Citric Acid Cycle Biomimic on a Carbon Electrode

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy

Biosensors and Bioelectronics 24 (2008) 945–950

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

journal homepage: www.elsevier.com/locate/bios

Citric acid cycle biomimic on a carbon electrode

Daria Sokic-Lazic, Shelley D. Minteer ∗

Department of Chemistry, Saint Louis University, 3501 Laclede Avenue, St. Louis, MO 63103, USA article info abstract

Article history: The citric acid cycle is one of the main metabolic pathways living cells utilize to completely oxidize Received 21 March 2008 biofuels to carbon dioxide and water. The overall goal of this research is to mimic the citric acid cycle Received in revised form 8 July 2008 at the carbon surface of an electrode in order to achieve complete oxidation of ethanol at a bioanode Accepted 22 July 2008 to increase biofuel cell energy density. In order to mimic this process, dehydrogenase enzymes (known Available online 3 August 2008 to be the electron or energy producing enzymes of the citric acid cycle) are immobilized in cascades at an electrode surface along with non-energy producing enzymes necessary for the cycle to progress. Six Keywords: enzymatic schemes were investigated each containing an additional dehydrogenase enzyme involved in Citric acid cycle Biofuel cell the complete oxidation of ethanol. An increase in current density is observed along with an increase in Ethanol power density with each additional dehydrogenase immobilized on an electrode, reﬂecting increased Naﬁon® electron production at the bioanode with deeper oxidation of the ethanol biofuel. By mimicking the Metabolic pathways complete citric acid cycle on a carbon electrode, power density was increased 8.71-fold compared to a single enzyme (alcohol dehydrogenase)-based ethanol/air biofuel cell. © 2008 Elsevier B.V. All rights reserved.

1. Introduction reducing the mediator, benzyl viologen. This dehydrogenase catalyzed methanol/dioxygen biofuel cell produced an open circuit Enzymatic biofuel cells are a type of fuel cell where chemical potential of 0.8 V and power density of 0.68 mW/cm2. Akers et al. energy is converted to electrical energy by employing enzymes as studied the two-step oxidation of ethanol to acetate using ADH the electrocatalysts. Most enzymatic biofuel cells in the literature and AldDH in a novel membrane assembly (MEA) configuration employ a single enzyme to do a partial oxidation of a specific fuel (Akers and Minteer, 2003; Akers et al., 2005). Since dehydrogenase (i.e. glucose, lactate, pyruvate, ethanol). In contrast, living systems enzymes are NAD+-dependent, a polymer-based electrocatalyst undergo metabolic processes such as the citric acid cycle through (poly(methylene green)) was used to regenerate NAD+ and to shut- which they are able to completely oxidize biofuels to carbon diox- tle electrons from NADH to the electrode (Thomas et al., 2003). ide and water. The citric acid cycle processes two carbon units from Bioanodes undergoing one-step oxidation were compared to bioan- carbohydrates, amino acids, and fatty acids in the form of acetyl- odes undergoing two-step oxidation. The ethanol/O2 biofuel cell CoA to generate reducing equivalents NADH and FADH2 for ATP undergoing only one-step oxidation with ADH immobilized in production by the electron transport chain. One of the key issues a tetrabutylammonium bromide (TBAB) modified Nafion® mem- in developing effective and efficient enzymatic biofuel cells is the brane has shown open circuit potentials ranging from 0.60 to 0.62 V successful immobilization of multi-enzyme systems that can com- and the average maximum power density of 1.16 ± 0.05 mW/cm2.In pletely oxidize the fuel to carbon dioxide in order to increase the contrast, the open circuit potential and the maximum power den- overall efficiency of the fuel cell. By doing so, the overall perfor- sity for the ethanol/O2 biofuel cell employing a mixture of both ADH mance of the biofuel cell is increased as well. The first biofuel-based and AldDH immobilized in a TBAB modified Nafion® membrane multistep oxidation of alcohols was demonstrated by Palmore et were 0.82 V and 2.04 mW/cm2, respectively. These bioanodes were al. (1998) who employed alcohol dehydrogenase (ADH), aldehyde able to function for more than 30 days and after 30 days of contin- dehydrogenase (AldDH) and formate dehydrogenase (FDH) to com- uous operation, the bioanode showed an 18.1% decrease in power pletely oxidize methanol to carbon dioxide and water. Since these output. Although the results were successful, the system permitted dehydrogenases are dependent upon NAD+ reduction, a fourth only a 33% oxidation of the ethanol fuel which represents a low fuel enzyme diaphorase was introduced to regenerate the NAD+ by utilization and therefore low energy density (Akers et al., 2005). In the last few decades, harvesting energy from renewable resources has become an important focus in order to eliminate our ∗ Corresponding author. Tel.: +1 314 977 3624. dependency on oil and other non-renewable resources necessary E-mail address: [email protected] (S.D. Minteer). as primary power sources (Davis and Higson, 2007; Kjeang et al.,

