Journal of Molecular Catalysis A: Chemical 387 (2014) 86–91
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
jou rna l h om epa ge: www.elsev ier .com/ locate /molcata
irect electrochemical regeneration of the cofactor NADH on bare Ti,i, Co and Cd electrodes: The influence of electrode potential andlectrode material
rshad Ali ∗, Tariq Khan, Sasha Omanovicepartment of Chemical Engineering, McGill University, 3610 University Street, Montreal, Quebec H3A 0C5, Canada
r t i c l e i n f o
rticle history:eceived 7 June 2013eceived in revised form 21 February 2014ccepted 25 February 2014vailable online 5 March 2014
a b s t r a c t
The regeneration of enzymatically-active reduced form of enzymatic cofactor nicotinamide adenine dinu-cleotide (1,4-NADH) from the oxidized form (NAD+) in a batch electrochemical reactor employing bare(non-modified) metal electrodes was investigated as a function of electrode potential and electrodematerial (Ti, Ni, Co and Cd). It was found that the regeneration of 1,4-NADH employing the electrodesis feasible; all the electrodes were capable of producing more than an 80% enzymatically-active product(1,4-NADH), reaching a 96% product purity on Ti. The product purity was found to be highly potential-,
eywords:lectrochemical 1,4-NADH regenerationon-modified metal electrodeslectrode potentialetal–hydrogen bond strengthydrogen surface coverage
and material dependant. The origin of the material/potential dependency was related to the strength ofthe metal–hydrogen (M Hads) bond, and thus to the potential dependence of the Hads electrode surfacecoverage. In contradiction to literature, bare (non-modified) metal electrodes were found to be goodcandidates for electrochemical regeneration of enzymatically-active 1,4-NADH, when the regenerationis performed at a specific overpotential.
. Introduction
Nicotinamide adenine dinucleotide NAD(H) is a cofactor used inedox enzymatic reactions [1–3]. NADH is a cellular fuel and is useduring cellular respiration involving redox enzymes to produceTP [4–6]. In biochemical reactions, NAD(H) serves as a proton andlectron transport molecule. Consequently, NAD(H) can be foundn two redox forms; in an oxidized, NAD+, and in a reduced, NADH,orm (there are several NADH isomers, but only the 1,4-NADHsomer is enzymatically-active). During the NAD+ reduction,he molecule accepts two electrons and a hydrogen, forming annzymatically-active reduced form 1,4-NADH. In industry, it isf importance in the field of chiral compounds preparation [7].t participates in enzymatic catalysis of industrially importantynthetic reactions where conventional chemical catalysts fails [8].t is also used in the production of high-value-added compoundse.g. expensive drugs), and for the development of biosensors andio-fuel cells [9–17]. In all enzymatic reactions, it is necessary to
rovide NAD(H) at stoichiometric quantities. However, due to itsery high cost (especially that of the reduced form, 1,4-NADH), itsndustrial use is very limited. For this reason, many research groups
have been developing methods for 1,4-NADH regeneration. Electro-chemical methods are of particular interest due to their simplicityand easy product isolation [18–22]. Alternatively, some researchgroups have been developing organo-metallic molecules that couldfully replace NADH and reduce the cost of NADH-based enzymaticprocesses [23,24].
The electrochemical regeneration of 1,4-NADH from NAD+ pro-ceeds in two steps, Fig. 1 [25]:
Step1 : NAD+ + e− → NAD• (1)
Step2a : NAD• + e− + H+ → NADH (2)
In Fig. 1, Step 1, NAD+ is reduced to give NAD-radical, which isfurther reduced and protonated in Step 2a to give NADH. Step 2ais considered to be slow due to the slow protonation of the NAD-radical [25–34]. This, in turns, results in very fast dimerization oftwo neighbouring NAD-radicals to produce enzymatically-inactivedimer, NAD2 (Step 2b):
Step2b : NAD• + NAD• → NAD2 (3)
Literature reports that on bare (unmodified) electrodes, thekinetics of Step 2b is indeed significantly faster than that ofStep 2a, and thus the major product of NAD+ reduction onthese electrodes is NAD2 [29]. In order to understand theNAD+ reduction kinetics better and try to control the 1,4-NADH
tive mass transport of electroactive species to/from the electrode
ig. 1. Reduction of NAD+ to NAD2 and enzymatically-active 1,4-NADH. = adenosine diphosphoribose.
egeneration reaction, many research groups have studied funda-ental aspects of the mechanisms and kinetics of NAD+ reduction,
sing mostly bare (non-modified) metallic electrodes, such as mer-ury [29,30,33,35–40] and a variety of carbon materials [41].
