Copper coordination in blue proteins.

The spectroscopic and electrochemical properties of blue copper proteins are strikingly different from those of inorganic copper complexes in aqueous solution. Over three decades ago this unusual behavior was ascribed to constrained coordination in the folded protein; consistent with this view, crystal structure determinations of blue proteins have demonstrated that the ligand positions are essentially unchanged on reduction as well as in the apoprotein. Blue copper reduction potentials are tuned to match the particular function of a given protein by exclusion of water from the metal site and strict control of the positions of axial ligands in the folded structure. Extensive experimental work has established that the reorganization energy of a prototypal protein, Pseudomonas aeruginosa azurin, is approximately 0.7 eV, a value that is much lower than those of inorganic copper complexes in aqueous solution. The lowered reorganization energy in the protein, which is attributable to constrained coordination, is critically important for function, since the driving forces for electron transfer often are low (approximately 0.1 eV) between blue copper centers and distant (>10 A) donors and acceptors.

In vitro membrane penetration of modified peptide nucleic acid (PNA).

Efficient cellular uptake is crucial for the success of any drug directed towards targets inside cells. Peptide nucleic acid (PNA), a DNA analog with a promising potential as a gene-directed drug, has been shown to display slow membrane penetration in cell cultures. We here used liposomes as an in vitro model of cell membranes to investigate the effect on penetration of a PNA molecule colvalently modified with a lipophilic group, an adamantyl moiety. The adamantyl attachment was found to increase the membrane-penetration rate of PNA three-fold, as compared to corresponding unmodified PNA. From the penetration behaviour of a number of small and large molecules we could conclude that passive diffusion is the mechanism for liposome-membrane passage. Flow linear dichroism (LD) of the modified PNA in presence of rod-shaped micelles, together with octanol-water distribution experiments, showed that the adamantyl-modified PNA is amphiphilic; the driving force behind the observed increased membrane-penetration rate appears to be an accumulation of the PNA in the lipid double layer.

The role of a single N-linked glycosylation site for a functional epitope of herpes simplex virus type 1 envelope glycoprotein gC.

A monoclonal antibody, B1C1, binding to an epitope of antigenic site II of the herpes simplex virus type 1 (HSV-1) glycoprotein gC-1, is a potent inhibitor of two important biological functions of gC-1: its binding to cell surface heparan sulfate and its binding to the receptor for complement factor C3b. Here, we have analyzed a B1C1-resistant HSV-1 variant (HSV-12762/B1C1B4.2), obtained after passage of wild type HSV-1 (HSV-12762) in the presence of high concentrations of B1C1. The transport of newly synthesized mutant gC-1 to the cell surface was comparable to that of wild type glycoprotein, but no binding of surface-associated mutant gC-1 to B1C1 was detected. However, mutant and wild type gC-1 bound equally well to other site II Mabs. Attachment of wild type but not mutant virus was inhibited by B1C1. Sequencing of the mutant gC-1 gene revealed only one nucleotide change, resulting in replacement of Thr150 by an Ile, in turn destroying an N-glycosylation site at Asn148. Loss of one complex type N-linked glycan was confirmed by endoglycosidase digestion and subsequent SDS-polyacrylamide gel electrophoresis. Circular dichroism analysis of purified gC-1 from cells infected with mutant or wild type virus did not reveal any difference in secondary structure between mutant and wild type gC-1. It was not possible to obtain a B1C1-resistant phenotype by nucleotide-directed mutagenesis of gC-1 where Asn148 was changed to a glutamine. These data demonstrated that the threonine of the glycosylation site and not the N-linked glycan in itself was essential for B1C1 binding

High-potential states of blue and purple copper proteins.

Electrochemical measurements show that there are high-potential states of two copper proteins, Pseudomonas aeruginosa azurin and Thermus thermophilus CuA domain; these perturbed states are formed in guanidine hydrochloride (GuHCl) solution in which the proteins are still blue (azurin) and purple (CuA). In each case, the high-potential state forms reversibly. Absorption (azurin, CuA), visible circular dichroism (azurin, CuA), resonance-Raman (CuA), and EPR (CuA) spectra indicate that the structure of the oxidized copper site of each high-potential form is very similar to that of the native protein. It is proposed that GuHCl perturbs one or more H-bonds in the blue or purple copper active site, thereby allowing Cu(I) to adopt a more favorable coordination structure than that in the rigid cavity of the native protein.

