Equilibrium Studies of Designed Metalloproteins.

Complete thermodynamic descriptions of the interactions of cofactors with proteins via equilibrium studies are challenging, but are essential to the evaluation of designed metalloproteins. While decades of studies on protein-protein interaction thermodynamics provide a strong underpinning to the successful computational design of novel protein folds and de novo proteins with enzymatic activity, the corresponding paucity of data on metal-protein interaction thermodynamics limits the success of computational metalloprotein design efforts. By evaluating the thermodynamics of metal-protein interactions via equilibrium binding studies, protein unfolding free energy determinations, proton competition equilibria, and electrochemistry, a more robust basis for the computational design of metalloproteins may be provided. Our laboratory has shown that such studies provide detailed insight into the assembly and stability of designed metalloproteins, allow for parsing apart the free energy contributions of metal-ligand interactions from those of porphyrin-protein interactions in hemeproteins, and even reveal their mechanisms of proton-coupled electron transfer. Here, we highlight studies that reveal the complex interplay between the various equilibria that underlie metalloprotein assembly and stability and the utility of making these detailed measurements.

Metalloprotein and redox protein design.

Metalloprotein and redox protein design are rapidly advancing toward the chemical synthesis of novel proteins that have predictable structures and functions. Current data demonstrate a breadth of successful approaches to metallopeptide and metalloprotein design based on de novo, rational and combinatorial strategies. These sophisticated synthetic analogs of natural proteins constructively test our comprehension of metalloprotein structure/function relationships. Additionally, designed redox proteins provide novel constructs for examining the thermodynamics and kinetics of biological electron transfer.

Hydrophobic modulation of heme properties in heme protein maquettes.

We have investigated the properties of the two hemes bound to histidine in the H10 positions of the uniquely structured apo form of the heme binding four-helix bundle protein maquette [H10H24-L6I,L13F](2), here called [I(6)F(13)H(24)](2) for the amino acids at positions 6 (I), 13 (F) and 24 (H), respectively. The primary structure of each alpha-helix, alpha-SH, in [I(6)F(13)H(24)](2) is Ac-CGGGEI(6)WKL.H(10)EEF(13)LKK.FEELLKL.H(24)EERLKK.L-CONH(2). In our nomenclature, [I(6)F(13)H(24)] represents the disulfide-bridged di-alpha-helical homodimer of this sequence, i.e., (alpha-SS-alpha), and [I(6)F(13)H(24)](2) represents the dimeric four helix bundle composed of two di-alpha-helical subunits, i.e., (alpha-SS-alpha)(2). We replaced the histidines at positions H24 in [I(6)F(13)H(24)](2) with hydrophobic amino acids incompetent for heme ligation. These maquette variants, [I(6)F(13)I(24)](2), [I(6)F(13)A(24)](2), and [I(6)F(13)F(24)](2), are distinguished from the tetraheme binding parent peptide, [I(6)F(13)H(24)](2), by a reduction in the heme:four-helix bundle stoichiometry from 4:1 to 2:1. Iterative redesign has identified phenylalanine as the optimal amino acid replacement for H24 in the context of apo state conformational specificity. Furthermore, the novel second generation diheme [I(6)F(13)F(24)](2) maquette was related to the first generation diheme [H10A24](2) prototype, [L(6)L(13)A(24)](2) in the present nomenclature, via a sequential path in sequence space to evaluate the effects of conservative hydrophobic amino acid changes on heme properties. Each of the disulfide-linked dipeptides studied was highly helical (>77% as determined from circular dichroism spectroscopy), self-associates in solution to form a dimer (as determined by size exclusion chromatography), is thermodynamically stable (-DeltaG(H)2(O) >18 kcal/mol), and possesses conformational specificity that NMR data indicate can vary from multistructured to single structured. Each peptide binds one heme with a dissociation constant, K(d1) value, tighter than 65 nM forming a series of monoheme maquettes. Addition of a second equivalent of heme results in heme binding with a K(d2) in the range of 35-800 nM forming the diheme maquette state. Single conservative amino acid changes between peptide sequences are responsible for up to 10-fold changes in K(d) values. The equilibrium reduction midpoint potential (E(m7.5)) determined in the monoheme state ranges from -156 to -210 mV vs SHE and in the diheme state ranges from -144 to -288 mV. An observed heme-heme electrostatic interaction (>70 mV) in the diheme state indicates a syn global topology of the di-alpha-helical monomers. The heme affinity and electrochemistry of the three H24 variants studied identify the tight binding sites (K(d1) and K(d2) values <200 nM) having the lower reduction midpoint potentials (E(m7.5) values of -155 and -260 mV) with the H10 bound hemes in the parent tetraheme state of [H10H24-L6I,L13F](2), here called [I(6)F(13)H(24)](2). The results of this study illustrate that conservative hydrophobic amino acid changes near the heme binding site can modulate the E(m) by up to +/-50 mV and the K(d) by an order of magnitude. Furthermore, the effects of multiple single amino acid changes on E(m) and K(d) do not appear to be additive.