946 D. Sokic-Lazic, S.D. Minteer / Biosensors and Bioelectronics 24 (2008) 945–950

2006; Winter and Brodd, 2004). In an attempt to do so, biofuel cell 1.1. Modified Nafion® membrane systems have been studied extensively, because of their ability to convert energy derived from biofuels to electrical energy by means Enzymes are extremely efficient biocatalysts when immobilized of the catalytic activity of microorganisms and/or their enzymes. at the electrode surface. When immobilizing an enzyme on an When building a fuel cell, maximizing both energy density (per- electrode surface, it is important to choose a method of attach- centage of total available electrons gained during the oxidation ment that will prevent loss of enzyme’s activity, but not change process times the energy density of the fuel (typically measured the chemical nature or reactive groups in the binding site of that in W h/L)) and power density (the operating voltage times the cur- enzyme. The immobilization technique employed for this biomimic rent per unit electrode area (typically measured in cm2)) is of crucial involves entrapping the respective enzyme(s) in a hydrophobically importance. Microbial biofuel cells reported in the literature have modified Nafion® polymer. Nafion® is a micellar cation exchange been successful in attaining high energy densities due to the capa- polymer with transport channels that deliver substrate or fuel to the bility of completely (100%) oxidizing complex biofuels (i.e. glucose); immobilized enzyme. Unmodified Nafion® is hydrophilic, acidic, however, their power densities have been low due to slow mass and has a micellar pore size of 4 nm, which is too small to effec- transport of the fuel across the cell wall (Karube et al., 1981; Lovely, tively immobilize the enzymes of the citric acid cycle. Quaternary 2006; Matsunaga and Suzuki, 1980; Suzuki and Karube, 1983). On ammonium bromide salt modified Nafion® membranes modify the the other hand, enzymatic biofuel cells have eliminated the fuel micro-environment of the pore to a near neutral pH that resists transport problems that caused low power densities in microbial a decrease in pH, because quaternary ammonium cations have a fuel cells; however, their energy densities have been low due to much higher affinity for the sulfonic acid site than protons due to incomplete oxidation of fuel (Austin, 1967; Nagy, 2003). In order the hydrophobicity of the quaternary ammonium cation (Thomas to build an optimal biofuel cell, both energy density and power et al., 2003). Previous research in our group has shown quaternary density need to be addressed and optimized. This work focuses on ammonium bromide modified Nafion® membranes have enlarged maximizing both the power and energy density through the use of micellar pores and lower proton exchange capacity. This modifi- enzymatic cascades; thereby, maintaining the high power densities cation creates micellar pores that can accommodate the chemical of enzymatic biofuel cells while increasing the energy density of the and biochemical needs of an enzyme while retaining the electrical biofuel cell through the ability to completely oxidize the biofuel properties of Nafion®. ethanol. Most living organisms acquire energy through metabolic 1.2. Poly(methylene green) enzyme pathways such as the citric acid cycle in order to completely oxidize most biological substrates. The citric acid cycle takes place Most of the enzymes employed in this study are NAD+- in the mitochondria of eukaryotic cells where it oxidizes acetyl- dependent dehydrogenase enzymes. During the oxidation of fuel, CoA, a key metabolic junction, to carbon dioxide and water. Several this coenzyme (NAD+) is reduced to NADH. NAD+ is regener- substrates can be fed into the citric acid cycle via this key metabolic ated by the oxidation of