The chemical biology of copper.

Major progress was made in 1997 in the understanding of the biological transport of copper. Blue copper and CuA sites have very low electron transfer reorganization energies. The mechanisms of copper-containing oxygenases and oxidases have been clarified by recent crystal structure determinations. Protein folding has been shown to tune the reduction potentials of blue copper proteins by hydrophobic encapsulation of the active sites and strict control of the axial ligation.

Pathways for electron tunneling in cytochrome c oxidase.

Warburg showed in 1929 that the photochemical action spectrum for CO dissociation from cytochrome c oxidase is that of a heme protein. Keilin had shown that cytochrome a does not react with oxygen, so he did not accept Warburg's view until 1939, when he discovered cytochrome a3. The dinuclear cytochrome a3-CuB unit was found by EPR in 1967, whereas the dinuclear nature of the CuA site was not universally accepted until oxidase crystal structures were published in 1995. There are negative redox interactions between cytochrome a and the other redox sites in the oxidase, so that the reduction potential of a particular site depends on the redox states of the other sites. Calculated electron-tunneling pathways for internal electron transfer in the oxidase indicate that the coupling-limited rates are 9 x 10(5) (CuA-->a) and 7 x 10(6) s(-1) (a-->a3); these calculations are in reasonable agreement with experimental rates, after corrections are made for driving force and reorganization energy. The best CuA-a pathway starts from the ligand His204 and not from the bridging sulfur of Cys196, and an efficient a-a3 path involves the heme ligands His378 and His376 as well as the intervening Phe377 residue. All direct paths from CuA to a3 are poor, indicating that direct CuA-->a3 electron transfer is much slower than the CuA-->a reaction. The pathways model suggests a means for gating the electron flow in redox-linked proton pumps.

Effect of redox state on the folding free energy of a thermostable electron-transfer metalloprotein: the CuA domain of cytochrome oxidase from Thermus thermophilus.

The unfolding of the CuA domain of cytochrome oxidase from the thermophilic bacterium Thermus thermophilus, induced by guanidine hydrochloride (GuHCl)1 at different temperatures, has been monitored by CD as well by electronic absorption (with the oxidized protein) and by fluorescence (with the reduced protein). The same unfolding curves were obtained with the different methods, providing evidence for a two-state model for the unfolding equilibrium. This was also supported by the shape of the unfolding equilibrium curves and by the observed refolding of the unfolded, oxidized protein on dilution of the denaturant. The oxidized protein cannot be unfolded by GuHCl at room temperature, and it was found to be thermally very stable as well, since, even in the presence of 7 M GuHCl, it is not fully unfolded until above 80 degrees C. For the reduced protein at room temperature, the unfolding equilibrium curve yielded a folding free energy of -65 kJ/mol. The corresponding value for the oxidized protein (-85 kJ/mol) could be estimated indirectly from a thermodynamic cycle connecting the folded and unfolded forms in both oxidation states and the known reduction potentials of the metal site in the folded and unfolded states; the potential is increased on unfolding, consistent with the higher folding stability of the oxidized form. The difference in folding stability between the oxidized and reduced proteins (20 kJ/mol) is exceptionally high, and this is ascribed to the unique structure of the dinuclear CuA site. The unfolded, reduced protein was found to refold partially on oxidation with ferricyanide.

The effect of the metal ion on the folding energetics of azurin: a comparison of the native, zinc and apoprotein.

