Redox Properties of Electron Transfer

The factors that affect the potentials of electron transfer Adrian W. Bott, Ph.D. Bioanalytical Systems, Inc. metalloproteins are discussed, together with specific examples. 2701 Kent Avenue West Lafayette, IN 47906-1382

E-mail: [email protected]

The ability of transition metals to thermodynamic data (i.e., redox po- in spectroelectrochemical experi- exist in more than one stable oxida- tentials). However, early voltammet- ments; instead, small redox-active tion state makes them suitable cata- ric studies of metalloproteins were (mediators) are used to lysts for biological processes that not successful (1,2). This lack of suc- transport electrons between the elec- require transfer of electrons. Conse- cess was attributed to a number of trode surface and the active site of the quently, transition metals are found factors, including adsorption and de- . Typical mediators 3+ at the active sites of a large number naturation of the proteins at the elec- include quinones (4), Ru(NH3)6 of proteins. The processes catalyzed trode surface and inaccessibility of (5), and methyl viologen (6). by such proteins can require the the coordinated metal ion (which is More recently, it has been shown transfer of both electrons and pro- typically buried in the interior of the that cyclic voltammograms of metal- tons to a substrate bound to the met- protein). Therefore, other techniques loproteins in the absence of media- al, or can simply involve the transfer were originally used to measure the tors (i.e., direct ) of an electron between proteins (i.e., redox potentials of metalloproteins, can be achieved under certain condi- as part of an electron transfer chain, including spectroelectrochemistry tions (1,2). For example, it was such as those involved in respiration and potentiometric titration. These shown that a quasi-reversible cyclic or photosynthesis). Due to the physi- involve the equilibration of the oxi- voltammogram could be obtained ological importance of electron dized and reduced forms of the ana- for cytochrome c at a gold electrode transfer metalloproteins, there has lyte in a spectroscopic cell at covered by a monolayer of adsorbed been much effort devoted to under- different potentials (the potentials bipyridine (7). It was proposed that standing their electrochemical prop- are determined either by addition of the hydrogen bonding between erties. The methods used for these appropriate chemical redox reagents bipyridine molecules adsorbed to the + studies, and the results obtained, are or by applying a constant potential to electrode surface and lysine (-NH3 ) discussed in this article. the cell). Once equilibrium has been groups on the surface of the protein Cyclic voltammetry is widely attained, the concentrations of the in the area where the active site is used to characterize the redox prop- oxidized and reduced species can be exposed hold the protein close to the erties of transition metal complexes. measured spectroscopically (e.g., electrode surface in an orientation This technique can provide informa- using a thin-layer cell (3)). It is im- that allows rapid electron transfer be- tion about the kinetics of the electron portant to note that there is typically tween the electrode and the active transfer reactions and of any coupled no direct electron transfer between site of the protein (F1). This idea of chemical reactions, in addition to the electrode surface and the analyte using functional groups (promoters)

47 Current Separations 18:2 (1999) F1 is that the group of cytochrome N N Schematic diagram of the H c is coordinated to cysteine residues interaction between the N N modified electrode surface H N+ in the protein, whereas that of cyto- and the protein that is H chrome b is not. However, in spite of required for electron N N transfer (adapted from their apparent structural similarities, H reference 1). gold N N the redox potentials for the H N+ H (III)/iron(II) couple are signifi- N N cantly different: +5 mV for cyto- chrome b5 (4), and +260 mV for N N Fe cytochrome c (5). Indeed, cyto-

N N chrome c from different sources ex-

H hibits a wide range of redox potential N N H N+ values (9). Consequently, a major fo- H N N cus of the studies of cytochromes (and other electron transfer proteins N N Cytochrome c (vide infra)) has been the investiga- tion of the factors that affect the re- F2 dox potential. In this section, only Molecular structures for cytochrome c and cytochrome b5 from cytochrome b will be discussed, since these are the and cytochrome c. two best characterized cytochromes. One method that has been used for examining the effect of structural variations on redox potentials has been the study of model compounds; that is, small molecular complexes (e.g., porphyrins (10, 11)) that con- tain some of the features of the active site under investigation. It should be stressed that model compounds can- not reproduce exactly the complex environment inside a protein—they b are used for systematic variation of b c specific structural features. For ex- ample, one difference between cyto- on the electrode surface to bind pro- a mercaptopropionate/polycation chrome c and cytochrome b5 is the teins on the electrode (using electro- complex (this complex must have a axial ligation of the iron center—in static interactions) in an orientation net positive charge, since cyto- cytochrome c, it is coordinated to suitable for electron transfer has chrome b5 has a net negative charge). one methionine residue (S coordina- been extended to other proteins (1, Redox potential values of -102 mV tion) and one histidine residue (N 2). It is important to note that the and -78 mV (Vs. NHE (8)), respec- coordination), whereas in cyto- promoters are not mediators; that is, tively, were recorded for the two chrome b5, it is coordinated to two they do not act as intermediates for methods (6). histidine residues. Comparison of electron transport. It should also be the redox potentials of tetra- noted that redox potentials measured Cytochromes phenylporphyrin derivatives with for the same protein using the two the appropriate axial coordination approaches should not be expected Cytochromes are part of a larger (using imidazoles and thioethers for to be the same, since the interactions group of proteins in which a heme the axial ligands) shows that chang- between the promoter and the pro- prosthetic group is the active site. ing from N/N axial coordiation to tein can affect the redox potential. The heme group consists of iron(III) N/S axial coordination leads to a For example, the redox potential of coordinated to a porphyrin group, shift of +170 mV (12); that is, the outer mitochondrial cytochrome b5 and the heme groups for cyto- iron(III) is desta- has been measured spectroelectro- chromes b and c (the two most bilized. Another significant property chemically using both methyl violo- widely studied cytochromes) are of these model compounds was the 3+ gen and Ru(NH3)6 as mediators, shown in F2. It should be noted that insensitivity of the axial bond and directly by cyclic voltammetry the two hemes are very similar—the lengths to the oxidation state of the using a gold electrode modified with only difference shown in the figure iron center; that is, there is little

