Methodologies for “Wiring” Redox Proteins/Enzymes to Electrode Surfaces

Abstract The immobilization of redox proteins or enzymes onto conductive surfaces has application in the analysis of biological processes, the fabrication of biosensors, and in the development of green technologies and biochemical synthetic approaches. This review evaluates the methods through which redox proteins can be attached to electrode surfaces in a “wired” configuration, that is, one that facilitates direct electron transfer. The feasibility of simple electroactive adsorption onto a range of electrode surfaces is illustrated, with a highlight on the recent advances that have been achieved in biotechnological device construction using carbon materials and metal oxides. The covalent crosslinking strategies commonly used for the modification and biofunctionalization of electrode surfaces are also evaluated. Recent innovations in harnessing chemical biology methods for electrically wiring redox biology to surfaces are emphasized.


Introduction to biological redox chemistry
The proteins that facilitateb iological electron transfer processes are referred to as "redox proteins." These molecules play essentialr oles in processesr anging from photosynthesis to respiration, from bioluminescence to nitrogenf ixation, and from nucleic acid biosynthesis to apoptosis. [1,2] The thermodynamics and kinetics of the biological electron transfer reactions are determined by the nature of the redox centers within the participating proteins. These redox-activec enters can be either organic cofactors (e.g.,q uinones and flavins) [3] or metal centers (e.g.,i ron sulfur clusters and Cu sites), [1] as exemplified by Figure 1. Redox enzymesa re as ubset of redox-active proteins that catalyzet he oxidation or reductiono fs ubstrate molecules at ar edox-active center.T he approximately 1.5 Vp otential window which such biological redox centers span (see Figure 2) is wider than the thermodynamic stability window of water,s incep roton reduction to hydrogen (E(2H + /H 2 ) = À0.41 Va tp H7)a nd water oxidation to oxygen (E(O 2 /H 2 O) = + 0.82 Va tp H7)a re both processes which have been occurring in biology for millennia. [4] Recent work on azurin, as inglecopper electron-transfer protein, has elegantly demonstrated how the reduction potential of biological redox centers is tunedb yt he interplay between both the redox centera rchitecture andt he surrounding protein structure. [5,6] Research interesti nb iological redox chemistry is inspired by more than just an academic curiosity in understandingt he biochemicalr eactions of life. Enzymes play an essential role in the production of biofuels, and redox-active metalloenzymes play ap articularly vital role in hydrogen generation, [7][8][9] methane production, [10] and arecently discovered role in cellulose breakdown. [11] Other redox-enzymeb ased applications range from the development of novel biocatalysts fors olving challenging synthetic problems, [12,13] to the sequestering of atmospheric CO 2 . [14,15] As both redox-activep roteins and enzymes can be used to elicit an electronic response from ab iological stimulus, there is also av ast range of literature exploring the use of such molecules for developingn ew sensort echnologies. [16] One of the most famous examples is the blood glucoses ensor, ad evice that helps billions of people worldwide by monitoring the concentration of glucose in the bloodstream throught he electrochemical response of glucose oxidase. [17,18]  This Review focuses on strategies for stably attaching proteins and enzymeso ntos urfaces. Regardless of whether proteins or enzymesa re redox-active or not, the analyticals tudy and commercial utilization of such biomolecules is aided by such immobilization methodologies. For example, surfacep lasmon resonance (SPR)d etectiono fd rug molecule bindingi si nherently reliant on the attachment of proteins or enzymes onto sensor chips [19] or nanoparticles. [20] In industrial catalysis, the localization of enzymes on the surface of as olid support can also help overcome high operation costs, improving the ease of separation of enzymef rom product, the lifetimea nd reusability of the enzyme, and potentially enhancing the thermostability. [21] In the case of redoxp roteins and enzymes, immobilization onto ac onducting surface provides ar oute for the delivery or removalo fe lectrons (Figure 3). Such a" wired" biomolecule-surfacec onfiguration can either be utilized in electroanalytical measurements that probe the biological redox process, or for constructing electrodes for biotechnological applicationss uch as medicalb io-sensing, [22] solar fuel production, [23,24] or cofactor regeneration systems. [25] 2. Electroactive protein adsorption onto unmodified conducting surfaces The orientation of redox proteins or enzymes onto electrode surfaces in as o-called "electroactive" configuration, that is, one that permits direct electron transfer between the surface and the biomolecule, is ap rerequisite for protein filme lectrochemistry,often referred to as PFE. [16,18,[26][27][28][29] This technique quantifies the thermodynamic and kinetic parameters of the electrochemicalr eactions of redox proteins and enzymes whichf orm a" film" on the surface of the workinge lectrode that is interrogated using as tandard three-electrode electrochemical setup. [16,18,[26][27][28][29] Thew ealth of detailed mechanistic PFE studies conducted using aw ide range of different proteins or enzymes directly adsorbed onto electrodes demonstrates the feasibility of immobilizing such redox-active macromolecules onto solid surfaces throughn oncovalent interactions. [16,18,[26][27][28][29] When successful, such an immobilization strategy clearly represents the simplesta pproach for achieving electroactive films of redox protein or enzyme.
In living systems, the exchange of electrons between soluble redox proteins is dependento nt he two proteins "docking" so that the electron-donor and electron-acceptor redox centers are brought into close enough approach to facilitater apid, direct electront ransfer. [30] These interactions are often mediated by areas of complementary polarity on the donor and acceptor proteins. [30] Thus, asimple model for understanding successfuld irect electroactive protein adsorption onto electrode surfaces is to envisage the electrode surface polarity complementing ar egion of oppositely charged residues on thep rotein surface that is proximal to the electron entry/exit redox center ( Figure 3). [31] This meanst hat direct electron transfer between redox proteins and electrode surfaces is mosteasily achieved when the electron entry/exit redox centeri sc lose to the protein surface. [32] The adsorption and orientation of proteins onto surfaces can be influenced by the solution electrolyte conditions. As described below,i onic strengthand pH are both important variables, and the entropically disfavored process of adsorption is also favored at lower temperatures. [29] Detailed work by Harry Gray and co-workersh as demonstratedt hat electrons tunnel through the protein structure which separates electron-donor and electron-acceptor partner redox-active centers, and therefore distance plays ac rucial role in determining the rate of electron transfer. [33,34] Ah elpful ruleof-thumb provided by Dutton and co-workers is that within metalloenzyme structuresatunneling distance of lesst han 14 between redox active sites appears to support electron transfer rates that are sufficiently fast to avoid limiting the rate of redox catalysis. [31] Ideally,a ll protein or enzyme molecules would therefore orient on the electrode with the same sub-14 distance between the redox-activec enter and the conductings urface. However, in some cases the electrochemical response of otherwise identical protein molecules differs, and this has been attributed to a" dispersion" in protein/enzyme orientation. [7,29,35] Redox proteins/enzymes can also become adsorbed onto surfaces in configurations whichd on ot facilitated irect electron transfer at all,a si llustrated in Figure 3. Alternatively,t he biomolecules may remain in solution,w ith the slow rate of diffusion of thesem acromoleculesi mpeding solution electrochemistry,a nd electron transfer to the electrode instead relying on the introduction of redox mediators. Such mediated bioelectrochemistry is extremely useful in sensord evelopment, [36][37][38] and careful design of electron-transfer polymer gels can even permit simultaneouse ntrapment of enzymes on the electrode and modification of the reactivity.F or example, aH 2 enzymew as recently made functional in O 2 -saturated solution through use of av iologen-polymer net. [39] However,a sn oted above,h erein we focus our attention on protein/enzyme-electrode immobilizations trategies that permit direct, unmediated electron transfer.
No tools are currently availablet op redict the likelihood that ar edox protein or enzyme of interest will becomea dsorbed in an electroactive configuration on as olid support, and screening for as uccessful protein-surfacec ombination remains an empirical process. [29] The electrode surfaces most commonly used for electroactive protein/enzyme electrode immobilization are briefly reviewed below.