To address the above-mentioned problems, many differentpproaches such as chemically modified electrodes [42–47] andnzyme-mediated electrodes [48,49,10,50–54] have been usedn designing electrode systems for in situ regeneration of active,4-NADH. Although many groups demonstrated success in devel-ping electrodes that can give some appreciable (relative to NAD2)mounts of 1,4-NADH (note that most of papers do not explic-tly report this percentage), many of these electrodes are complex,xpensive and/or not stable for a long-term use. Therefore, forndustrial purposes, there is still a need to develop a stable andheap electrode surface that would enable direct electrochemicalegeneration of 1,4-NADH at high recovery (i.e. product purity, rel-tive to 1,4-NADH) [55,56]. Obviously, bare (non-modified) metallectrodes would be the best candidates, but as already mentionedbove, many research groups concluded that these are not capablef regenerating highly pure 1,4-NADH, and the major product isnzymatically-inactive NAD2.
However, one can see that the kinetics of Step 2a depends onhe concentration of H+ at the reaction site. This H+ can reactith the surface (electrode)-adsorbed NAD-radical, according to
he Eley–Rideal mechanism, or it can first adsorb on the electrodeurface as M Hads and then react with the neighbouring NAD-adical, following the Langmuir–Hinshelwood mechanism [57]. Ifhe reaction follows the latter mechanism, then the kinetics of theeaction (Eq. (2)), and thus the amount of enzymatically-active 1,4-ADH produced (relative to NAD2), would be dependent on theydrogen (Hads) surface coverage. The latter, in turns, depends onoth the electrode material and electrode potential.
However, these two effects have not yet been investigated byther research groups in relation to the 1,4-NADH regenerationeaction. We recently demonstrated that when the reduction ofAD+ is performed in a batch electrochemical reactor using a barelassy carbon (GC) and carbon nano-fibre (CNF) cathodes, the purityf active 1,4-NADH regenerated depended strongly on the appliedlectrode potential [26,58]. We showed that an increase in elec-rode potential to more negative values resulted in an increasedecovery of 1,4-NADH from NAD+, reaching a 99–100% at −1.7 VNHE26,58]. The origin of the behaviour was put in relation to thenhancement of the NAD-radical protonation kinetics (Fig. 1; Step
a, Eq. (2)) by providing more ‘active’ hydrogen adsorbed on thearbon electrode surface, Hads, at more negative electrode poten-ials.
is A: Chemical 387 (2014) 86–91 87
With this in mind, we now hypothesize that the recovery(purity) of 1,4-NADH regenerated from NAD+ is dependent not onlyon the electrode potential, but also on the strength of the metalelectrode–hydrogen bond (M Hads). Namely, a metal that bindshydrogen more strongly (e.g. Ti) should attain a higher hydrogensurface coverage at more positive electrode potentials (lower NAD+
reduction overpotentials). Consequently, the higher the hydrogensurface coverage, the faster the kinetics of Step 2a (Eq. (2)), andhigher the purity of 1,4-NADH recovered from NAD+. In an attemptto prove the hypothesis, several selected bare (non-modified)metal electrodes: M Ti, Ni, Co and Cd (in addition to the alreadyinvestigated carbon electrodes [26]) were investigated, in a batchelectrochemical reactor, towards the efficiency in regeneration of1,4-NADH from NAD+. The respective M Hads bond strengths are348.3, 195.9, 192.6 and 119.7 kJ mol−1 [59]. Hence, it was hypothe-sized that Ti would give a maximum 1,4-NADH recovery percentage(i.e. product purity) at the most positive electrode potentials (rel-ative to other metals), while Cd would give it at more negativecathodic potentials. As it will be shown later in the manuscript,this turned out to be the case.
It will be shown that the 1,4-NADH regeneration kinetics andpurity indeed depend on the 1,4-NADH regeneration potential andelectrode material, both predominantly controlling the Hads surfacecoverage and the M Hads bond strength. Electrode potential wasconveniently controlled externally (using a potentiostat), while thestrength of the M Hads bond was varied by choosing different elec-trode materials (M Ti, Ni, Co and Cd).
2. Experimental
Electrochemical regeneration of enzymatically-active 1,4-NADH, from a 1 mM NAD+ solution in 0.1 M phosphate buffer(pH = 5.8), was performed at 295 K under the potentiostatic condi-tions in a three-electrode/two-compartment batch electrochemicalreactor (cell). The buffer was prepared by dissolving KH2PO4 (ACSgrade, BioShop PPM 302) in ultra-pure deionized water (resistivity18.2 M� cm). To adjust pH, 1 N NaOH (Caledon Laboratories 7861-6) was used.
