Yeast surface display of dehydrogenases in microbial fuel-cells

    Yeast surface display of dehydrogenases in microbial fuel-cells Idan Gal, Orr Schlesinger, Liron Amir, Lital Alfonta PII: DOI: Reference:

S1567-5394(16)30090-1 doi: 10.1016/j.bioelechem.2016.07.006 BIOJEC 6964

To appear in:

Bioelectrochemistry

Received date: Revised date: Accepted date:

15 March 2016 17 July 2016 17 July 2016

Please cite this article as: Idan Gal, Orr Schlesinger, Liron Amir, Lital Alfonta, Yeast surface display of dehydrogenases in microbial fuel-cells, Bioelectrochemistry (2016), doi: 10.1016/j.bioelechem.2016.07.006

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.

ACCEPTED MANUSCRIPT Research Article Yeast surface display of dehydrogenases in microbial fuel-cells Idan Gal,a,1 Orr Schlesingera, Liron Amira and Lital Alfontaa,* a

T

Department of Life Sciences and Ilse Katz Institute for Nanoscale Science and Technology, Ben-Gurion University of the Negev, P.O. Box 653, Beer-Sheva 84105, Israel

IP

Abstract

1.

NU

SC R

Two dehydrogenases, cellobiose dehydrogenase from Corynascus thermophiles and pyranose dehydrogenase from Agaricus meleagris, were displayed for the first time on the surface of Saccharomyces cerevisiae using the yeast surface display system. Surface displayed dehydrogenases were used in a microbial fuel cell and generated high power outputs. Surface displayed cellobiose dehydrogenase has demonstrated a midpoint potential of -82 mV (vs. Ag/AgCl) at pH=6.5 and was used in a mediator-less anode compartment of a microbial fuel cell producing a power output of 3.3 μW cm-2 using lactose as fuel. Surface-displayed pyranose dehydrogenase was used in a microbial fuel cell and generated high power outputs using different substrates, the highest power output that was achieved was 3.9 μW cm-2 using D-xylose. These results demonstrate that surface displayed cellobiose dehydrogenase and pyranose dehydrogenase may successfully be used in microbial bioelectrochemical systems.

Introduction

the enzyme and the electrode. Second, GOx is limited to the oxidation of only one possible anomeric form of glucose (β). That means that it may utilize only up to 64% of the total soluble glucose content in aqueous solutions [15]. Third, GOx oxidizes glucose only at its C-1 position, releasing only two electrons per one molecule of glucose, resulting in a low Coulombic efficiency. Forth, GOx produces hydrogen peroxide in the presence of oxygen, allowing oxygen to compete continuously with mediators in the solution and eventually leading to a decrease in coulombic efficiency. Cellobiose dehydrogenase (CDH, EC 1.1.99.18) is an extracellular sugar oxidoreductase produced by various wood degrading fungi [16]. CDH is a monomeric N-glycosylated peptide which has a distinguishable structure as the only known extracellular flavocytochrome. CDH is composed of two domains, the small cytochrome domain (CYT), carrying a heme b redox cofactor, which is located in the N-terminus and connected by a long flexible linker to the flavodehydrogenase domain (DH), where a flavin adenine dinucleotide (FAD) is used as the redox cofactor [17]. CDH can oxidise the β anomers of mono-, di- and oligosaccharides such as cellobiose, lactose, maltose and glucose on their C-1 position, with a preference towards cellobiose and lactose [18,19]. In the saccharide oxidative reaction, two electrons are obtained in the DH domain and can be transferred to a two electron acceptor only by the DH domain or to a one electron acceptor by either the DH or the CYT domain after an internal electron transfer (IET) from the DH domain to the CYT domain [20]. In addition, CDH can transfer the electrons directly to an anode through DET using the CYT domain, which acts as a built-in mediator and eliminates the need for soluble redox mediator in bioelectrochemical systems [21,22]. CDH from Corynascus thermophiles (CtCDH) is a very attractive dehydrogenase, which can operate under diverse pH values, with pH optima of the DH domain between 5.0-6.0 and pH optima of CYT domain at 7.5 with a broad activity peak over the range of pH 6.0-9.0. The latter indicating that IET between DH and CYT domains is considerably favored at neutral pH, which is an important trait for future applications such as implantable devices [20,23]. Pyranose dehydrogenase (PDH, EC 1.1.99.29) is an extracellular sugar oxidoreductase that can be found in "litter-decomposing" fungi which grow mostly on lignocellulose-rich forest litter [24]. The native

AC

CE P

TE

D

MA

Biofuel cells are electrochemical devices that spontaneously convert chemical energy into electrical energy using biochemical pathways and redox enzymes as the biocatalysts. Biofuel cells can be classified into enzymatic fuel cells (EFCs) and microbial fuel cells (MFCs) that use either isolated redox enzymes or whole living microorganisms as their respective biocatalysts [1,2]. EFCs can only partially oxidize the fuel and have limited lifetime owing to relatively short term enzyme stability, while MFCs are typically stable for longer periods and can catalyze full oxidation of different fuels by using the microorganism’s entire metabolism for power production [3,4]. Nevertheless, the major factor influencing power production in MFCs relies on the biocatalyst ability to communicate efficiently with the electrode [5]. Hence, it is important to optimize the microorganism itself, using genetic engineering tools, for fuel and energy production in order to improve electrobiocatalysis in MFCs [6]. Electrons may flow between the biocatalyst active site and the electrode either by mediated electron transfer (MET) using small redox active molecules, redox mediators, or by direct electron transfer (DET), where the biocatalyst is able to communicate directly with the electrode. DET possesses some important advantages over MET. First, mediators are often toxic and their use leads to potential losses arising from the potential difference between the redox potential of the enzyme active site and that of the mediator. Second, a mediator-less system is less prone to interfering reactions. Third, DET allows the possibility of modulating the desired properties of the system by protein modification using genetic and chemical engineering tools [7,8]. In our recent studies, an anode containing glucose oxidase (GOx) from aspergillus niger displayed on the surface Saccharomyces cerevisiae (S. cerevisiae) cell surface, using D-glucose as a fuel, was described [9–11]. The surface-displayed GOx allows bypassing of the cellular membrane [12], while offering the possibility of in situ regeneration of the active enzyme by the living organism [13]. Although GOx has been widely used in biofuel cells, it still has a number of caveats in bio-fuel cells (BFCs) applications. First, GOx belongs to a group of redox enzymes that harbor a catalytic center buried deep inside the protein matrix [14]. This prevents DET between 1

Current address: Department of Biotechnology and Molecular Microbiology, Tel-Aviv University, Tel-Aviv, Israel *Corresponding author E-mail address: [email protected] (L. Alfonta)