Carbon bulk materials
Carbon is an extremelyp opular material for constructinge lectrodesf or the electrochemical interrogation of small molecules. [40] As ah ighly conductive allotrope of carbon,g raphite electrodes are common. [40] Either pyrolyticg raphite edge (PGE) or basal plane graphite (BPG) electrodes can be fabricated from cutting highly ordered pyrolytic graphite (HOPG) substrate in perpendicular directions, across or parallel to the graphite sheets, respectively ( Figure 4). [41] For PFE, PGE has provedt ob et he most successful carbon electrode materialf or the electroactive adsorption of redox proteins and enzymes. [16, 18, 26-29, 42, 43] This has been attributed to the PGE surface components, including ad iverser ange of aro-  . Non-specific protein adsorption outcomes. a) Electrostatic attraction of oppositelycharged protein residues and electrode surface facilitates the immobilization of the protein in an electroactive orientation, facilitating direct electron transfer between ar edox center and the electrode. b) Protein becomes immobilized in an orientation thatdoes not facilitate direct electron transfer.c)Proteindoes not adsorb to the electrode surface. matic, hydrophilic( i.e. phenolic), and carboxylate functionalities that are present as defects on the edge plane ( Figure 5), yieldingagenerally negatively charged surfacet hat will electrostatically attractr egions of complementary positive polarity on the protein surface. [44,45] Electrode surface polishing/abrasion processes using emery paper or pastes of diamond or alumina are often used to actively increase the surfacer oughness and thus increase the number of defect sites. [29] Alumina and diamondp olishing materials can remain on the electrode surface even after rinsing and sonicating the electrodes, so it is also possible that the presence of polishingm aterials contributes to the performance of PGE electrodes. [43] The combination of the chemicalh eterogeneity and the topological roughnesso ft he PGE surfaceh as also been credited with making it particularly suitable for electroactive protein/enzymei mmobilization. [42] The chemical heterogeneity allowsm ultiple and varied favorable contacts to be madeb etween the protein and the electrode surface. [29] The roughness of the surface can ensure that ar ange of immobilized protein orientations are electroactive, [46] as even if the face through which the protein is adsorbed to the electrode surface is distant from an electron entry/exit site, rapid electron transfer may still be feasible because this site is close to another part of the electrode surface( Figure 5). [29] Such orientational flexibility may also explain as tatistical variationi nt he electrochemical reactionp arameters. For example, in H 2 -enzymev oltammetry modellings tudies, the need to include ar ange of different interfacial electron-transfer kinetic rate constantsinthe calculations is attributed to dispersion in the distance between the electrode surface and the electron entry/exit site in the protein. [16,35,47] In studies on azurin, variations in the apparent midpoint potentialofthe biological electron transfer were attributed to different protein-surface orientations/environments. [48,49] In cases where the electron entry/exit site of the protein or enzymei sl ocated within ar egion of negative charge, the polarity of the PGE surface may not helpf acilitatee lectroactive adsorption,a nd may insteadp romote the desorption of adsorbedp roteins. [45,51] Alleviating the electrostatic repulsion between protein ande lectrode can be achieved through mild The different potential configurations of HOPG in disk electrodes, either:i )with the basal plane exposed, or ii)the edge plane (often denoted pyrolytic graphite edge or "PGE" electrodes). acidification, or throught he co-adsorption of protein with polycationic hydrophilic compoundss uch as aminocyclitols, [32] polylysine, [32] polymyxin [16,29,51] or polyethyleneimine. [45,51] Polycationics pecies have an affinity fort he PGE surface, and are thought to mediate protein adsorption through the formation of ternary salt bridges between areas of negative charge on the protein and electrode surface ( Figure 6). [32,29] Aside from PGE, other carbon surfaces have also proveds uccessfulf or producing electroactive films of redox proteins or enzymes. Carbon felt is comprised of an amorphous tangle of smooth carbon fibers. [52] The high surfacea rea, high conductivity,l arge void spaces and low cost of this materialm ake it suitable for application in redox-enzyme biofuel devices. [52] Carbon felt electrodes of small geometrics urface area can accommodate and directly exchange electrons with large quantities of enzymes, with ad iiron hydrogenase used in ab io-H 2 device. [53] Such porousm aterials can be less usefuli nm echanistic studies of redoxe nzymes since the diffusion rates of substrate, product, or inhibitor through the materialm ay limit the rate of reactivity.T his would mean that electrochemical current cannot be used to monitor the inherent maximum turnoverr ate of the enzyme. However,i ne nzyme fuel cell developments, where the focus is to maximize the enzyme current per unit surfacea rea, such porous materials are very useful, and have enabledo rder-of-magnitude powerincreases. [54] In solution-state electrochemical studies of small redoxactive molecules, commonc arbon-based electrode substrates include boron-doped diamond( BDD) and glassy carbon (GC). [55] BDD consists of diamond in which approximately one atom in at housand has been replaced by boron,g iving the material p-types emiconductive properties and yielding the hardest carbon material used for electrodes. [41] The very low capacitance of BDD minimizes background current, effectively enhancing the sensitivity of the electrochemical measurement. [41] However, BDD is not widely utilized in PFE, presumably indicating that the surface electrostatics do not facilitate protein adsorption.T he structure of GC consistso fi nterwoven graphite ribbons,r eminiscent of three-dimensionalc hainmail. [41] GC is much harder than HOPG, and contains hydropho-bic basal-like and hydrophilic edge-like regions within the same plane.T his complexs urface can facilitate the adsorption of some proteins onto the bare GC surface, [16,56] but much of the recent literature using GC electrodes for direct immobilization of redox proteins describes the functionalization of the GC surfacew ith nanomaterials, such as carbonn anotubes (CNTs), [57][58][59][60] carbon black, [61] and even silicon dioxide nanoparticles. [62]

