Tandem mass spectrometry of protein-protein complexes: cytochrome c-cytochrome b5.

An improved method to interpret triple quadrupole MS/MS experiments of complexes of large ions is presented and applied to a study of the complex formed by the proteins cytochrome c and cytochrome b5. Modeling of the activation and dissociation process shows that most of the reaction occurs near the collision cell exit where ions have the highest internal energies. Experiments at different collision cell pressures or with different collision gases (Ne, Ar, Kr) are interpreted with a previously proposed collision model (Chen et al., Rapid Commun. Mass Spectrom. 1998, 12, 1003-1010) to calculate the internal energy added to ions to cause dissociation. Small but systematic differences under different experimental conditions are attributed to different times available for reaction. A method to correct for this is presented. Ne, Ar, and Kr are found to have similar energy transfer efficiencies. Complexes of cytochrome c and cytochrome b5 are detected in ESI mass spectra but with abundances less than expected from the solution equilibrium. Dissociation of the cytochrome c-cytochrome b5 complexes with charge k gives as the most abundant fragments, cytochrome b5(+3) and cytochrome c+(k-3). Adding charges to the complex destabilizes it. A series of cytochrome c variants with Lys residues thought to be involved in solution binding replaced by Ala showed no differences in the energy required to induce dissociation of the gas phase complex. The implications for the binding of the gas phase ions are inconclusive.

A DNA oligonucleotide-hemin complex cleaves t-butyl hydroperoxide through a homolytic mechanism.

Both electron paramagnetic resonance (EPR) and electronic absorption spectroscopy have been employed to investigate the reaction of a guanine-rich DNA nucleotide-hemin complex (PS2.M-hemin complex) and organic peroxide (t-Bu-OOH). Incubation of the PS2.M-hemin complex with t-Bu-OOH resulted in the time-dependent decrease in the heme Soret with concomitant changes to the visible bands of the electronic absorbance spectrum for the PS2.M-hemin complex. Parallel EPR studies using the spin trap 5,5-dimethyl-1-pyrroline N-oxide (DMPO) combined with spectral simulation demonstrated the presence of tert-butyloxyl, carbon-centered methyl, and methyl peroxyl radicals as well as a simple nitroxide (triplet) signal. Experiments, performed by maintaining a constant ratio of t-Bu-OOH/PS2.M-hemin complex ( approximately 35 mol/mol) while varying DMPO concentration, indicated that the relative contributions of each radical adduct to the composite EPR spectrum were significantly influenced by the DMPO concentration. For example, at DMPO/PS2.M-hemin of 10-50 mol/mol, a complex mixture of radicals was consistently detected, whereas at high trapping efficiency (i.e., DMPO/PS2.M-hemin of approximately 250 mol/mol) the tert-butyloxyl-DMPO adduct was predominant. In contrast, at relatively low DMPO/PS2.M-hemin complex ratios of < or =5 mol/mol, a simple nitroxide three-line EPR signal was detected largely in the absence of all other radicals. Together, these data indicate that tert-butyloxyl radical is the primary radical likely formed from the homolytic cleavage of the O-O peroxy bond of t-Bu-OOH, while methyl and methyl peroxyl radicals result from beta-scission of the primary tert-butyloxyl radical product.

Reaction of human myoglobin and peroxynitrite: characterizing biomarkers for myoglobin-derived oxidative stress.