NADH, however, NADH oxidation and junction and each substrate will require different enzyme cascades. regeneration on platinum and carbon electrodes occurs at large One of these substrates is glucose which can be oxidized through overpotential and typically passivates bare electrode surfaces another metabolic pathway called glycolysis to pyruvate, which (Blaedel and Jenkins, 1975). Therefore, a polymer-based electrocat- then gets oxidized to acetyl-CoA in presence of pyruvate dehy- alyst, poly(methylene green), was used to regenerate NAD+ and to drogenase. Besides, glucose and pyruvate, lactate can also be fed shuttle electrons from the NADH to the electrode. Cyclic voltam- into the citric acid cycle by incorporating lactate dehydrogenase metric studies of poly(methylene green)-coated glassy carbon and oxidizing it to the key metabolic junction, acetyl-CoA. In addi- electrodes have shown that poly(methylene green) is an electro- tion, ethanol can also be the substrate of choice after incorporating catalyst for NADH (Moore et al., 2004; Zhou et al., 1996). ADH, AldDH, and S-acetyl CoA synthetase and oxidizing ethanol to acetyl CoA, which has been done in the research presented in this paper. Nevertheless, the citric acid cycle uses acetyl-CoA 2. Experimental as the substrate and undergoes eight enzymatic reactions out of which four are electron producing dehydrogenases. The electron 2.1. Materials producing enzymes of the citric acid cycle are NAD-dependent dehydrogenases except for succinate dehydrogenase (SDH), which Methylene green (Sigma), sodium nitrate (Fisher), sodium is a FAD-dependent dehydrogenase. The redox couples NAD+/NADH borate (Fisher), sodium phosphate (Sigma), sodium hydroxide ® and FAD/FADH2 are two electron electrochemical processes and (Sigma), Nafion 1100 EW suspension (Aldrich), ethanol (Sigma) their regeneration can be catalyzed by poly(methylene green). In and tetrabutylammonium bromide (TBAB) (Sigma) were purchased this citric acid cycle biomimic, which uses ethanol as the starting and used as received. In addition, Tris–HCl, NaCl, dl-dithiothreitol, point, expanding from a single enzyme system to a multi-enzyme EDTA, BSA, and cysteine were purchased from Sigma. Russet pota- system allows for additional electrons to be generated which con- toes were purchased from a grocery store. Enzymes employed tributes to the overall current and power density of the biofuel include: alcohol dehydrogenase (ADH) (E.C.1.1.1.1; Sigma), aldehyde cell. dehydrogenase (AldDH) (E.C.1.2.1.5; Roche), S-acetyl-coenzyme A In this research paper, in order to mimic the citric acid cycle synthetase (E.C.6.2.1.1; Sigma), citrate synthase (CS)-(E.C.4.1.3.7; at the carbon surface of an electrode, all the enzymes employed Sigma), aconitase (Aco.) (E.C.4.2.1.3; Sigma), isocitric dehydroge- for this biomimic were immobilized in a quaternary ammonium nase (IDH) (E.C.1.1.1.42; Sigma), ␣-ketoglutarate dehydrogenase bromide salt modified Nafion® membrane layer. Dehydrogenase (KDH) (E.C.1.2.4.2; Sigma), succinyl CoA synthetase (E.C.6.2.1.4; enzymes along with non-electron producing enzymes and cofac- Bioacatalytics), fumarase (E.C.4.2.1.2; Sigma), and malic dehydro- tors were immobilized in cascades for the cycle to progress. The genase (MDH) (E.C.1.1.1.37; Sigma). All enzymes purchased were starting point for this biomimic was ethanol. ADH, AldDH, and stored at −20 ◦C except AldDH, citrate synthase, fumarase, and MDH S-acetyl-CoA synthetase were incorporated along with all the which were stored at +2 to +8 ◦C. The necessary cofactors used enzymes and cofactors of the citric acid cycle. The entire ethanol were all purchased from Sigma and they include ␤-nicotinamide metabolic path can be seen in Fig. 1. adenine dinucleotide hydrate (NAD+), flavin adenine disodium salt Author's personal copy