Carbon nanomaterials
There are two classes of CNT:s ingle-wall carbon-nanotubes (SWCNTs )a nd multi-wall carbon nanotubes (MWCNTs). [63,64] SWCNTsh ave ac ylindrical nanostructure, and can be thought of as as ingle graphite sheet rolled up into at ube, [63] whereas MWCNTsc omprise severall ayers of SWCNTsc oncentrically arrangedl ike rings in at ree trunk. [63] The ability of CNTst om ediate directE Ti sa ttributed to the combination of high surface area, high conductivity,a nd the polarities of the surfaces they present;t he side walls of CNT likelyh ave properties similar to those of the basal plane of HOPG,w hereas the ends of the tubes likely have properties akin to PGE. [63,65] The walls of these nanotubes are capable of forming strong p-p interactions to small molecule species, such as pyrene. [64] There are av ariety of methods for structuring CNT/redox protein assemblies on electrode surfaces, and these methods have been comprehensively reviewed. [63,66,67] Simple approaches include the evaporation of ad ropleto fr edox protein/CNT dispersion onto aG Celectrode surface, followed by the addition of as mall amount of Nafion membrane to act a binding agent, [60] or the filling of microcavities in the bulk electrode surfacew ith CNTs. [68] Such methodsh ave resulted in facile direct ET being establishedb etween the electrode surface and ar ange of proteins, including hemoglobin, [60] horseradish peroxidase [60] and, remarkably,g lucoseo xidase; [60,68] a protein for whicht he establishment of direct ET is infamously difficult owing to the coenzymef lavin adenine dinucleotide (FAD) unit of GOx being deeply embedded within the protein structure. [69] More advanced techniques, such as the construc- www.chemeurj.org tion of "CNT forests" (i.e.,s hort SWCNTsa rranged orthogonally to an electrode surface by self-assembly [70] )p rovide high surface area assemblies into which redox proteins can be spontaneously incorporated, [65] or atop which redox enzymes can be covalently wired. [71] Graphene can be formulated as ah ighly conductive carbon nanomaterial, that can also be used to make electrode surfaces amenable for PFE. The attachment of graphene to as upporting electrode can be achieved throughs imple electrode treatments,s uch as the application of graphene suspensions to GC, which promotes the formation of as table thin film owing to p-p stacking interactions. [72] Alternatively,c omposite mixtures of chitosana nd graphene can be applied to carbon electrode surfaces as thin films which promote the physisorption of redox proteins. [73][74][75][76] Ar eview of the uses of graphene in electrochemical sensors and biosensors has been compiled by Shao et al. [77] The functionalization of electrode surfaces with highc onductivity carbon black (CB) nanomaterials, such as Ketjen Black powder, [78] can also promote direct ET to redox proteins or enzymes. [78,79] The affinity between CB and protein surfaces has been attributed to hydrophobic-hydrophobic interactions, high porosity,a nd high surface-area-to-volume ratio. [80] The electroactive immobilization procedurei so ften performed by evaporation of suspension/slurries of CB particleso nto carbon electrodes. [61,78,79,81] More complex hybrid bio-synthetic catalytic systems can be generated by combining CB particles with redox enzymesa nd other nanoparticles. For example, Matteo Duca and co-workers showedt hat an itrate reductase from E. coli could be immobilizedo nto carbon black, and the co-deposition onto aP GE surface of these bio-modified particlesa long with Pt or Rh nanoparticles yieldedasystem capable of the electrocatalysis of nitrate to ammonia at neutralp H. [82] In the absence of enzyme, the slow reduction of nitrate by the noble metal catalysts alone significantly limitedt he rate of denitrification, whereas the enzyme-containing system may be applicable for wastewater treatment. [82]

Metal oxide semiconductors
Electrodes constructedo fm etal oxide semiconductors have become increasingly important in both PFE studies and metalloenzyme biotechnological device development. In particular, n-type metal oxide semiconductors such as TiO 2 , [83] indium tin oxide (ITO) [84] and CdS [85] were used fors olarf uel applications [24] and NADH recycling. [25] TiO 2 electrode surfaces are rough, porous structures consisting of aggregated nanoparticles. [24] TheC dS surface topology is similar,c omprising ah ighly poroust hree-dimensional network of CdS sheets. [24,46] ITOe lectrodes with porousa rchitecturess uitable for redox-protein immobilization can also be constructed [84,86] and, along with PGE [24,45] and TiO 2 , [87,88] present negatively chargedo xide functionalitiesf or adsorbing protein or enzyme moleculesa tn eutral pH. [43,89] The rough/porous natureo ft hese electrode materials is thought to aid in electroactivee nzyme immobilization, as described for PGE. [46,86] Indeed, PFE of aH 2 -producing [FeFe]-hydrogenase from Clostridium acetobutylicum was re-cently demonstrated using aT iO 2 electrode, [90,91] whereas previously immobilization of [FeFe]-hydrogenases on native electrode surfaces hado nly been achieved using rough carbon electrode substrates, such as PGE [92,93] or carbon felt. [53] Unlike PGE, ITO is transparent and the porosity of such metal oxide electrode surfaces is also readily tunable. [84,86] An especially high-surface-areah ierarchically structured ITOe lectrode with am icroporous inverse opal architecture and am esoporous skeleton was recently developed by Reisner and coworkers. [84] Immobilization of high quantities of the enzymes photosystem II and a[ NiFeSe]-hydrogenase onto ap hotoanode and acathode, respectively,yielded aphotoelectrochemical solar-water-splitting enzymec ell (Figure 7). [84] This device is capable of yielding al ight-to-hydrogen conversion efficiency of as much as 5.4 %. [84] Alternatively,u sing photosystem I, cytochromec and human sulfite oxidase, Lisdat and co-workers have demonstrated the possibility of using ITO as as upport for light-driven bio-sensing redox enzymedevices. [94] As with PGE, nonspecific adsorption of protein to as emiconductor can be facilitated by considering the effect of pH. For example, the isoelectricp oint (pI) of aT iO 2 surface was found to be 6.2, [95] whereas the pI values of ac arbon monoxide dehydrogenase [96] and a[ NiFeSe]-hydrogenase [83] were found to be 5.5 and5 .4, respectively.B oth enzymes could be adsorbed to TiO 2 nanoparticlesa tp H6, [83,97,98] and this has been rationalized by considering that under these conditions the net surface chargeo ft he enzymes is negative whereas that of the TiO 2 is positive. Similarly,t he work of Emmanuel To poglidisa nd co-workers [95] has shown that the adsorption to TiO 2 of cytochromecand hemoglobinw as greater at pH 7t han at pH 6. [95] Likewise, this was explained by considering that at pH < 7.5, the proteins would be positivelyc harged whereas the TiO 2 surface is negatively charged at pH 7b ut not at pH 6. [95] Figure 7. The water-splitting photoelectrochemicalcelldeveloped by Erwin Reisner and co-workers, utilizing photosystemIIand hydrogenase enzymes immobilized onhierarchically structured ITO electrodes. [84] 3. Common electrode functionalization strategies to promote electroactive surface adsorption In this section we outline surface functionalization strategies that make electrode surfaces amenable to electroactive redoxprotein and redox-enzyme adsorption. The generalm erit of all such electrode modification strategiesi st hat they do not require changes to be made to the protein structure. Instead, the surface-protein interactions should ideally mimic those which underpin electron exchange between the biological molecule and its redox partner(s) in vivo. Covalent bonding strategies that aim to make single, site-specific linkages between electrodes and proteins or enzymes will be discussed in Section5.