Current Separations 18:2 (1999) 48 F3 two possible isomers with different Molecular structures of redox potentials (+230 mV for the iron sulfur prosethetic S-Cys Cys-S S S-Cys Met-80-His-18 isomer, and +150 groups found in iron sulfur proteins. mV for the His-82-His-18 isomer), Cys-S Fe Fe Fe cyclic voltammograms of this modi- S-Cys Cys-S S S-Cys Cys-S fied protein were shown by digital simulation to be consistent with a square scheme redox mechanism. Smaller variations in the redox potentials have been found in modi- Cys-S S-Cys Cys-S S-Cys fied versions of cytochrome c in S Fe S Fe which the axial coordination of the Fe S Fe S iron center is not altered (17-21). These have been attributed to Fe S Fe S changes in the following factors: the S S Fe Cys-S Cys-S S-Cys hydrophobicity of the interior of the protein, the solvent accessibility of the active site, and electrostatic and hydrogen bonding interactions be- F4 tween amino acid residues and the iron center (22-25). Based on com- Range of redox potentials 4Fe cluster in [8Fe-8S] Ferredoxins 4Fe cluster in HiPIP observed for iron sulfur 4Fe cluster in [7Fe-8S] Ferredoxins parison of the redox potentials of proteins (relative to the native cytochrome c with unfolded normal hydrogen 4Fe cluster in [4Fe-8S] Ferredoxins electrode) (adapted cytochrome c and microperoxidase from reference 31). (a proteolytic fragment of cyto- 3Fe cluster in [7Fe-8S] Ferredoxins chrome c containing the heme group 3Fe cluster in [3Fe-4S] Ferredoxins and a small number of amino acids), it has been suggested that encapsula- tion of the heme group in a hydro- 2Fe cluster in [2Fe-4S] Ferredoxins phobic environment causes a positive shift of 240 mV in the redox potential (26) (i.e., the iron(III) oxi- Rubredoxins dation state is destabilized). Since the porphyrin group in cy- -700 -600 -500 -400 -300 -200 -100 0 100 200 300 400 500 tochrome b5 is not directly bound to Eo ’ (mV) the enzyme, the heme group can be removed, modified, and then re- structural reorganization required is reflected by the shift of -240 mV placed in the active site. Modifica- for electron transfer, which facili- in the redox potential (i.e., stabiliza- tions of the porphyrin that have been tates the electron transfer reaction. tion of the iron(III) center). It was studied include esterification of the It is also possible to change spe- also shown that this modified (mu- propionate groups (21,27) (which cific amino acid residues in the pro- tant) protein could bind a substrate destabilizes the iron(III) state), re- tein to investigate the effect of this in the axial site and could catalyze placement of the propionate groups particular residue on the structure oxidative reactions of this substrate, with methyl groups (28) (which sta- and function of the protein (site spe- similar to those catalyzed by cyto- bilizes the iron(III) state), replace- cific mutagenesis). The largest chrome P450 enzymes, in which ment of the vinyl groups with alkyl changes are obtained by changing there are vacant axial sites. A change groups (28,29) (which stabilizes the the residue that is involved in axial in axial coordination can also be in- iron(III) state), variation of the ori- coordination to the iron center duced by changing a residue that is entation of the heme group in the (methione (Met)-80 and histidine not directly coordinated to the iron active site (30), and substitution of (His)-18 for cytochrome c, and His- center. Replacement of phenylalan- porphyrin by chlorin analogues (31). 63 and His-39 for cytochrome b5) ine(Phe)-82 with histidine generates (5,13-15). For example, replacement a mutant cytochrome c in which Iron-Sulfur Proteins of His-39 in cytochrome b5 with Met-80 and His-18 are coordinated methionine generates a protein in in the iron(II) complex, whereas His- There are a wide range of differ- which one of the axial sites is occu- 82 and His-18 are coordinated in the ent iron-sulfur electron transfer pro- pied by a water (13). This iron(III) state (16). Since there are teins, which vary in the number of