D. Sokic-Lazic, S.D. Minteer / Biosensors and Bioelectronics 24 (2008) 945–950 947

Fig. 1. Schematic of the complete biofuel cell. Ethanol is oxidized serving as the fuel source at the anode (dark red lettering represents dehydrogenase enzymes, whereas the light red/pink lettering represents other non-energy producing enzymes). Oxygen is reduced to water at the 20% Pt on carbon GDE cathode. Potentiostat is used to measure open circuit potential and linear sweep polarization curves. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.) hydrate (FAD), guanosine 5-diphosphate sodium salt (GDP), coen- suspension. Mixture-cast membranes were cast in a weigh boat zyme A disodium salt (CoA), adenosine 5-triphosphate (ATP), and and allowed to air dry overnight. The next day 18 M water was oxaloacetic acid. All solutions were prepared with 18 M cm (Barn- added to the weigh boat to remove the excess HBr and quaternary stead Nanopure) water. The anode material used was TGPH-060 ammonium bromide salts. The second step was to resuspend and Toray® carbon paper with a typical thickness of 0.17 mm purchased then recast the membrane after quaternary ammonium bromide from E-TEK. The cathode material implemented was an ELAT gas and HBr salts have been extracted from the original membranes. diffusion electrode (GDE) that contains 20% Pt on carbon with a Following the HBr salt extraction, the films were resuspended in proprietary amount of Nafion® ion exchange resin and Teflon wet- ethanol. proofing purchased from E-TEK. 2.2.3. Anode fabrication 2.2. Methods NAD-dependent anodes are fabricated using 1 cm2 pieces of Toray® paper (E-TEK, TGPH-060). The anodes are electropolymer- 2.2.1. Succinate dehydrogenase isolation ized in a methylene green solution. The thin film of poly(methylene Succinate dehydrogenase was isolated following the method green) is formed by performing cyclic voltammetry from −0.3 to described by Laurence Marechal-Drouard (Plant Molecular Biology 1.3 V for 6 scans at a scan rate of 0.05 V/s in a solution of 0.4 mM Reporter 19: 67a–67h, March 2001), which uses high ionic strength methylene green, 10 mM sodium borate, and 0.1 M sodium nitrate. medium to isolate mitochondria from potato tubers and extract After polymerized, the anodes are rinsed and allowed to dry for succinate dehydrogenase. 24 h. After this point, anodes are further modified by casting one of six different enzyme/tetrabutylammonium bromide modified 2.2.2. Membrane preparation Nafion® suspensions directly on the poly(methylene green) coated For the sake of our enzymatic system involving dehydrogenases, carbon electrode. This process ensured no gap between the enzyme tetrabutylammomium bromide (TBAB) was the quaternary ammo- immobilization layer and the electropolymerized poly(methylene nium bromide salt employed in modifying the Nafion® membrane. green). Different enzyme solutions are made in a 1 mg:1 mL ratio The modified Nafion® membranes were formed in a two-step pro- of each enzyme to pH 7.4 phosphate buffer. Out of these different cess. The first step was to cast a suspension of Nafion® with TBAB enzyme solutions, 50 ␮L of each is used. The enzymes are immobi- salt dissolved in suspension. The mixture casting solution was pre- lized in TBAB modified Nafion® membrane. The amount of each pared so the concentration of TBAB salt is in a threefold excess of necessary cofactors used is 1 mg. The final mixture for the first the concentration of sulfonic acid sites in the 5% by wt. Nafion® anode contains 50 ␮L of ADH, 100 ␮L of TBAB modified Nafion®, Author's personal copy