Thiol self-assembled monolayerso ngold
As ignificant amount of literature describes the immobilization of redox proteins onto surface-modified gold nanoparticles and surface-modified macroscopic gold surfaces. [43] The requirementf or surface modification does not arise because proteins cannot bind to gold surfaces;c omputational evidence suggestst he alcoholm oieties of serinea nd threonine amino acid residues can bind to crystalline Au (111)s urfaces. [99] The problem is that such interactions can induce protein unfolding. [43] The functionalization of gold surfaces with alkanethiol based self-assembled monolayers (SAMs) is thus commonpractice as it offers the dual opportunity to both mask the gold atoms [100] and presentareactive headgroupi nto solution that will induce orientation of the protein in an electroactive configuration ( Figure 8). [101][102][103][104][105][106][107][108][109] SAM formation is generally achieved by immersing ac lean gold substrate into ad ilute solution of the desired thiol in ethanol, whereupon the thiol functionalities chemisorb to the gold, spontaneously forming SÀAu bonds. [110] The "self-assembled" nature of the monolayer arises owing to the hydrophobic effect which drives the spontaneousv erticala lignment of the alkane chains, yielding au niform monolayer of densely packed alkanethiols (Figure 8). [111] Theg old cleaningp rocess is necessaryt or emove any oxide coating and/or adsorbed organic moieties on the gold surface. [112] Metals other than gold also form strong-enough thiol bonds to enable alkanethiol SAM formation.T his is relatively trivial for palladium, silver,a nd mercury that, like gold, do not form stable oxide layers. [113,114] However,i ti sm ore challenging to form high-quality SAMs on copper, [113] anda ccordingly such surfacem odificationsa re more poorly understood than those constructed on other coinage metals. [111] The biggest limitation for using gold-thiol based SAM systems in redox protein/enzyme electrochemical applications is that they have al imited electrode potential window over which they are stable. This window has been reporteda sb etween À0.9 and + 1.0 Vv ersus standard hydrogen electrode (SHE) at ambient temperature, [112,115] but am ore conservative estimate further limits this range to between À0.4 and + 0.6 V versus SHE. [43] At as ufficiently negative potential, reduction of the gold-thiol bond causes the SAM to detach from the surface, whereas over-oxidization leads to SAM detachmentattributed to the generation of sulfur oxides. [43] This inherentS AM redox activity prevents the use of gold-thiols in some bioelectrochemical applications, [43] forexample the constructionofen-zymaticC O 2 reduction or H 2 Oo xidation systems. SAMs also often have poor long-term storage stability,o wing to air-induced oxidationo ft he metal-thiolateb ond. [112] As exemplified below,t his has not prevented the use of Au-SAMs in as ignificant number of analytical bioelectrochemical studies, but potentially introduces the requirement for more stable electrode modification routes for the development of commercial technological devices.

Single-component SAMs on gold
Alkanethiol SAMs are frequently used to tailor the polarity of a metal electrode to complement that of the target protein, mediating immobilization through non-specific interactions, as described in Section 2. [102][103][104][105][106][107][108][109]114] Azurin, ab lue type-I copper protein, has been immobilized as monolayers or submonolayers using simple SAMs of different length, such as pentanethiol [103] and decanethiol. [102] Such non-functionalized (i.e. alkane headgroup) alkanethiolsa re thought to facilitated irecte lectron transfer between ag old electrode and azurin because the protein has ap atch of hydrophobic surfacer esidues proximal to the redox-activec opper center. [116,117] The stabilityo fa zurin on such alkanethiol SAMs has been put to particularly good use in the quantification of kinetic and thermodynamic dispersion, through the coupling of fluorescence monitoring of the copperr edox state with electrochemical control of the redox potential. [118,119] The immobilization of proteins which interact well with negativelyc harged PGE electrodes has been achieved throught he use of carboxylic-acid-terminated SAMs. [105][106][107][108][109]120] This has been probedi nd etail using cytochrome c, ap rotein thought to transfer electrons throughi nteraction with redox partners that are attracted to the positively chargeds urface lysine moieties close to the redox-activeh aem group. [121,29,37] When aS AM with SO 3 Hh eadgroupsw as used insteado faCOOH-terminated SAM, electroactive electrode immobilization was still achieved. [122] Cytochromec has also been used in experimentst o probe the impact of alkane chain length on the rate of electron transfer, k ET ,b etween ag old electrode and ap rotein sitting atop aS AM.W hen shorter alkanethiols( 6c arbon atoms) are used, k ET is independent of the alkane chain length, indicat- ing that the electrochemistry is reporting on the inherent maximum rate of the Fe 3 + + 1e À ÐFe 2 + biological redox process of interest. [123][124][125] However, k ET decreases exponentially with the length of the alkanethiol when molecules of more than nine carbon atoms are used. [123][124][125] This indicates that the tunneling of the electron through the SAM has become the rate-limiting step in electron transfer. [123][124][125] For redox proteins bearing negative surface charges closet o the electron entry/exit site, such as plastocyanin or ferredoxins, [32] amino-terminated SAMs can support direct electron transfer in as imilarw ay. [37] Alternatively,a sw ith PGE (Figure 6), the treatment of acid-terminated alkanethiol SAMsw ith poly-llysine allows for electroactive immobilization of negatively chargedproteins such as cytochrome b5, [126] avidin, [127] and glucose oxidase, [128] with the cationic poly-amine again acting as an electrostatic "glue" betweent he negative protein and SAM surfaces. [126] Alkanethiolsa re not the only molecules which can be used for the formation of SAMs that support electron transfer to an immobilized redox protein or enzyme. Short peptides have been used to form SAMs that permitt he electrochemical assay of cytochrome b562 from E. coli [129] and am ethane monooxygenasefrom Methylococcus capsulatus. [130]

