Copper is a biologically essential trace element, which functions at the active sites of enzymes involved in electron transfer, processing of dioxygen, and in manipulation of certain nitrogen oxides. My group’s interest has focused on chemistry related to the latter two functions. Reversible copper(I)/dioxygen interactions occur in hemocyanin proteins, which are blood dioxygen carriers in molluscs and arthropods. Monooxygenase enzymes “activate” O2, promoting oxygen atom incorporation into biological substrates. Study of these enzymes or models for their action are also of interest in developing practical reagents or catalysts for oxidative transformations. Oxidases couple the reduction of O2 (to either hydrogen peroxide or water), while effecting one or two-electron substrate (e.g., alchohols, amines, other) oxidations. Copper ions also react with reduced dioxygen species such as superoxide (O2-), in Cu-Zn superoxide dismutases, and may also be involved in copper-mediate oxdiative damage in biological media, including possibly in Alzheimer’s disease.
A primary focus in recent years has been the study of reversible dioxygen binding with copper complexes, to elucidate possible structures of Cu-O2species, their associated spectroscopy (UV-Vis, resonance Raman, etc.), and subsequent reactivity. To illustrate both the type of approach we use in our research, and some of the results obtained, it is instructive to provide some details of our work on a compound which provided the first x-ray crystallographic description for a copper-dioxygen adduct, synthetic or protein. [{(TMPA)CuII}2(O2-)]2+ (1), a Cu/O2 2:1 adduct, with PF6- or ClO4- counter-anions, forms by reversible oxygenation of the copper(I) precursor complex [(TMPA)CuI(RCN)]+ (1a, R = Me, Et, TMPA = tris[2-pyridylmethyl]amine). Product (1) is only stable at -80 C in solution or as a solid.
Physical Properties of {[(TMPA)CuII}2(O22-)]2+ (1)
We have shown that the binding of O2 is reversible, and this can be illustrated by the spectroscopic experiments shown, wherein we can remove the bound O2 by application of a vacuum, and cycle back and forth between unoxygenated copper(I) complex (1a; lmax = 340 nm and (1; lmax = 525 nm, 600 nm) .
Models for Heme-Copper Oxidase and Related Enzymes
Heme-copper oxidases, including cytochrome c oxidase (CcO), catalyze the four-electron, four-proton reduction of dioxygen to water, while also ‘pumping’ four protons (per O2 molecule reduced) across the cell membrane. The electrochemical potential gradient generated by this process ultimately provides the driving force for ATP synthesis. The key chemistry for O2-binding, reduction, and coupled proton translocation occurs at the binuclear heme a3- CuB active-site. Recent protein x-ray structures show the copper ion ligated to three imidazole (from histidine) N-donor groups, with the high spin heme a3 being bound to an imidazole from histidine which is distal to copper (Figure).
The goals of our model studies for heme-copper oxidases are to develop chemistry relevant to (i) the O2-binding and reduction chemistry, (ii) the effects of protonation and/or addition of electrons (as in the enzyme), and (iii) the numerous studies carried out on oxidized resting state of the enzyme, which may give Fe-X-Cu species (X = oxide, hydroxide, cyanide, azide, carboxylate).
In our initial attempts to delve into this type of chemistry, we chose to use TMPA-copper complexes, with known O2-chemistry (see above). When (3) and (1a) react with O2 (1/2 equivalent), [(F8-TPP)FeIII-(O2-)-CuII(TMPA)]+ (4) is isolated. This chemistry already represents a crude functional model for CcO, since O2 has been reduced to the oxidation state level of water; isotope labeling studies indicate the bridging oxo group in (4) comes from O2. Interestingly, the acid-base reaction of equimolar amounts of (5), (6), and Et3N gives the same complex, (4).
The x-ray structure of (7) reveals a linear Fe(III)-oxo-Cu(II) coordination (angle of Fe-O-Cu = 178.2?, with very short Fe-O and Cu-O bond lengths. The complex has also been characterized by UV-Vis, IR, Mssbauer, and NMR spectroscopies, as well as magnetic susceptibility measurements. The 1H-NMR spectrum is very rich, with resonances ranging from +65 ppm downfield to -104 ppm upfield. A detailed NMR study has been published (ref. # 146) and the unusual behavior is ascribed to the S = 2 ground state with strong antiferromagnetic coupling between Fe(III) and Cu(II) ions.
Protonation of (4) by addition of one equivalent of triflic acid produces the hydroxo-bridged complex, [(F8-TPP)Fe-(OH)-Cu(TMPA)]2+ (7). Structural insights from x-ray absorption spectroscopy reveal elongated Fe-OH and Cu-OH bonds and a bending such that angle of Fe-O-Cu at 157? Complex (7) is a plausible model for the resting (‘as-isolated) oxidized enzyme, with its supposed Fe-X-Cu antiferromagnetic coupling. Since protonation of enzyme turnover intermediates leads to product water and/or translocated protons, it is of interest that the oxo moiety in (4) is basic; Protonation experiments suggest that pKa ~ 8 ?2.5 in water.