Biotin synthase contains two distinct iron-sulfur cluster binding sites: chemical and spectroelectrochemical analysis of iron-sulfur cluster interconversions.

Biotin synthase is an iron-sulfur protein that utilizes AdoMet to catalyze the presumed radical-mediated insertion of a sulfur atom between the saturated C6 and C9 carbons of dethiobiotin. Biotin synthase (BioB) is aerobically purified as a dimer that contains [2Fe-2S](2+) clusters and is inactive in the absence of additional iron and reductants, and anaerobic reduction of BioB with sodium dithionite results in conversion to enzyme containing [4Fe-4S](2+) and/or [4Fe-4S](+) clusters. To establish the predominant cluster forms present in biotin synthase in anaerobic assays, and by inference in Escherichia coli, we have accurately determined the extinction coefficient and cluster content of the enzyme under oxidized and reduced conditions and have examined the equilibrium reduction potentials at which cluster reductions and conversions occur as monitored by UV/visible and EPR spectroscopy. In contrast to previous reports, we find that aerobically purified BioB contains ca. 1.2-1.5 [2Fe-2S](2+) clusters per monomer with epsilon(452) = 8400 M(-)(1) cm(-)(1) per monomer. Upon reduction, the [2Fe-2S](2+) clusters are converted to [4Fe-4S] clusters with two widely separate reduction potentials of -140 and -430 mV. BioB reconstituted with excess iron and sulfide in 60% ethylene glycol was found to contain two [4Fe-4S](2+) clusters per monomer with epsilon(400) = 30 000 M(-)(1) cm(-)(1) per monomer and is reduced with lower midpoint potentials of -440 and -505 mV, respectively. Finally, as predicted by the measured redox potentials, enzyme incubated under typical anaerobic assay conditions is repurified containing one [2Fe-2S](2+) cluster and one [4Fe-4S](2+) cluster per monomer. These results indicate that the dominant stable cluster state for biotin synthase is a dimer containing two [2Fe-2S](2+) and two [4Fe-4S](2+) clusters.

Proof of principle in a de novo designed protein maquette: an allosterically regulated, charge-activated conformational switch in a tetra-alpha-helix bundle.

New understanding of the engineering and allosteric regulation of natural protein conformational switches (such as those that couple chemical and ionic signals, mechanical force, and electro/chemical free energy for biochemical activation, catalysis, and motion) can be derived from simple de novo designed synthetic protein models (maquettes). We demonstrate proof of principle of both reversible switch action and allosteric regulation in a tetra-alpha-helical bundle protein composed of two identical di-helical subunits containing heme coordinated at a specific position close to the disulfide loop region. Individual bundles assume one of two switch states related by large-scale mechanical changes: a syn-topology (helices of the different subunits parallel) or anti-topology (helices antiparallel). Both the spectral properties of a coproporphyrin probe appended to the loop region and the distance-dependent redox interaction between the hemes identify the topologies. Beginning from a syn-topology, introduction of ferric heme in each subunit (either binding or redox change) shifts the topological balance by 25-50-fold (1.9-2.3 kcal/mol) to an anti-dominance. Charge repulsion between the two internal cationic ferric hemes drives the syn- to anti-switch, as demonstrated in two ways. When fixed in the syn-topology, the second ferric heme binding is 25-80-fold (1.9-2.6 kcal/mol) weaker than the first, and adjacent heme redox potentials are split by 80 mV (1.85 kcal/mol), values that energetically match the shift in topological balance. Allosteric and cooperative regulation of the switch by ionic strength exploits the shielded charge interactions between the two hemes and the exposed, cooperative interactions between the coproporphyrin carboxylates.