1

ACCEPTED MANUSCRIPT Research Article for 2.5 min. samples were precipitated right away at 40C for 20 minutes and the supernatant was filtered using a 0.45 μm PVDF filter (MillexHV) followed by 4 centrifuge cycles with a 50 kDa cutoff VIVASPIN 20 filter (Sartorius stedim biotech GmbH, Goettingen, Germany) at 40C for 15-20 minutes. After each cycle samples were re-suspended in acetate buffer 0.1M pH 5.5. Afterwards, 3.5 kDa cutoff cellulose tubular dialysis membranes (Membrane Filtration Products, Seguin, TX, USA) were filled with samples and placed in a stirred 1L beaker, containing acetate buffer 0.1M pH 5.5, at 40C for 24-48hr. The beaker buffer was replaced, and incubated again for 24-48hr, after which samples were collected for CtCDH activity assay. This procedure was based on previously reported protocols [31–34]. CtCDH activity assay CtCDH activity was determined using a modified procedure by spectrophotometrically following the lactose-dependent reduction of 2,6 dichlorophenol-indophenol (DCIP, Sigma-Aldrich) to DCIPH2 at 520 nm (ε520= 6.8 mM-1 cm-1) for 1hr at 37°C [20,23,35]. 100μL of each sample were transferred to a flat-bottom 96-well plate (Nunclon Delta Surface, Thermo scientific, Denmark) in triplicates. To each sample, 100μL of 0.6 mM DCIP and 60 mM Lactose (Sigma-Aldrich) in 0.1M acetate buffer (pH 5.0) were added. Absorbance measurements at 520nm were taken every 1min using a microplate reader (BioTek instruments, Winoosky, VT, USA), and a plot of the absorbance as a function of time was plotted. The linear range of the slope was extracted to calculate CtCDH activity using Beer-Lambert’s law. One unit of enzyme activity is defined as the amount of enzyme reducing 1 μmol of DCIP per min under the above reaction conditions. Solutions of DCIP were prepared by dissolving 1.6 mg in 2.0 mL of DDW containing 10% v/v ethanol. Absorbance varied only about 3% between pH=3.0 and 8.0 [23]. Error bars are based on at least three measurements for each sample. CtCDH’s CYT domain activity assay CtCDH activity was determined using a modified procedure by spectrophotometrically following the lactose-dependent reduction of Cyt c at 550nm (ε550= 19.6 mM-1 cm-1) for 5hr at 37°C [20,23]. S. cerevisiae expressing CtCDH on their surface were grown to an OD600=1.0, centrifuged and re-suspended in 0.1M phosphate buffer (PB, pH 6.5). 100μL of each sample were transferred to a flat-bottom 96-well plate (Nunclon Delta Surface, Thermo scientific, Denmark) in triplicates. To each sample, 100μL of 180 μM Cyt c (Sigma-Aldrich) and 60 mM lactose in 0.1M phosphate buffer (PB, pH 6.5) were added. An absorbance measurement at 550nm was taken every min using a microplate reader (BioTek instruments, Winoosky, VT, USA), and a plot of the absorbance as a function of time was plotted. The linear range of the slope was extracted to calculate CtCDH activity using Beer-Lambert’s law. One unit of enzyme activity was defined as the amount of enzyme reducing 1 μmol of Cyt c per min under the above reaction conditions. Error bars are based on at least three measurements for each sample. AmPDH activity assay AmPDH activity was determined using a modified procedure by spectrophotometrically following the D-glucose-dependent reduction of the ferricenium ion (Fc+) to ferrocene (Fc) at 300nm (ε300= 4.3 mM-1 cm-1) for 6hr at 37°C [27,36,37]. S. cerevisiae expressing PDH on their surface were grown to an OD600=1.0, centrifuged and resuspended in 0.1M PB, pH 6.5. 100μL of each sample were transferred to a flat-bottom UV 96-well plate in triplicates (Grenier bio-one, Frickenhausen, Germany). To each sample, 100μL of 0.4 mM ferricemium hexafluorophosphate (Fc+PF6-, Sigma-Aldrich, Saint Louis, USA) and 200mM D-Glucose in PB were added. An absorbance measurement at 300nm was taken every min in a microplate reader, and a graph of the absorbance as a function of time was plotted. The

CE P

TE

D

MA

NU

SC R

IP

T

enzyme is a monomeric glycosylated polypeptide with one covalently bound FAD molecule acting as the redox cofactor. PDH has a broad substrate tolerance due to its excellent ability to oxidize different aldopyranoses including mono-, di- and oligosaccharides [15]. PDH can oxidize its substrate at the C-1, C-2 or C-3, as well as a double oxidation at C-1,2, C-2,3 and C-3,4 by acting on the CH–OH groups of a non-phosphorylated sugars, converting them to the corresponding aldonolactones (C-1), 2/3-dehydro sugars or di-dehydro sugars [15,25]. PDH ability for double oxidation increases current densities by gaining four electrons per one molecule of substrate [15] thus has a potential to increase MFCs power outputs. In addition, PDH lacks an anomeric specificity which can increase the number of oxidized substrate molecules in a solution, improving oxidation reaction efficiency. The preferred saccharides of PDH from Agaricus meleagris (AmPDH), with the highest catalytic efficiencies, are D-glucose (double oxidation), Dgalactose (C2 oxidation), L-arabinose (C2 oxidation), D-xylose (double oxidation) and cellobiose [26–28]. AmPDH has a broad pH stability between 4.0-10.0, with maximum stability at pH 7.0 [26]. It is an important trait for applications such as biofuel cells proposed for power generation in remote locations. Herein, we wanted to characterize the activity of two novel surface-displayed dehydrogenases in MFCs. CtCDH and AmPDH displayed on the surface of S. cerevisiae, using the a-agglutinin yeast surface display (YSD) system [29,30]. Schematic 1 describes the two different surface displayed enzymes used in our system. S. cerevisiae displaying CtCDH were used in a mediatorless anode compartment of an MFC. S. cerevisiae displaying AmPDH were used in an MFC, fed with different sugars as fuels while methylene blue (MB) was used as the redox mediator. The ability of AmPDH to oxidize various fuels while still performing double oxidation, enable an improved MFC performance while rendering the MFC more suitable for wastewater treatment applications. Schematic 1. Schematic presentation of CtCDH and AmPDH MFCs anodes. Left: mediator-less MFC anode containing S. cerevisiae surface displaying CtCDH. Right: MFC anode containing S. cerevisiae surface displaying AmPDH.

Materials and Methods

AC

2.