Mixtures of human myoglobin (Mb) (or the Y103F variant of human Mb), authentic peroxynitrite (ONOO(-), ONOO(-):protein 2 mol/mol), and 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) gave radicals adducts at cysteine-110 (DMPO-C110) that are detected directly by electron paramagnetic magnetic spectroscopy (EPR). DMPO-C110 was detected exclusively over a range of DMPO concentrations (DMPO:protein ratios 25-100 mol/mol). Treatment of human Mb (or Y103F Mb) with the ONOO(-) generator 5-amino-3-(4-morpholinyl)-1,2,3-oxadiazolium (SIN-1) chloride (ONOO(-):protein 5 mol/mol) yielded a cross-linked Mb dimer as judged by SDS-PAGE analyses. Addition of DMPO or carbonate effectively eliminated the cross-linked product. Mass analyses of samples containing human Mb (or Y103F Mb), carbonate, and ONOO(-) indicated that nitration occurs exclusively at Y103. Thus, reaction of human Mb and ONOO(-) yields specific products that depend on the presence or absence of physiological concentrations of carbonate. These products may serve as biomarkers for the participation of Mb-derived radicals in the oxidative damage associated with myocardial reperfusion injury.

The peroxidase activity of a hemin--DNA oligonucleotide complex: free radical damage to specific guanine bases of the DNA.

A specific DNA oligonucleotide--hemin complex (PS2.M--hemin complex) that exhibits DNA-enhanced peroxidative activity was studied by EPR and UV--visible spectroscopy and by chemical probing analysis. EPR data obtained from low-temperature experiments on the PS2.M--hemin complex showed both a low-field g approximately 6 and a high-field g approximately 2 signal. These EPR signals are typical of high-spin ferric heme with axial symmetry as judged by the EPR spectrum of six-coordinate heme iron in acidic Fe(III)-myoglobin. This similarity is consistent with the presence of two axial ligands to the heme iron within the PS2.M--hemin complex, one of which is a water molecule. Optical analyses of the acid-base transition for the hemin complex yielded a pK(a) value for the water ligand of 8.70 +/- 0.03 (mean +/- SD). Low-temperature EPR analysis coupled with parallel spin-trapping investigations following the reaction of the PS2.M--hemin complex and hydrogen peroxide (H(2)O(2)) indicated the formation of a carbon-centered radical, most likely on the PS2.M oligonucleotide. Chemical probing analysis identified specific guanine bases within the PS2.M sequence that underwent oxidative damage upon reaction with H(2)O(2). These and other experimental findings support the hypothesis that the interaction of specific guanines of PS2.M with the bound hemin cofactor might contribute to the superior peroxidative activity of the PS2.M--hemin complex.

Reaction of human myoglobin and H2O2. Electron transfer between tyrosine 103 phenoxyl radical and cysteine 110 yields a protein-thiyl radical.

The sequence of human myoglobin (Mb) is similar to that of other species except for a unique cysteine at position 110 (Cys(110)). Adding hydrogen peroxide (H(2)O(2)) to human Mb affords Trp(14)-peroxyl, Tyr(103)-phenoxyl, and Cys(110)-thiyl radicals and coupling of Cys(110)-thiyl radicals yields a homodimer through intermolecular disulfide bond formation (Witting, P. K., Douglas, D. J., and Mauk, A. G. (2000) J. Biol. Chem. 275, 20391-20398). Treating a solution of wild type Mb and H(2)O(2) with 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) at DMPO:protein /= 100 mol/mol only DMPO-Tyr(103) radicals were present. The DMPO-dependent decrease in DMPO-Cys(110) was matched by a near 1:1 stoichiometric increase in DMPO-Tyr(103). In contrast, reaction of the Y103F human Mb with H(2)O(2) gave no DMPO-Cys(110) at DMPO:protein /= 100 mol/mol (i.e. conditions that consistently gave DMPO-Tyr(103) in the case of wild type Mb). No detectable homodimer was formed by incubation of the Y103F variant with H(2)O(2). However, the homodimer was detected in a mixture of both the Y103F and C110A variants of human Mb upon treatment with H(2)O(2) (C110A:Y103F:H(2)O(2) 2:1:5 mol/mol/mol); the yield of this homodimer increased with increasing ratios of C110A:Y103F. Together, these data suggest that addition of H(2)O(2) to human Mb can produce Cys(110)-thiyl radicals through an intermolecular electron transfer reaction from Cys(110) to a Tyr(103)-phenoxyl radical.

Characterization of BphF, a Rieske-type ferredoxin with a low reduction potential.