Heme redox potential control in de novo designed four-alpha-helix bundle proteins.

The effects of various mechanisms of metalloporphyrin reduction potential modulation were investigated experimentally using a robust, well-characterized heme protein maquette, synthetic protein scaffold H10A24 [¿CH(3)()CONH-CGGGELWKL.HEELLKK.FEELLKL.AEERLKK. L-CONH(2)()¿(2)](2). Removal of the iron porphyrin macrocycle from the high dielectric aqueous environment and sequestration within the hydrophobic core of the H10A24 maquette raises the equilibrium reduction midpoint potential by 36-138 mV depending on the hydrophobicity of the metalloporphyrin structure. By incorporating various natural and synthetic metalloporphyrins into a single protein scaffold, we demonstrate a 300-mV range in reduction potential modulation due to the electron-donating/withdrawing character of the peripheral macrocycle substituents. Solution pH is used to modulate the metalloporphyrin reduction potential by 160 mV, regardless of the macrocycle architecture, by controlling the protonation state of the glutamate involved in partial charge compensation of the ferric heme. Attempts to control the reduction potential by inserting charged amino acids into the hydrophobic core at close proximity to the metalloporphyrin lead to varied success, with H10A24-L13E lowering the E(m8.5) by 40 mV, H10A24-E11Q raising it by 50 mV, and H10A24-L13R remaining surprisingly unaltered. Modifying the charge of the adjacent metalloporphyrin, +1 for iron(III) protoporphyrin IX or neutral for zinc(II) protoporphyrin IX resulted in a loss of 70 mV [Fe(III)PPIX](+) - [Fe(III)PPIX](+) interaction observed in maquettes. Using these factors in combination, we illustrate a 435-mV variation of the metalloporphyrin reduction midpoint potential in a simple heme maquette relative to the about 800-mV range observed for natural cytochromes. Comparison between the reduction potentials of the heme maquettes and other de novo designed heme proteins reveals global trends in the E(m) values of synthetic cytochromes.

Self-assembly of heme A and heme B in a designed four-helix bundle: implications for a cytochrome c oxidase maquette.

