Enhanced Shewanella oneidensis MR-1 anode performance by adding fumarate in microbial fuel cell

Accepted Manuscript Enhanced Shewanella onedensis MR-1 anode performance by adding Fumarate in microbial fuel cell Peng Zhang, Jia Liu, Youpeng Qu, Yujie Feng PII: DOI: Reference:

S1385-8947(17)31153-1 http://dx.doi.org/10.1016/j.cej.2017.07.008 CEJ 17273

To appear in:

Chemical Engineering Journal

Received Date: Revised Date: Accepted Date:

24 April 2017 2 July 2017 3 July 2017

Please cite this article as: P. Zhang, J. Liu, Y. Qu, Y. Feng, Enhanced Shewanella onedensis MR-1 anode performance by adding Fumarate in microbial fuel cell, Chemical Engineering Journal (2017), doi: http://dx.doi.org/ 10.1016/j.cej.2017.07.008

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Date: April 24, 2017 Submitted to: Chemical Engineering Journal

Enhanced Shewanella onedensis MR-1 anode performance by adding Fumarate in microbial fuel cell Peng Zhanga, Jia Liua*, Youpeng Qub, Yujie Fenga* a

State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology. No 73 Huanghe Road, Nangang District, Harbin 150090, China b School of Life Science and Technology, Harbin Institute of Technology. No. 2 Yikuang Street, Nangang District, Harbin 150080, China *Corresponding Author: E-mail: [email protected]; phone: (+86) 451-86287017; Fax: (+86) 451-86287017 *Co-Corresponding Author: E-mail: [email protected]

1

Abstract The anode biofilm plays an important role in the microbial fuel cell (MFC) performance, which relies on the catalytic function of the anodic biofilm to transform the chemical energy into electricity energy. In this study, fumarate, as a kind of electron acceptor, is added into the anode system of MFC to figure out its effect on the biofilm formation of S. onedensis MR-1. With fumarate addition, more bacteria are observed on the anode surface, and the power density has been promoted by 2.41 times than that without fumarate. Mechanism analysis shows that the fumarate addition could impair the secretion of riboflavin and inhibit the indirect electron transfer process, by finishing the respiration process with fumarate as electron acceptor, which promoted the bacteria proliferation and the anode biofilm formation. Keywords: Shewanella; Fumarate; Biofilm; Microbial fuel cell; Riboflavin;

1. Introduction Microbial fuel cell (MFC), an emerging technology for simultaneously treating wastewater and recovering electricity energy, has drawn lots of attentions in recent years [1-4]. MFCs are able to generate electricity by oxidizing a variety of organic substrates, including cellulose [5], organic acids [6-8] and pollutants [2, 9]. In MFCs, electrons are transferred to anode by bacteria, which possess certain electron transfer pathways that electrically connect intracellular catabolic reactions with extracellular electron acceptors [10, 11]. In general, the electricity-generating bacteria could form a layer of biofilm on the anode, which could determine the electricity generation performance of MFC technology. 2

In MFCs, Shewanella species, one of the most widely studied exoelectrogens, have all the known bacterial extracellular electron transfer strategies, i.e. indirect electron transfer (IET) via self-secreted electron mediators and direct electron transfer (DET) via outer membrane cytochrome c and nanowire [12, 13]. The performance of Shewanella biofilm could be affected by many diverse factors [14], including pH, dissolved oxygen concentration, temperature and electrode material [15, 16]. Shewanella, as facultative exoelectrogens, can propagate and generate electricity under both anoxic and aerobic situations [17-19]. S. oneidensis MR-1 is an extensively investigated strain of Shewanella species [20]. This bacterium could respire using a wide variety of substrates as electron acceptor, including oxygen, fumarate, nitrate, nitrite, thiosulface, Fe(Ⅲ), Cr(VI), dimethyl sulphoxide (DMSO) and trimethylamino oxide (TMAO) [21-23]. Even though as a competing electron acceptor to electrode, oxygen is reported to be able to promote the biofilm formation of S. putrefaciens CN32 [14]. For S. oneidensis MR-1, oxygen could also increase the biomass proliferation, resulting in an overall increase in electricity generation [15]. A possible reason has been affirmed that, oxygen could promote the bacteria growth and therefore increase the bacteria attachment on anode. Moreover, further research pointed out that oxygen exposure always promotes biomass growth and impedes per-cell extracellular electron transfer (EET) rate. Thus, when the increase of biomass overcome the decrease of per-cell EET ability at a certain level of oxygen concentration, higher current generation could be achieved. Fumarate is another kind of electron acceptor for Shewanella bacteria. The 3

