuration and operation of MFCs to shorten their startup time and to improve their sensitivity of electrical signals.