The volume of electrolyte in the reactor was 80 mL. NAD+ solu-tions were prepared by dissolving a proper amount of �-NAD+
(sodium salt, purity 98%, Sigma N0632) in phosphate buffer.Pure (non-modified) electrodes (Ti, Ni, Co or Cd) served as work-
ing electrodes in the batch electrochemical reactor. These elec-trodes were ordered from ESPI Metals (0.152 cm × 2.5 cm × 2.5 cm).The total geometric area of all the working electrodes was 12.5 cm2.Before each measurement, the surface of the working electrodewas carefully wet-polished with polishing paper (grid #600 fol-lowed by grid #1200) until a mirror finish was obtained, followedby degreasing with ethanol and sonication for 5 min in ethanol inorder to remove polishing residues. The counter electrodes weretwo graphite rods, which were, prior to each use, sonicated for30 min in ethanol, followed by thorough rinsing with water. Therods were placed opposite of the two sides of the working electrode,to ensure the uniform electric field. A mercury/mercurous sulphateelectrode (MSE; +0.60 V vs. NHE) was used as a reference electrode,but all the potentials in this paper are referred to the NHE. Whenreferring to literature values, the corresponding potential valuesare also converted with respect to the NHE and quoted as “VNHE”.
In order to maintain an oxygen-free electrolyte, argon (99.998%pure) was purged through the electrolyte prior to, and duringelectrochemical measurements. The Ar purge also ensured convec-
surface. The electrochemical reactor was connected to a potentio-stat, which was used to apply a constant potential to the workingelectrode.
Fig. 2. Linear voltammograms of a Ti electrode recorded in 0.1 M phosphate buffersolution in the absence of NAD+ (dashed line) and in the presence of 4 mM NAD+
(solid line). Scan rate, sr = 10 mV s−1. Temperature, T = 295 K.
90011001300150017001900
Wavenumber / cm-1
∆R
/R
0.0005
1130
1540
1654
Fig. 3. PM-IRRAS spectrum of a NAD+ layer adsorbed on a Ti surface. The layer
8 I. Ali et al. / Journal of Molecular C
To determine the enzymatic activity of the regenerated NADH,ctivity tests were made according to the regular Sigma Qualityontrol Test Procedure (EC 1.8.1.4) which was further modified forhis purpose using lipoamide dehydrogenase (5.3 U/mg, Calzymeaboratories, Inc. 153A0025) as an enzyme and dl-6,8-thioctic acidmide (Fluka T5875) as a substrate [25–28]. First, a volume of 0.2 mLf substrate and 0.1 mL of EDTA (Sigma ED4S) were added into.6 mL of regenerated 1,4-NADH in a cuvette. The absorbance ofhe solution at 340 nm was then monitored using a UV–vis spec-rophotometer, until reaching a steady state value. Then, 0.1 mLf the enzyme was injected into the cuvette while the absorbanceas recorded until reaching a final constant value, signifying that
he entire active 1,4-NADH formed during the electrolysis wasonsumed by the enzymatic reaction. This reaction requires a stoi-hiometric quantity of 1,4-NADH:
Hence, one can expect to see a decrease in absorbance at 340 nmuring the occurrence of reaction (4) due to the oxidation of 1,4-ADH to NAD+. Finally, taking into account the initial and finalbsorbance at 340 nm, the purity of enzymatically-active 1,4-NADHroduced (regenerated) by electrolysis was calculated:
here A stands for absorbance. The assay was first calibrated usingommercially available NADH that contains 98% of enzymatically-ctive 1,4-NADH.
. Results and discussion
.1. Reduction of NAD+ on Ti electrode
In order to determine a potential region in which NAD+
ndergoes electrochemical reduction, linear voltammetry (LV)easurements were first performed using Ti as working electrode.ur previous studies on GC, Au and GC-Ru electrodes [25,26,28,31]
howed that a well-defined cathodic peak was observed at ca.0.9 VNHE in a NAD+ containing solution, indicating that the reduc-
ion of NAD+ occurs in this potential region. Therefore, linearoltammograms on the Ti electrode were also recorded in thisotential region. Fig. 2 shows the response of a Ti electrode recorded
n the absence (dashed line) and presence (solid line) of NAD+ inhe electrolyte.