BphF is a small, soluble, Rieske-type ferredoxin involved in the microbial degradation of biphenyl. The rapid, anaerobic purification of a heterologously expressed, his-tagged BphF yielded 15 mg of highly homogeneous recombinant protein, rcBphF, per liter of cell culture. The reduction potential of rcBphF, determined using a highly oriented pyrolytic graphite (HOPG) electrode, was -157+/- 2 mV vs the standard hydrogen electrode (SHE) (20 mM MOPS, 80 mM KCl, and 1 mM dithiothreitol, pH 7.0, 22 degrees C). The electron paramagnetic resonance spectrum of the reduced rcBphF is typical of a Rieske cluster while the close similarity of the circular dichroic (CD) spectra of rcBphF and BedB, a homologous protein from the benzene dioxygenase system, indicates that the environment of the cluster is highly conserved in these two proteins. The reduction potential and CD spectra of rcBphF were relatively independent of pH between 5 and 10, indicating that the pK(a)s of the cluster's histidinyl ligands are not within this range. Gel filtration studies demonstrated that rcBphF readily oligomerizes in solution. Crystals of rcBphF were obtained using sodium formate or poly(ethylene glycol) (PEG) as the major precipitant. Analysis of the intermolecular contacts in the crystal revealed a head-to-tail interaction that occludes the cluster, but is very unlikely to be found in solution. Oligomerization of rcBphF in solution was reversed by the addition of dithiothreitol and is unrelated to the noncovalent crystallographic interactions. Moreover, the oligomerization state of rcBphF did not influence the latter's reduction potential. These results indicate that the 450 mV spread in reduction potential of Rieske clusters of dioxygenase-associated ferredoxins and mitochondrial bc(1) complexes is not due to significant differences in their solvent exposure.

Reaction of human myoglobin and nitric oxide. Heme iron or protein sulfhydryl (s) nitrosation dependence on the absence or presence of oxygen.

The amino acid sequence of human myoglobin (Mb) is similar to other mammalian Mb except for a unique cysteine residue at position 110 (Cys(110)). Anaerobic treatment of ferrous forms of wild-type human Mb, the C110A variant of human Mb or horse heart Mb, with either authentic NO or chemically derived NO in vitro yields heme-NO complexes as detected by electron paramagnetic resonance spectroscopy (EPR). By contrast, no EPR-detectable heme-NO complex was observed from the aerobic reactions of NO and either the ferric or oxy-Mb forms of wild-type human or horse heart myoglobins. Mass analyses of wild-type human Mb treated aerobically with NO indicated a mass increase of approximately 30 atomic mass units (i.e., NO/Mb = 1 mol/mol). Mass analyses of the corresponding apoprotein after heme removal showed that NO was associated with the apoprotein fraction. New electronic maxima were detected at A(333 nm) (epsilon = 3665 +/- 90 mol(-)(1) cm(-)(1); mean +/- S.D.) and A(545 nm) (epsilon = 44 +/- 3 mol(-)(1) cm(-)(1)) in solutions of S-nitrosated wild-type human Mb (similar to S-nitrosoglutathione). Importantly, the sulfhydryl S-H stretch vibration for Cys(110) measured by Fourier transform infrared (nu approximately 2552 cm(-)(1)) was absent for both holo- and apo- forms of the wild-type human protein after aerobic treatment of the protein with NO. Together, these data indicate that the reaction of wild-type human Mb and NO yields either heme-NO or a novel S-nitrosated protein dependent on the oxidation state of the heme iron and the presence or absence of dioxygen.

Cytochrome c peroxidase-cytochrome c complex: locating the second binding domain on cytochrome c peroxidase with site-directed mutagenesis.