The unfolding by guanidine hydrochloride (GuHCl) and the refolding on dilution of zinc and apoazurin have been monitored by far-UV circular dichroism (CD). With the native protein, the unfolding was followed by CD and optical absorption in the visible spectral region. With the zinc protein, the reversible unfolding has also been followed by tryptophan fluorescence and NMR. The zinc and Cu2+ metal ions remain associated with the protein in the unfolded state. When the unfolding of the native protein is followed by CD, the initial, reversible transition due to unfolding is followed by a slow change associated with the reduction of Cu2+ by the thiol group of the ligand Cys112. The unfolding of apoazurin displays two CD transitions, which evidence suggests represent different folding domains, the least stable one including the metal-binding site and the other one the rest of the beta-sheet structure. Both occur at a lower GuHCl concentration than the unfolding of the native protein. The CD titrations also demonstrate that zinc azurin has a lower stability than the copper protein. Unfolding of zinc azurin followed by tryptophan fluorescence occurs at a much lower GuHCl concentration than the CD changes, and NMR spectra show that there is no loss of secondary and tertiary structure at this concentration, whereas the CD-detected loss of secondary structure correlates with the NMR changes. Thus, the fluorescence change is ascribed to a small local perturbation of the structure around the single tryptophan residue. The differences in stability of the three forms of azurin are discussed in terms of the rack mechanism. A bound metal ion stabilizes the native fold, and this stabilization is larger for Cu(II) than for Zn(II), reflecting the higher affinity of the protein for Cu(II).

Effects of folding on metalloprotein active sites.

Experimental data for the unfolding of cytochrome c and azurin by guanidinium chloride (GuHCl) are used to construct free-energy diagrams for the folding of the oxidized and reduced proteins. With cytochrome c, the driving force for folding the reduced protein is larger than that for the oxidized form. Both the oxidized and the reduced folded forms of yeast cytochrome c are less stable than the corresponding states of the horse protein. Due to the covalent attachment of the heme and its fixed tetragonal coordination geometry, cytochrome c folding can be described by a two-state model. A thermodynamic cycle leads to an expression for the difference in self-exchange reorganization energies for the folded and unfolded proteins. The reorganization energy for electron exchange in the folded protein is approximately 0.5 eV smaller than that for a heme in aqueous solution. The finding that reduced azurin unfolds at lower GuHCl concentrations than the oxidized protein suggests that the coordination structure of copper is different in oxidized and reduced unfolded states: it is likely that the geometry of CuI in the unfolded protein is linear or trigonal, whereas CuII prefers to be tetragonal. The evidence indicates that protein folding lowers the azurin reorganization energy by roughly 1.7 eV relative to an aqueous Cu(1, 10-phenanthroline)22+/+ reference system.

A flash-photolysis study of the reactions of a caa3-type cytochrome oxidase with dioxygen and carbon monoxide.

The time course of absorbance changes following flash photolysis of the fully-reduced carboxycytochrome oxidase from Bacillus PS3 in the presence of O2 has been followed at 445, 550, 605, and 830 nm, and the results have been compared with the corresponding changes in bovine cytochrome oxidase. The PS3 enzyme has a covalently bound cytochrome c subunit and the fully-reduced species therefore accommodates five electrons instead of four as in the bovine enzyme. In the bovine enzyme, following CO dissociation, four phases were observed with time constants of about 10 microseconds, 30 microseconds, 100 microseconds, and I ms at 445 nm. The initial, 10-microsecond absorbance change at 445 nm is similar in the two enzymes. The subsequent phases involving heme a and CuA are not seen in the PS3 enzyme at 445 nm, because these redox centers are re-reduced by the covalently bound cytochrome c, as indicated by absorbance changes at 550 nm. A reaction scheme consistent with the experimental observations is presented. In addition, internal electron-transfer reactions in the absence of O2 were studied following flash-induced CO dissociation from the mixed-valence enzyme. Comparisons of the CO recombination rates in the mixed-valence and fully-reduced oxidases indicate that more electrons were transferred from heme a3 to a in PS3 oxidase compared to the bovine enzyme.

Site-directed mutagenesis of residues lining a putative proton transfer pathway in cytochrome c oxidase from Rhodobacter sphaeroides.