The response of the Ti electrode in the absence of NAD+ is asxpected (background response, dashed line); with an increasen cathodic potential past ca. −0.4 V, the resulting current alsoncreases, which is due to the increase in the kinetics of the hydro-en evolution reaction (HER) [25–27]. Surprisingly, the response ofhe electrode in the NAD+-containing solution was quite differenthan expected. Namely, while on GC, Au and GC-Ru electrodes aell-defined NAD+ reduction peak was observed at ca. −0.9 VNHE
25,26,28,31], the peak is absent in Fig. 2. In addition, the NAD+
eduction current/peak on the GC, Au and GC-Ru electrodes isignificantly higher than the background current recorded in thebsence of NAD+, while the response on the Ti electrode (Fig. 2,olid line) is quite opposite – the current recorded in the presencef NAD+ in the electrolyte (solid line) is significantly lower than that
ecorded in the absence of NAD+ in the electrolyte (the backgroundurrent, dashed line). A possible reason for the observed behaviourould be the adsorption of NAD+ on the Ti electrode surface; thisdsorbed NAD+ layer blocks the electrode surface and thus inhibits
was adsorbed from a 4 mM solution of NAD+ in 0.1 M phosphate buffer, by linearlypolarizing the Ti electrode surface from 0.0 V to −1.3 V at a scan rate of 10 mV s−1
(Fig. 2).
the HER, which results in a very low current recorded at potentialsnegative of −0.4 V (Fig. 2, solid line). Also, the absence of a currentpeak related to NAD+ reduction indicates that the NAD+ reductionreaction does not seem to be mass-transport controlled. It shouldbe noted that a similar behaviour to that presented in Fig. 2, wasalso obtained on the other three electrodes (Cd, Co, and Ni).
To investigate if NAD+ indeed adsorbs on the Ti electrode surfaceunder the experimental conditions employed in the experimentpresented in Fig. 2, the Ti sample that underwent linear polarization(Fig. 2, solid line) was characterized ex situ by polarization modula-tion infrared reflection absorption spectroscopy (PM-IRRAS), afterthorough rinsing of the surface with deionized water; the resultingspectrum is presented in Fig. 3.
Several characteristics peaks are visible in Fig. 3. The well-pronounced peak at 1130 cm−1 corresponds to the response of theribose moiety of NAD+, while the peak at 1540 cm−1 representsthe C N stretching vibration in the nicotinamide moiety [60]. Thevibration at 1654 cm−1 corresponds to the bending of NH2 in theadenine moiety. Thus, the PM-IRRAS measurements confirmed thatNAD+ indeed adsorbs on the Ti electrode surface during the linearpolarization in the potential region in Fig. 2, confirming and that
the decrease in the current in Fig. 2 is due to the blocking of theelectrode surface by adsorbed NAD+.
In order to regenerate enzymatically-active 1,4-NADH, long-term potentiostatic (electrolysis) experiments were done at several
I. Ali et al. / Journal of Molecular Catalys
E / VNHE
-1.8-1.6-1.4-1.2-1.0-0.8
1,4-
NA
DH
reco
very
/ %
60
70
80
90
10050
60
70
80
90
100(a)
(b)
Fig. 4. The percentage of enzymatically-active 1,4-NADH recovered on (a) Ti and(b) Cd electrodes, obtained by reduction of 1 mM NAD+ in a batch electrochemicalr
s−tfapua
rt(pdwTs(tTbaea
wi
visible in Fig. 4b. The increase in electrode cathodic potential past−1.5 V has a negative impact on the recovery of 1,4-NADH, which
eactor operating at different electrode potentials.
elected potentials in the potential region of NAD+ reduction (from0.8 to −1.7 V) in a batch electrochemical reactor using a Ti elec-
rode. The percentage of enzymatically-active 1,4-NADH recoveredrom NAD+ was determined by the enzymatic assay, and presenteds a function of regeneration potential in Fig. 4a (note that theercentage refers to the relative amount of 1,4-NADH in the prod-ct mixture, which could also contain enzymatically-inactive NAD2nd other NADH isomers).
Fig. 4a demonstrates that the percentage of active 1,4-NADHecovered is strongly dependent on the electrode potential. Athe lowest cathode potential (−0.8 V), a maximum recovery96.0 ± 2.6%) of 1,4-NADH was obtained, and by increasing theotential to more negative values the recovery decreased graduallyown to 54.0 ± 2.2% (at −1.7 V). This behaviour is quite consistentith what was expected. Namely, since the electronic structure of
i is such that the 3d orbital has only two unpaired electrons andince hydrogen is an electron donor, the Ti Hads bond is strong348.3 kJ mol−1), and one can thus expect that at low HER elec-rode potentials (−0.8 V in Fig. 4a), the surface coverage of Hads oni is relatively high, as compared to metals that form weak M Hadsonds. Since the NAD-radical protonation (Fig. 1; Step 2a, Eq. (2)) is
strong function of “active” hydrogen concentration, one can thenxpect to obtain a relatively high recovery (purity) of 1,4-NADHlready at −0.8 V, as indeed demonstrated in Fig. 4a.