Cytochrome c peroxidase (CcP) can bind as many as two cytochrome c (Cc) molecules in an electrostatic complex. The location of the two binding domains on CcP has been probed by photoinduced interprotein electron transfer (ET) between zinc-substituted horse cytochrome c (ZnCc) and CcP with surface charge-reversal mutations and by isothermal titration calorimetry (ITC). These results, which are the first experimental evidence for the location of domain 2, indicate that the weak-binding domain includes residues 146-150 on CcP. CcP(E290K) has a charge-reversal mutation in the tight-binding domain, which should weaken binding, and it weakens the 1:1 complex; K(1) decreases 20-fold at 18 mM ionic strength. We have employed two mutations to probe the proposed location for the weak-binding domain on the CcP surface: (i) D148K, a "detrimental" mutation with a net (+2) change in the charge of CcP, and (ii) K149E, a "beneficial" mutation with a net (-2) change in the charge. The interactions between FeCc and CcP(WT and K149E) also have been studied with ITC. The CcP(D148K) mutation causes no substantial change in the 2:1 binding but an increase in the reactivity of the 2:1 complex. The latter can be interpreted as a long-range influence on the heme environment or, more likely, the enhancement of a minority subset of binding conformations with favorable pathways for ET. CcP(K149E) has a charge-reversal mutation in the weak-binding domain that produces a substantial increase in the 2:1 binding constant as measured by both quenching and ITC. For the 1:1 complex of CcP(WT), DeltaG(1) = -8.2 kcal/mol (K(1) = 1.3 x 10(6) M(-)(1)), DeltaH(1) = +2.7 kcal/mol, and DeltaS(1) = +37 cal/K.mol at 293 K; for the second binding stage, K(2) < 5 x 10(3) M(-)(1), but accurate thermodynamic parameters were not obtained. For the 1:1 complex of CcP(K149E), DeltaG(1) = -8.5 kcal/mol (K(1) = 2 x 10(6) M(-)(1)), DeltaH(1) = +2. 0 kcal/mol, and DeltaS(1) = +36 cal/K.mol; for the second stage, DeltaG(2) = -5.5 kcal/mol (K(1) = 1.3 x 10(4) M(-)(1)), DeltaH(2) = +2.9 kcal/mol, and DeltaS(2) = +29 cal/K.mol.

Characterization of an alkaline transition intermediate stabilized in the Phe82Trp variant of yeast iso-1-cytochrome c.

In general, mutation of the phylogenetically conserved residue Phe82 in yeast iso-1-cytochrome c destabilizes the native conformation of the protein by facilitating the ligand exchange reactions that are associated with the alkaline conformational transitions of the ferricytochrome. Of the Phe82 variants surveyed thus far, Phe82Trp is unique in that it adopts a thermodynamically stable, high-spin conformation at mildly alkaline pH. This species exhibits spectroscopic features that can only be detected transiently in other ferricytochromes c within the first 100 ms immediately after a pH-jump from neutrality to pH >10. Spectroscopic characterization of this high-spin reaction intermediate suggests that in addition to an obligatory pentacoordinate heme iron, a group within the heme pocket coordinates the heme iron but is then replaced either by Met80, to revert to the native conformation, or by Lys73 or Lys79, to yield one of the conventional alkaline conformers. Evidence is presented to suggest that this group is either a hydroxide ion or Tyr67 rather than a loosely bound Met80.

Hemin is kinetically trapped in cytochrome b(5) from rat outer mitochondrial membrane.

Cytochrome b(5) from the outer mitochondrial membrane of rat liver (OM cyt b(5)) is substantially more stable to thermal and chemical denaturation than cytochrome b(5) from the endoplasmic reticulum of bovine liver (microsomal, or Mc cyt b(5)). In contrast, the corresponding apoproteins have similar stability, suggesting stronger interactions between hemin and the polypeptide in OM cyt b(5). Whereas complete transfer of hemin from bovine Mc cyt b(5) to apomyoglobin at pH 5.2 takes less than 1 h, hemin transfer from OM cyt b(5) is unmeasurably slow. Coupled with the previously reported 1:1 ratio of hemin orientational isomers in OM cyt b(5), this finding suggests that the cofactor is kinetically trapped under physiologically relevant conditions. This conclusion is confirmed by (1)H NMR studies which show that the hemin isomeric ratio changes when the protein is incubated for several hours at 68 degrees C. Interestingly, the orientational isomer favored in OM cyt b(5) is the form less favored in all other known cytochromes b(5).