49 Current Separations 18:2 (1999) iron centers (32,33). Four different amide groups close to the clusters there is one Fe4S4 cluster, and one iron-sulfur structures are observed and the resulting cluster-amide inter- Fe3S4 cluster (a 7Fe ferredoxin). n n (F3):Fe(SR)4 ,Fe2S2(SR)2 , actions (36,37), including differ- Azotobacter vinelandii ferredoxin I n n Fe3S4(SR)2 , and Fe4S4(SR)4 (the ences in the number of hydrogen (AvFdI) is one example of this type bridging ligands are sulfides (for- bonding interactions (35) (it was also of ferredoxin (44). The redox poten- mally S2-), and the terminal ligands noted that the cluster environment in tials for the two clusters are signifi- are RS- (where R = cysteine for pro- HiPIPs is more hydrophobic than it cantly different (45) (-425 mV for tein-bound iron-sulfur clusters)). is in Fe4S4 ferredoxins (35)). The the Fe3S4 cluster, and -650 mV for Since iron can readily exist in the +3 variation among Fe4S4 ferredoxins the Fe4S4 cluster). or +2 oxidation state, the following has been attributed to differences in Much effort has been devoted to oxidation states are available for the solvent accessibility and cluster rationalizing the differences in the FexSy cluster core: +2, +1, 0 (Fe2S2); solvation (36-38), whereas the vari- redox potentials of the three struc- +1, 0,-1 (Fe3S4); +4, +3, +2, +1, 0 ations among HiPIPs has been attrib- tures described, and site-specific (Fe4S4). However, a given iron sulfur uted to variations in cluster solvation mutagenesis has again been exten- cluster in a protein typically only (39) and the net charge on the protein sively used. In one study, selected exhibits one redox process when (39-41). amino acids in AvFdI were changed studied by cyclic voltammetry. For It is also possible to have two to those at the same position in the Fe2S2 proteins, the two observable iron-sulfur clusters in the same pro- amino acid sequence of PaFd (38) to cluster oxidation states are +2 and +1 tein. Three different classes of these examine the effect on the redox po- (2Fe(III) and Fe(III)Fe(II)), and proteins have been identified. “Clos- tential of the Fe4S4 cluster. These Fe3S4 clusters can exist as +1 and 0 tridial” 8Fe ferredoxins contain two mutations involved substitution of (3Fe(III) and 2Fe(III)Fe(II)) (al- Fe4S4 clusters with similar ligating three negatively charged surface though the formation of “hyper-re- amino acid sequences (Cys-X-X- amino acids for neutral ones (aspar- duced” 3Fe(II) Fe3S4 clusters has Cys-X-X-Cys-X-X-X-Cys-Pro, tate→asparagine, glutamate→serine, recently been reported (34)). Two where Cys = cysteine, Pro = proline, and glutamate→alanine), one neutral different classes of iron-sulfur Fe4S4 and X = other amino acid). One surface residue for a negatively proteins have been observed; those structurally characterized example is charge residue (histidine→aspar- with oxidation states of +3 and +2 Peptococcus aerogenes ferredoxin tate), and two hydrophobic residues in (3Fe(III)Fe(II) and 2Fe(III)2Fe(II)), (PaFd) (35). The redox potentials of the interior of the protein (phenylalan- and those with oxidation states of +2 the two clusters are essentially iden- ine→tyrosine and phenylalanine→ and +1 (2Fe(III)2Fe(II) and tical at around -400 mV (33). “Chro- isoleucine). These amino acids were Fe(III)3Fe(II)). The former group is matial” 8Fe ferredoxins also have chosen because of their proximity to referred to as high-potential iron- two Fe4S4 clusters, but the ligating the Fe4S4 cluster. None of the resi- sulfur proteins (HiPIPs), whereas the sequences are different; one se- dues had any effect on the redox latter group is referred to as ferre- quence is similar to that found in potential of the Fe3S4 cluster (which doxins (the term ferredoxin is also clostridial ferredoxins (the clos- is to be expected, since none of the used to describe Fe2S2 and Fe3S4 tridial sequence), whereas in the mutations was close to this site), and clusters, whereas FeS proteins are other sequence five or six additional only one mutation (phenylalan- referred to as rubredoxins). The amino acids have been inserted be- ine→isoleucine) altered the redox range of redox potentials exhibited tween the second and third cysteines potential of the Fe4S4 cluster (by by iron-sulfur clusters in proteins is (the chromatial sequence). This about 20 mV). It was concluded from shown in F4 (33). structure is shown by the Chroma- these results that surface charges had Structural analyses of iron-sulfur tium vinosum ferredoxin (CvFd) little effect on the redox potentials, have shown that there is little vari- (42). The redox potentials of these and that the large difference between ation in the structure of the iron-sul- two clusters differ by about 200 mV the redox potentials of the Fe4S4 fur clusters of a given type in (43) (-460 and -655 mV). In the third clusters in the two proteins was as- different proteins (33,35). There- class, the ligating sequences are also sociated with solvent accessibility. fore, the wide variation in measured different; one sequence is again A larger change in the redox po- redox potentials has again been at- similar to that found clostridial fer- tential of the Fe4S4 cluster was tributed to variations in the cluster redoxins, whereas in the other se- achieved by using a cysteine→alan- environment within the protein (e.g., quence the second cysteine has been ine mutation at position 20 (45). hydrogen bonding, electrostatic in- substituted by a different amino acid, Since this residue is coordinated to teractions, solvent accessibility, and or two additional amino acids have the cluster, this mutation forces a dif- hydrophobicity). The different oxi- been inserted between the second ferent coordination geometry on the dation states of Fe4S4 ferredoxins and third cysteine. These changes peptide chain, leading to a change in and HiPIPs has been attributed to decrease the number of coordinating the redox potential from -650 to -750 differences in the configuration of cysteines from four to three; that is, mV (as in the previous example, the