Further, the decrease in the percentage of 1,4-NADH recovered
ith an increase in electrode potential to more negative values seen
n Fig. 4a can be related to the competitive behaviour of the NAD-
is A: Chemical 387 (2014) 86–91 89
protonation and H2 evolution (HER) reactions, both starting withthe formation Ti Hads [61]:
Ti + H+ + e− → Ti Hads (6)
Due to the strong affinity of hydrogen for adsorption on tita-nium (because of the almost empty 3d orbitals in Ti), the kineticsof reaction (6) is fast even at low negative (cathodic) potentials inthe HER region (Fig. 2). The second step in the HER is the forma-tion of molecular hydrogen through one (or both) of the followingreactions [61]:
Ti Hads + H+ + e− → H2 + Ti (7)
2Ti Hads → H2 + 2Ti (8)
The kinetics of the second step increases with an increase inelectrode potential to more negative values and, thus the totalHER kinetics, which is evident in Fig. 2 as a continuous increase incathodic current (dashed line). Contrary, at more positive cathodicpotentials, the total HER kinetics is slow. Taking into account thehigh strength of the Ti Hads bond, this yields high surface coverageby hydrogen, Ti Hads (Eq. (6)), at positive potentials and, conse-quently, the fast NAD-radical protonation kinetics (Fig. 1; Step 2a,Eq. (2)). This enables a high purity 1,4-NADH to be recovered fromNAD+, Fig. 4a. However, with the increase in electrode potentialto more negative values, the HER kinetics increases relative to theNAD-radical protonation kinetics, enabling the formation of NAD2.As the result, there is a decrease in the percentage of 1,4-NADHrecovered from NAD+, i.e. an increase in the percentage of NAD2formed, as seen in Fig. 4a.
3.2. Reduction of NAD+ on Cd electrode
Similarly to the Ti electrode, 1,4-NADH regeneration experi-ments were also performed with a Cd electrode at various electrodepotentials. The rationale for using Cd next was that the Cd Hadsbond is significantly weaker (119.7 kJ mol−1) than the Ti Hads bond(348.3 kJ mol−1), and a behaviour opposite to that presented inFig. 4a was expected to be seen. Indeed, Fig. 4b demonstrates thatthis is the case. Unlike on Ti (Fig. 4a), the lowest relative recov-ery (64.0 ± 1.4%) of 1,4-NADH was obtained at the most positivecathode potential (−0.8 V). However, by increasing the regener-ation potential to more negative values the percentage recoveryincreased gradually, reaching a maximum of 93.0 ± 1.4% at −1.5 V,and then decreasing to 86.0 ± 3.6% at −1.7 V. As explained earlier,the percentage of 1,4-NADH regenerated from NAD+ (relative toNAD2) is dependent not only on the electrode potential, but alsoon the strength of the metal electrode–hydrogen bond (M Hads).Since, the Cd Hads bond is weak (119.7 kJ mol−1), at low HERelectrode potentials (−0.8 V) the Cd surface coverage by Hads is,therefore, relatively low, as compared to Ti that forms a strongM Hads bond. Therefore, the kinetics of the hydrogen adsorptionreaction (6) and, thus the formation of the ‘active’ hydrogen (Hads) isslow at low negative (cathodic) potentials. Consequently, the lowerthe hydrogen surface coverage, the slower the kinetics of Step 2a,Eq. (2), and the lower the percentage of 1,4-NADH recovered fromNAD+. On the other hand, by increasing the electrode potential tomore cathodic values the surface coverage of Hads on the Cd surfaceincreases, and thus the kinetics of Step 2a, Eq. (2). This, in turns,results in an increase in recovery of 1,4-NADH, which is indeed
is due to the increase in the kinetics of the competitive HER (Eqs.(7) and/or (8)).
90 I. Ali et al. / Journal of Molecular Catalysis A: Chemical 387 (2014) 86–91
E / VNHE
-1.8-1.6-1.4-1.2-1.0-0.8
1,4-
NA
DH
reco
very
/ %
20
40
60
80
50
60
70
80
90
(a)
(b)
Fig. 5. The percentage of enzymatically-active 1,4-NADH recovered on (a) Co and( +
r
3
bispS((iibp
tii1(Caortit(
Table 1Dependence between the electrode material, M Hads bond strength and electrolysispotential (Emax,NADH) at which a maximum recovery of enzymatically-active 1,4-NADH was obtained.