Materials and Reagents CtCDH and AmPDH genes originated from C. thermophiles and A. meleagris, respectively. Both genes were cloned into YSD vector and transformed to S. cerevisiae (Strain EBY100) yielding CtCDH modified S. cerevisiae and AmPDH modified S. cerevisiae strains. 2,6 dichlorophenol-indophenol (DCIP), Cyt c, ferricemium hexafluorophosphate (Fc+PF6-), 2,2'-azino-bis(3-ethylbenzothiazoline6-sulfonic acid) (ABTS), methylene blue (MB), D-xylose, L-arabinose and lactose were purchased from sigma-aldrich (Rehovot, Israel). Dglucose and D-cellobiose were purchased from Chem-Impex international Inc (Wood dale, USA) and D-Galactose was purchased from Acros Organics (Geel, Belgium). Dionized water (DW) were purified using a milli-Q water system (18.2 MΩ cm, Millipore, Bedford, MA) to result in DDW. Biochemical activity assays CtCDH separation from S. cerevisiae CtCDH transformed S. cerevisiae were induced by standard induction protocols (SI section), concentrated to an OD600=2.0 and resuspended in 0.1M acetate buffer, pH 5.5. 20.0mL of S. cerevisiae expressing CtCDH were treated with 200 mM β-Mercaptoethanol (Biorad) and protease inhibitor set IV-for yeast (Calbiochem) at 400C

2

ACCEPTED MANUSCRIPT Research Article unmodified EBY100 S. cerevisiae and a buffer containing only lactose with no S. cerevisiae. multi-walled carbon nanotubes (MWCNTs) adsorbed on graphite plate anodes After the graphite plate was sanded, cleaned with ethanol and DDW and left to dry, both sides were coated with 100µL of 1.7 mg/mL MWCNT (sigma Aldrich, O.D. х L ,6-9 nm х 5 μm) in ethanol (Carlo Erba Reagents, Val-de-Reuil, France) and were left to dry. After drying, the graphite plates were used as anodes in the fuel cell. Air cathode MFCs The MFC used was a custom-made, single-chamber air cathode fuel cell made of polymethyl 2-methylpropenoate (PERSPEX). A graphite plate (10 cm2), sanded and cleaned with ethanol and DDW, served as the anode in a single cubical chamber (4 cm long by 2.7 cm diameter; cell volume of 25mL). The custom-made cathode, consisted of a carbon cloth covered with a carbon black layer and four layers of polytetrafluoroethylene (Teflon) on the air side and a Pt/C catalyst layer on the solution side [38,39]. The anode was wired with a stainless-steel wire (0.1 cm in diameter), and both the anode and cathode were connected to the outer circuit using copper wires. Generally, the anode compartment consisted of yeast surface displayed AmPDH, 0.1 M variable sugar substrates and 1 mM MB (sigma-aldrich, MO, USA) in 0.1 M PB (pH 6.5). The anode negative controls used were unmodified EBY100 S. cerevisiae (WT) and a buffer containing MB and variable types of sugars with no S. cerevisiae. All yeast cultures were used at an OD600=2.0. The anode compartment was kept under anaerobic atmosphere.

T

linear range of the slope was extracted to calculate AmPDH activity using Beer-Lambert’s law. One unit of enzyme activity is defined as the amount of enzyme reducing 1 μmol of Fc+ per min under the above mentioned reaction conditions. Solutions of Fc+PF6- were prepared by dissolving 3.3 mg in 20 mL of 5 mM HCl. Error bars are based on at least three measurements for each sample.

IP

Electrochemical measurements

D

MA

NU

SC R

Electrode preparation CtCDH-displaying S. cerevisiae-modified electrodes serving as working electrodes were made of a glassy carbon (GCE) (3mm in diameter, ALS, Tokyo, Japan). The surface of the GCE was prepared first by polishing with 0.05μm polishing alumina slurry (ALS), then by thoroughly rinsing with DDW, sonicating for 5-10 minutes and drying with Argon. Electrode preparation for CV measurements CtCDH-displaying S. cerevisiae were induced by standard protocols (SI), concentrated to an OD600 of 2.0 and re-suspended in 0.1 M PB, pH 6.5. Working electrode was prepared by placing an aliquot of 5.0µL of CtCDH-displaying S. cerevisiae solution on the electrode surface. The solution was air-dried at room temperature for 1hr and then incubated overnight, 40C. Prior to use, the electrodes were rinsed with DDW to remove weakly adsorbed S. cerevisiae. Cyclic Voltammetry (CV) CV measurements were performed using a PalmSens potentiostat (Palm Instruments, Houten, Netherlands) in a three electrode standard electrochemical cell, using GCE as the working electrode, a graphite rod as the counter electrode, and Ag|AgCl|KCl 3M reference electrode was employed (ALS, Tokyo, Japan). Measurements were conducted in a 8.0 mL cell, 0.1M PB, pH 6.5. Scan rate was 10100 mV s-1, at a potential range of -0.3–0.3V. All measurements were carried out under an ambient temperature and anaerobic conditions – argon was purged into the electrochemical cell for 20 min prior to measurements and was continuously purged above the solution during measurements, for all samples. Background currents of electrode capacitance were subtracted by measuring capacitance of a bare GCE. Every experiment was conducted at least three times.

TE

3.

Results and Discussion

AC

CE P

CtCDH-displaying S. cerevisiae Surface dispalyed CtCDH oxidoreductase demonstrated high expression levels as shown by FACS results shown in Fig. S1. Next, biochemical activity assay was conducted to verify that surfacedisplayed CtCDH is active. In order to do so, separation of CtCDH from the S. cerevisiae surface was performed, to detect CtCDH activity exclusively, avoiding any background reactions from S. cerevisiae metabolism. For the separation, we reduced the 2 disulfide bonds, connecting Aga2p and Aga1p subunits which are part of the YSD system, using β-mercaptoethanol (β-ME). Following enzyme purification, an activity assay was conducted, using DCIP as an electron acceptor.