Several putative proton transfer pathways have been identified in the recent crystal structures of the cytochrome oxidases from Paracoccus denitrificans [Iwata et al. (1995) Nature 376, 660-669] and bovine [Tsukihara (1996) Science 272, 1138-1144]. A series of residues along one face of the amphiphilic transmembrane helix IV lie in one of these proton transfer pathways. The possible role of these residues in proton transfer was examined by site-directed mutagenesis. The three conserved residues of helix IV that have been implicated in the putative proton transfer pathway (Ser-201, Asn-207, and Thr-211) were individually changed to alanine. The mutants were purified, analyzed for steady-state turnover rate and proton pumping efficiency, and structurally probed with resonance Raman spectroscopy and FTIR difference spectroscopy. The mutation of Ser-201 to alanine decreased the enzyme turnover rate by half, and was therefore further characterized using EPR spectroscopy and rapid kinetic methods. The results demonstrate that none of these hydrophilic residues are essential for proton pumping or oxygen reduction activities, and suggest a model of redundant or flexible proton transfer pathways. Whereas previously reported mutants at the start of this putative channel (e.g., Asp-132-Asn) dramatically influence both enzyme turnover and coupling to proton pumping, the current work shows that this is not the case for all residues observed in this channel.

Redox-linked conformational changes in cytochrome c oxidase.

The CD spectrum of oxidized cytochrome oxidase in the wavelength region 185-260 nm shows that the secondary structure of the protein consists of close to 60% alpha-helix and slightly less than 20% beta-structure. CD spectra of oxidized cytochrome oxidase, of half and fully reduced carboxycytochrome oxidase as well as of fully reduced cytochrome oxidase, have been recorded in the wavelength region 200-260 nm. The results demonstrate a conformational change on going from the oxidized to the half reduced state in carboxycytochrome oxidase; no further change occurs on full reduction. A conformational change is also seen in the fully reduced enzyme without bound CO. The conformational transitions are suggested to be part of the proton pumping mechanism of cytochrome oxidase.

Water-soluble, recombinant CuA-domain of the cytochrome ba3 subunit II from Thermus thermophilus.

Recently, the genes of cytochrome ba3 from thermus thermophilus [Keightley, J.A., et al. (1995) J. Biol. Chem. 270, 20345-20358], a homolog of the heme-copper oxidase family, have been cloned. We report here expression of a truncated gene, encoding the copper A (CuA) domain of cytochrome ba3, that is regulated by a T7 RNA polymerase promoter in Escherichia coli. The CuA-containing domain is purified in high yields as a water-soluble, thermostable, purple-colored protein. Copper analysis by chemical assay, mass spectrometry, X-ray fluorescence, and EPR spin quantification show that this protein contains two copper ions bound in a mixed-valence state, indicating that the CuA site in cytochrome ba3, is a binuclear center. The absorption spectrum of the CuA site, free of the heme interference in cytochrome ba3, is similar to the spectra of other soluble fragments from the aa3-type oxidase of Parachccus denitrificans [Lappalainen, P., et al. (1993) J. Biol Chem. 268, 26416-26421] and the caa3-type oxidase of Bacillus subtilis [von Wachenfeldt, C. et al. (1994) FEBS Lett. 340, 109-113]. There are intense bands at 480 nm (3100 M(-1) cm(-1)) and 530 nm (3200 M(-1) cm(-1)), a band in the near -IR centered at 790 nm (1900 M(-1) cn(-1)), and a weaker band at 363 nm (1300M(-1) cm(-1)). The visible CD spectrum shows a positive-going band at 460 nm and a negative-going band at 527 nm, the opposite signs of which may result from the binuclear nature of the site. The secondary structure prediction from the far-UV CD spectrum indicates that this domain is predominantly beta-sheet, in agreement with the recent X-ray structure reported for the complete P. denitrificans cytochrome aa3 molecule [Iwata, S., et al. (1995) Nature 376, 660-669] and the engineered, purple CyoA protein [Wilmanns, M., et al. (1996) Proc. Natl Acad. Sci. U.S.A. 92, 11955-11959]. However, the thermostability of the fragment described here (Tm approximately 80 degrees C) and the stable binding of copper over a broad pH range (pH 3-9) suggest this protein may be uniquely suitable for detailed physical-chemical study.

A ligand-exchange mechanism of proton pumping involving tyrosine-422 of subunit I of cytochrome oxidase is ruled out.