Current Separations 18:2 (1999) 50 F5 referred to as “normal”). Type 3 cop- per sites contain two interacting cop- Molecular structures of (a)S (b) SMet active site of blue copper Met per centers. As discussed above, blue proteins. a) , copper proteins are one class of elec- b) azurin, c) cucumber 2.9 Å 3.1 Å basic protein, tron transfer proteins. N NHis d) stellacyanin (adapted His Blue copper proteins are differ- from reference 53). Cu Cu ent from the other two classes of 2.1 Å electron transfer proteins discussed N SCys His N S His Cys in this article in that there is no pros- 3.1 Å thetic group associated with the cop- per, and the copper center is directly O Glu coordinated to amino acid residues. A number of blue copper proteins have been characterized by x-ray (c) (d) X SMet crystallography (47,48), and these have similar copper(II) coordination

NHis 2.6 Å NHis sites (F5) (although there are impor- Cu Cu tant differences). The common fea- ture in all the copper sites is a

SCys SCys NHis NHis trigonal planar arrangement of two N ligands (each from histidine) and one S (thiolate) ligand (from cyste- ine), with strong interactions be- tween the copper center and all these

redox potential of the Fe3S4 cluster is clostridial sequence and led to a shift ligands. The copper sites from differ- not affected by this mutation). Re- in the redox potential of cluster II ent proteins differ in the number and moval of the two additional amino from -460 to -400 mV; that is, a value strength of the axial interactions (i.e., acids between the second and third similar to those found for clostridial above and below the N2S plane), and cysteine residues of the sequence that ferredoxins. However, the clostridial the position of the copper center rela- ligates the Fe3S4 cluster (i.e., conver- ligating sequence was also found for tive to the N2S plane. In amicyanin sion to a clostridial sequence) leads cluster I, which had a more negative (49) and plastocyanin (50) (F5a), to the formation of a Fe4S4 cluster redox potential. It was observed that there is a weak axial bond to a rather than a Fe3S4 cluster at this site; substitution of valine by the less thioether (from methionine) with a that is, AvFdI has been converted bulky glycine at position 13 caused Cu-S distance of about 2.90 Å, and from a 7Fe ferredoxin to an 8Fe fer- a positive shift of 50 mV in the redox the copper center lies 0.3 - 0.35 Å redoxin (46). The redox potential of potential, which is consistent with above the N2S plane (i.e., a distorted the new Fe4S4 cluster was found to the proposed role of solvent accessi- tetrahedron). In azurin (51) (F5b), be -466 mV. The redox potential of bility in determining redox potential there are two axial ligands, the the other Fe4S4 cluster was -612 mV; (36,37). thioether from methionine at a dis- that is, there are two Fe4S4 clusters tance of 3.1 Å, and a carbonyl oxy- in the same protein with similar li- Blue Copper Proteins gen from glycine, also at a distance gating sequences, but with very dif- of 3.1 Å, with the copper center ferent redox potentials. There are a number of different much closer (0.08 Å) to the N2S As discussed above (vide supra), coordination modes for copper in plane (i.e., a distorted trigonal the ligating sequences of the two metalloproteins, and these have been bipyramid). The copper sites of cu- Fe4S4 clusters in CvFd are different, labeled Type 1, Type 2, and Type 3 cumber basic blue protein (plantacy- and the redox potentials of these two based on the spectroscopic proper- anin) (52) and pseudoazurin (53) centers are very different (-460 and ties of the protein (47,48). Type 1 (F3c) are similar to that of plastocy- -655 mV). Mutagenesis was used to copper sites are characterized by an anin, but the axial Cu-S interaction assign the redox potentials to the two intense blue color, due to an absorp- is stronger (Cu-S = 2.62 and 2.69 Å, clusters; mutations around cluster I tion at 600 nm, and hence proteins respectively) and there is a tetragonal affected the more negative redox po- containing Type 1 copper are often distortion relative to the plastocy- tential, whereas mutations around referred to as “blue” copper proteins, anin copper site that leads to differ- cluster II affected the more positive whereas Type 2 copper sites have ences in the spectroscopic redox potential (43). One particular spectroscopic properties similar to characteristics of the copper sites mutation converted the chromatial those found for square-planar coor- (54-56). In stellacyanin (57) (F3d), sequence around cluster II to the dination complexes (and hence are the copper site is again a distorted