MFCs characterization

Voltage and current generated by the biofuel cells were measured by a 2700 multimeter (TES, Taipei, Taiwan) using a resistance decade box (Lutron Electronic Enterprise, Taipei, Taiwan). Measurements were carried out at an ambient temperature. Mediator-less U-Shaped MFCs The biofuel cell was custom-made from two curved glass compartments (cathode and anode) with clamping ledges, O-ring and a Nafion® membrane (1.17mm thick, Alfa Aesar, USA) between them to give an overall ‘U’-shape. Each compartment had a volume of 8-10 mL, and was loaded with 5mL of electrolyte, containing all the cell’s components. Graphite plates (4 cm2, Fuel Cell Store, CA, USA) were cut into rods, then sanded and cleaned with ethanol followed by DDW. After drying, the graphite rods were used both as anodes and cathodes in the fuel cell. The anode half-cells were kept under a deoxygenated atmosphere and contained CtCDH-expressing S. cerevisiae at an OD600= 2.9 supplemented with 0.1M Lactose in 0.1M PB, pH 6.5. The cathode half-cell contained 1mg/mL commercial laccase from Trametes Versicolor (≥0.5 U/mg, Sigma-Aldrich, MO, USA) suspended in 0.1M acetate buffer, pH 5.0, supplemented with 5mM 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS, SigmaAldrich) in an oxygen saturated environment. The anode half-cell’s negative controls used were induced GOx-expressing S. cerevisiae or

Figure 1. CtCDH activity assay with lactose.

As can be seen from both Fig.1 and Fig. S2, CtCDH oxidoreductase shows high activity levels of 13.65 mU/mL, ca. 30-fold increase compared to WT yeast. CtCDH activity was also tested under a range of pH values and showed an optimum of activity at pH 5.0 (Fig. S3). This result is in a good agreement with previously reported values [20]. Next, biochemical activity assay was conducted to verify that the CYT domain, is active and could communicate with the electrode through DET. This assay was based on a one electron acceptor, the protein Cyt c, which is exclusively reduced at the CYT domain, thus reflects the DET activity of intact CtCDH and the IET between DH and CYT domains [20]. Figure 2. CtCDH CYT domain activity with lactose.

As can be seen from Fig. 2, CtCDH-displaying S. cerevisiae have shown activity levels of 0.36 mU/mL, ca. 12 and 36-fold higher than GOx-displaying S. cerevisiae and WT yeast, respectively, serving as negative controls that are not known for their DET capabilities. In

3

ACCEPTED MANUSCRIPT Research Article addition, CYT domain activity was tested under a range of pH values, with a pH optimum at pH 6.5 (Fig. S4). These results are in partial agreement with literature showing a pH optimum at pH 7.5 and 80% of the pH optimum activity at pH 6.5 [20].

power output than for GOx-displaying S. cerevisiae (Fig. 4A, curve b) and unmodified S. cerevisiae (Fig. 4A, curve c), generating 0.28 and 0.27 μW cm-2, respectively. In addition, CtCDH-displaying S. cerevisiae reached an open circuit voltage (OCV) value of 860 mV (curve a), ca. 2 fold higher OCV than GOx-dispalying S. cerevisiae (curve c) and unmodified S. cerevisiae (curve d), MFCs that reached only 445 mV and 569 mV, respectively. We attribute the significant differences in OCV values to a poor electrode communication of both control experiments (i.e. GOx-yeast and unmodified yeast) in the absence of a redox mediator. In contrast, CtCDH-displaying S. cerevisiae displayed better power outputs and OCVs, emphasizing its efficient communication with the anode through DET. Even though, higher power outputs have been achieved by CtCDH-displaying S. cerevisiae through DET, they are still lower than we anticipated. This is a drawback of enzyme surface displaying MFCs compared to EFCs. While MFCs use the entire microorganism along with expressed biocatalysts with a micrometer size scale of the microorganism, EFCs use only the enzyme as biocatalyst (nanometer size scale), which can dramatically increase the power output, when using adsorbed MWCNTs that are of the same dimensions of enzymes (nanometer scale). In addition, the polarization curves (Fig. 4b) indicate ohmic losses caused by the concentrated S. cerevisiae solution which are insulating in nature, nafion® membrane and the large distance between anode and cathode. Long term performance of this MFC is shown in Fig. S7. To the best of our knowledge, this is the first attempt to assemble an MFC using CtCDH-displaying S. cerevisiae. Comparison with a study of a mediator-less gold-nanoparticle (AuNP) EFC using CtCDH and myrothecium verrucaria bilirubin oxidase (MvBOx) as anodic and cathodic bioelements, respectively and using lactose as the fuel, generated a power output of 15 μW cm-2 [41]. The power output generated in this study is ca. 5 times lower than reported in the literature however, it is still in a good agreement when taking into account the three-dimensional AuNP electrodes and much larger surface area. Considering this, more work is required to dramatically increase the available interface between surface displaying microorganisms and porous carbonaceous materials [42]. DsCDH has the most efficient DET, among all CDHs, generating a 4.4 times higher specific current density at pH 5.0, compared to CtCDH at physiological pH [23]. Still, between the 6 ascomycetous class II CDHs examined in that paper, CtCDH and DsCDH have the most efficient DET, and even though DsCDH has better DET properties, CtCDH is more attractive due to its optimal activity under physiological pH of 7.4.

D

MA

NU

SC R

IP

T

Electrochemical characterization of adsorbed CtCDH-displaying S. cerevisiae DET of adsorbed CtCDH-displaying S. cerevisiae was electrochemically characterized, using GCE as the working electrode. The CV shown in Fig. 3 curve a, demonstrates clear oxidation and reduction peaks for CtCDH-displaying S. cerevisiae compared to GOxdisplaying S. cerevisiae and unmodified S. cerevisiae (Fig. 3 curves b, c), respectively. The voltammogram shows that surface-displayed CtCDH is able to deliver electrons directly to a bare GCE electrode without any electrode modification or the presence of a mediator. It was demonstrated to be a surface confined DET as can be seen from Fig. S5 where peak currents were plotted against scan rate and exhibited a linear dependence (Fig. S5B). which is an indication of a surface controlled process. To support this claim, SEM image of S. cerevisiae cells adsorbed on a graphite electrode have shown that the cells are in direct contact with the electrode (Fig. S6). CtCDH-displaying S. cerevisiae midpoint potential was about -82 mV (vs. Ag/AgCl), at pH=6.5, which is different than reported in the literature for this specific enzyme [21,23,40]. The difference probably stems from the use of a different working electrode, a different maturation process that the enzyme undergoes when surface displayed on S. cerevisiae, in addition to the process being an average of an entire metabolic process and not just a single enzymatic reaction.

TE

Figure 3. CVs of (a) adsorbed CtCDH-displaying S. cerevisiae; (b) GOx-displaying S. cerevisiae; (c) unmodified S. cerevisiae; (d) bare GCE. Reference electrode: Ag|AgCl, scan rate: 10 mV s-1.

CE P

Figure 4. (A) Power outputs and (B) polarization curves of MFCs containing an anode with adsorbed MWCNT and constructed with (a) CtCDH-displaying S. cerevisiae with lactose; (b) GOx-displaying S. cerevisiae with D-glucose; (c) unmodified S. cerevisiae with lactose; (d) background solution with lactose.