electrons, derived from the substrate metabolism inside of the cell, are transferred to the periplasmic fumarate reductase FccA, where fumarate was reduced to succinate [24]. It was reported that fumarate respiration could induce permeability of S. decolorationis S12 cell membranes, which could be repaired by switching fumarate-respiration to electrode-respiration [25]. The current output and overall performance of a S. oneidensis MR-1 MFC could be modulating the lactate and fumarate concentration, tipping the balance between cells inside and outside the electrode environment [26]. Thus, as another competitive electron acceptor to electrode, fumarate may also promote the Shewanella biofilm formation on the anode by increasing the biomass proliferation. However, it was rarely reported the current generation performance of MFCs with pure cultures in the present of fumarate as electron acceptor. In this study, we added fumarate in the anode chamber with S. oneidensis MR-1 as exoelectrogen to figure out whether the existence of fumarate would facilitate anode bacteria attachment and promote electricity generation. A two-chamber MFC system has been setup, and the electricity generation and electrochemical performance of the reactors with and without fumarate addition were compared.

2. Experiments 2.1 MFC operation and bacteria culture Two-chamber MFCs, with volume of 300 mL for each chamber, was separated by proton exchange membrane (Nafion 117 PEM, from Hesen, Shanghai) and used in this experiment (Figure 1). Carbon cloth (WOS1002, CeTech Co., Ltd) with a size of 4

1*2 cm and a size of 2 *4 cm were used as anode and cathode, respectively. The anode and cathode were connected to the external circuit via titanium wire. (Figure 1) All MFCs were operated under batch mode with a consistent resistance of 1000 Ω. For the control experiment, the anode medium was 18 mM sodium lactate dissolved in 50 mM phosphate buffer solution (PBS) (Na2HPO4·12H2O, 10.32 g/L; NaH2PO4·2H2O, 3.32 g/L; NH4Cl, 0.31 g/L; KCl, 0.13 g/L; trace minerals and vitamins). For the experimental groups, despite the same ingredient as the control one, 9 mM fumarate was also added in to the anode electrolyte. The cathode solution was 50 mM K3[Fe(CN)6] dissolved in the PBS solution. All the tests were operated in triplicates. S. oneidensis was cultured at 30 °C in LB medium (25 g/L) with shaking at 150 rpm for the chemostat until the optical density at 600 nm (OD 600) reached about 1.5 [27, 28]. The cells were collected by centrifugation (5000 rpm, 10 min) and washed with PBS for three times. Then, the cells were re-suspended with electrolyte (95% PBS and 5% LB, 18 mM lactate) to desired concentration (OD 600 = 1) [27]. And 10 mL of the electrolyte were inoculated in the anode chamber of MFC. After a batch operation ended, both the anode and cathode solution were refreshed to start the next batch operation. After the start-up, all the MFC were refreshed with anolyte without fumarate. 2.2 Electrochemical measurements The voltage of the external resistance connected to the MFCs was recorded at 30 min intervals by a data acquisition system (PISO-813, ICP DAS Co., Ltd.) connected 5