Determinants of cytochrome c pro-apoptotic activity. The role of lysine 72 trimethylation.

Cytochrome c released from vertebrate mitochondria engages apoptosis by triggering caspase activation. We previously reported that, whereas cytochromes c from higher eukaryotes can activate caspases in Xenopus egg and mammalian cytosols, iso-1 and iso-2 cytochromes c from the yeast Saccharomyces cerevisiae cannot. Here we examine whether the inactivity of the yeast isoforms is related to a post-translational modification of lysine 72, N-epsilon-trimethylation. This modification was found to abrogate pro-apoptotic activity of metazoan cytochrome c expressed in yeast. However, iso-1 cytochrome c lacking the trimethylation modification also was devoid of pro-apoptotic activity. Thus, both lysine 72 trimethylation and other features of the iso-1 sequence preclude pro-apoptotic activity. Competition studies suggest that the lack of pro-apoptotic activity was associated with a low affinity for Apaf-1. As cytochromes c that lack apoptotic function still support respiration, different mechanisms appear to be involved in the two activities.

Modulation of the activities of catalase-peroxidase HPI of Escherichia coli by site-directed mutagenesis.

Catalase-peroxidases have a predominant catalatic activity but differ from monofunctional catalases in exhibiting a substantial peroxidatic reaction which has been implicated in the activation of the antitubercular drug isoniazid in Mycobacterium tuberculosis. Hydroperoxidase I of Escherichia coli encoded by katG is a catalase-peroxidase, and residues in its putative active site have been the target of a site directed-mutagenesis study. Variants of residues R102 and H106, on the distal side of the heme, and H267, the proximal side ligand, were constructed, all of which substantially reduced the catalatic activity and, to a lesser extent, the peroxidatic activity. In addition, the heme content of the variants was reduced relative to the wild-type enzyme. The relative ease of heme loss from HPI and a mixture of tetrameric enzymes with 2, 3, and 4 hemes was revealed by mass spectrometry analysis. Conversion of W105 to either an aromatic (F) or aliphatic (I) residue caused a 4-5-fold increase in peroxidatic activity, coupled with a >99% inhibition of catalatic activity. The peroxidatic-to-catalatic ratio of the W105F variant was increased 2800-fold such that compound I could be identified by both electronic and EPR spectroscopy as being similar to the porphyrin cation radical formed in other catalases and peroxidases. Compound I, when generated by a single addition of H(2)O(2), decayed back to the native or resting state within 1 min. When H(2)O(2) was generated enzymatically in situ at low levels, active compound I was evident for up to 2 h. However, such prolonged treatment resulted in conversion of compound I to a reversibly inactivated and, eventually, to an irreversibly inactivated species, both of which were spectrally similar to compound I.

Reaction of human myoglobin and H2O2. Involvement of a thiyl radical produced at cysteine 110.

The human myoglobin (Mb) sequence is similar to other mammalian Mb sequences, except for a unique cysteine at position 110. Reaction of wild-type recombinant human Mb, the C110A variant of human Mb, or horse heart Mb with H(2)O(2) (protein/H(2)O(2) = 1:1.2 mol/mol) resulted in formation of tryptophan peroxyl (Trp-OO( small middle dot)) and tyrosine phenoxyl radicals as detected by EPR spectroscopy at 77 K. For wild-type human Mb, a second radical (g approximately 2. 036) was detected after decay of Trp-OO( small middle dot) that was not observed for the C110A variant or horse heart Mb. When the spin trap 5,5-dimethyl-1-pyrroline N-oxide (DMPO) was included in the reaction mixture at protein/DMPO ratios /=1:25 mol/mol, DMPO-tyrosyl radical adducts were detected. Mass spectrometry of wild-type human Mb following reaction with H(2)O(2) demonstrated the formation of a homodimer (mass of 34,107 +/- 5 atomic mass units) sensitive to reducing conditions. The human Mb C110A variant afforded no dimer under identical conditions. Together, these data indicate that reaction of wild-type human Mb and H(2)O(2) differs from the corresponding reaction of other myoglobin species by formation of thiyl radicals that lead to a homodimer through intermolecular disulfide bond formation.