Electrode material M Hads bondstrength(kJ mol−1)
Emax,NADH
(VNHE)NADHrecovery(%)
Ti 348.3 −0.8 96Ni 195.9 −1.3 92
further shown that a 96% product purity (1,4-NADH) can be regen-
b) Ni electrodes, obtained by reduction of 1 mM NAD in a batch electrochemicaleactor operating at different electrode potentials.
.3. Reduction of NAD+ on Co and Ni electrodes
The previous two sections demonstrated that when a M Hadsond is strong (such as Ti Hads) the highest recovery of 1,4-NADH
s obtained at low cathodic potentials (Fig. 4a), since the electrodeurface is at this potentials already covered by Hads which is areferable species for the protonation of the NAD-radical (Fig. 1;tep 2a, Eq. (2)), as opposed to the hydrogen ion from the solutionH+). The opposite is obtained in the case of a weak M Hads bondsuch as Cd Hads, Fig. 4b). It would now be interesting to examinef the intermediate M Hads bond strength would yield a maximumn 1,4-NADH recovery somewhere between these two cases, i.e.etween the two extreme potentials (−0.8 and −1.7 V). For thaturpose, Co and Ni electrodes were tested.
Fig. 5a shows the corresponding results obtained with a Co elec-rode. The maximum in the 1,4-NADH recovery (82.0 ± 1.9%) wasndeed obtained at a slightly more negative potentials (−0.9 V)n comparison to Ti (−0.8 V). The Co Hads bond strength is92.6 kJ mol−1, which is weaker than the Ti Hads bond strength348.3 kJ mol−1) and stronger than that of Cd Hads (119.7 kJ mol−1).onsequently, a high Hads surface coverage of Co could be expectedt potentials more negative than that on Ti, but more positive thann Cd. Accordingly, the kinetics of the NAD-radical protonationeaction (Fig. 1; Step 2a, Eq. (2)) could also be expected to followhe same behaviour, which is indeed shown in Fig. 5a. A decrease
n the 1,4-NADH percentage recovery at potentials more negativehan −0.9 V (Fig. 5a) is due to the increase in the HER kinetics (Eqs.7) and/or (8)).
Co 192.6 −0.9 82Cd 119.7 −1.5 93
If we now take nickel, the Ni Hads bond strength is195.94 kJ mol−1, which is very close to the Co Hads bond strength,and one can thus expect to see a maximum in 1,4-NADH recov-ery also somewhere between −0.8 V and −1.7 V, as actually seen inFig. 5b. The maximum value obtained was 92.0 ± 1.4%. The trendin Fig. 5b could be described in the same manner as those inFigs. 4a and b and 5a, and will thus not be repeated here.
Table 1 summarizes the findings presented so far, by presentingthe relationship between the electrode material used for NADHregeneration, the corresponding metal–hydrogen bond strengthand the NADH regeneration potential that yielded a maximumrecovery of enzymatically-active 1,4-NADH (the correspondingpercentage is also shown).
The first important contribution of the present work is thedemonstration of the possibility of obtaining a high recovery(purity) of enzymatically-active 1,4-NADH on non-modified elec-trodes, which has not yet been reported in the literature. The secondimportant contribution of the work is the demonstration of therelationship between the M Hads bond strength, applied regen-eration potential and the recovery (purity) of produced 1,4-NADH.However, since there is no published systematic research except[27] (which is also our laboratory’s work) on the influence of elec-trode potential on the percentage recovery of 1,4-NADH producedby electrolysis of NAD+, direct comparison of the data presented inthis manuscript, to the literature data is not quite possible. Never-theless, some general comparison still can be made. The literaturesurvey shows that the amount of 1,4-NADH regenerated on a plat-inum electrode ranges from 30 to 50% [40,42,62]. On mercury,it ranges from 10 to 76% [30,37,39,63–65], a tin oxide electrodeyielded 10% pure 1,4-NADH [5], and on a reticulated vitreous car-bon electrode it was only 0.6% [43], while on gold-amalgam itwas 10% [42]. However, the reader should note that, unfortunately,some research groups have reported yields of 1,4-NADH regener-ated electrochemically, on the basis of only considering the 340 nmpeak in the UV/vis [66,67], instead of testing the purity by theenzymatic assays. However, this peak corresponds not only to theamount of 1,4-NADH produced, but also to the presence of pro-duced NADH isomers and the dimer, NAD2, the latter two beingenzymatically-inactive. Thus, these values cannot be trusted.