Heme A, a prosthetic group of cytochrome c oxidase [EC 1.9.3.1], has been introduced into two de novo designed four helix bundle proteins, [H10A24](2) and [H10H24](2), known to bind 2-4 equiv of heme B, respectively [Robertson, D. E., Farid, R. S., Moser, C. C., Mulholland, S. E., Pidikiti, R., Lear, J. D., Wand, A., J., DeGrado, W. F., and Dutton, P. L. (1994) Nature 368, 425-432]. [H10A24](2), [Ac-CGGGELWKL x HEELLKK x FEELLKL x AEERLKK x L-CONH(2)](2)(2), binds two heme A molecules per four-helix unit via bis-histidine ligation at the 10,10' positions with measured K(d) values of <0.1 and 5 nM, values much lower than those measured for heme B (K(d) values of 50 and 800 nM). The heme A-protein complex, [heme A-H10A24](2), exhibits well-defined absorption spectra in both the ferric and ferrous states, and an electron paramagnetic resonance spectrum characteristic of a low spin heme in the ferric form. A single midpoint redox potential (E(m8)) was determined for [heme A-H10A24](2) at -45 mV (vs SHE), which is significantly higher than that of the protein bound heme B (-130 and -200 mV). The observation of a single midpoint redox potential for [heme A-H10A24](2) and a pair of midpoints for [heme B-H10A24](2) indicates that the di-alpha-helical monomers are oriented in an anti topology (disulfides on opposite sides of bundle) in the former (lacking heme-heme electrostatic interaction) and syn in the latter. A mixture of global topologies was indicated by the potentiometric titration of the related [heme A-H10H24](2) which possess two distinct reduction potentials of +41 (31%) and -65 mV (69%). Self-assembly of the mixed cofactor heme A-heme B-[H10A24](2) was accomplished by addition of a single equivalent of each heme A and heme B to [H10A24](2). The single midpoint redox potential of heme B, E(m8) = -200 mV, together with the split midpoint redox potential of heme A in heme A-heme B-[H10A24](2), E(m8) = +28 mV (33%) and -65 mV (67%), indicated the existence of both syn and anti topologies of the two di-alpha-helical monomers in this four helix bundle. Synthesis of the mixed cofactor [heme A-heme B-H10H24](2) was accomplished by addition of a 2 equiv of each heme A and heme B to [H10H24](2) and potentiometry indicated the pair of hemes B resided in the 10,10' sites and heme A occupied the 24,24' sites. The results indicate that heme peripheral structure controls the orientation of the di-alpha-helical monomers in the four-helix bundle which are interchangeable between syn and anti topologies. In the reduced form, [heme A-H10A24](2), reacts quantitatively to form [carbonmonoxy-heme A-H10A24](2) as evidenced by optical spectroscopy. The synthetic [heme A-H10A24](2) can be enzymatically reduced by NAD(P)H with natural reductases under anaerobic conditions, and reversibly oxidized by dioxygen to the ferric form.

Iron-sulfur cluster interconversions in biotin synthase: dissociation and reassociation of iron during conversion of [2Fe-2S] to [4Fe-4S] clusters.

Biotin synthase catalyzes the insertion of a sulfur atom into the saturated C6 and C9 carbons of dethiobiotin. This reaction has long been presumed to occur through radical chemistry, and recent experimental results suggest that biotin synthase belongs to a family of enzymes that contain an iron-sulfur cluster and reductively cleave S-adenosylmethionine, forming an enzyme or substrate radical, 5'-deoxyadenosine, and methionine. Biotin synthase (BioB) is aerobically purified as a dimer of 38 kDa monomers that contains two [2Fe-2S](2+) clusters per dimer. Maximal in vitro biotin synthesis requires incubation of BioB with dethiobiotin, AdoMet, reductants, exogenous iron, and crude bacterial protein extracts. It has previously been shown that reduction of BioB with dithionite in 60% ethylene glycol produces one [4Fe-4S](2+/1+) cluster per dimer. In the present work, we use UV/visible and electron paramagnetic resonance spectroscopy to show that [2Fe-2S] to [4Fe-4S] cluster conversion occurs through rapid dissociation of iron from the protein followed by rate-limiting reassociation. While in 60% ethylene glycol the product of dithionite reduction is one [4Fe-4S](2+) cluster per dimer, the product in water is one [4Fe-4S](1+) cluster per dimer. Further, incubation with excess iron, sulfide, and dithiothreitol produces protein that contains two [4Fe-4S](2+) clusters per dimer; subsequent reduction with dithionite produces two [4Fe-4S](1+) clusters per BioB dimer. BioB that contains two [4Fe-4S](2+/1+) clusters per dimer is rapidly and reversibly reduced and oxidized, suggesting that this is the redox-active form of the iron-sulfur cluster in the anaerobic enzyme.

Primary steps in the energy conversion reaction of the cytochrome bc1 complex Qo site.