A cytochrome c variant resistant to heme degradation by hydrogen peroxide.

BACKGROUND: Cytochrome c has peroxidase-like activity and can catalyze the oxidation of a variety of organic substrates, including aromatic, organosulfur and lipid compounds. Like peroxidases, cytochrome c is inactivated by hydrogen peroxide. During this inactivation the heme prosthetic group is destroyed. RESULTS: Variants of the iso-1-cytochrome c were constructed by site-directed mutagenesis and were found to be more stable in the presence of hydrogen peroxide than the wild type. No heme destruction was detected in a triple variant (Tyr67-->Phe/Asn52-->Ile/Cys102-->Thr) with the catalytic hydrogen peroxide concentration of 1 mM, even following the loss of catalytic activity, whereas both double variants Tyr67-->Phe/Cys102-->Thr and Asn52-->Ile/Cys102-->Thr showed a greater rate of peroxide-induced heme destruction than observed with the wild-type protein. CONCLUSIONS: Heme destruction and catalytic inactivation are two independent processes. An internal water molecule (Wat166) is shown to be important in the heme destruction process. The absence of a protein radical in the resistant variant suggests that the protein radical is necessary in the heme destruction process, but presumably is not involved in the reactions leading up to the protein inactivation.

Investigation of the role of a surface patch in the self-association of Chromatium vinosum high potential iron-sulfur protein.

The role of a flattened, relatively hydrophobic surface patch in the self-association of Chromatium vinosum HiPIP was assessed by substituting phenylalanine 48 with lysine. The reduction potential of the F48K variant was 26 mV higher than that of the wild-type (WT) recombinant (rc) HiPIP, consistent with the introduction of a positive charge close to the cluster. Nuclear magnetic resonance spectroscopy (NMR) revealed that the electronic structure of the oxidized cluster in these two proteins is very similar at 295 K. In contrast, the electron transfer self-exchange rate constant of F48K was at least 15-fold lower than that of the WT rcHiPIP, indicating that the introduction of a positive charge at position 48 diminishes self-association of the HiPIP in solution. Moreover, the substitution at position 48 abolished the fine structure in the g(z) region of the electron paramagnetic resonance (EPR) spectrum of oxidized C. vinosum rcHiPIP recorded in the presence of 1 M sodium chloride. These results support the hypothesis that the flattened, relatively hydrophobic patch mediates interaction between two molecules of HiPIP and that freezing-induced dimerization of the HiPIP mediated by this patch is responsible for the unusual fine structure observed in the EPR spectrum of the oxidized C. vinosum HiPIP.

The structural and functional role of lysine residues in the binding domain of cytochrome c in the electron transfer to cytochrome c oxidase.