AC

The generated oxidation and reduction current density of 0.11 and 0.06 μA cm-2 (Fig. 3, curve a), respectively, is lower compared to other reported values of oxidation and reduction current densities of about 2.75 and 4.5 μA cm-2, respectively, generated by recombinant CtCDH (rCtCDH) purified from pichia pastoris, and adsorbed on thioglycerol-modified gold electrodes in high enzymatic concentrations [40]. Surface coverage of rCtCDH was reported to be 80 mmolcm-2 while surface coverage in this study could not be determined for certain. It is important to note that in the rCtCDH case it is a purified enzyme that is adsorbed on the gold electrode compared to CtCDHdisplaying S. cerevisiae adsorbed on a GCE electrode in this study. Based on this information, we assumed that the reported surface coverage of rCtCDH free enzyme is ca. two orders of magnitude higher than CtCDH-displaying S. cerevisiae by the comparison of current densities.

AmPDH-displaying S. cerevisiae Displayed on the S. cerevisiae surface, AmPDH oxidoreductase demonstrated high expression levels (Fig. S9). Next, biochemical activity assay was conducted to verify that AmPDH oxidoreductase is active. As can be seen in Fig. 5, AmPDH-displaying S. cerevisiae showed activity of 0.74 mU/mL, ca. 6 times increase than unmodified WT yeast. However, the activity of the reaction mixture was only 3 times lower than AmPDH-displaying S. cerevisiae and twice as high as unmodified S. cerevisiae cells. Through the activity assay experiment, reduction of the electron acceptor (Fc+) is observed by a decrease in the Fc cation absorbance. Hence, it seems like the reaction medium is able to reduce Fc+ somewhat even in the absence of an active enzyme (Fig. 5). Nevertheless, we reason that, AmPDH-displaying S. cerevisiae activity originated from AmPDH activity itself and not from the reaction mixture’s non-specific activity. This can be explained by the fact that the reaction mixture activity is higher than that of unmodified S. cerevisiae, this means that yeast cells compete with the reaction medium reducing abilities and can oxidize the reduced ferrocene (Fc)

MFC performance Two compartments MFC was assembled, comprising mediator-less anode compartment which contained CtCDH-displaying S. cerevisiae and lactose as fuel. The cathode compartment contained commercial laccase from Trametes versicolor and ABTS as the redox mediator. A graphite plate was used as an electrode in each 5.0 mL anode and cathode compartment, separated by a nafion® membrane. On the surface of the anode, multi-walled carbon nanotubes (MWCNT) were adsorbed in order to increase electrode surface area. As can be seen from Fig. 4A curve a, the maximum power output of CtCDHdisplaying S. cerevisiae MFC was 3.3 μW cm-2, ca. 12-fold higher

4

ACCEPTED MANUSCRIPT Research Article to Fc+. This phenomenon was observed by the increase in absorbance at the beginning of every activity assay, by unmodified S. cerevisiae cells, instead of the anticipated decrease only. Hence, when using AmPDH-displaying S. cerevisiae, even though living S. cerevisiae cells are oxidizing Fc to Fc+, AmPDH reduces all Fc+ introduced into the reaction mixture which is a good indication for high activity of the enzyme.

T

source, therefore only a minimal increase in S. cerevisiae population is expected in the MFC for its long term operation [43]. On the other hand, when glucose is used in the MFC, the S. cerevisiae population increases constantly, without any newly expressed AmPDH enzyme on newly budded cells (D-glucose is a YSD system inhibitor). As a consequence, the internal resistance of D-glucose based MFCs (as opposed to D-xylose MFCs) was higher, which lead to lower power densities. L-arabinose activity compared to D-glucose was 1.6 fold lower. In addition, L-arabinose has the highest catalytic efficiency and the lowest KM among tested sugars (table 1). Hence, in spite of undergoing only a two electrons oxidation reaction, L-arabinose MFCs produced significant power densities of 3.2 μW cm-2, 0.1 μW cm-2 higher than with D-glucose and 0.7 μW cm-2 lower than with D-xylose, both of which can undergo a four electrons oxidation reaction.

IP

Figure 5. AmPDH activity with D-glucose.

SC R

Next, AmPDH activity was tested under a range of pH values, using D-glucose as a substrate, with the highest measured activity at pH 6.5 (Fig. S10). In addition, AmPDH activity assay was conducted using different substrates to find out an optimal substrate for surface displayed AmPDH (table 1). Table 1: Activity and kinetic parameters comparison between various AmPDH substrates. D-xylose

Larabinose

Dcellobiose

Dgalactose

mU/mL

0.64

0.24

0.4

0.4

0.32

Vmax (μmol min-1 mg-1) KM (mM)

41.4

39.1

33.5

34.5

43.7

0.82 0.03

1.93 0.17

0.54 0.08

6.82 0.2

1.05 0.05

kcat (s-1)

45.9 0.3 57.5

43.4 0.8 22.9

37.2 0.6 62.1

38.4 0.3 5.6

48.5 0.6 46.2

Ref.

Present study

MA

AmPDH-displaying S. cerevisiae, using D-cellobiose as a fuel, generated 2.2 μW cm-2 even though D-cellobiose can also undergo a 4 electron oxidation by AmPDH. The MFC activity was 1.6 fold lower than with D-glucose and among the tested sugars, it had the lowest catalytic efficiency with the highest KM value (table 1). In addition, AmPDH exhibits anomeric preference to β-D-cellobiose [28], which is another reason for the lower power output. AmPDH-displaying S. cerevisiae, using D-galactose as a fuel, generated 1.4 μW cm-2. D-galactose undergoes only a 2 electron oxidation reaction by AmPDH and the reported catalytic efficiency using D-galactose compared to that with D-glucose is lower with a higher KM value (native AmPDH, table 1). Relative activity with Dgalactose compared to that with D-glucose is 99% according to reports [28], while in this study, its activity is 2-fold lower than with Dglucose. Despite the lower activity with D-galactose’s, it is the second best carbon source for S. cerevisiae and a YSD system inducer. This way the MFC S. cerevisiae population can expand while being constantly induced for in situ PDH expression. Due to all of these factors, we expected a better performance of the MFC, however, AmPDH induction is inconsistent and occurs at 370C compared to MFCs operation at room temperature. Hence, even though the population of S. cerevisiae will grow inside the MFC, this does not guarantee more biocatalysts. Under these conditions, not only did the number of AmPDH enzymes not increase dramatically, but the impact of MFC population growth increased the internal resistance and constituted a drawback for the MFC. This phenomenon can be seen clearly in the polarization curve, with a steep drop in voltage (Fig. 6B, curve e). The long term performance of this MFC with different fuels is shown in figure S11. The highest power output of unmodified S. cerevisiae was 0.8 μW cm-2 (Fig.6A, curve g) and was achieved using D-xylose as a fuel. In addition, using all tested sugars, power output of unmodified S. cerevisiae dropped very fast, resulting in low power outputs (data not shown). The higher power densities presented in the MFCs results are attributed mainly to AmPDH’s unique ability for a 4 electron oxidation process per one molecule of potential fuel (D-glucose, D-xylose, Dcellobiose or maltose using AmPDH) compared to a 2 electron oxidation process. Higher current densities achieved by AmPDH modified S. cerevisiae are crucial to enhancing MFC power densities. Moreover, S. cerevisiae-displaying-PDH ability to catalyze the oxidation of various sugars while still producing high power outputs, as can be seen in this research, could be of an advantage in wastewater