51 Current Separations 18:2 (1999) tetrahedron, but the axial methionine pointed out that there is an another lacyanin, it is glycine (coordinated thioether has been replaced by the Jahn-Teller distortion available for carbonyl oxygen). The coordination carbonyl oxygen of a glycine (Cu-O removing the degeneracy of the tet- of oxygen rather than sulfur may sta- = 2.2 Å). rahedral geometry; lengthening one bilize the copper(II) oxidation state Comparison of the geometries of of the bonds lowers the symmetry and lower the redox potential. This copper(I) sites with those of the cop- from Td to C3v, and shortening one idea has been tested by replacing the per(II) sites from the same protein of the remaining bonds further low- coordinating methionine (Met-121) shows that the ligand geometry does ers the symmetry to Cs (55). This in azurin with glutamine (75). It was not vary with the oxidation state of distorted geometry is achieved in the shown by X-ray crystallography that the copper (58-61) (as previously copper coordination site of plastocy- the carbonyl oxygen of glutamine mentioned, this facilitates electron anin, which has a long Cu-S bond for 121 was coordinated in the axial site. transfer, since the activation energy coordination to the methionine resi- However, the redox potential only required for structural changes is due, and a short Cu-S for coordina- decreased by 20 mV (from +288 mV negligible). The ligand geometries tion to the cysteine residue. It was to +268 mV). This was attributed to also remain unaffected when the proposed that the copper(II) geome- changes in the copper coordination copper center is removed (62-64) or try found in plastocyanin is not a that occurred for the mutant protein when it is substituted by mercury high energy structure as had been upon reduction that stabilized the (65) or cadmium (66). It was con- previously proposed, and therefore copper(I) oxidation state (similar cluded from these results that the the geometry imposed by the protein changes did not occur in the wild geometry at the copper site is con- on the metal center is not the reason type protein). trolled by the protein, and this ge- for the destabilization of the cop- Oxygen coordination in the axial ometry is imposed upon the copper per(II) oxidation state. The high re- site has also been obtained for azurin center (67). The unusual spectro- dox potential was attributed to the by substituting Met-121 with gluta- scopic properties of blue copper pro- weak interaction between the copper mate, which can coordinate through teins have been attributed to this center and the axial methionine li- the carboxylate group (76). How- constrained geometry (54-56). The gand, since the poor electron dona- ever, the properties of this mutant potentials for the Cu(II)/Cu(I) redox tion from this fourth ligand increases protein are strongly dependent upon couple are of particular interest, the favorability of the lower (cop- pH, due to the protonation/deproto- since they are higher (more positive) per(I)) oxidation state (55). nation of the coordinated carboxy- than typically observed for copper Although the structures of the late group. At pH 4, the carboxylate coordination complexes (68) (i.e., copper coordination sites are similar, is protonated, and the spectroscopic the copper(I) oxidation state is more there is a wide variation in the values properties are typical of a blue cop- favored). This behavior was ex- of the redox potentials (similar to per protein. However, at pH 8, the plained by noting that, electroni- those discussed above for cyto- carboxylate is deprotonated, and the cally, copper(II) favors a square chromes and iron-sulfur proteins). change in axial coordination gives planar or tetragonal geometry over a Typical values include +680 mV for rise to significant spectroscopic tetrahedral geometry. This is due to rusticyanin (47), +390 mV for plas- changes. The redox potential also d orbital degeneracy in a d9 tetrahe- tocyanin (70), +320 mV for cucum- changes from +370 mV at pH 4 to dral geometry, which is removed by ber basic blue protein (71), +310 mV +184 mV at pH 8. These changes the tetragonal distortion (Jahn-Teller for azurin 72, + 280 mV for were attributed to the stronger inter- distortion) towards a square planar pseudoazurin (73), and +184 mV for action with the lone pair on the de- geometry. In contrast, the d10 cop- stellacyanin (47). Considerable ef- protonated oxygen donor. Similar per(I) center has no electronic pref- fort has been devoted to elucidating behavior has also been reported for erences. Therefore, forcing the the factors that cause these wide vari- the histidine-121 mutant (77,78). copper center into a distorted tetra- ations, and, again, site-directed mu- Met-121 can also be substituted hedral geometry increases the rela- tagenesis has been extensively used by amino acids that cannot coordi- tive stability of the copper(I) (74). Some selected examples are nate to the copper center (e.g., those oxidation state. This increased sta- discussed below. with aliphatic side chains, such as bility has been illustrated using a glycine, valine, alanine, and tetradentate imidazole ligand to con- Substitution of the Amino leucine). These mutants typically strain copper(II) in a tetrahedral ge- Acids Directly Coordinated to have redox potentials higher than the ometry (69); the redox potential of the Copper Center wild type protein (72,79), and this this model complex was several hun- has been attributed to the increased dred millivolts more positive than As discussed above (vide supra), hydrophobicity of the copper coordi- the redox potentials of CuN4 com- the axial (fourth) ligand in most blue nation site (the decreased electron plexes with a square planar geome- copper proteins is methionine (coor- donation to the copper center may try. However, it has recently been dinated thioether), whereas in stel- also need to be considered (vide su-