The interactions of yeast iso-1 cytochrome c with bovine cytochrome c oxidase were studied using cytochrome c variants in which lysines of the binding domain were substituted by alanines. Resonance Raman spectra of the fully oxidized complexes of both proteins reveal structural changes of both the heme c and the hemes a and a3. The structural changes in cytochrome c are the same as those observed upon binding to phospholipid vesicles where the bound protein exists in two conformers, B1 and B2. Whereas the structure of B1 is the same as that of the unbound cytochrome c, the formation of B2 is associated with substantial alterations of the heme pocket. In cytochrome c oxidase, the structural changes in both hemes refer to more subtle perturbations of the immediate protein environment and may be a result of a conformational equilibrium involving two states. These changes are qualitatively different to those observed for cytochrome c oxidase upon poly-l-lysine binding. The resonance Raman spectra of the various cytochrome c/cytochrome c oxidase complexes were analyzed quantitatively. The spectroscopic studies were paralleled by steady-state kinetic measurements of the same protein combinations. The results of the spectra analysis and the kinetic studies were used to determine the stability of the complexes and the conformational equilibria B2/B1 for all cytochrome c variants. The complex stability decreases in the order: wild-type WT > J72K > K79A > K73A > K87A > J72A > K86A > K73A/K79A (where J is the natural trimethyl lysine). This order is not exhibited by the conformational equilibria. The electrostatic control of state B2 formation does not depend on individual intermolecular salt bridges, but on the charge distribution in a specific region of the front surface of cytochrome c that is defined by the lysyl residues at positions 72, 73 and 79. On the other hand, the conformational changes in cytochrome c oxidase were found to be independent of the identity of the bound cytochrome c variant. The maximum rate constants determined from steady-state kinetic measurements could be related to the conformational equilibria of the bound cytochrome c using a simple model that assumes that the conformational transitions are faster than product formation. Within this model, the data analysis leads to the conclusion that the interprotein electron transfer rate constant is around two times higher in state B2 than in B1. These results can be interpreted in terms of an increase of the driving force in state B2 as a result of the large negative shift of the reduction potential.

Replacement of the proximal histidine iron ligand by a cysteine or tyrosine converts heme oxygenase to an oxidase.

The H25C and H25Y mutants of human heme oxygenase-1 (hHO-1), in which the proximal iron ligand is replaced by a cysteine or tyrosine, have been expressed and characterized. Resonance Raman studies indicate that the ferric heme complexes of these proteins, like the complex of the H25A mutant but unlike that of the wild type, are 5-coordinate high-spin. Labeling of the iron with 54Fe confirms that the proximal ligand in the ferric H25C protein is a cysteine thiolate. Resonance-enhanced tyrosinate modes in the resonance Raman spectrum of the H25Y.heme complex provide direct evidence for tyrosinate ligation in this protein. The H25C and H25Y heme complexes are reduced to the ferrous state by cytochrome P450 reductase but do not catalyze alpha-meso-hydroxylation of the heme or its conversion to biliverdin. Exposure of the ferrous heme complexes to O2 does not give detectable ferrous-dioxy complexes and leads to the uncoupled reduction of O2 to H2O2. Resonance Raman studies show that the ferrous H25C and H25Y heme complexes are present in both 5-coordinate high-spin and 4-coordinate intermediate-spin configurations. This finding indicates that the proximal cysteine and tyrosine ligand in the ferric H25C and H25Y complexes, respectively, dissociates upon reduction to the ferrous state. This is confirmed by the spectroscopic properties of the ferrous-CO complexes. Reduction potential measurements establish that reduction of the mutants by NADPH-cytochrome P450 reductase, as observed, is thermodynamically allowed. The two proximal ligand mutations thus destabilize the ferrous-dioxy complex and uncouple the reduction of O2 from oxidation of the heme group. The proximal histidine ligand, for geometric or electronic reasons, is specifically required for normal heme oxygenase catalysis.

Biological electron-transfer reactions.

A wide range of biological processes makes extensive use of electron-transfer reactions. Rigorous characterization of a biological electron-transfer reaction requires a combination of kinetic, thermodynamic, structural and theoretical methods. The rate of electron transfer from an electron donor to an electron acceptor through a protein is dependent on the difference in reduction potential of the electron acceptor and electron donor and the distance over which electron transfer occurs. The manner in which the rate of electron transfer also depends on the structure of the protein located between the electron donor and acceptor sites remains an active topic of investigation. Diverse protein-engineering strategies are providing new insights into fundamental mechanistic considerations regarding electron-transfer properties of biological molecules, and they can provide novel means by which insights concerning biological electron-transfer reactions can be employed to develop new and useful types of chemistry.