MulticomponentSAMs on gold
Mixing two or more different alkanethiol molecules together enablest he formation of multicomponent SAMs. For example, myoglobin has been stabilized by forming am ulticomponent SAM using alkanethiols with OH headgroups and alkanethiols with COOH headgroups. [104] In certain electroanalytical applications, mixed SAM systemsm ay prove superior to single-component SAM modifications. The standard rate constantfor electron transfer, k ET ,t oc ytochrome ci mmobilizedo nam ulticomponent SAM of composition8 :2 mercaptoundecanoic acid (MUA) to decanethiol was about fivet imes greater than that on asingle-component SAM of MUA at pH 7. [131,122] This was attributedt ot he notion that deprotonation of the headgroups of aS AM formed from just COOH-terminateda lkanethiolsi ntroducess uch ah ighc oncentration of negative charge on the surfaceo ft he electrode that immobilized proteins are induced to adopt an orientationt hat is not optimized for rapid electron transfer. [122] In ap articularly elegant example of biological mimicry, the incorporation of further self-assembling layers on top of multicomponent SAMs can be used to fabricate structures that mimic biological membranes (Figure9). [132,133] Such electrodeconfined tethered bilayer lipid membranes are constructed by first creatingamulticomponent SAM using am ixture of specially designed lipid tethers and small alkanethiol molecules, such as 6-mercaptohexanol ( Figure 9). [132,133] Owing to the mismatchesi nc hain length and polarity between these two species, they form nanoscale phase-separated domainso nt he gold surface.T he lipid tethers bind to the electrode surface through an Au-Sb ond, while their headgroups (often cholesterol lipids) induce the self-assembly of phospholipid layers on top of them. Phospholipid bilayers are formed to span across the alkanethiol spacerd omains that sit between the lipid tetherd omains, and transmembrane proteins can be embedded into these bilayer regions ande lectrochemically interrogated, often throught he mediation of electron transfer by quinone molecules that are incorporatedi nto the bilayer,s uch as ubiquinone (Figure 9). [132,133] This technique has been applied to study proteins ranging from the relatively small cytochrome bo3 from Escherichia coli [132] to the very large [NiFe]-hydrogenase of Ralstoniaeutropha. [133,134]

Long-length conducting SAMs
As described in 3.1.1, slow electron transfer through longlength alkanethiols( > 9c arbon atoms) can introduce an artefact into biological electrochemistry experiments,w ith the limiting rate of the redox process reflecting the interfacial electrode-to-protein electron-transfer rate insteado ft he speed of the biological reaction. [123][124][125] This can be overcome by using more electrically conductiveS AMs. [135] For example, the use of aS AM containing ah ighly conjugated diarylethene moiety for modification of ag old electrode enabled fast electron transfer to the small blue copper protein azurin. [135] The redox chemistry rate constant was higher (3-27 times faster) than obtained when using SAMs formed from alkanethiols of as imilar length. [135] Alternatively, redox-actives o-called electron transfer "mediator" units can be built into SAMs. An example of such ac on-ductingS AM precursor molecule is 1-(10-mercaptodecyl)-1'benzyl-4,4'-bipyridinium dibromide, which was synthesized for immobilization of aH 2 -producing [FeFe] hydrogenase. [136] Unfortunately the enzymatic activity was only approximately 2.5 %o ft hat expected based on solution-state experiments, illustrating the complexityi no ptimizing such aS AM-enzyme system. [136]

Aryl diazonium salt reduction
The reductiono fa ryl diazonium salts for the functionalization of electrodes has been demonstrated on av arietyo fd ifferent materials including all conducting allotropes of carbon, [137][138][139] silicon, [140] ITO, [141] andarange of metals including gold, platinum, and copper. [142] As urface-to-carbonb ond is formed via the one-electron reductive formation of an aryl radical, which subsequently attacks the electrode surface, as illustrated in Scheme 1. [137,143,144] Electrode functionalization using aryl diazonium salts is therefore electrochemically controllable (Figure 10), and can often be performed in aqueous or organic electrolyte. [137,143,144] Either isolated aryl diazonium salts can be utilized, or they can be generated in situ using an aniline or nitrophenyl derivative and as ource of the NO + cation, such as NaNO 2 /HCl or NOBH 4 (Scheme1). [145,146] Given the range of commercially available aniline and nitrophenyld erivatives, the scope of chemical functionalities that can be introduced onto the surface using diazonium chemistry is comparable to that which can be accessedu sing commercial alkanethiol derivatives for SAM formation.U nlike SAM formation, this methodology is theoretically applicable to the covalent functionalization of any conducting surface, not just those that form as table bond to sulfur.T he redox stability of the electrode-carbon bond does not restrictt he electrochemical window of biological experiments,a nd such surfacem odifications are also more amenable to long-term storaget han SAMmodified gold surfaces. [144,[147][148][149][150] Diazonium electrode modification is not entirely without challenges. Multilayer formation can occur when further aryl radicals attack the unsaturated bonds of the aromatic p systems of the originalm onolayer,r esulting in carbonÀcarbon bonds. [144,[151][152][153] Alternatively, multilayersc an arise from diazonium cations coupling to surfacep henylg roups througha zo bond formation. [144,[151][152][153] Both modes of multilayer formation can contribute to the build-up of an amorphous, organic, insulating layer on the surface of the electrode. [144,[151][152][153] Methodologies to prevent or minimalize multilayer formation have been reported, such as the use and subsequent cleavage of bulky protecting groups, [154] sterically hindering the 3,5-positionso f the aryl diazonium salt, [155] and addition of the radical scavenger 2,2-diphenyl-1-picrylhydrazyl (DPPH) to quenche xcess aryl radicals. [152,153] In the contexto fb ioelectrochemistry,d iazonium electrode modifications can be used to induce protein adsorption through non-covalent interactions in as imilar manner to that achieved using unmodified PGE or SAMs on gold. Table 1s ummarizes some examples that have utilized different diazonium electrode functionalization methods. Thef lexibility of the method is illustrated by the literature precedence of the use of the same diazonium-protein immobilization strategy on a range of different electrode surfaces to immobilize ar ange of redox proteins.T he coupling of dialdehydes to aryl amine groups,i ntroduced through diazonium cation electrografting ( Table 1, entry d) was used to immobilize several redox enzymeso nb oth carbon [156,157] and gold [158,159] electrodes.
Scheme1.Electrochemical reduction of aryl diazonium salts resulting in the formation of ap acifying multilayer film.