to a computer. Current density (I) was calculated as I = V (voltage)/R (external resistance)/A (anode surface area), and the power density was calculated as P =V*I [29]. Power density and polarization curves were obtained using an electrochemical working station (Metrohm, Autolab, B.V) by varying the applied voltage. Cyclic voltammetry (CV) was conducted by an electrochemical workstation (Metrohm, Autolab, B.V.) with Ag/AgCl electrode and Pt wire as the reference electrode and counter electrode, respectively. The CVs were tested at a scan rate of 1 mV/s and a potential window of -0.6 V to 0.2 V at turnover condition. The Columbic efficiency (CE) of the MFC was calculated as E = (Cp/CT) × 100%, where Cp is the total coulombs calculated by integrating the current over the time for lactate consumption, and CT is the theoretical amount of coulombs that can be produced from the metabolism of lactate, calculated as

where F is Faraday’s constant (96485 C/mol-electrons), b is the number of electrons potentially transferred from the nutrient, mollactate is the added moles of sodium lactate. The number of potential electrons used for lactate was 12 [17, 30] 2.3 Data Measurement The concentration of riboflavin in the anode electrolyte was analyzed by using ultra performance liquid chromatography (UPLC) according to the study before [27]. In brief, 2 mL of electrolyte from the electrochemical reactor was centrifuged (10000 rpm, 3 min) and the supernatant was filtrated with 0.22 μm cellulose membrane, and then subjected to UPLC system for analysis. UPLC analysis was performed at Agilent 6

HPLC system (CA, USA) equipped with an Agilent C18 analytical column and a photodiode array detector. Riboflavin was separated in a solvent system of water-methanol-acetic acid (68:32:0.1 v/v) at a flow rate of 0.1 mL/min, the UPLC profile was monitored at 254 nm. The concentration of lactate was also determined by the Agilent HPLC system with the mobile phase of 0.5 mM sulfuric acid solution and flow rate of 0.6 mL/min [31]. Quantitative biofilm formation analyze was conducted based on the protein mass on the surface on carbon cloth anode.[32] Briefly, after the experiment, the carbon cloth anode with certain area was cut down and soaked in 0.1 M KOH solution and heated for half an hour at 90 °C. The concentrations of the protein in the solution and on the anode were determined by BCA method. 2.4 Morphology and structural characterization The morphology of anode and biofilm was examined by a scanning electron microscope (SEM). The pretreatment process of all the samples with microorganisms was as follows: (1) samples were primarily fixed overnight in the fixative containing 2.5% glutaraldehyde at 4 oC, then washed with 0.1 M phosphate buffer solution for three times; (2) The samples were dehydrated in increasing concentrations of ethanol solution (50, 70, 90 and 100%). (3) The samples were washed with the mixture of ethanol and tert-butyl alcohol (1:1) and pure tert-butyl alcohol once respectively for dehydration; (4) samples were finally sputter coated with 10 nm of gold. All the SEM images were taken by a field emission scanning electron microscope (FEI XL30 Sirion SEM). 7

3. Results and discussion 3.1 Enhancement of MFCs performance by fumarate addition (Figure 2) Fumarate, as an intracellular electron acceptor, have an important impact on the activity of S. oneidensis MR-1, and may affect the performance of MFCs with S. oneidensis MR-1 as exoelectrogen. As showed in Figure 2, the addition of fumarate into the anode chamber exhibited enhanced effect on the voltage outputs at both start-up period and stable period. As showed in Figure 2a, the output voltage increased gradually during the start-up period, indicating that more bacteria may propagate on the anode or in the solution gradually. In Figure 2a (start-up period), the peak voltage of MFC with fumarate addition reached about 19 mV, while that of the MFC without fumarate showed no obvious peak voltage below 4 mV. In the second batch, the maximum voltage of the MFC with fumarate reached 37 mV, which was 2.7 times higher than the control one (12 mV). The same trend was also observed in the third batch, in which the peak voltage of MFC with fumarate was 117 mV, more than 2 times higher than that of the control one (59 mV). At last, in the fourth batch, the peak voltage of MFC with fumarate reached 367 mV; while, that of the control was just 128 mV. Considering the same potassium ferricyanide cathode condition, the enhanced voltage output should be due to the enhancement of anode performance. It could be concluded that, with the fumarate addition, the anode electricity generation ability was greatly increased. (Figure 3) 8