[26]

D

[26] [26]

TE

kcat/KM (mM-1 s-1)

[26]

NU

Dglucose

Substrate

Figure 6. (A) Power outputs and (B) polarization curves of MFCs containing AmPDH-displaying S. cerevisiae with (a) D-xylose; (b) L-arabinose; (c) D-glucose; (d) D-cellobiose and (e) D-galactose; (f) unmodified S. cerevisiae with D-xylose; (g) background solution with D-xylose.

MFC performance

AC

CE P

Single compartment air cathode MFCs were assembled comprising of an anode which contained AmPDH-displaying S. cerevisiae, different sugars as a fuel and MB as a redox mediator. A graphite plate was used in the anode compartment while a modified carbon cloth was used as the air cathode. As can be seen from Fig. 6A and table 2, AmPDH-displaying S. cerevisiae generated high power outputs using various fuels. Table 2: Power outputs of AmPDH-displaying S. cerevisiae MFCs using various fuels. Substrate power output (μW cm-2)

D-glucose

3.1

D-xylose

L-arabinose

D-cellobiose

D-galactose

3.9

3.2

2.2

1.4

Power densities using D-xylose, D-glucose and L-arabinose were the highest performing fuel cells, showing ca. 4-5-fold increase compared to unmodified S. cerevisiae. Among these sugars, D-xylose and D-glucose are known to undergo 4 electron oxidation process by AmPDH, and indeed resulted in high current and power densities. This is an indication of the direct correlation between the number of electrons extracted from a fuel molecule for a biocatalytic oxidation event to higher current densities achieved. In the literature [26], catalytic parameters (Vmax and kcat) of native AmPDH using D-glucose and D-xylose as substrates are similar except that D-glucose’s KM value is 2 times lower than that for D-xylose (table 1), hence its catalytic efficiency is approx. 3-folds higher. In this study, AmPDH’s activity using D-glucose was also approx. 3-fold higher compared to D-xylose, confirming that D-glucose is a better substrate for this enzyme also when surface displayed. Higher power density produced when using D-xylose compared to D-glucose, can be explained by the long periods of MFC characterization. Namely, S. cerevisiae yeast cannot grow on xylose and its oxidation products as a sole carbon

5

ACCEPTED MANUSCRIPT Research Article treatment applications. In addition, despite the fact that only 5 substrates were examined in this study, multiple other substrates can be oxidized efficiently by AmPDH. To the best of our knowledge, this is the first attempt to assemble an MFC using AmPDH-displaying S. cerevisiae. This surface displayed enzyme could reach 3.9 μW cm-2 of power output. It is the highest power density in this study but not the highest generated compared to different membrane-less EFCs reported in the literature. In one report, AmPDH was wired and entrapped in a low-potential Os-complexmodified redox-polymer hydrogel immobilized on a low density graphite electrode and combined with a cathode based on MvBOx to generate a power output of 13 μW cm-2, using glucose as substrate at pH 7.4 [44]. In another report, an AmPDH modified electrode was embedded in an Os-redox polymer hydrogel together with SWCNT at the anode with a Pt black electrode as a cathode and generated 35 μW cm-2, using glucose as a substrate at pH 7.4 [18]. It is important to note that only AmPDH enzyme is adsorbed on the electrodes in the above mentioned studies, compared to surface displayed AmPDH used in this research. In spite of the relative lower power outputs reported here, AmPDH-displaying S. cerevisiae MFC is simple to assemble and reinduction of the biocatalyst using the YSD system enables the fuel cell to regenerate continuously when a decrease in cell performance is observed. In contrast, EFCs that use purified enzymes, demand laborious and expensive purification processes and suffer from a considerable disadvantage of loss in enzyme activity over time.

Appendix A. Supplementary data

4. Conclusions CtCDH was successfully expressed on S. cerevisiae surface and by using cyclic voltammetry showed that it can deliver electrons to a bare GCE without the need for a mediator, using its “built in” mediator, the CYT domain. Surface-displayed CtCDH allowed us to demonstrate a new generation of mediator-less anode compartment MFC, efficiently communicating with the anode through DET. This new approach is a one more step to render MFCs more simple for use and ecofriendly while still generating power. Using adsorbed MWCNTs, maximum power output of 3.3 μW cm-2 with an OCV of 860 mV was achieved by CtCDH-displaying S. cerevisiae mediator-less MFC. AmPDH expression on S. cerevisiae surface was inconsistent and further research is required to achieve reproducibility as well as to increase the copy number of surface displayed AmPDH. Even though, surface-displayed AmPDH succeeded to achieve good power densities (compared to other reported surface displayed enzymes in fuel cells), with the highest in this study: 3.9 μW cm-2 using D-xylose. This power output is attributed to PDH’s unique ability for a four electron oxidation process per one molecule of potential fuel. In addition, catalyzing the oxidation of various fuels, while still producing high power densities, turns AmPDH-displaying S. cerevisiae MFCs an excellent choice for applications of power recovery from waste water streams. Despite these results, we also observed, that one of the most performance-limiting factors is the anode surface coverage. When this will be improved by electrode as well as microorganisms engineering approaches, we believe that these MFCs performance will significantly improve. Still the major advantage of surface displayed enzyme is the relative ease of renewability of the system, compared to purified enzymes and the fact that the microorganisms take part in the metabolic pathway of fuel oxidation, following enzymatic activity as we have shown in our previous studies [9].

[9]

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bioelechem.