Covalent coupling of electrodeston ative proteins
Attempts to physically adsorb proteins onto surfaces in an electroactive configuration are not alwayss uccessful and, as describeda bove,e ven when they do work, the adsorption strategies may be strongly dependento nt he pH of the electrolyte solution. Alternatively,t he film of molecules may only be transiently stabilized, with either misfolding [160] or possible desorption processes leading to as teady decreasei nr edox activity.T oa void such problems the covalenta ttachment of proteins to electrode surfaces is desirable, particularly in biotechnological device development. Such ac ovalent coupling approach often requires ac omplementarys urface functionalization strategy,s oe ither thiol self-assembly or diazonium modification is often used to introduce surfaceg roups that will react with protein moieties. [154][155][156][161][162][163][164]

Entry
Surface functionalizations trategy [a] References a) i) R = COOH [167] ii)R= CH 2 COOH [168] iii)R= CH=CHCOOH [169] iv) R = NO 2 [69,[170][171][172]175] v) R = NH 2 [176] b) [69,[170][171][172]175] c) [69] d) [156][157][158][159] e) [156] f) [ with electrochemical reduction being used to reduce the nitro "headgroups" into the desired amine functionalities in ap ostdiazonium crosslinking step. [151,158,159] The introduction of more reactive alkyl amine groups to electrode surfaces through diazonium modification can be achieved through the use of the 4-aminoethylbenzenediazoniumc ation, [178] or phthalimide-protected alkylamine functionalities. [154] Amide bond formation strategies have been used in the fabrication of many mediator-free biosensors;E DC/NHS-activated tyrosinase was crosslinkedt oa minophenyl groups on BDD electrodes and used to detect phenolic compounds. [170] The EDC/NHS activation and crosslinking of horseradish peroxidase or cytochrome P450 enzymes to amine moieties on carbon electrodes has been used in the fabrication of biosensors for the detection of as eries of pharmaceutically relevant drugs. [157,171,172] Horseradish peroxidase could be used to sense levetiracetam, [171] and the specific cytochrome P450 enzymes could be used to detect phenobarbital [157] and codeine. [172] The immobilizationo fa no xygen-tolerant hydrogenase onto pyrene-modified multiwalled carbon nanotubes coated onto PGE electrodes was also achievedt hrough EDC/NHS coupling. [174] The resultant derivatized PGE electrode was utilized as the anode in the fabrication of an enzyme H 2 /O 2 fuelc ell, which resulted in significantly improved current density and stabilityw hen compared to af uelc ell containing ah ydrogenase electrode fabricated using simple adsorptionp rocedures. [174] The most significant limitation of such approaches is that regardlesso fw hether carboxylic acid residues or lysine groups are targeted (multiple occurrences of such amino acid side chains on the surface of the redox protein or enzyme of interest are often present), significant dispersion in the orientation of the immobilizedb iomolecule commonly results( Figure 11). Careful genetic engineering of the target protein can overcome this problem.Arecent publication by Lalaouie tal. [179] reports the site specific immobilizationo falaccase onto CNTs through the generation of av ariant enzyme that only contains as ingle surface-accessible lysine residue that is located proximal to the electron entry/exit type 1c opperc enter. [179]

Imine tethering
Redox proteins or enzymes can also be covalently crosslinked to surfaces through imine bond formation between electrodesurfacea ldehyde moieties and protein-surface lysine residues (Scheme 2). For example, diazonium electrografting methods have been used to introduce aldehyde functionality onto electrodes( Ta ble 1, entry d) that have subsequently been modified with enzymes, including acetylecholinesterase [156,180] horseradish peroxidase, [175] and tyrosinase. [158] Analogously,t he reaction of glutaraldehyde witha mine terminated SAMs yields an aldehyde-functionalizeds urface that can be used to attach proteins through their surfacel ysine residues. [164,181,182] To generate more stable covalentl inkages, the imine bonds can be reduced to aminelinkages using reagents such as sodium cyanoborohydride. [183] As with amide bond formation between an electrode and surface-lysine residues on ap rotein or enzyme, the same limitation remains;c rosslinking electrodes to lysine residues that are not within close approach of the electron entry/exit site in ap rotein or enzymew ill not yield electroactively bound biomolecules. Additionally,t he presence of multiple surface lysine residues couldr esult in dispersion in the orientation of the protein or enzyme on the electrode surface, as illustrated in Figure 11. Figure 11. Amidebondformation betweens urfaces and protein residues, catalyzedthrough EDC/NHS activation.a )Activation of electrode-surface carboxylic acid groups and reaction with proteinl ysine residues. b) Activation of carboxylic acid groups on the protein surface and reaction with electrode surface amine groups. c) Owing to the presence of many amine/carboxylic acid moieties on proteins urfaces;immobilization through EDC/NHS activationoften leads to ad ispersion in immobilized protein orientation.

Crosslinking strategies for site-specifically connecting proteins to electrodes
In theory,a ne xcellent method to generate au niform configuration of proteins or enzymes on as urface, with each biomolecule attached through the same single point, is to develop site-selective covalentc rosslinking strategies, as illustrated in Figure 12. This is often ac omplex process which usually requires ac ombination of genetic manipulation and surface chemistry to ensure that there is as ingle amino acid residue suitable for selectiver eactionw ithac omplementary surface moiety.T he advantage of modifying electrodes rather than non-conductive solid substrates is that redox-activated reactions such as diazonium salt electroreduction can be utilized in the surface chemistry (Section3.2). However,t his is tempered by the disadvantage that for direct electron transfer between the electrode surfacea nd ar edox protein or enzyme to be feasible the target amino acid reaction site mustb es ufficiently close to the electron entry/exit site (Section 2).