Even at the stable period (Figure 2b), the maximum voltage achieved by the MFC with fumarate was about 0.339 V, 2.75 times higher than the control one (0.123 V). The fluctuation of the peak voltages may due to the refresh operation inaccuracy. The maximum power density and polarization curves were determined with an electrochemical workstation. As showed in the Figure 3a, the maximum power density of MFC with fumarate addition reached to 87 mW/m2, which was 2.41 times higher than the control one (36 mW/m2). The open circuit voltage of the MFC with fumarate addition was 0.65 V, which was also higher than the control one (0.59 V). Considering the same condition of ferricyanide cathode, the higher open circuit voltage of the MFC with fumarate should be due to the lower anode potential. CE of the reactor with fumarate was 9.12%, slightly higher than that without fumarate (8.77%), indicating that more electrons were transferred to the anode. The fumarate addition also consumed certain electrons from lactate, which may lead to the slight difference of CE between the reactor with and without fumarate. 3.2 Enhanced anodic bacterial attachment by fumarate addition As showed in Figure 3b, the area of the CV curves of anode with fumarate addition was obviously larger than the control one, indicating that more bacteria adhered on the surface of the anode [33]. Meanwhile, a pair of redox peaks centered at about -0.25 V, corresponding to the c-type cytochromes on the outer membrane of S. oneidensis MR-1, was found for both anodes. Yu et al. once reported that, for the anode with S. oneidensis MR-1, a pair of redox peaks centered at a similar position (-0.28 V) was observed [28]. In addition, the redox peak currents of anode with 9

fumarate were much higher than the control one. The oxidation and reduction peaks of anode with fumarate were 0.27 mA and -0.16 mA, respectively, while these of the control were just 0.11 mA and -0.068 mA, respectively. This result indicated that, for the anode cultured with fumarate addition, more c-type cytochromes were involved in the electron transfer process, which should be due to more S. oneidensis MR-1 bacteria adhesion on the anode surface. (Figure 4) The result is affirmed by the SEM test and biomass concentration analyze. As showed in Figure 4, more bacteria adhesion was observed on the anode with fumarate addition than that without fumarate addition. For the control one, S. oneidensis MR-1 cells just scattered on the carbon fiber and no obvious biofilm was observed on the anode surface. While, for the anode with fumarate addition, much thicker biofilm appeared on some part of the anode, which may be caused by the higher S. oneidensis MR-1 cell concentration in the solution. The SEM results were also supported by the biomass analysis result. The protein concentration of the anode with fumarate was 131.72 μg/cm2, almost two times higher than the control one (74.61 μg/cm2). 3.3 Cell and riboflavin concentration variation During an electricity generation period (Figure S1), the control one with no fumarate addition reached the maximum voltage after 17 days, while the MFC with fumarate addition obtained the peak voltage of 120 mV after just 13 days, indicating that the addition of fumarate did not only increased the peak voltage but also shortened the time reaching the maximum voltage. OD600 was used to evaluate the 10