4. Conclusions
The influence of electrode potential and electrode material onthe purity of regenerated 1,4-NADH was investigated. It was foundthat the regeneration of 1,4-NADH from NAD+ in a batch electro-chemical reactor employing non-modified electrodes (Ti, Ni, Co andCd) is feasible. It was demonstrated that the 1,4-NADH regenerationkinetics and product purity depends on the 1,4-NADH regenerationpotential and electrode material, both predominantly controllingthe Hads surface coverage and the M Hads bond strength. It was
erated on a Ti electrode polarized at a (low) cathodic potential of−0.8 V. The origin of the material/potential dependency is relatedto the strength of the metal-hydrogen (M Hads) bond, and thus
atalys
ta
A
EaP
R
[
[
[
[
[[
[[[[[[
[[
[[[[[
[
[[[[
[[[
[[
[
[
[[[[
[[[[[
[
[[[
[
[[[[
[[[
[[[[
I. Ali et al. / Journal of Molecular C
o the potential dependence of the Hads electrode surface cover-ge.
cknowledgements
The authors would like to acknowledge the Natural Sciences andngineering Research Council of Canada (grant number 202887)nd the University of Engineering and Technology, Peshawar,akistan for providing the support for this research.
[2] A.M. Klibanov, A.V. Puglisi, Biotechnol. Lett. 2 (1980) 445–450.[3] M. Musameh, J. Wang, A. Merkoci, Y. Lin, Electrochem. Commun. 4 (2002)
743–746.[4] A. Liese, K. Seelbach, C. Wandrey, Industrial Biotransformations, Wiley-VCH,
Weinheim, 2006.[5] Y.H. Kim, Y.J. Yoo, Enzyme. Microb. Technol. 44 (2009) 129–134.[6] A. Schmid, J.S. Dordick, B. Hauer, A. Kiener, M. Wubbolts, B. Witholt, Nature 409
1011–1016.[8] S.H. Lee, J. Ryu, D.H. Nam, C.B. Park, Chem. Commun. 47 (2011) 4643–4645.[9] P. Belenky, K.L. Bogan, C. Brenner, Trends Biochem. Sci. 32 (2007) 12–19.10] K. Delecouls-Servat, A. Bergel, R. Basseguy, Bioprocess Biosyst. Eng. 26 (2004)
205–215.11] R. Devaux-Basseguy, A. Bergel, M. Comtat, Enzyme. Microb. Technol. 20 (1997)
248–258.12] F. Pariente, F. Tobalina, M. Darder, E. Lorenzo, H.D. Abruna, Anal. Chem. 68
(1996) 3135–3142.13] A.C. Pereira, M.R. Aguiar, A. Kisner, D.V. Macedo, L.T. Kubota, Sens. Actuators B
124 (2007) 269–276.14] N. Pollak, C. Dolle, M. Ziegler, Biochem. J. 402 (2007) 205–218.15] R.A. van Santen, P.W.N.M. van Leeuwen, J.A. Moulijn, B.A. Averill, Catalysis An
Integrated Approach, Elsevier, Amsterdam, 1999.16] A. Rongvaux, F. Andris, F. Van Gool, O. Leo, BioEssays 25 (2003) 683–690.17] A.A. Sauve, J. Pharmacol. Exp. Ther. 324 (2008) 883–893.18] H.K. Chenault, G.M. Whitesides, Appl. Biochem. Biotechnol. 14 (1987) 147–197.19] M.D. Leonida, Curr. Med. Chem. 8 (2001) 345–369.20] W. Liu, P. Wang, Biotechnol. Adv. 25 (2007) 369–384.21] J. Rocha-Martin, D. Vega, J.M. Bolivar, C.A. Godoy, A. Hidalgo, J. Berenguer, J.M.
Guisan, F. Lopez-Gallego, BMC Biotechnol. 11 (2011) 101–111.22] W.A. Van Der Donk, H. Zhao, Curr. Opin. Biotechnol. 14 (2003) 421–426.23] F. Hollmann, P.C. Lin, B. Witholt, A. Schmid, J. Am. Chem. Soc. 125 (2003)
8209–8217.
24] A.K. Udit, F.H. Arnold, H.B. Gray, J. Inorg. Biochem. 98 (2004) 1547–1550.25] A. Azem, F. Man, S. Omanovic, J. Mol. Catal. A: Chem. 219 (2004) 283–299.26] I. Ali, B. Soomro, S. Omanovic, Electrochem. Commun. 13 (2011) 562–565.27] A. Damian, K. Maloo, S. Omanovic, Chem. Biochem. Eng. Q. 21 (2007) 21–32.28] A. Damian, S. Omanovic, J. Mol. Catal. A: Chem. 253 (2006) 222–233.