Current Separations 18:2 (1999) 52 pra)). The vacant axial coordination tuted by alanine or isoleucine) in 8. All potentials quoted in this article are relative to the NHE. site can be occupied by exogeneous both oxidation states were eluci- 9. M.A. Cusanovich, T.E.Meyer, and G. ligands (e.g., water, azide, and thio- dated using X-ray crystallography Tollin in “Advances in Inorganic cyanate); coordination of these li- (85). These showed that the substitu- Biochemistry: Heme Proteins” G.L. gands stabilizes the copper(II) tion by alanine created a surface Eichhorn and L.G. Marzelli (eds.), Elsevier (1988) 37. oxidation state, leading to lower re- pocket containing a water molecule, 10. L.A. Bottomley, L. Olsen, and K.M. dox potentials (72,80,81). and electrostatic calculation showed Kadish in “Electrochemical and The coordination sphere of cop- that the presence of this additional Spectroelectrochemical Studies of per has also been altered by substi- water molecule altered the solvation Biological Redox Components” (K.M. Kadish, ed.), ACS Adv. Chem. tuting the histidine and cysteine energy at the copper center. The sub- Ser. 201 (1982) 279. amino acids that coordinate in the stitution by isoleucine created more 11. F.A.Walker, J.A. Barry, V.L.Balke, trigonal plane. Substitution with protein flexibility at the copper bind- G.A. McDermott, M.Z. Wu, and P.F. non-coordinating amino acids gen- ing site that allowed a more trigonal Linde in “Electrochemical and Spectroelectrochemical Studies of erally leads to an increase in the re- geometry to be adopted by the cop- Biological Redox Components” dox potential (82) and conversion of per(I) oxidation state. Both these (K.M. Kadish, ed.), ACS Adv. Chem. the copper coordination site from changes stabilize the copper(I) oxi- Ser. 201 (1982) 378. Type 1 to Type 2 (83,84) (as judged dation state, which is consistent with 12. T. Mashiko, C.A.Reed, K.J. Haller, M.A. Kastner, and W.R. Scheidt, J. by changes in spectroscopic proper- the increase in the redox potential Am. Chem. Soc. 103 (1981) 5758. ties). Type 1 coordination can be re- from +270 mV for the wild type 13. S.G. Sligar, K.D. Egeberg, J.T. Sage, stored by the addition of an protein to +409 mV in the alanine D. Morikis, and P.M. Champion, J. appropriate ligand (e.g., imidazole). mutantand+450mVinthe Am. Chem. Soc. 109 (1987) 7896. isoleucine mutant. 14. C.J.A. Wallace and I. Clark-Lewis, J. Biol. Chem. 267 (1992) 3852. Substitution of Other The examples given above for the 15. J.C. Rodriquez and M. Rivera, Amino Acids three different classes of electron Biochemistry 37 (1998) 13082. transfer proteins illustrate how the pro- 16. B.A. Feinberg, X. Liu, M.D. Ryan, A. It has been noted in previous sec- tein environment is used to modulate Schejter, C. Zhang, and E. tions of this article that the redox the redox potentials over a wide range. Margoliash, Biochemistry 37 (1998) 13091. potential of metal centers can be in- However, although much progress has 17. S.P.Rafferty, L.L. Pearce, P.D. fluenced by a variety of factors, in- been made in determining the effects Barker, J.G. Guillemette, C.M. Kay, cluding electrostatic interactions of protein environment, considerable M. Smith, and A.G. Mauk, 29 between the metal center and point work is still required (i.e., relating (1990) 9365. charges and dipoles in the protein, changes in the redox potential to 18. W.D. Funk, T.P.Lo, M.R. Mauk, G.D. Brayer, R.T.A.MacGillivray, and hydrogen bonding between the met- changes in the rate of electron transfer A.G. Mauk, Biochemistry 29 (1990) al center and the protein, and the between proteins (86)). 5500. hydrophobicity and solvent accessi- 19. K.K. Rodgers and S.G. Sligar, J. Am. bility of the metal center. Therefore, References Chem. Soc. 113 (1991) 9419. changes in amino acids not directly 20. A.L. Burrows, L.-H. Guo, H.A.O. Hill, coordinated to the metal center can 1. F.A.Armstrong, H.A.O. Hill, and N.J. G. McLendon, and F.Sherman, Eur. Walton, Acc. Chem. Res. 21 (1988) J. Biochem. 202 (1991) 543. affect the redox potential. The redox 407. 21. M. Rivera, R. Seetharaman, D. potentials of blue copper proteins 2. F.A.Armstrong, Struct. Bond, 72 Girdhar, M. Wirtz, X. Zhang, W. can be further influenced through (1990) 137. Wang, and S.White, Biochemistry 38 (1998) 1485. modulations in the tertiary structure 3. W.R. Heineman, C.W. Anderson, H.B. brought about by changes in electro- Halsall, M.M. Hurst, J.M. Johnson, 22. R.J. Kassner, Proc. Nat. Acad. Sci. G.P.Kreishman, B.J. Norris, M.J. USA 69 (1972) 2263. static and hydrogen bonding interac- Simone, and C.-H. Su in 23. A.K. Churg and A. Warshel, tions that accompany variations in “Electrochemical and Biochemistry 25 (1986) 1675. Spectroelectrochemical Studies of non-coordinated amino acids. The 24. G.R. Moore, G.W. Pettigrew, and Biological Redox Components” N.K. Rodgers, Proc. Natl. Acad. Sci magnitude and direction of changes (K.M. Kadish, ed.), ACS Adv. Chem. 83 (1986) 4998. in the redox potential depend on the Ser. 201 (1982) 1. 25. A.G. Mauk and G.R. Moore, J. Biol. 4. L.S. Reid, V.T.Taniguchi, H.B. Gray, site of substitution and changes in Inorg. Chem. 2 (1997) 119. charge, polarity, etc. (72,74). Since and A.G. Mauk, J. Am. Chem. Soc. 104 (1982) 7516. 26. F.A.Tezcan, J.R. Winkler, and H.B. there are so many variables to con- Gray, J. Am. Chem. Soc. 120 (1998) 5. A.L. Raphael and H.B. Gray, 13383. sider, structural studies on the mu- J. Am. Chem. Soc. 113 (1991) 1038. 27. L.S. Reid, M.R. Mauk, and A.G. tant proteins are generally required 6. M. Rivera, M.A. Wells, and F.A. Mauk, J. Am. Chem. Soc.106 Walker, Biochemistry 33 (1994) when trying to rationalize changes in (1984) 2182. 2161. redox potentials. For example, the 28. K.-B. Lee, E. Jun, G.N. La Mar, I, 7. M.J. Eddowes and H.A.O. Hill, solid state structures of two mutants Rezzano, R.K. Pandey, K.M. Smith, J. Chem. Soc., Chem. Commun. F.A.Walker, and D.H. Buttlaire, J. of pseudoazurin (proline-40 substi- (1977) 3154. Am. Chem. Soc. 113 (1991) 3576.

