Being widely reported its natural ability to generate electricity, Shewanella oneidensis MR-1 was fed with lactate in the MFC to quantify the performance of our MFC device and to be used as the experimental control for our engineered Shewanella system. While Shewanella can utilize different carbon sources such as lactate, pyruvate and acetate, lactate is selected in our experiment because it has shown that lactate is an energy-favorable carbon substrate for this strain^[1]. As described from literature^[2], it is desirable to inoculate Shewanella culture until it reaches an early stationary phase. Therefore, we decided to characterize the stationary OD₆₀₀ of Shewanella in LB by its growth curve since we used LB to proliferate the cell. To get a full understanding of the cell population of Shewanella in the designed condition of MFC, we also stimulated the anaerobic bacterial growth in M9 minimal medium. "Anaerobic condition is chosen because Shewanella will have a better performance in generating electricity."

Growth curve of Shewanella oneidensis MR-1

From the graph, the stationary OD600 value was observed at around 1.8-1.9. In addition, the doubling time was 0.981h. Since a stable cell population is desired for a stable electricity generation in the MFC, we attempted to stimulate the anaerobic bacterial growth.

Stationary Shewanella MR-1 in LB was washed and transferred to M9 anaerobic medium at time zero. Although no significant cell growth was observed, the OD₆₀₀ remained at 1.45-1.5 for 24 hours. After the decreasing trend was observed, we decided to repeat the measurement, but allowing aeration to the cell culture after shifting from LB to M9 for better cell adaptation to the new medium.

Stationary Shewanella MR-1 in LB was washed and transferred to M9 aerobic medium at time zero. At time t = 30h, M9 medium was renewed. The drop of OD₆₀₀ is likely due to the loss of cell during medium replacement. Similar to figure 2, the OD600 stayed at 1.79-1.84 with slight fluctuation. While no significant cell growth was observed again, the OD₆₀₀ sustained for 24 hours after shifting to anaerobic condition when 20mM lactate was added.

The MFC was constructed as described in Notebook. As a fast and convenient method, we scaled down the microbial fuel cell in the experimental set-up as shown in figure 4. We constructed it with eppendorfs, which are easily obtained from the laboratory while less reagents and materials are required. It was sealed with epoxy and paraffin to prevent liquid leakage and provide anaerobic condition.

Inspired by Bielefeld 2013 iGEM Team, we also attempted to use 3D-printing to make our MFC prototype as our product. For convenient, our prototypes as shown in figure 5 and 6 are designed to print as a whole such that no assembly of individual component is required.

MFC experiment

The control has the same construction as the sample except no cells were added. 20mM lactate was added as carbon source. Despite the fluctuation shown from time 20 to 35 minutes of three individual trials, the MFC potential increases over time and reach the similar plateau. Besides, the curves of three trials are significantly higher than the control curve, while the control is staying at the low voltage over the whole experimental period. We could conclude that the difference between the control and MFC curves is most likely due to the exogenic properties of the Shewanella and the increase of voltage could be attributed to the growth of the Shewanella in the MFC.

After we built the mfc circuit as shown in figure 10, we could obtain the current over time when the variable resistor is fixed. Dividing the measured current by contact surface area of the electrodes, we could obtain the above graph. Figure 11 showed a constant increase of the current density over time during the experimental period.

Therefore, it is possible to determine the internal resistance by plotting the above graph, given that the emf is a constant. However, emf is not a constant for mfc as well as the internal resistance.

As known, the internal resistance of a microbial fuel cell is affected by the emf of itself. To determine its relationship, different emf with corresponding internal resistance was recorded over time. Since the above graph is only valid in constant emf, we could only approximate the constant emf by obtaining the current-resistance relationship in a very short period of time so that the change of the emf could be neglected.

The graph above showed the relationship of internal resistance and the output voltage of the microbial fuel cell across the first 90 minutes after it was built. 20mM lactate was added as the sole carbon source. The curve indicates that the internal resistance increases with the output voltage more rapidly in the early stage, but reaches its maximum as the voltage further increases. From this result, it could be concluded that the internal loss of the MFC will be gradually minimised as long as the voltage continue to increase.

>

Since we have obtained the relationship of voltage against the current, which could be evaluated to the terminal power output, we want to tentatively evaluate the actual energy output of our MFC. As we learnt from the figure 13, the power will fluctuate at the beginning, but it will develop an apparent increasing trend after the experiment had been carried out for 50 minutes. To further visualise the actual energy output of the MFC, we calculate the total work done from the data of power and it gives the figure 14, which still abide to the increasing trend of the power. From the two figure above, we could conclude that our designed MFC could be applicable if further adjustments were made to enhance the output.

MFC Efficiency

Assuming the decrease of cell population in the growth curve was because of using up the 20mM lactate after 24 hours in the M9 anaerobic medium, we estimated the average lactate uptake flux from the wet mass of cell, then we search for the theoretical electron generating flux through the FBA model described in figure 4 of modelling pages.

The wet mass was converted to dry mass by a factor of 0.3, assuming 70% water content in the mass. As described in modelling, the corresponding maximum DET flux was evaluated based on the assumption of 5% of biomass growth flux. The efficiency was evaluated by comparing the measured current with the maximum theoretical current.

Transformation

We first made the chemically competent cells and attempted to transform the Shewanella by chemical method. After several unsuccessful attempts by increasing both cell and DNA concentration, we turned to use electroporation for higher transformation efficiency. Instead of RFP in the competency kit plate, GFP is chosen as reporter for competency check to avoid confusion with pink Shewanella oneidensis MR-1. However, we cannot get any transformed colony yet. Literature also showed low transformation efficiency of pSB1C3-GFP in Shewanella, 4 colonies were observed on the transformed plate after electroporation^[3]. We hypothesized that the competent cell concentration is not high enough to achieve observable transformed colonies on plate.

[1]J. Myers and C. Myers, "Role for Outer Membrane Cytochromes OmcA and OmcB of Shewanella putrefaciens MR-1 in Reduction of Manganese Dioxide", Applied and Environmental Microbiology, vol. 67, no. 1, pp. 260-269, 2001.
[2]E. Marsili, D. Baron, I. Shikhare, D. Coursolle, J. Gralnick and D. Bond, "Shewanella secretes flavins that mediate extracellular electron transfer", Proceedings of the National Academy of Sciences, vol. 105, no. 10, pp. 3968-3973, 2008.
[3]I. Ng, Y. Guo, Y. Zhou, J. Wu, S. Tan and Y. Yi, "Turn on the Mtr pathway genes under pLacI promoter in Shewanella oneidensis MR-1", Bioresources and Bioprocessing, vol. 5, no. 1, 2018.