The molecular mechanism by which proton pumping is coupled to electron transfer in cytochrome c oxidase has not yet been determined. However, several models of this process have been proposed which are based on changes occurring in the vicinity of the redox centers of the enzyme. Recently, a model was described in which a well-conserved tyrosine residue in subunit I (Y422) was proposed to undergo ligand exchange with the histidine ligand (H419) of the high-spin heme a3 during the catalytic cycle, allowing both residues to serve as part of a proton transporting system. Site-directed mutants of Y422 have been constructed in the aa3-type cytochrome c oxidase of Rhodobacter sphaeroides to test this hypothesis (Y422A, Y422F). The results demonstrate that Y422 is not an essential residue in the electron transfer and proton pumping mechanisms of cytochrome c oxidase. However, the results support the predicted proximity of Y422 to heme a3, as now confirmed by crystal structure. In addition, it is shown that the pH-dependent reversed electron transfer between heme a and heme a3 is normal in the Y422F mutant. Hence, these data also demonstrate that Y422 is not the residue previously postulated to interact electrostatically with heme a3, nor is it responsible for the unique EPR characteristics of heme a in this bacterial oxidase.

EPR studies of wild-type and several mutants of cytochrome c oxidase from Rhodobacter sphaeroides: Glu286 is not a bridging ligand in the cytochrome a3-CuB center.

Wild-type and several mutants of cytochrome c oxidase from Rhodobacter sphaeroides were characterized by EPR spectroscopy. A pH-induced g12 signal, seen previously in mammalian cytochrome oxidase and assigned to the presence of a bridging carboxyl ligand in the bimetallic cytochrome a3-CuB site, is found also in the bacterial enzyme. Mutation of glutamate-286 to glutamine inactivates the enzyme but does not affect this signal, demonstrating that the carboxyl group of this residue is not the bridging ligand. Three mutants, M106Q, located one helix turn below a histidine ligand to cytochrome a, and T352A as well as F391Q, located close to the bimetallic center, are shown to affect dramatically the low-spin heme signal of cytochrome a. These mutants are essentially inactive, suggesting that these three mutations result in alterations to cytochrome a that render the oxidase non-functional.

Phospholipid membrane permeability of peptide nucleic acid.

Phospholipid vesicles (liposomes) as membrane models have been used to study the penetration properties of peptide nucleic acid (PNA), a new DNA analog in which the nucleobases are attached to a pseudo-peptide backbone. The liposomes were characterised by carboxyfluorescein efflux, light-scattering and cryogenic transmission electron microscopy. The liposome structure was found not to be affected by the incorporation of PNA or an oligonucleotide. Two 10-mer fluorescein-labelled PNAs were found to have low efflux rates (half-times of 5.5 and 11 days), comparable to that of a 10-mer oligonucleotide (half-time of 7 days). We conclude that passive diffusion of unmodified PNA over the lipid membrane is not likely to be an effective way of transport into biological cells.

The CuA center of cytochrome-c oxidase: electronic structure and spectra of models compared to the properties of CuA domains.

The electronic structure and spectrum of several models of the binuclear metal site in soluble CuA domains of cytochrome-c oxidase have been calculated by the use of an extended version of the complete neglect of differential overlap/spectroscopic method. The experimental spectra have two strong transitions of nearly equal intensity around 500 nm and a near-IR transition close to 800 nm. The model that best reproduces these features consists of a dimer of two blue (type 1) copper centers, in which each Cu atom replaces the missing imidazole on the other Cu atom. Thus, both Cu atoms have one cysteine sulfur atom and one imidazole nitrogen atom as ligands, and there are no bridging ligands but a direct Cu-Cu bond. According to the calculations, the two strong bands in the visible region originate from exciton coupling of the dipoles of the two copper monomers, and the near-IR band is a charge-transfer transition between the two Cu atoms. The known amino acid sequence has been used to construct a molecular model of the CuA site by the use of a template and energy minimization. In this model, the two ligand cysteine residues are in one turn of an alpha-helix, whereas one ligand histidine is in a loop following this helix and the other one is in a beta-strand.