microbial bacteria concentration in the anode solution. As showed in Figure 5a, the OD600 value of both MFCs with and without fumarate increased along with the electricity generation due to the proliferation of S. oneidensis MR-1 cells. And the OD600 value of MFC with fumarate increased to a value of about 0.25, which was obviously higher than the control one (about 0.15), indicating that the fumarate addition greatly enhanced the bacteria growth in the anode solution. We also approximately calculated the relative amount of the biomass both on the anode and in the solution in the supporting information. Based on these empirical equations, with fumarate addition, approximately, 13.22% of the cells were attached on the anode; while, only 9.29% of the cells were found on the anode. Thus, fumarate addition could not only enhance the bacteria proliferation, but also increase the ratio of the cell on the anode. (Figure 5) However, interestingly, riboflavin concentrations of the MFCs showed opposite variation trend to the bacteria concentration. Although, more bacteria were observed in the anode solution, the concentration of riboflavin of anode solution with fumarate addition was obviously lower than that of the control (Figure 5b). For the control, about 0.25 mg/L riboflavin was detected in the anode solution, while, that with fumarate addition was just about 0.10 mg/L. Riboflavin, as the electron transfer mediator for IET, reflected the activity of mediated electron transfer process [34, 35]. Thus, the result indicated that the fumarate addition decreased the concentration of riboflavin and depressed the mediated electron transfer process. Previous study once 11

reported that oxygen could inhibit the riboflavin secretion, and thus IET process [15]. The mechanism of impeded per-cell riboflavin extraction by fumarate should be studied in the future work. 3.4 Mechanism analysis (Figure 6) As showed in Figure 6, all three kinds of electron transfer mechanism for S. oneidensis MR-1 respiration existed in the system with fumarate addition. The first kind of electron transfer pathway is DET: the electrons are transferred to the anode through the cytochromes on the outer membrane, which connected to the anode surface directly. Thus, DET process occurred just on the S. oneidensis MR-1 cells attached on the anode surface. The second kind of electron transfer pathway is IET: the electrons are transferred to the anode through the mediator of riboflavin, relying on the transformation between the oxidation and reduction status of this kind of shuttle. And the last electron transfer pathway is the electron transfer with fumarate as electron acceptor (FET): the electrons are transferred to the reductase FccA on the periplasmic space, where fumarate is reduced to succinate. Without fumarate, the S. oneidensis MR-1 cells wandering in the solution fulfill their respiration just through the IET process. With fumarate addition, S. oneidensis MR-1 cells could transfer their electrons to riboflavin through cytochromes or fumarate through the reductase. The lower riboflavin concentration with fumarate addition indicated the depressed IET respiration process for cell respiration. Meanwhile, as showed in the OD600 result, more S. oneidensis MR-1 cells were 12

detected in the anode solution with fumarate addition, indicating that the proliferation rate was greatly increased by transform the main respiration electron transfer process from IET to FET. Thus, it was concluded that S. oneidensis MR-1 cells tended to prefer FET process to IET process to accomplish the electron transfer chain with higher respiration rate. If the cells attached on the anode surface, they could transfer electrons through DET process. Thus, the cells on the carbon cloth anode surface should derived from both the proliferation of cells on the anode and the newly attachment of cells in the solution. So, more S. oneidensis MR-1 cells attachment on the anode attributed to the higher cell concentration in the solution, due to faster proliferation speed through FET process. Our results confirmed that fumarate leads to higher concentration of S. oneidensis MR-1 on electrodes, which enhanced the faster formation of anode biofilm for current production.

4. Conclusion Fumarate, as a competing electron acceptor to electrode, was added into the anode chamber, resulting in higher electricity generation and thicker biofilm formation. The power density was increased by 2.41 times with fumarate addition. The mechanism analysis indicating that the fumarate addition promoted the S. oneidensis MR-1 cell proliferation in the solution by preferring FET process to IET process. Fumarate addition resulted in more cells attached on the surface; however, it impaired the IET process by inhibiting the riboflavin secretion.