The primary energy conversion (Qo) site of the cytochrome bc1 complex is flanked by both high- and low-potential redox cofactors, the [2Fe-2S] cluster and cytochrome bL, respectively. From the sensitivity of the reduced [2Fe-2S] cluster electron paramagnetic resonance (EPR) spectral g(x)-band and line shape to the degree and type of Qo site occupants, we have proposed a double-occupancy model for the Qo site by ubiquinone in Rhodobacter capsulatus membrane vesicles containing the cytochrome bc1 complex. Biophysical and biochemical experiments have confirmed the double occupancy model and from a combination of these results and the available cytochrome bc1 crystal structures we suggest that the two ubiquinone molecules in the Qo site serve distinct catalytic roles. We propose that the strongly bound ubiquinone, termed Qos, is close to the [2Fe-2S] cluster, where it remains tightly associated with the Qo site during turnover, serving as a catalytic cofactor; and the weaker bound ubiquinone, Qow, is distal to the [2Fe-2S] cluster and can exchange with the membrane Qpool on a time scale much faster than the turnover, acting as the substrate. The crystallographic data demonstrates that the FeS subunit can adopt different positions. Our own observations show that the equilibrium position of the reduced FeS subunit is proximal to the Qo site. On the basis of this, we also report preliminary results modeling the electron transfer reactions that can occur in the cytochrome bc1 complex and show that because of the strong distance dependence of electron transfer, significant movement of the FeS subunit must occur in order for the complex to be able to turn over at the experimental observed rates.

Effect of inhibitors on the ubiquinone binding capacity of the primary energy conversion site in the Rhodobacter capsulatus cytochrome bc(1) complex.

A key issue concerning the primary conversion (Q(O)) site function in the cytochrome bc(1) complex is the stoichiometry of ubiquinone/ubihydroquinone occupancy. Previous evidence suggests that the Q(O) site is able to accommodate two ubiquinone molecules, the double occupancy model [Ding, H., Robertson, D. E., Daldal, F., and Dutton, P. L. (1992) Biochemistry 31, 3144-3158]. In the recently reported crystal structures of the cytochrome bc(1) complex, no electron density was identified in the Q(O) site that could be ascribed to ubiquinone. To provide further insight into this issue, we have manipulated the cytochrome bc(1) complex Q(O) site occupancy in photosynthetic membranes from Rhodobacter capsulatus by using inhibitor titrations and ubiquinone extraction to modulate the amount of ubiquinone bound in the site. The nature of the Q(O) site occupants was probed via the sensitivity of the reduced [2Fe-2S] cluster electron paramagnetic resonance (EPR) spectra to modulation of Q(O) site occupancy. Diphenylamine (DPA) and methoxyacrylate (MOA)-stilbene are known Q(O) site inhibitors of the cytochrome bc(1) complex. Addition of stoichiometric concentrations of MOA-stilbene or excess DPA to cytochrome bc(1) complexes with natural levels of ubiquinone elicits the same change in the [2Fe-2S] cluster EPR spectra; the g(x)() resonance broadens and shifts from 1. 800 to 1.783. This is exactly the same signal as that obtained when there is only one ubiquinone present in the Q(O) site. Furthermore, addition of MOA-stilbene or DPA to the cytochrome bc(1) complex depleted of ubiquinone does not alter the [2Fe-2S] cluster EPR spectral line shapes, which remain indicative of one ubiquinone or zero ubiquinones in the Q(O) site, with broad g(x)() resonances at 1. 783 or 1.765, respectively. The results are quite consistent with the Q(O) site double occupancy model, in which MOA-stilbene and DPA inhibit by displacing one, but not both, of the Q(O) site ubiquinones.

Histidine placement in de novo-designed heme proteins.