948 D. Sokic-Lazic, S.D. Minteer / Biosensors and Bioelectronics 24 (2008) 945–950

1mgofNAD+, and 550 ␮L of pH 7.4 phosphate buffer. Out of this are required as well to complete each particular step of the path- mixture, 50 ␮L is applied to the anode and the anode is allowed way. These non-electron producing enzymes consist of kinases, to dry overnight. The final mixture for the second anode contains responsible for transferring functional groups from one substrate 50 ␮L of ADH, 50 AldDH, 100 ␮L of TBAB modified Nafion®,1mg to another; isomerases, responsible for transitioning a substrate of NAD+, and 500 ␮L of pH 7.4 phosphate buffer. Out of this mix- (byproduct) to a chemical isomer of that substrate; ligases, respon- ture, 50 ␮L is applied to the anode and the anode is allowed to dry sible for joining together two molecules; and lyases, responsible overnight. The rest of the anodes are prepared in the same way for non-hydrolytically adding or removing groups from substrates. with the amount of buffer decreasing by 50 ␮L with the addition of Therefore, as an example, for a three dehydrogenase-based bioan- each enzyme. The amount for the TBAB modified Nafion® and the ode, besides ADH, AldDH, and isocitrate dehydrogenase (IDH), three amount of necessary cofactors is the same for each anode prepared. other enzymes are incorporated as well: S-acetyl CoA synthetase, In addition to the two previously mentioned enzymes, the third CS, and Aco. anode contains S-acetyl-CoA synthetase (50 ␮L), citrate synthase In this biomimic, as has been mentioned previously, most of the (50 ␮L), aconitase (50 ␮L), isocitric dehydrogenase (IDH) (50 ␮L), enzymes employed are enzymes of the citric acid cycle with the 100 ␮L of TBAB modified Nafion®, oxaloacetate (50 ␮L), 250 ␮Lof exception of ADH, AldDH, and S-acetyl CoA synthetase. It has been pH 7.4 phosphate buffer, and all the necessary cofactors (NAD+, proposed in the literature that enzymes in many metabolic path- NADP+, CoA). The fourth anode contains KDH (50 ␮L) in addition ways, including the citric acid cycle, are physically associated to to everything that is immobilized onto the third anode with the facilitate substrate channeling and overcome diffusive barriers. This exception that the amount of buffer used is 200 ␮L. The fifth anode metabolic organization in cells has been an important and unre- contains all the enzymes and cofactors immobilized onto the fourth solved general problem in cellular biochemistry. Although, there anode plus succinyl-CoA-synthetase, SDH and cofactors GDP and has not been direct evidence supporting the view of the cell as a FAD. The last anode contained immobilized fumarase and MDH in bag of uniformly dispersed enzymes, there has been a large body of addition to the same mixture immobilized onto the fifth anode. experimental data supporting the concept of an intracellular organization of enzymes and channeling of intermediates (Morgunov and Srere, 1998; Ovadi, 1995; Robinson et al., 1987; Robinson and 2.2.4. Electrochemical measurements Srere, 1985; Srere, 1987; Srere et al., 1973; Sumegi and Srere, 1984; All data were collected and analyzed for the test cell with a Velot et al., 1997). In 1985, Paul Srere used the term metabolon to CH instrument 610 potentiostat interfaced to a PC computer. The refer to this type of organization (Srere, 1985). Many of the individ- bioanode is allowed to equilibrate in the 100 mM ethanol fuel solu- ual enzymes of the citric acid cycle are inhibited by their reactants, tion with 1 mM NAD+ for few hours and open circuit potential products, intermediates or even cofactors involved in the cycle. is recorded after it has been stable for 1 h. During this equilibra- However, when all of the enzymes are present, they function as a tion time, surface adsorbed enzyme may leach into solution. Linear unit so that even if a certain