Redox-center targeted binding
The easiest way to avoid the need for geneticm anipulation of the target redox protein or enzyme is to devise an electrode binding strategy that anchors the biomoleculet ot he conducting surface through an on-amino acid functionality.Ano bvious choice of center for such linking strategies is the electron entry/exit redox-active cofactor of the protein/enzyme, since anchoring to the electrode surface through such ag roup will ensure that the biomoleculei sc rosslinked to the electrode in an electroactivec onfiguration.W ed escribe an umber of approaches that have used this understanding of biological structure and function to rationally design bespoke wiring strategies for attaching proteins or enzymes to electrodes. The most obviouslimitation of such cofactor-targeted surface binding strategies is that biology utilizes aw ide range of different redox-active cofactors, as illustrated by Figure 1. Anchoring different classes of redox proteins or enzymes through ar edoxcenter-targeted binding strategy therefore requirest he design and optimization of many differentc hemical strategies:anontrivial synthetic task. In the case of enzymes such as lytic polysaccharide monooxygenases, [11] the fact they contain as ingle redox site where the substrate must bind also introduces the challenge of whether linkersc an be designed that do not hinders ubstrate binding and catalysis.

Cofactor ligation
In some instances, redox-active cofactors can be synthesized and incorporated into so-calledc ofactor-free "apo-proteins." This offers ar oute to generating redox proteins containing modified redox cofactors that have chemical functionalities complementary to those which can be added to the electrode surface. For example, incorporation of an azide-functionalized heme group into cytochrome b562 enabled copper(I)-catalyzed azide-alkyne cycloaddition to an alkyne-functionalized CNT immobilized onto aG Ce lectrode. [184] Alternatively,f or proteins containing metal-electron entry/exit sites that have multiple ligands,g enetic removal of an amino acid ligand residue offers the opportunity for structuralr econstitution of the redoxp rotein with an external ligand that is tethered to the electrode surface. This strategy has been demonstrated for an azurin variant. [188] The copper-coordinating histidine residue (His117) was replaced with ag lyciner esidue throughg enetic manipulation. [188] This openedu pt he coordination sphere aroundt he redox-active metal anda llowed ap yridine headgroupt ethered to an electrode surface to coordinate directly to the copper center, immobilizingt he azurin in an orientation suitable for facile direct electron transfer and mimicking the native copper ligationt hat is afforded by His117. [188] Surfacea ttachment of the pyridine group was enabled by the synthesis of at hiol-terminatedl inker that covalently bound to gold surfaces. [188]

Substrate electrode tethers
With redox enzymes that internally transfer electrons from the oxidation of as ubstrate in one binding pocket to reduce a second substrate in as econd binding site, surfacea ttachment can be achieved based on the "lock andk ey" model [189] of siteselectivee nzyme-substrate binding.F or example, multicopper oxidases (that have evolvedt oc ouple organic-substrate oxidation at one copper site to oxygen reduction at another copper site) can be immobilized for use as Pt-free, low-overpotential O 2 -reduction electrocatalysts throught he use of surface-attached organic-substrate mimics ( Figure 13). [185,186] The highly conjugated natureo fa na nthracenee lectrode linker,i mmobilized onto graphite using diazonium chemistry,w as shown to ensure rapid electron transfer from the electrode surface to laccase. [185,186] Using as imilars trategy,s urface naphthoic acid moieties were effective in the immobilization of bilirubin oxidase from Myrothecium verrucaria. [190] O 2 reduction by this enzyme was externally wired to ah ydrogenase-coated electrode to construct an all-enzyme, membrane-free H 2 /O 2 fuel cell where H 2 -oxidation is used as the source of electrons for O 2 reduction ( Figure 13). [187] In as imilarv ein, the surfaceb inding of DNA is used to immobilize redox proteins for the electrochemical interrogation of the redox reactions that may underpin DNA translation and repair in vivo. [191]

Cysteine-based surface ligation
As the sole thiol-containing canonical amino acid, cysteine presentsaunique chemical functionality that can be harnessed in the design of biochemical ligationmethodologies that selectively target only cysteiner esidues. [192][193][194][195] This chemical selectivity is complemented by the fact that, relative to other amino acids, cysteines are rarely presento np rotein surfaces. [196] Thus, it can be relativelyt rivial to use site-directed mutagenesis and chemicalb iology conjugation methodst oe ngineer proteins and enzymes with single, covalently modified surface-cysteine residues [192][193][194][195] Such strategies are of enormous value in the development of new biopharmaceutical therapies. [192,195] Surface-attachment strategies have been developed along similar lines, with the added consideration that for direct electron transfer between ac onducting surfacea nd ar edox protein or enzyme, the cysteine residuem ust serve as at ethering site that holds the redox protein/enzyme in an electroactiveo rientation. [43,89,197,198] The most significant limitations to the use of cysteine residues for enzyme electrode "wiring" applications arise from the potentialf or these residues to form intermolecular disulfide bonds, [199] or to cause misfolding through the formation of non-native disulfide bond formation, [200] or through the accidental introduction of an extra metal-ligand residue to am etalloprotein or enzyme. For example, iron-sulfur cluster incorporation into ap rotein structure is dependent on metal cluster binding to ah ighly conserved sequence of cysteines, [1] and additiono fe xtra residues can be used to convert a[ Fe 3 S 4 ]c enter into [Fe 4 S 4 ]. [201] To avoid the issue of unwanted disulfide bond formation,p roteins displaying free surface-cysteine residues can be kept under reducingc onditions through addition of dithiothreitol (DTT).H owever,b ecause DTT contains thiol groups, this reducing agent must be removed before surface bioconjugation is attempted, to avoid unwanted reactions between DTT and the electrode surface. [202] 5.2.1 Directi mmobilization onto gold As described in Section 3.1, the formation of SAMs onto gold electrodes is facilitatedb yt he generation of gold-sulfur bonds.A na nalogousa pproachi st otherefore graft surfacecysteine-containingproteins or enzymesdirectly onto gold surfaces. [203,204] The ability of this approach to immobilize redox proteins in chosen orientationsh as been definitively demonstratedu sing ac ytochrome b562 engineered to presentc ysteine residues on either the long axis or short axis. [203] The resultant orthogonal orientations of these different protein variants on atomically flat gold was then observed using STM imaging. [203] As with SAMso ng old, as ignificant limitation of this methodi st he redox instabilityo fs uch covalente lectrodeprotein modifications( Section 3.1). The fact that electroreduction can be used to break AuÀSb onds at relativelyh ighr educing potentials impedes the use of this methodology for studying important biofuel reactions such as hydrogen production.