The effects of histidine residue placement in a de novo-designed four-alpha-helix bundle are investigated by placement of histidine residues at coiled coil heptad a positions in two distinct heptads and at each position within a single heptad repeat of our prototype heme protein maquette, [H10H24]2 [[Ac-CGGGELWKL x HEELLKK x FEELLKL x HEERLKK x L-CONH2]2]2 composed of a generic (alpha-SS-alpha)2 peptide architecture. The heme to peptide stoichiometry of variants of [H10H24]2 with either or both histidines on each helix replaced with noncoordinating alanine residues ([H10A24]2, [A10H24]2, and [A10A24]2) demonstrates the obligate requirement of histidine for biologically significant heme affinity. Variants of [A10A24]2, [[Ac-CGGGELWKL x AEELLKK x FEELLKL x AEERLKK x L-CONH2]2]2, containing a single histidine per helix in positions 9 to 15 were evaluated to verify the design based on molecular modeling. The bis-histidine site formed between heptad positions a at 10 and 10' bound ferric hemes with the highest affinity, Kd1 and Kd2 values of 1.5 and 800 nM, respectively. Placement of histidine at position 11 (heptad position b) resulted in a protein that bound a single heme with moderate affinity, Kd1 of 9.5 microM, whereas the other peptides had no measurable apparent affinity for ferric heme with Kd1 values >200 microM. The bis-histidine ligation of heme to [H10A24]2 and [H11A24]2 was confirmed by electron paramagnetic resonance spectroscopy. The protein design rules derived from this study, together with the narrow tolerances revealed, are applicable for improving future heme protein designs, for analyzing the results of randomized heme protein combinatorial libraries, as well as for implementation in automated protein design.

Determination of nonligand amino acids critical to [4Fe-4S]2+/+ assembly in ferredoxin maquettes.

The prototype ferredoxin maquette, FdM, is a 16-amino acid peptide which efficiently incorporates a single [4Fe-4S]2+/+ cluster with spectroscopic and electrochemical properties that are typical of natural bacterial ferredoxins. Using this synthetic protein scaffold, we have investigated the role of the nonliganding amino acids in the assembly of the iron-sulfur cluster. In a stepwise fashion, we truncated FdM to a seven-amino acid peptide, FdM-7, which incorporates a cluster spectroscopically identical to FdM but in lower yield, 29% relative to FdM. FdM-7 consists solely of the. CIACGAC. consensus ferredoxin core motif observed in natural protein sequences. Initially, all of the nonliganding amino acids were substituted for either glycine, FdM-7-PolyGly (.CGGCGGC.), or alanine, FdM-7-PolyAla (.CAACAAC.), on the basis of analysis of natural ferredoxin sequences. Both FdM-7-PolyGly and FdM-7-PolyAla incorporated little [4Fe-4S]2+/+ cluster, 6 and 7%, respectively. A systematic study of the incorporation of a single isoleucine into each of the four nonliganding positions indicated that placement either in the second or in the sixth core motif positions,.CIGCGGC. or.CGGCGIC., restored the iron-sulfur cluster binding capacity of the peptides to the level of FdM-7. Incorporation of an isoleucine into the fifth position,.CGGCIGC., which in natural ferredoxins is predominantly occupied by a glycine, resulted in a loss of [4Fe-4S] affinity. The substitution of leucine, tryptophan, and arginine into the second core motif position illustrated the stabilization of the [4Fe-4S] cluster by bulky hydrophobic amino acids. Furthermore, the incorporation of a single isoleucine into the second core motif position in a 16-amino acid ferredoxin maquette resulted in a 5-fold increase in the level of [4Fe-4S] cluster binding relative to that of the glycine variant. The protein design rules derived from this study are fully consistent with those derived from natural ferredoxin sequence analysis, suggesting they are applicable to both the de novo design and structure-based redesign of natural proteins.

Ubiquinone binding capacity of the Rhodobacter capsulatus cytochrome bc1 complex: effect of diphenylamine, a weak binding QO site inhibitor.

Diphenylamine (DPA), a known inhibitor of polyene and isoprene biosynthesis, is shown to inhibit flash-activatable electron transfer in photosynthetic membranes of Rhodobacter capsulatus. DPA is specific to the QO site of ubihydroquinone:cytochrome c oxidoreductase, where it inhibits not only reduction of the [2Fe-2S]2+ cluster in the FeS subunit and subsequent cytochrome c reduction but also heme bL reduction in the cytochrome b subunit. In both cases, the kinetic inhibition constant (Ki) is 25 +/- 10 microM. A novel aspect of the mode of action of DPA is that complete inhibition is established without disturbing the interaction between the reduced [2Fe-2S]+ cluster and the QO site ubiquinone complement, as observed from the electron paramagnetic resonance (EPR) spectral line shape of the reduced [2Fe-2S] cluster, which remained characteristic of two ubiquinones being present. These observations imply that DPA is behaving as a noncompetitive inhibitor of the QO site. Nevertheless, at higher concentrations (>10 mM), DPA can interfere with the QO site ubiquinone occupancy, leading to a [2Fe-2S] cluster EPR spectrum characteristic of the presence of only one ubiquinone in the QO site. Evidently, DPA can displace the more weakly bound of the two ubiquinones in the site, but this is not requisite for its inhibiting action.