enzyme is inhibited by its product like sweep polarization curves were recorded for each electrode by con- citrate synthase as an example, there are other enzymes present trolling potential and using the data obtained, power curves were that will take the product, oxidize it further and by that minimize generated. Each type of bioanode was tested in triplicate on the its inhibition on a particular enzyme. same day and fuel solution replaced daily. The temperature in the The ethanol/air biofuel cell experiments were performed in a lab ranged between 20 and 25 ◦C. standard single compartment (anolyte) air-breathing cathode test cell as described in reference Arechederra et al. (2007). In the anodic 3. Results and discussion compartment, a solution of 100 mM ethanol and 1 mM NAD+ dissolved in phosphate buffer was introduced to the anode, while Every fuel has a theoretical energy density and that energy the commercial platinum-based cathode is allowed to “breathe” density provides a first approximation of the maximum energy den- air in order to get sufficient oxygen. The commercial platinum- sity of a biofuel cell employing that fuel, assuming the electrodes, based cathode is 4.5 cm2 in area. Polarization curves were taken for wiring, and casing have negligible weight and volume and the sys- six different anodes each containing an additional dehydrogenase. tem is employing pure fuel without dilution. Theoretical energy Triplicate measurements were performed and the data are shown density for the complete oxidation of pure, undiluted ethanol is in Table 1. It is important to note that these complex electrodes have 6320 W h/L (Weast and Lide, 1990). The complete oxidation of lower stability. The maximum power density decreased by 55% over ethanol using this biomimic yields a total of 12 electrons as a result a 24-h period. In addition, representative power curves can be seen of six dehydrogenases employed, versus only two electrons for sin- in Fig. 2. Results obtained show an increase in power density with gle step oxidation of ethanol. In terms of theoretical energy density, each additional dehydrogenase immobilized on the electrode sur- a single step oxidation of ethanol yields a theoretical energy density face as expected. The power density ranges from 1.16 × 10−4 W/cm2 of only 1053 W h/L, because only 2 out of the total of 12 electrons produced in the complete oxidation of ethanol (2/12th or 16.67% of the total 6320 W h/L) are produced from the single step oxida- Table 1 tion of ethanol with alcohol dehydrogenase. The theoretical energy Enzyme cascade polarization data for 100 mM ethanol/air biofuel cells employing density is expected to increase to 2107 W h/L when two dehydroge- different multi-enzyme cascades from a single dehydrogenase to the six dehydrogenases required for complete oxidation of ethanol nases (ADH and AldDH, which are responsible for the production of 4 out of the total of 12 electrons produced in the complete oxidation Dehydrogenase Current density (A/cm2) Power density (W/cm2) of ethanol (4/12th or 33% of the total 6320 W h/L)) are employed, cascades then to 3160 W h/L when three dehydrogenases are employed, and ADH 3.77 ± 1.97 × 10−4 1.16 ± 0.88 × 10−4 ± × −4 ± × −4 it is expected to keep increasing following the same trend until ADH, AldDH 5.30 2.50 10 1.30 0.52 10 ADH, AldDH, IDH 8.02 ± 1.98 × 10−4 1.63 ± 0.37 × 10−4 complete oxidation occurs. Therefore, it is extremely important to ADH, AldDH, IDH, KDH 9.84 ± 6.47 × 10−4 2.02 ± 0.83 × 10−4 do complete/deep oxidation of a fuel in order to maintain high ADH, AldDH, IDH, KDH, 1.62 ± 0.35 × 10−3 2.86 ± 0.36 × 10−4 energy density. SDH − − In order to fabricate some of these bioanodes, such as the bioan- ADH, AldDH, IDH, KDH, 3.60 ± 0.23 × 10 3 1.01 ± 0.01 × 10 3 ode containing three dehydrogenases, other non-redox enzymes SDH, MDH Author's personal copy