Acknowledgements This work was supported by the International Cooperating Project between China 13

and European Union (Grant No. 2014DFE90110), National Natural Science Fund of China (Grant No. 51408156 and 51209061) and the Fundamental Research Funds for the Central Universities"(Grant No.HIT.MKSTISP.201614 and HIT. NSRIF. 2015090)

Reference [1] Y. Qiao, S.-J. Bao, C.M. Li, Electrocatalysis in microbial fuel cells—from electrode material to direct electrochemistry, Energy & Environmental Science, 3 (2010) 544-553. [2] S.K. Chaudhuri, D.R. Lovley, Electricity generation by direct oxidation of glucose in mediatorless microbial fuel cells, Nature biotechnology, 21 (2003) 1229-1232. [3] B.E. Logan, Exoelectrogenic bacteria that power microbial fuel cells, Nature Reviews Microbiology, 7 (2009) 375-381. [4] J. Liu, G.M. Geise, X. Luo, H.J. Hou, F. Zhang, Y.J. Feng, M.A. Hickner, B.E. Logan, Patterned ion exchange membranes for improved power production in microbial reverse-electrodialysis cells, Journal Of Power Sources, 271 (2014) 437-443. [5] S. Ishii, T. Shimoyama, Y. Hotta, K. Watanabe, Characterization of a filamentous biofilm community established in a cellulose-fed microbial fuel cell, BMC microbiology, 8 (2008). [6] M.D. Yates, P.D. Kiely, D.F. Call, H. Rismani-Yazdi, K. Bibby, J. Peccia, J.M. Regan, B.E. Logan, Convergent development of anodic bacterial communities in microbial fuel cells, Isme Journal, 6 (2012) 2002-2013. [7] J. Liu, Y.J. Feng, X. Wang, Q. Yang, X.X. Shi, Y.P. Qu, N.Q. Ren, The effect of water proofing on the performance of nickel foam cathode in microbial fuel cells, Journal Of Power Sources, 198 (2012) 100-104. [8] J. Liu, Y.J. Feng, X. Wang, X.X. Shi, Q. Yang, H. Lee, Z.H. Zhang, N.Q. Ren, The use of double-sided cloth without diffusion layers as air-cathode in microbial fuel cells, Journal Of Power Sources, 196 (2011) 8409-8412. [9] J. Yu, J. Seon, Y. Park, S. Cho, T. Lee, Electricity generation and microbial community in a submerged-exchangeable microbial fuel cell system for low-strength domestic wastewater treatment, Bioresource technology, 117 (2012) 172-179. [10] L. Shi, H. Dong, G. Reguera, H. Beyenal, A. Lu, J. Liu, H.-Q. Yu, J.K. Fredrickson, Extracellular electron transfer mechanisms between microorganisms and minerals, Nature Reviews Microbiology, (2016). [11] J. Liu, F. Zhang, W.H. He, X.Y. Zhang, Y.J. Feng, B.E. Logan, Intermittent contact of fluidized anode particles containing exoelectrogenic biofilms for continuous power generation in microbial fuel cells, Journal Of Power Sources, 261 (2014) 278-284. [12] E. Marsili, D.B. Baron, I.D. Shikhare, D. Coursolle, J.A. Gralnick, D.R. Bond, Shewanella secretes flavins that mediate extracellular electron transfer, Proceedings of the National Academy of Sciences, 105 (2008) 3968-3973. [13] L. Na, Z. Shungui, N. Jinren, Mechanism of energy generation of microbial fuel cells, Progress in Chemistry, 20 (2008) 1233-1240. [14] C. Wu, Y.Y. Cheng, H. Yin, X.N. Song, W.W. Li, X.X. Zhou, L.P. Zhao, L.J. Tian, J.C. Han, H.Q. Yu, Oxygen promotes biofilm formation of Shewanella putrefaciens CN32 through a diguanylate cyclase