[
[
is A: Chemical 387 (2014) 86–91 91
29] P.J. Elving, W.T. Bresnahan, J. Moiroux, Z. Samec, Bioelectrochem. Bioenerg. 9(1982) 365–378.
30] H. Jaegfeldt, Bioelectrochem. Bioenerg. 8 (1981) 355–370.31] F. Man, S. Omanovic, J. Electroanal. Chem. 568 (2004) 301–313.32] J. Moiroux, S. Deycard, T. Malinski, J. Electroanal. Chem. 194 (1985) 99–108.33] C.O. Schmakel, K.S.V. Santhanam, P.J. Elving, J. Am. Chem. Soc. 97 (1975)
5083–5092.34] K. Takamura, A. Mori, F. Kusu, Bioelectrochem. Bioenerg. 8 (1981) 229–238.35] W.T. Bresnahan, P.J. Elving, J. Am. Chem. Soc. 103 (1981) 2379–2386.36] W.T. Bresnahan, J. Moiroux, Z. Samec, P.J. Elving, Bioelectrochem. Bioenerg. 7
625–634.40] M. Studnickova, H. Paulova-Klukanov, J. Turanek, J. Kovar, J. Electroanal. Chem.
252 (1988) 383–394.41] J. Moiroux, P.J. Elving, J. Electroanal. Chem. 102 (1979) 93–108.42] S.H. Baik, C. Kang, I.C. Jeon, S.E. Yun, Biotechnol. Tech. 13 (1999) 1–5.43] M. Beley, J.P. Collin, J. Mol. Catal. 79 (1993) 133–140.44] A.A. Karyakin, O.A. Bobrova, E.E. Karyakina, J. Electroanal. Chem. 399 (1995)
179–184.45] Y.T. Long, H.Y. Chen, J. Electroanal. Chem. 440 (1997) 239–242.46] K. Warriner, S. Higson, P. Vadgama, Mater. Sci. Eng. C5 (1997) 91–99.47] Y. Shimizu, A. Kitani, S. Ito, K. Sasaki, Denki Kagaku 61 (1993) 872–873.48] K. Cheikhou, T. Tzedakis, AIChE J. 54 (2008) 1365–1376.49] X. Chen, J.M. Fenton, R.J. Fisher, R.A. Peattie, J. Electrochem. Soc. 151 (2004)
5607–5610.51] S. Kim, S.-E. Yun, C. Kang, Electrochem. Commun. 1 (1999) 151–154.52] S. Kim, S.E. Yun, C. Kang, J. Electroanal. Chem. 465 (1999) 153–159.53] S.B. Sobolov, M.D. Leonida, A. Bartoszko-Malik, K.I. Voivodov, F. McKinney, J.
Kim, A.J. Fry, J. Org. Chem. 61 (1996) 2125–2128.54] K.I. Voivodov, S.B. Sobolov, M. Draganoiu Leonida, A.J. Fry, Bioorg. Med. Chem.
Lett. 5 (1995) 681–686.55] F. Hollmann, A. Schmid, Biocatal. Biotransform. 22 (2004) 63–88.56] C. Kohlmann, W. Maerkle, S. Luetz, J. Mol. Catal. B: Enzym. 51 (2008) 57–72.57] P. Atkins, Elements of Physical Chemistry, W. H. Freeman, New York, 2009.58] I. Ali, M. McArthur, N. Hordy, S. Coulombe, S. Omanovic, Int. J. Electrochem. Sci.
7 (2012) 7675–7683.59] S. Trasatti, J. Electroanal. Chem. 39 (1972) 163–184.60] A. Damian, S. Omanovic, Langmuir 23 (2007) 3162–3173.61] E. Navarro-Flores, Z. Chong, S. Omanovic, J. Mol. Catal. A: Chem. 226 (2005)
179–197.62] J. Cantet, A. Bergel, M. Comtat, Bioelectrochem. Bioenerg. 27 (1992) 475–486.63] M. Aizawa, R.W. Coughlin, M. Charles, Biotechnol. Bioeng. 18 (1976) 209–215.64] E. Bojarska, B. Czochralska, Bioelectrochem. Bioenerg. 16 (1986) 287–300.65] Y. Nakamura, S.I. Suye, J.I. Kira, H. Tera, I. Tabata, M. Senda, Biochimica et
Biophysica Acta – General Subjects 1289 (1996) 221–225.66] G. Rahman, J.Y. Lim, K.D. Jung, O.S. Joo, Int. J. Electrochem. Sci. 6 (2011)
2789–2797.67] G. Rahman, J.Y. Lim, K.D. Jung, O.S. Joo, Electrochem. Commun. 12 (2010)