D. Sokic-Lazic, S.D. Minteer / Biosensors and Bioelectronics 24 (2008) 945–950 949

Fig. 2. Representative power curves of ethanol/air biofuel cells with different enzy- Fig. 3. Averages power densities of the complete (six dehydrogenase enzymatic cas- matic cascades at the bioanode. All solutions are 100 mM ethanol and 1 mM NAD+ cade) citric acid cycle biomimic-based ethanol/air biofuel cell in a 100 mM ethanol in pH 7.5 phosphate buffer and all measurements were made at room temperature. anolyte solution at different phosphate buffer pHs. Anode electrode area is 1 cm2.

in the range of optimum pH value for the majority of the enzymes for single dehydrogenase (ADH) to 1.01 × 10−3 W/cm2 for all six employed in this biomimic. It is important to note that this data dehydrogenases. Although, it can be observed that the power den- shows that the biomimic is extremely pH dependent. A pH change sity increases with the addition of each dehydrogenase, it does not of one unit from pH 6.5 to 7.5 results in a 25.8-fold increase in the increase as expected. As has been seen in other research, the cur- power density produced. rent density typically scales with the amount of catalyst present. The reason for this deviation is believed to be due to these enzymes being inhibited by the build up of reaction product when present 4. Conclusions separately. For instance, the product of the first reaction of this biomimic is acetaldehyde, a very toxic compound that inhibits The citric acid cycle can be mimicked at the electrode surface ® ADH. However, when other enzymes are co-immobilized on the as presented in this study. The TBAB modified Nafion membrane same electrode, acetaldehyde is further oxidized and its inhibit- has an ability to immobilize multi-enzyme systems. In addition, ing effects on ADH are minimized, because there is not a build up methylene green, when polymerized, serves as a successful elec- of concentration of acetaldehyde. However, a build up of acetate trocatalyst for multi-enzyme systems. Due to the addition of more occurs in a two-enzyme system and acetate has its own inhibitory electron generating enzymes, there is an increase in power density effects on enzymes of the citric acid cycle. Therefore, once all six for a multi-enzyme system compared to single enzyme system. All dehydrogenases are immobilized on the same electrode along with enzymes/cofactors/coenzymes of the citric acid cycle can be immo- ® other necessary enzymes and cofactors, the previously mentioned bilized onto the electrode in a TBAB modified Nafion membrane −3 2 inhibiting effects on the enzymes are minimized and the biomimic producing power density of 1.01 × 10 W/cm . This multi-enzyme is functioning most effectively by increasing the power output system can completely oxidize ethanol to carbon dioxide. 8.71-fold compared to a single dehydrogenase system. Although it is theoretically expected to see a sixfold increase going from a Acknowledgements single-dehydrogenase to a six-dehydrogenase system, the 8.71-fold increase in power density and the 9.55-fold increase in current den- The authors would like to acknowledge Air Force Office of sity further shows the point of inhibition playing a role. Also, the Scientific Research for the funding and Plamen Atanassov of the peak shape of the power curve changed with the increase in the University of New Mexico for critical conversations. number of dehydrogenases added to the enzymatic cascades at the bioanodes. The peaks became more parabolic as additional dehy- References drogenase enzymes were added to the enzymatic cascades. This shows that as additional dehydrogenases enzymes are added, con- Akers, N.L., Minteer, S.D., 2003. Preprints of Symposia—American Chemical Society, centration polarization is less noticeable. Another interesting piece Division of Fuel Chemistry 48(2), 895–896. of information is that there is no statistical difference in open cir- Akers, N.L., Moore, C.M., Minteer, S.D., 2005. Electrochim. Acta 50 (12), 2521–2525. Arechederra, R.L., Treu, B.L., Minteer, S.D., 2007. J. Power Sources 173, 156–161. cuit potential between going from a single dehydrogenase system Austin, L.G., 1967. Fuel Cells: A Review of Government-Sponsored Research. Office of to a two dehydrogenase system. The main reason for this is due to Technology Utilization National Aeronautics and Space Administration, Wash- the fact that the open circuit potential observed is not due to the ington, DC. enzyme(s) system present, but rather due to the potential of the Blaedel, W.J., Jenkins, R.A., 1975. Anal. Chem. 47 (8), 1337–1343. Davis, F., Higson, S.P.J., 2007. Biosens. Bioelectron. 22, 1224–1235. + NAD /NADH electrochemistry at the poly(methylene green) layer. Karube, I.S., Matsunaga, S.T., Kuriyama, S., 1981. Ann. N.Y. Acad. Sci., 91. Literature shows that most of the enzymes’ optimum pH is Kjeang, E., Sinton, D., Harrington, D.A., 2006. J. Power Sources 158, 1–12. between 7.3 and 7.6. In order to determine the optimum pH for Lovely, D.R., 2006. Microbe 1, 323. Matsunaga, T.K., Suzuki, S., 1980. Biotechnol. Bioenerg., 22. the whole biomimic system, a pH study was conducted. Six differ- Moore, C.M., Akers, N.L., Hill, A.D., Johnson, Z.C., Minteer, S.D., 2004. Biomacro- ent pH values were tested and they included 6.5, 7.0, 7.25, 7.5, 7.75, molecules 5 (4), 1241–1247. and 8.0. Fig. 3 shows the results obtained. The average power den- Morgunov, I., Srere, P.A., 1998. J. Biol. Chem. 273 (November), 29540–29544. Nagy, Z., 2003. Electrochemistry Encyclopedia. Center for Electrochemical Science sity values obtained are presented along with standard deviations. and Engineering, Chicago. The maximum power density was obtained at a pH of 7.5 which is Ovadi, J., 1995. Cell Architect. Metab. Channel.. Author's personal copy