Isolation and characterization of a two-subunit cytochrome b-c1 subcomplex from Rhodobacter capsulatus and reconstitution of its ubihydroquinone oxidation (Qo) site with purified Fe-S protein subunit.

The presence of a two-subunit cytochrome (cyt) b-c1 subcomplex in chromatophore membranes of Rhodobacter capsulatus mutants lacking the Rieske iron-sulfur (Fe-S) protein has been described previously [Davidson, E., Ohnishi, T., Tokito, M., and Daldal, F. (1992) Biochemistry 31, 3351-3358]. Here, this subcomplex was purified to homogeneity in large quantities, and its properties were characterized. As expected, it contained stoichiometric amounts of cyt b and cyt c1 subunits forming a stable entity devoid of the Fe-S protein subunit. The spectral and thermodynamic properties of its heme groups were largely similar to those of a wild-type bc1 complex, except that those of its cyt bL heme were modified as revealed by EPR spectroscopy. Dark potentiometric titrations indicated that the redox midpoint potential (Em7) values of cytochromes bH, bL, and c1 were very similar to those of a wild-type bc1 complex. The purified b-c1 subcomplex had a nonfunctional ubihydroquinone (UQH2) oxidation (Qo) site, but it contained an intact ubiquinone (UQ) reductase (Qi) site as judged by its ability to bind the Qi inhibitor antimycin A, and by the presence of antimycin A sensitive Qi semiquinone. Interestingly, its Qo site could be readily reconstituted by addition of purified Fe-S protein subunit. Reactivated complex exhibited myxothiazol, stigmatellin, and antimycin A sensitive cyt c reductase activity and an EPR gx signal comparable to that observed with a bc1 complex when the Qo site is partially occupied with UQ/UQH2. However, a mutant derivative of the Fe-S protein subunit lacking its first 43 amino acid residues was unable to reactivate the purified b-c1 subcomplex although it could bind to its Qo site in the presence of stigmatellin. These findings demonstrated for the first time that the amino-terminal membrane-anchoring domain of the Fe-S protein subunit is necessary for UQH2 oxidation even though its carboxyl-terminal domain is sufficient to provide wild-type-like interactions with stigmatellin at the Qo site of the bc1 complex.

Non-inhibiting perturbation of the primary energy conversion site (Qo site) in Rhodobacter capsulatus ubihydroquinone: cytochrome c oxidoreductase (cytochrome bc1 complex).

Ethanol added to Rhodobacter capsulatus chromatophore membranes containing the cytochrome bc1 complex effectively uncouples the sensitivity of the [2Fe-2S] cluster EPR spectrum to the number and redox state of ubiquinone/ubihydroquinone within the Qo site. Ethanol has no effect upon the rate of catalysis, leading to a non-inhibiting perturbation of cytochrome bc1 function. We suggest that displacement occurs by ethanol acting from the aqueous phase to successfully compete with the Qo site ubiquinones and water to hydrogen bond the N(epsilon)H atom(s) of the coordinating [2Fe-2S] cluster histidines.

Effect of four helix bundle topology on heme binding and redox properties.