[2]

[3]

NU

[4]

J. A. Cracknell, K. A. Vincent, F. A. Armstrong, Enzymes as working or inspirational electrocatalysts for fuel cells and electrolysis, Chem. Rev. 108 (2008) 2439–2461. doi:10.1021/cr0680639. K. Rabaey, W. Verstraete, Microbial fuel cells: novel biotechnology for energy generation., Trends Biotechnol. 23 (2005) 291–298. doi:10.1016/j.tibtech.2005.04.008. S.D. Minteer, B.Y. Liaw, M.J. Cooney, Enzyme-based biofuel cells, Curr. Opin. Biotechnol. 18 (2007) 228–234. doi:10.1016/j.copbio.2007.03.007. F. Davis, S.P.J. Higson, Biofuel cells-Recent advances and applications, Biosens. Bioelectron. 22 (2007) 1224–1235. doi:10.1016/j.bios.2006.04.029. A. Ramanavicius, A. Ramanaviciene, Hemoproteins in design of biofuel cells, Fuel Cells. 9 (2009) 25–36. doi:10.1002/fuce.200800052. L. Alfonta, Genetically Engineered Microbial Fuel Cells, Electroanalysis. 22 (2010) 822–831. doi:10.1002/elan.200980001. M. Falk, Z. Blum, S. Shleev, Direct electron transfer based enzymatic fuel cells, Electrochim. Acta. 82 (2012) 191–202. doi:10.1016/j.electacta.2011.12.133. L. Gorton, A. Lindgren, T. Larsson, F.D. Munteanu, T. Ruzgas, I. Gazaryan, Direct electron transfer between heme-containing enzymes and electrodes as basis for third generation biosensors, Anal. Chim. Acta. 400 (1999) 91–108. S. Fishilevich, L. Amir, Y. Fridman, A. Aharoni, L. Alfonta, Surface display of redox enzymes in microbial fuel cells., J. Am. Chem. Soc. 131 (2009) 12052–12053. doi:10.1021/ja9042017. K. Bahartan, L. Amir, A. Israel, R.G. Lichtenstein, L. Alfonta, In Situ fuel processing in a microbial fuel cell., ChemSusChem. 5 (2012) 1820–1825. doi:10.1002/cssc.201200063. K. Behartan, J. Gun, S. Sladkevich, P. Prikhodchenko, O. Lev, L. Alfonta, Encapsulation of Yeast Displaying Glucose Oxidase on their Surface in a Graphene Oxide Hydrogel Scaffolding and its Bioactivation, Chem. Commun. (2012) 11957–11959. doi:10.1039/c2cc36959a. H.W. Wang, Q.L. Lang, L. Li, B. Liang, X.J. Tang, L.R. Kong, et al., Yeast Surface Displaying Glucose Oxidase as Whole-Cell Biocatalyst: Construction, Characterization, and Its Electrochemical Glucose Sensing Application, Anal. Chem. 85 (2013) 6107–6112. doi:Doi 10.1021/Ac400979r. A. Szczupak, D. Kol-Kalman, L. Alfonta, A hybrid biocathode: surface display of O2-reducing enzymes for microbial fuel cell applications., Chem. Commun. (Camb). 48 (2012) 49–51. doi:10.1039/c1cc16207a. R. Wilson, Review article Glucose oxidase : an ideal enzyme, Biosens. Bioelectron. 7 (1992) 165–185. M.N. Zafar, F. Tasca, S. Boland, M. Kujawa, I. Patel, C.K. Peterbauer, et al., Wiring of pyranose dehydrogenase with osmium polymers of different redox potentials., Bioelectrochemistry Amsterdam Netherlands. 80 (2010) 38–42. doi:10.1016/j.bioelechem.2010.04.002. G. Henriksson, G. Johansson, G. Pettersson, A critical review of cellobiose dehydrogenases., J. Biotechnol. 78 (2000) 93–113. http://www.ncbi.nlm.nih.gov/pubmed/10725534. R. Ludwig, W. Harreither, F. Tasca, L. Gorton, Cellobiose dehydrogenase: a versatile catalyst for electrochemical applications., Chemphyschem. 11 (2010) 2674–2697. doi:10.1002/cphc.201000216. F. Tasca, L. Gorton, M. Kujawa, I. Patel, W. Harreither, C.K. Peterbauer, et al., Increasing the coulombic efficiency of glucose biofuel cell anodes by combination of redox enzymes., Biosens. Bioelectron. 25 (2010) 1710– 1716. doi:10.1016/j.bios.2009.12.017. R. Ludwig, R. Ortiz, C. Schulz, W. Harreither, C. Sygmund, L. Gorton, Cellobiose dehydrogenase modified electrodes: advances by materials science and biochemical engineering., Anal. Bioanal. Chem. 405 (2013) 3637–3658. doi:10.1007/s00216-012-6627-x. W. Harreither, C. Sygmund, M. Augustin, M. Narciso, M.L. Rabinovich, L. Gorton, et al., Catalytic properties and classification of cellobiose dehydrogenases from ascomycetes., Appl. Environ. Microbiol. 77 (2011) 1804–1815. doi:10.1128/AEM.02052-10. F. Tasca, M.N. Zafar, W. Harreither, G. Nöll, R. Ludwig, L. Gorton, A third generation glucose biosensor based on cellobiose dehydrogenase from Corynascus thermophilus and single-walled carbon nanotubes., Analyst. 136 (2011) 2033–2036. doi:10.1039/c0an00311e. C. Schulz, R. Kittl, R. Ludwig, L. Gorton, Direct Electron Transfer from the FAD Cofactor of Cellobiose Dehydrogenase to Electrodes, ACS Catal. 6 (2016) 555–563. doi:10.1021/acscatal.5b01854. W. Harreither, P. Nicholls, C. Sygmund, L. Gorton, R. Ludwig, Investigation of the pH-Dependent Electron Transfer Mechanism of

SC R

[1]

[5] [6]

MA

[7]

D

[8]

TE

[10]

CE P

[11]

[12]

AC

[13]

[14] [15]

[16]

[17]

[18]

[19]

[20]

[21]

Acknowledgements [22]

Research was supported by an Israel Science Foundation (ISF) Program (232/13), L.A. The helpful assistance of Mr Alon Szczupak and Mr Yehonatan Ravenna is gratefully acknowledged.

