1. Biolog Sample Preparation and Analysis
For studying biofilm metabolic pathway activity change, community level physiological profiling (CLPP) using Biolog AN plates (Biolog, Hayward, CA, USA) was employed. At the end of each stage, two MFCs from both groups were opened and one layer of the anodes was removed aseptically for the Biolog analysis. The anode sample was transferred into 40 mL of sterile phosphate buffered saline (PBS, Sigma-Aldrich, UK) solution and re-suspended by rigorous vortex mixing for 3 min. The 96 wells of a Biolog AN plate were inoculated with 150 µL of each sample per well. Then the microplates were incubated anaerobically (10 % CO2 in oxygen free N2) in a portable container at 30 °C to allow utilisation reactions to proceed along with tetrazolium colour changes, and the changes in colour intensity were measured at 590 nm every 24 hours up to 120 h, using a Biolog Microstation in accordance with the Biolog operating protocol.
Average well colour development (AWCD) was calculated according to Garland and Mills (Garland and Mills, 1991), i.e., AWCD = ∑(C − R)/n where C is the optical density of each well measured at 590 nm, R is the absorbance value of the control well (A1), and n is the number of substrates (n = 95). In order to compare a specific carbon source (acetate in this case) utilisation of the two groups at different stages, the raw difference data of the well containing acetic acid was divided by the AWCD of the plate, i.e. (C - R)/AWCD. As a measure of the degree of substrate utilisation (substrate richness) and diversity of extent of particular substrates utilisation (substrate evenness), the Shannon-Wiener index (SI) was used: H = -∑_(i=1)^n▒〖p_i (〖ln⁡p〗_i)〗 where p_i is the proportion of a microbial activity on a particular substrate (ODi) to the total microbial activity (∑▒〖OD〗_i ) and N is the number of substrates on a plate (Frąc et al., 2012; Zak et al., 1994). Plate readings at 24 h of inoculation were used to calculate AWCD and SI.

2. Indirect Measurement of Bacterial Population
Viable counts were performed on non-selective nutrient agar (Oxoid, UK) and the number of colony forming units per sample (cfu/mL for effluent and cfu/mm2 for anodic biofilm) was calculated. A 1 mL volume of each sample was serially diluted to 10-6 and 100 µL from sample dilution 10-4, 10−5 and 10−6 spread onto the non-selective recovery medium. All plates were incubated in an anaerobic cabinet (MK3 anaerobic workstation, Don Whitley, Shipley, UK) at 37 °C for 5 days.
The optical density at 600 nm wavelength of each undiluted 1 mL sample was measured using a spectrophotometer (model name: 6700, Jenway, Staffordshire, UK). 

3. Polarisation Measurement and Power Output Calculations
Power output of the MFCs was monitored in real time in volts (V) against time using an ADC-24 Channel Data Logger (Pico Technology ltd., Cambridgeshire, UK). Polarisation experiments were performed weekly by connecting a decade variable resistor box (Centrad Boite A Decades De Resistances DR07, ELC, France) between the anode and cathode electrodes and changing the external resistance from 30 kΩ to 10 Ω in 5-minute intervals, after the MFCs had established a steady-state open circuit voltage at the start of the experiment. Maximum power output (PMAX) and internal resistance (RINT) were calculated from the power curves. The current (I) in amperes (A) was determined using Ohm’s law.