We have designed two alternative four helix bundle protein scaffold topologies for maquette construction to examine the effect of helix orientation on the heme binding and redox properties of our prototype heme protein maquette, (alpha-SS-alpha)2, previously described as H10H24 [Robertson, D. E., Farid, R. S., Moser, C. C., Mulholland, S. E., Pidikiti, R., Lear, J. D., Wand, A. J., DeGrado, W. F., and Dutton, P. L. (1994) Nature 368, 425]. Conversion of the disulfide-bridged di-alpha-helical monomer of (alpha-SS-alpha)2 into a single polypeptide chain results in topological reorientation of the helix dipoles and side chains within a 62 amino acid helix-loop-helix monomer, (alpha-l-alpha), which self-associates to form (alpha-l-alpha)2. Addition of an N-terminal cysteine residue to (alpha-l-alpha) with subsequent oxidation yields a 126 amino acid single molecule four helix bundle, (alpha-l-alpha-SS-alpha-l-alpha). Gel permeation chromatography demonstrated that (alpha-SS-alpha)2 and (alpha'-SS-alpha')2, a uniquely structured variant of the prototype, as well as (alpha-l-alpha)2 and (alpha'-l-alpha')2 assemble into distinct four helix bundles as designed, whereas (alpha-l-alpha-SS-alpha-l-alpha) elutes as a monomeric four alpha-helix bundle. Circular dichroism (CD) spectroscopy proves that these peptides are highly alpha-helical, and incorporation of four hemes has little effect on the helical content of the secondary structure. Four heme dissociation constants were evaluated by UV-visible spectroscopy and ranged from the 15 nM to 25 microM range for each of the peptides. The presence of Cotton effects in the visible CD illustrated that the hemes reside within the protein architecture. The equilibrium redox midpoint potentials (Em8) of the four bound hemes in each peptide are between -100 and -280 mV, as determined by redox potentiometry. The heme affinity and spectroelectrochemical properties of the hemes bound to (alpha-l-alpha)2 and (alpha-l-alpha-SS-alpha-l-alpha) are similar to those of the prototype, (alpha-SS-alpha)2, and to bis-histidine ligated b-type cytochromes, regardless of the global architectural changes imposed by these topological rearrangements. The hydrophobic cores of these peptides support local electrostatic fields which result in nativelike heme chromophore properties (spectroscopy, elevated reduction potentials, heme-heme charge interaction, and reactivity with exogenous diatomics) illustrating the utility of these non-native peptides in the study of metalloproteins.

A designed cavity in the hydrophobic core of a four-alpha-helix bundle improves volatile anesthetic binding affinity.

The structural features of protein binding sites for volatile anesthetics are being explored using a defined model system consisting of a four-alpha-helix bundle scaffold with a hydrophobic core. Earlier work has demonstrated that a prototype hydrophobic core is capable of binding the volatile anesthetic halothane. Exploratory work on the design of an improved affinity anesthetic binding site is presented, based upon the introduction of a simple cavity into a prototype (alpha 2)2 four-alpha-helix bundle by replacing six core leucines with smaller alanines. The presence of such a cavity increases the affinity (Kd = 0.71 +/- 0.04 mM) of volatile anesthetic binding to the designed bundle core by a factor of 4.4 as compared to an analogous bundle core lacking such a cavity (Kd = 3.1 +/- 0.4 mM). This suggests that such packing defects present on natural proteins are likely to be occupied by volatile general anesthetics in vivo. Replacing six hydrophobic core leucine residues with alanines results in a destabilization of the folded bundle by 1.7-2.7 kcal/mol alanine, although the alanine-substituted bundle still exhibits a high degree of thermodynamic stability with an overall folded conformational delta GH2O = 14.3 +/- 0.8 kcal/mol. Covalent attachment of the spin label MTSSL to cysteine residues in the alanine-substituted four-alpha-helix bundle indicates that the di-alpha-helical peptides dimerize in an anti orientation. The rotational correlation time of the four-alpha-helix bundle is 8.1 +/- 0.5 ns, in line with earlier work on similar peptides. Fluorescence, far-UV circular dichroism, and Fourier transform infrared spectroscopies verified the hydrophobic core location of the tryptophan and cysteine residues, showing good agreement between experiment and design. These small synthetic proteins may prove useful for the study of the structural features of small molecule binding sites.