[23]

6

IP

References:

T

Keywords: Cellobiose dehydrogenase • Pyranose dehydrogenase • Yeast Surface Display • Microbial Fuel Cells • Enzymatic biofuel Cells

ACCEPTED MANUSCRIPT Research Article

[31] [32]

[33] [34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

T IP SC R

[30]

NU

[29]

MA

[28]

D

[27]

TE

[26]

CE P

[25]

AC

[24]

Ascomycetous Class II Cellobiose Dehydrogenases on Electrodes., Langmuir. (2012) 6714-6723. doi:10.1021/la3005486. R. Kittl, C. Sygmund, P. Halada, J. Volc, C. Divne, D. Haltrich, et al., Molecular cloning of three pyranose dehydrogenase-encoding genes from Agaricus meleagris and analysis of their expression by real-time RT-PCR., Curr. Genet. 53 (2008) 117–127. doi:10.1007/s00294-007-0171-9. C.K. Peterbauer, J. Volc, Pyranose dehydrogenases: biochemical features and perspectives of technological applications., Appl. Microbiol. Biotechnol. 85 (2010) 837–848. doi:10.1007/s00253-009-2226-y. C. Sygmund, R. Kittl, J. Volc, P. Halada, E. Kubátová, D. Haltrich, et al., Characterization of pyranose dehydrogenase from Agaricus meleagris and its application in the C-2 specific conversion of D-galactose., J. Biotechnol. 133 (2008) 334–342. doi:10.1016/j.jbiotec.2007.10.013. J. Volc, E. Kubátová, G. Daniel, Screening of basidiomycete fungi for the quinone-dependent sugar C-2 / C-3 oxidoreductase , pyranose dehydrogenase , and properties of the enzyme from Macrolepiota rhacodes, Arch Microb. 176 (2001) 178–186. doi:10.1007/s002030100308. P. Sedmera, P. Halada, E. Kub, New biotransformations of some reducing sugars to the corresponding ( di ) dehydro ( glycosyl ) aldoses or aldonic acids using fungal pyranose dehydrogenase, J. Mol. Catal. 41 (2006) 32–42. doi:10.1016/j.molcatb.2006.04.004. T.E. Boder, D.K. Wittrup, Yeast surface display for screening combinatorial polypeptide libraries, Nat. Biotechnol. 15 (1997) 553-557. E.T. Boder, K.D. Wittrup, Yeast surface display for directed evolution of protein expression, affinity, and stability., Methods Enzymol. 328 (2000) 430–444. http://www.ncbi.nlm.nih.gov/pubmed/11075358. C. Anfinsen, E. Haber, Studies on the reduction and re-formation of protein disulfide bonds, J. Biol. Chem. 236 (1961) 1361–1363. C.B. Anfinsen, E. Haber, M. Sela, F.H. White, The kinetics of formation of native ribonuclease during oxidation of the reduced polypeptide chain., Proc. Natl. Acad. Sci. U. S. A. 47 (1961) 1309–1314. doi:10.1073/pnas.47.9.1309. L. Kirley, Inactivation of (Na+,K+)-ATPase by beta-Mercaptoethanol, J. Biol. Chem. 265 (1990) 4227–4232. Y. Diao, W. Liu, C.C.L. Wong, X. Wang, K. Lee, P. Cheung, et al., Oxidation-induced intramolecular disulfide bond inactivates mitogenactivated protein kinase kinase 6 by inhibiting ATP binding., Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 20974–20979. doi:10.1073/pnas.1007225107. U. Baminger, B. Nidetzky, K.D. Kulbe, D. Haltrich, A simple assay for measuring cellobiose dehydrogenase activity in the presence of laccase, J. Microbiol. Methods. 35 (1999) 253–259. I. Pisanelli, M. Kujawa, D. Gschnitzer, O. Spadiut, B. Seiboth, C. Peterbauer, Heterologous expression of an Agaricus meleagris pyranose dehydrogenase-encoding gene in Aspergillus spp. and characterization of the recombinant enzyme., Appl. Microbiol. Biotechnol. 86 (2010) 599–606. doi:10.1007/s00253-009-2308-x. M. Kujawa, J. Volc, P. Halada, P. Sedmera, C. Divne, C. Sygmund, et al., Properties of pyranose dehydrogenase purified from the litter-degrading fungus Agaricus xanthoderma., FEBS J. 274 (2007) 879–894. doi:10.1111/j.1742-4658.2007.05634.x. S. Cheng, H. Liu, B.E. Logan, Increased performance of single-chamber microbial fuel cells using an improved cathode structure, Electrochem. Commun. 8 (2006) 489–494. doi:10.1016/j.elecom.2006.01.010. J. Middaugh, S. Cheng, W. Liu, R. Wagner, How to Make Cathodes with a Diffusion Layer for Single-Chamber Microbial Fuel Cells, Avilable http://www.engr.psu.edu/ce/enve/logan/bioenergy/pdf/Cathode_093008.pd f (2006). W. Harreither, A.K.G. Felice, R. Paukner, L. Gorton, R. Ludwig, C. Sygmund, Recombinantly produced cellobiose dehydrogenase from Corynascus thermophilus for glucose biosensors and biofuel cells., Biotechnoly. J. 7 (2012) 1359–1366. doi:10.1002/biot.201200049. X. Wang, M. Falk, R. Ortiz, H. Matsumura, J. Bobacka, R. Ludwig, et al., Mediatorless sugar/oxygen enzymatic fuel cells based on gold nanoparticle-modified electrodes, Biosens. Bioelectron. 31 (2012) 219–225. doi:10.1016/j.bios.2011.10.020. L. Xia, Y. Ravenna, L. Alfonta, Layer-by-layer assembly of a redox enzyme displayed on the surface of elongated bacteria into a hierarchical artificial biofilm based anode, Chem. Commun. Chem. Commun. 51 (2015) 2633–2636. doi:10.1039/c4cc09781e. D.J. Wohlbach, A. Kuo, T.K. Sato, K.M. Potts, A.A. Salamov, K.M. Labutti, et al., Comparative genomics of xylose-fermenting fungi for enhanced biofuel production, J. Am. Chem. Soc. 108 (2011) 13212–13217. doi:10.1073/pnas.1103039108//DCSupplemental.www.pnas.org/cgi/doi/10.1073/pnas.1103039108. M. Shao, M.N. Zafar, M. Falk, R. Ludwig, C. Sygmund, C.K. Peterbauer, et al., Optimization of a membraneless glucose/oxygen enzymatic fuel cell based on a bioanode with high coulombic efficiency and current density., Chemphyschem. 14 (2013) 2260–2269. doi:10.1002/cphc.201

7

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

8

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

9

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

10

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

11

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

12

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

13

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

14

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

15

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Research Article

16

ACCEPTED MANUSCRIPT Research Article

AC

CE P

TE

D

MA

NU

SC R

IP

T

Highlights  Yeast surface displayed CDH communicates with the anode through DET.  Yeast surface displayed CDH possess a middle point potential of -82 mV vs. Ag/AgCl.  Yeast surface displayed CDH yielded 3.3 μW cm-2 in a mediator-less MFC.  Yeast surface displayed PDH yielded 3.9 μW cm-2 in an air-cathode MFC.

17