53 Current Separations 18:2 (1999) 29. L.S. Reid, A.R. Lim, and A.G. Mauk, 51. E.N. Baker, J. Mol. Biol. 203 (1988) 70. F.N.Büchi, A.M. Bond, R. Codd, L.N. J. Am. Chem. Soc. 108 (1986) 8197. 1071. Huq, and H.C. Freeman, Inorg. 30. F.A.Walker, D. Emrick, J.E. Rivera, 52. J.M. Guss, E.A Merritt, R.P. Chem. 13 (1992) 5007. B.J. Hanquet, and D.H. Buttlaire, J. Phizackerley, B. Hedman, M. 71. G. Battistuzzi, M. Borsari, and M. Am. Chem. Soc. 110 (1988) 6234. Murata, K.O. Hodgson, and H.C. Sola, J. Inorg. Biochem. 69 (1998) 31. S.A. Martinis, C. Sotiriou, C.K. Freeman, Science 241 (1988) 806. 97. Chang, and S.G. Sligar, 53. E.T.Adman, S. Turley, R. Bramson, 72. T.Pascher, B.G. Karlsson, M. Biochemistry 28 (1989) 879. K. Petratos, D. Banner, D. Nordling, B.G. Malmström, and T. 32. R. Cammack, Adv. Inorg. Chem. 38 Tsernoglou, T.Beppu, and H. Vanncard, Eur. J. Biochem. 212 (1992) 281. Watanbe, J. Biol. Chem. 264 (1989) (1993) 289. 87. 33. F.Capozzi, S. Curli, and C. Luchinat, 73. T. Sakurai, O. Ikeda, and S. Suzuki, Struct. Bonding 90 (1998) 127. 54. R.H. Holm, R. Kennephol, and E.I. Inorg. Chem. 29 (1990) 4715. Solomon, Chem. Rev. 96 (1996) 74. G.W. Canters and G. Gilardi, FEBS 34. J.L.C. Duff, J.L.J. Breton, J.N. Butt, 2239. F.A.Armstrong, and A.J. Thomson, Lett. 325 (1993) 39. J. Am Chem. Soc. 118 (1996) 8593. 55. J.A. Guckert, M.D. Lowery, and E.I. 75. A. Romero, C.W.G. Hoitkin, H. Nar, Solomon, J. Am. Chem. Soc. 117 R. Huber, A. Messerschmidt, and 35. G. Backes, Y.Mino, T.M.Loehr, T.E. (1995) 2817. Meyer, M.A. Cusanovich, W.V. G.W. Canters, J. Mol. Biol. 229 Sweeney, E.T.Adman, and J. 56. L.B. LaCroix, D.W. Randall, A.M. (1993) 1007. Sanders-Loehr, J. Am. Chem. Soc. Nersissian, C.W.G. Hoitink, G.W. 76. B.G. Karlsson, L.C. Tsai, H. Nar, J. 113 (1991) 2055. Canters, J.S. Valentine, and E.I. Sanders-Loehr, N. Bonander, V. Solomon, J. Am. Chem. Soc. 120 Langer, and L. Sjolin, Biochemistry 36. R. Langen, G.M. Jensen, U. Jacob, (1998) 9621. P.J. Stephens, and A. Warshel, 267 36 (1997) 4089. (1992) 25625. 57. P.J. Hart, A.M. Nersissian, R.G. 77. S.J. Kroes, C.W. G. Hoitink, and G.W. Herrmann, R.M. Nalbandyan, J.S. 37. G.M. Jensen, A. Warshel, and P.J. Canters, Eur. J. Biochem. 240 Valentine, and D. Eisenberg, (1996) 342. Stephens, Biochemistry 33 (1994) Protein Sci. (1996) 2175. 10911. 78. A. Messerschmidt, L. Prade and 58. W.E.B. Shepard, B.F.Anderson, D.A. 38. B.Shen, D.R. Jollie, C.D. Stout, T.C. G.W. Canters, Proc. Natl.Acad. Sci. Lewandowski, G.A. Norris, and E.B. USA 95 (1998) 3443. Diller, F.A.Armstrong, C.M. Gorst, Baker, J. Am. Chem. Soc. 112 G.N. La Mar, P.J. Stephens, and (1990) 7817. 79. A.J. Di Bilio, T.K.Chang, B.G. B.K. Burgess, J. Biol. Chem. 269 Malmström, H.B. Gray, B.G. (1994) 8564. 