Thiol-Michaela dditionclick reactions
In recent years, the reaction between cysteineresidues and unsaturated p systems through Michael addition has been used as ag eneral tool for the chemical modification of many proteins, extendingw ell beyonde lectrode-protein-surface ligation strategies. [205][206][207][208] Numerous different methodologies using different p systems have been optimized for differenta pplica- Figure 13. Membrane-free H 2 /O 2 fuels cells canbef abricated by coupling the redox activity of hydrogenasestoo xidases. [185][186][187] The orientation of the multicopper oxidaseT rametes versicolor laccase III (PDB code:1 KYA) onto an electrode surface for O 2 -reduction catalysis can be achieved through the modification of the electrodesurface with anthracenes ubstrate mimics, thereby anchoring the enzymeb yt he bindingp ocket and allowing facile direct electron transfer. [185,186] Chem. tions. [205] In the field of electrode ligation, surfaces have been functionalized with maleimideg roups,t he most reactive of the commonlya vailablev inyl Michael acceptors. [205] Between pH 6.5 and 7.5, maleimideg roups react selectively with thiols, as within this pH range aminesr emainp rotonated and are thus not of ah igh enough nucleophilicity to partake in competing side reactions (Scheme 3). [198] Maleimide groups can be introduced onto electrode surfaces using av ariety of techniques, including the use of speciallyd esigned alkanethiol SAMs, [206,209] diazonium cation electrografting (Table 1, entry f), [177] and sequential electrochemicala nd solid-phase preparation. [198] The reactionb etween surface-maleimide groups and one of the two thiol groupst hat naturally occur near the heme cofactor of cytochrome cr esults in the immobilizationo ft his protein in an ear-site-specific orientation that is suitable for direct electron transfer (Scheme 3). [177]
For UAA mutagenesis to serve as an immobilization methodology,f unctionalities complementary to those of the UAA must also be introduced to the electrode surface. Azide-alkyne cycloaddition click reactions therefore represent an attractive approachs ince methods for the introductiono ft hese moieties onto electrode surfaces have been developed for other applications such as DNA sensor development. [219] Should the use of ac opper catalystf or activation of the cycloaddition reaction be undesirable, copper-free reactions can be performed through the use of ring-straineda lkynes (Figure14). [220] Surprisingly,b ased on the robust nature of this chemical ligation strategy, [221] the only known example using azide-alkyne UAA reactions for the site-specific linkage of ar edox protein or enzymet oa ne lectrode is the immobilization of the 4-azido-lphenylalanine (1,Figure14) containing laccase from Streptomyces coelicolor onto aM WCNT-coated electrode functionalized with complementary cyclooctyne containing linkers ( Figure 14). [216] Interestingly,t he most effective orientation for direct electron transfer was found to be one that tethered the laccase at as ite distal from any redox centers but adjacent to aw ater channel;t he structured water molecules are thought to substantially enhancethe electron transfer rate between the Figure 14. Azide-alkyne cycloaddition click reactionsbetween surfaces and proteins. a) To p: copper-free non-catalyzed azide-alkyne cycloaddition click reaction, promoted by ar ing-strained alkyne. Bottom:copper-catalyzed azide-alkyne cycloaddition reaction.b )Owing to the precedence for unnatural amino acids bearing both alkyne and azide functionalities, it is possible to functionalize ap rotein with either azides or alkynes. c) The site-selectiveelectroactive immobilization of al accase onto aM WCNT using acopper-free non-catalyzedazide-alkyne cycloaddition click reaction between an azide-functionalized UAA and asurface-confined cyclooctyne, as described in ref. [216]. Residues from only one monomer are depicted (PDB ID:3CG8 [222] ).
Scheme3.Maleimide-thiol Michaela ddition reactionsb etween maleimide groupsi ntroducedo nto an electrode surface and cytochromecsurfacecysteine residues. [177] Chem. Eur.J.2018, 24,1 2164-12182 www.chemeurj.org 2018 The Authors. Published by Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim electrode and thel accase. [216] As imilar strategy has also been used to "wire" whole bacteria to electrodes. [217] Through the incorporation of UAA 1 into an alcohold ehydrogenase that is displayed on the surface of E. coli, [217] copper(I)-catalyzedc ycloaddition to an alkyne functionalized SAM linker was used to bond bacterial cells to agold surface. [217] Other,n on-azide-alkyne chemical ligations trategies can be realized throught he use of different UAA residues.T he incorporation of 3-amino-l-tyrosine( NH 2 Tyr) into myoglobin has been used to covalently attach the protein ontoagold surface derivatized with acryloyl moieties, courtesy of aD iels-Alderr eaction specific to NH 2 Tyr ( Figure 15). [218] The rate of electron transfer between the electrode and myoglobin was slow, which was attributed to the length of the anchoring tether (26.7 ). [218] 6. Summary and outlook Although highly informative reviewsh ave been written on the powerful bio-analysist echnique of protein film electrochemistry,t he fundamentals tep of protein film formationi so ften overlooked. This is understandable;t oe stablish the technique of PFE it has been necessary to provides ubstantial insight into electrochemical methodd esign and dataa nalysisa pproaches, as well as showcase the powerful insight which can be gained from conducting such experimentso nb iologically and biotechnologically important systems. To complement such papers, this review aims provide an up to date and broad ranging overview of the many different surfaces, surface modification strategies andp rotein conjugation approaches which can be used to "wire" redox-activemacro-biomolecules to electrodes.
Owing to the diversity of redox proteins and the wide range of possible usages, there are currently no universal surfaceconfinement approaches. By comparing and critiquing the different approaches we hope to provide the readerw ith ao nestop reference library that will aid selection of appropriate electroactive surface immobilization techniquesf or use in studying/harnessing new redox proteins and enzymes, or introduce them to this diverse field. For gifted bio-conjugation chemists, we hope to inspiret he need for further method development,h aving emphasized that we still lackaway to generate robust, site-selective bonds from anyp rotein or enzyme to any electrode.