14

and an adhesin, Scientific reports, 3 (2013). [15] M. Lu, S. Chan, S. Babanova, O. Bretschger, Effect of oxygen on the per-cell extracellular electron transfer rate of Shewanella oneidensis MR-1 explored in bioelectrochemical systems, Biotechnology and bioengineering, 114 (2017) 96-105. [16] P. Clauwaert, P. Aelterman, T.H. Pham, L. De Schamphelaire, M. Carballa, K. Rabaey, W. Verstraete, Minimizing losses in bio-electrochemical systems: the road to applications, Applied microbiology and biotechnology, 79 (2008) 901-913. [17] J.C. Biffinger, J.N. Byrd, B.L. Dudley, B.R. Ringeisen, Oxygen exposure promotes fuel diversity for Shewanella oneidensis microbial fuel cells, Biosensors & bioelectronics, 23 (2008) 820-826. [18] J.C. Biffinger, R. Ray, B.J. Little, L.A. Fitzgerald, M. Ribbens, S.E. Finkel, B.R. Ringeisen, Simultaneous Analysis of Physiological and Electrical Output Changes in an Operating Microbial Fuel Cell With Shewanella oneidensis, Biotechnology and bioengineering, 103 (2009) 524-531. [19] M. Rosenbaum, M.A. Cotta, L.T. Angenent, Aerated Shewanella oneidensis in Continuously Fed Bioelectrochemical Systems for Power and Hydrogen Production, Biotechnology and bioengineering, 105 (2010) 880-888. [20] C. Wu, Y.-Y. Cheng, B.-B. Li, W.-W. Li, D.-B. Li, H.-Q. Yu, Electron acceptor dependence of electron shuttle secretion and extracellular electron transfer by Shewanella oneidensis MR-1, Bioresource technology, 136 (2013) 711-714. [21] A. Kouzuma, T. Kasai, A. Hirose, K. Watanabe, Catabolic and regulatory systems in Shewanella oneidensis MR-1 involved in electricity generation in microbial fuel cells, Front Microbiol, 6 (2015). [22] S. Bencharit, M.J. Ward, Chemotactic responses to metals and anaerobic electron acceptors in Shewanella oneidensis MR-1, Journal Of Bacteriology, 187 (2005) 5049-5053. [23] D. Wu, D.F. Xing, L. Lu, M. Wei, B.F. Liu, N.Q. Ren, Ferric iron enhances electricity generation by Shewanella oneidensis MR-1 in MFCs, Bioresource technology, 135 (2013) 630-634. [24] D. Leys, A.S. Tsapin, K.H. Nealson, T.E. Meyer, M.A. Cusanovich, J.J. Van Beeumen, Structure and mechanism of the flavocytochrome c fumarate reductase of Shewanella putrefaciens MR-1, Nat Struct Biol, 6 (1999) 1113-1117. [25] Y. Yang, Y. Xiang, M. Xu, From red to green: the propidium iodide-permeable membrane of Shewanella decolorationis S12 is repairable, Scientific reports, 5 (2015). [26] N.P. Shroff, S. Xu, J. Acevedo, M.Y. El-Naggar, S.E. Finkel, A Microenvironment for Shewanella oneidensis MR-1 Exists within Graphite Felt Electrodes, Journal of The Electrochemical Society, 164 (2017) H3103-H3108. [27] Y.S. Xu, T. Zheng, X.Y. Yong, D.D. Zhai, R.W. Si, B. Li, Y.Y. Yu, Y.C. Yong, Trace heavy metal ions promoted extracellular electron transfer and power generation by Shewanella in microbial fuel cells, Bioresource technology, 211 (2016) 542-547. [28] Y.Y. Yu, C.X. Guo, Y.C. Yong, C.M. Li, H. Song, Nitrogen doped carbon nanoparticles enhanced extracellular electron transfer for high-performance microbial fuel cells anode, Chemosphere, 140 (2015) 26-33. [29] P. Zhang, Y. Qu, J. Liu, Y. Feng, A new design of activated carbon membrane air-cathode for wastewater treatment and energy recovery, RSC Adv., 6 (2016) 4587-4592. [30] B.R. Ringeisen, E. Henderson, P.K. Wu, J. Pietron, R. Ray, B. Little, J.C. Biffinger, J.M. Jones-Meehan, High power density from a miniature microbial fuel cell using Shewanella oneidensis DSP10, Environmental science & technology, 40 (2006) 2629-2634. [31] S. Luo, W. Guo, K.H. Nealson, X. Feng, Z. He, 13C pathway analysis for the role of formate in