59. J.M. Guss, P.R. Harrowell, M, Karlsson, M. Nordling, T.Pascher, Murata, V.A.Norris, and H.C. 39. H.A. Heering, Y.B.M. Bulsink, W.R. and L.G. Lundberg, Inorg. Chim. Freeman, J. Mol. Biol. 192 (1986) Acta 198-200 (1992) 145. Hagen, and T.E.Meyer, 361. Biochemistry 34 (1995) 14675. 80. M. Vidakovic and J.P.Germanas, 60. E. Vakoufari, K.S. Wilson, and K. 40. L. Banci, I. Bertini, S. Curli, C. Angew. Chem., Intl. Eng. Ed. 34 Patratos, FEBS Lett. 347 (1994) (1995) 1622. Luchinat, and R. Pierattelli, Inorg. 203. Chim. Acta. 24 (1995) 251. 81. L.M. Murphy, R.W. Strange, B.G. 61. C.A.P.Libeu, M. Kukimoto, M. 41. L. Banci, I. Bertini, G.G. Savellini, Karlsson, L.G. Lundberg, T. Nishiyama, S. Horinouchi, and E.T. Pascher, B. Reinhammar, and S.S. and C. Luchinat, Inorg. Chem. 35 Adman, Biochemistry 36 (1997) (1996) 4248. Hasnain, Biochemistry 32 (1993) 13160. 1965. 42. J.M. Moulis, L.C. Sieker, K.S., 62. T.P.J. Garrett, D.J. Clingeleffer, J.M. 82. G. van Pouderoyen, C.R. Andrew, Wilson, and Z. Dauter, Protein Sci. Guss, S.J. Rodgers, and H.C. 5 (1996) 1765. and G.W. Canters, Biochemistry 35 Freeman, J. Biol. Chem. 259 (1984) (1996) 1397. 43. P.Kyritsis, O.M. Hatzfeld, T.A.Link, 2822. and J.M. Moulis, J. Biol. Chem. 273 83. T. den Blaauwen, and G.W. Canters, 63. W.E.B. Shepard, R.L. Kinston, B.F. J. Am. Chem. Soc. 115 (1993) 1121. (1998) 15404. Anderson, and E.N. Baker, Acta 44. C.D. Stout, J. Mol. Biol. 205 (1989) Crystallogr. D49 (1993) 331. 84. T.J. Mizoguchi, A.J.Di Bilio, H.B. 545. Gray, and J.H. Richards, J. Am. 64. K.Z. Petratos, M. Papadovasilaki, Chem. Soc. 114 (1992) 10076. 45. S.E. Iismaa, A.E. Vazquez, G.M. and Z. Dauter, FEBS Lett. 368 Jensen, P.J. Stephens, J.N. Butt, (1995) 442. 85. C.A.P.Libeu, M. Kukimoto, M. Nishiyama, S. Horinouchi, and E.T. F.A.Armstrong, and B.K. Burgess, 65. W.B. Church, J.M. Guss, J.J. Potter, J. Biol. Chem. 266 (1991) 21563. Adman, Biochemistry 36 (1997) and H.C. Freeman, J. Biol. Chem. 13160. 46. M.A. Kemper, H.S. Gao-Sheridan, B. 261 (1986) 234. Shen, J.L.C. Duff, G.J. Tilley, F.A. 86. J.K. Hurley, A.M. Weber-Main, M.T. 66. K.A. Blackwell, B.F.Anderson, and Stankovich, M.W. Benning, J.B. Armstrong, and B.K. Burgess, E.N. Baker, Acta Crystallogr. D50 Biochemistry 37 (1998) 12829. Thoden, J.L. Vanhooke, H.M. (1994) 263. Holden, Y.K.Chae, B. Xia, H. 47. A.G. Sykes, Adv. Inorg Chem. 36 67. G. Malmström. Eur. J. Biochem. 223 Cheng, J.L. Markley, M. (1991) 377. (1994) 711. Martinez-Julez, C. Gomez-Moreno, 48. A. Messerschmidt, Struct. Bonding 68. N. Kitajima, Adv. Inorg. Chem. 39 J.L. Schmeits, and G. Trolin, 90 (1998) 37. (1992) 1. Biochemistry 36 (1997) 11100. 49. R. Durley, L. Chen, L.W. Louis, F.S. 69. S. Knapp, T.P.Kennan, Z. Xioahua, Mathews, and V.I.Davidson, R. Fikar, J.A. Potenza, and H.J. Protein Sci. 2 (1993) 739. Schugar, J. Am. Chem. Soc. 112 50. J.M. Guss and H.C. Freeman, J. Mol. (1990) 3452. Biol. 169 (1983) 521.

Current Separations 18:2 (1999) 54