Functional comparison of specifically cross-linked hemoglobins biased toward the R and T states.

Selected functional and spectroscopic properties of two human hemoglobin (HbA0) derivatives that were site-specifically cross-linked in the cleft between beta-chains where 2, 3-bisphosphoglycerate normally binds have been determined to assess the effects of the cross-linking on the behavior of the protein. Trimesoyl tris(3,5-dibromosalicylate) (TTDS) cross-links Hb between beta82Lys residues. The resulting TTDS-Hb exhibits a slower rate of oxygen dissociation and an increased rate of carbon monoxide association than observed for HbA0. The electron paramagnetic resonance (EPR) spectrum of TTDS-HbNO does not exhibit the hyperfine structure that is indicative of significant conformational change despite the fact that the 2,3-bisphosphoglycerate binding site is occupied by the cross-linking reagent. The reactivity of the beta93Cys residues of TTDS-Hb is only slightly decreased relative to that of HbA0. On the other hand, cross-linking Hb between Lys82 and the amino-terminal beta1Val group with trimesoyl tris(methyl phosphate) (TMMP) increases the rate of oxygen dissociation and reduces the rate of CO association relative to the rates observed for HbA0. In addition, the EPR spectrum of the TMMP-HbNO exhibits the three-line hyperfine structure that results from disruption of the proximal His-Fe bond of the alpha-chains, and the accessibility of the betaCys93 residues in this derivative is decreased fourfold. The present results are consistent with the conclusion that the quaternary structure of TTDS-Hb is shifted toward the R state whereas the quaternary structure of TMMP-Hb is shifted toward the T state and provides additional evidence that the identity of the residues involved in intramolecular cross-linking of hemoglobin within the 2,3-bisphosphoglycerate binding site between beta-chains can have a significant influence on the conformational and functional properties of the protein.

In vitro evolution of horse heart myoglobin to increase peroxidase activity.

Random mutagenesis and screening for enzymatic activity has been used to engineer horse heart myoglobin to enhance its intrinsic peroxidase activity. A chemically synthesized gene encoding horse heart myoglobin was subjected to successive cycles of PCR random mutagenesis. The mutated myoglobin gene was expressed in Escherichia coli LE392, and the variants were screened for peroxidase activity with a plate assay. Four cycles of mutagenesis and screening produced a series of single, double, triple, and quadruple variants with enhanced peroxidase activity. Steady-state kinetics analysis demonstrated that the quadruple variant T39I/K45D/F46L/I107F exhibits peroxidase activity significantly greater than that of the wild-type protein with k1 (for H2O2 oxidation of metmyoglobin) of 1. 34 x 10(4) M-1 s-1 ( approximately 25-fold that of wild-type myoglobin) and k3 [for reducing the substrate (2, 2'-azino-di-(3-ethyl)benzthiazoline-6-sulfonic acid] of 1.4 x 10(6) M-1 s-1 (1.6-fold that of wild-type myoglobin). Thermal stability of these variants as measured with circular dichroism spectroscopy demonstrated that the Tm of the quadruple variant is decreased only slightly compared with wild-type (74.1 degreesC vs. 76.5 degreesC). The rate constants for binding of dioxygen exhibited by the quadruple variant are identical to the those observed for wild-type myoglobin (kon, 22.2 x 10(-6) M-1 s-1 vs. 22.3 x 10(-6) M-1 s-1; koff, 24.3 s-1 vs. 24.2 s-1; KO2, 0.91 x 10(-6) M-1 vs. 0.92 x 10(-6) M-1). The affinity of the quadruple variant for CO is increased slightly (kon, 0.90 x 10(-6) M-1s-1 vs. 0.51 x 10(-6) M-1s-1; koff, 5.08 s-1 vs. 3.51 s-1; KCO, 1.77 x 10(-7) M-1 vs. 1.45 x 10(-7) M-1). All four substitutions are in the heme pocket and within 5 A of the heme group.