950 D. Sokic-Lazic, S.D. Minteer / Biosensors and Bioelectronics 24 (2008) 945–950

Palmore, G.T.R., Bertschy, H., Bergens, S.H., Whitesides, G.M., 1998. J. Electroanal. Sumegi, B., Srere, P.A., 1984. J. Biol. Chem. 259 (December), 15040–15045. Chem. 443 (1), 155–161. Suzuki, S.K., Karube, I.S., 1983. Appl. Biochem. Bioeng. 4 (281). Robinson, J.B., Inman, L., Sumegi, B., Srere, P.A., 1987. J. Biol. Chem. 262 (February), Thomas, T.J., Ponnusamy, K.E., Chang, N.M., Galmore, K., Minteer, S.D., 2003. J. Membr. 1786–1790. Sci. 213, 55–66. Robinson, J.B., Srere, P.A., 1985. J. Biol. Chem. 260 (September), 10800–10805. Velot, C., Mixon, M.B., Teige, M., Srere, P.A., 1997. Biochemistry 36 (47), 14271–14276. Srere, P.A., 1985. Trends Biochem. Sci. 10, 109–110. Weast, R.C., Lide, D.R., 1990. CRC Handbook of Chemistry and Physics, 70th ed. CRC Srere, P.A., 1987. Annu. Rev. Biochem. 56, 89–124. Press, Inc., Boca Raton, Florida. Srere, P.A., Mattiasson, B., Mosbach, K., 1973. Proc. Nat. Acad. Sci. U.S.A. 70 (9), Winter, M., Brodd, R.J., 2004. Chem. Rev. 104, 4245–4269. 2534–2538. Zhou, D., Fang, H., Chen, H., Ju, H., Wang, Y., 1996. Anal. Chim. Acta 329 (1/2), 41–48.