15

electricity generation by Shewanella Oneidensis MR-1 using lactate in microbial fuel cells, Scientific reports, 6 (2016). [32] X.P. Zhu, M.D. Yates, M.C. Hatzell, H.A. Rao, P.E. Saikaly, B.E. Logan, Microbial Community Composition Is Unaffected by Anode Potential, Environmental science & technology, 48 (2014) 1352-1358. [33] T. Liu, Y.Y. Yu, X.P. Deng, C.K. Ng, B. Cao, J.Y. Wang, S.A. Rice, S. Kjelleberg, H. Song, Enhanced Shewanella biofilm promotes bioelectricity generation, Biotechnology and bioengineering, (2015). [34] C. Lau, J. Roy, P. Atanassov, P. Chellamuthu, K. Nealson, Role of riboflavin in electrical communication and signaling by Shewanella oneidensis, Abstr Pap Am Chem S, 241 (2011). [35] Y.-C. Yong, Z. Cai, Y.-Y. Yu, P. Chen, R. Jiang, B. Cao, J.-Z. Sun, J.-Y. Wang, H. Song, Increase of riboflavin biosynthesis underlies enhancement of extracellular electron transfer of Shewanella in alkaline microbial fuel cells, Bioresource technology, 130 (2013) 763-768.

16

Figure Captions Figure 1. The schematic picture of the two-chamber MFC used in this experiment. Figure 2. The time course profiles of voltage outputs with and without fumarate addition at start-up period (a) and stable period (b); Arrow indicates the anode solution refresh. Figure 3. Power density and polarization curves of MFCs with and without fumarate addition; CV curves of the anodes with and without fumarate. Solid and hollow icons represent the polarization and power density curves, respectively. Figure 4. SEM images of carbon cloth anode cultured without (a) and with fumarate (b). Images are obtained with magnification of 1000. Figure 5. OD600 value (a) and riboflavin concentration (b) during the cycle in the start-up period; error bar represents the standard error of triplicate samples. Figure 6. Three kinds of electron transfer pathway existed in the system with fumarate. DET, direct electron transfer pathway; MET, mediated electron transfer pathway; FET, the electron transfer pathway with fumarate as electron acceptor.

17

Figure 1

18

Figure 2

(a)

0.4 Control With Fumarate

Voltage (V)

0.3

0.2

0.1

0.0

0

5

10

15

20

25

30

35

40

Time (day)

(b)

0.4

Voltage (V)

Control With Fumarate 0.3

0.2

0.1

0.0

0

5

10

15

20

Time (day)

19

25

30

Figure 3

(a) 0.8

2

Control With Fumarate

100

Voltage (V)

0.6 80

0.4

60 40

0.2 20

0.0

0

50

100

150

200

250

0 300

2

Current density (mA/m ) (b)

0.4 Control With Fumarate

Current (mA)

0.2

0.0

-0.2

-0.4

-0.6

-0.4

-0.2

0.0

0.2

Potential (V vs Ag/AgCl electrode)

20

Power density (mW/m )

120

Figure 4

21

Figure 5

(a)

0.30 Control With Fumarate

OD 600

0.25 0.20 0.15 0.10 0.05 0.00

0

5

10

15

20

Time (day)

Riboflavin (mg/L)

(b)

0.3

0.2

0.1 Control With Fumarate 0.0

0

5

10

Time(day)

22

15

20

Figure 6



Fumarate addition enhanced the electricity generation of Shewanella cells.



Fumarate could promote the bacteria propagation and the anode biofilm

formation. 

The riboflavin secretion of Shewanella cells was impaired with fumarate addition.

23