While the dioxygen chemistry and mechanism of formation of (4) from (1a) and (3) is of interest, we have chosen to emphasize related studies on binucleating ligands, where a tetradentate or tridentate donor for copper (or another metal ion) is synthetically tethered to the periphery of a porphyrin. Recent work with the 6L (see Figure) indicates analogous chemistry affords a di-iron species with 6L (ref. 136) and [(6L)FeIII-OH], with empty tether. Reaction of [(6L)FeIII-OH] with a Cu(II) salt in presence of a base results in the formation of the new oxo-bridged complex [(6L)FeIII-O-CuII]+, which has been characterized by UV-Vis and 1H-NMR spectroscopies, mass spectrometry, and X-ray diffraction (see below). Preliminary observations show that a reduced species [(6L)FeII-CuI]+ can be generated, and that its subsequent reaction with O2 leads to [(6L)FeIII-O-CuII]+. [(6L)FeIII-O-CuI]+ is prepared by reduction of [(6L)FeIII-OH] with dithionite followed by addition of a copper(I) salt. It is characterized by lambda(max) = 424 nm, and by 1H-NMR properties consistent with high-spin Fe(II) in tetrahydrofuran (THF) as solvent. Interestingly, when the oxygenation of this reduced compound is carried out at -80 �C in THF, a new spectroscopically distinct species is observed; this cleanly converts to [(6L)FeIII-O-CuII]+ after warming to room-temperature. These observations bode well for future investigations where we wish to probe the possibility of forming Fe-peroxo-Cu or other interesting intermediates which may have relevance to the processes of O2-binding, reduction, O-O cleavage and proton translocation by the Fe-Cu center in cytochrome c oxidases.
Other Research Projects or Activities
A number of other projects have been or are being studied in our laboratories. In some cases, these have arisen as natural extensions of chemistry uncovered from the investigations described above.
Nitrogen-Oxide Cu or Fe Chemistry
We have been off-and-on interested in the chemistry of nitrogen oxides (e.g., NO, N2O, NO2-) (see publications #77, #85, #135), since their chemistry is of environmental significance, nitrite (NO2-) and nitrous oxide (N2O) are natural substrates for copper enzymes (for example in detrifying bacteria), and (as mentioned above), NO(g) is a substrate for a heme/non-heme diiron active site in NO Reductase. Nitric oxide has also been used to probe the active site chemistry of iron and copper enzymes, since NO(g) is considered an O2 surrogate. We have found that NO(g) readily reacts with our copper(I) complexes, with evolution of N2O, and these reactions are being studied mechanistically. Structural, spectrsocopic and rectivity models for NO Reductase are also being investigated. We are also trying to design compounds which may be reactive enough to deoxygenate nitrous oxide, which is kinetically highly inert, but which is known to be reduced at a dicopper enzyme active center in Nitrous Oxide Reductase.
Hydrolysis
The actives sites of a variety of hydrolases, for example enzymes which carry out ester (e.g. phosphate esters) or amide (peptide) hydrolysis reactions, often contain binuclear or trinuclear metal active sites. The metal ion involved may be zinc, manganese, nickel, iron, magnesium, or even copper as a substitute has activity. A key feature of the chemistry appears to be a metal-hydroxide or M-(OH)-M moiety. In our copper-dioxygen chemistry studies, binuclear or trinuclear copper-hydroxide complexes have been generated and characterized. As a consequence, we have studies several examples of dicopper complex mediated hydrolysis, and have found our compounds are capable of hydrolysis of unactived amides (publication #113), esters or even nitriles (publication #144), and carbon dioxide (publication # 108). A novel case occurs when hydrolysis of amides can be effected through direct copper-dioxygen chemistry (#113 ). Continued study of hydrolysis chemistry with synthetically designed multimetal (Zn, Cu, Mn) complexes is planned.
Reactions of Copper Complexes with DNA/Proteins
The study of metal complexes which are capbable of oxidative chemistry and/or hydrolytic activity is of interest with respect to interactions with DNA or proteins. Metal complexes may be utilized as ‘footprinting’ agents to help probe DNA or protein structure, and/or DNA/protein complexes which are important in gene regulation. Copper complexes, especially a Cu-phenanthroline species, has been found to be of considerable biochemical or biotechnological interest. Thus, with the many types of redox and hydrolytically active compounds studied in our group, we have a small but active ongoing effort to survey our complexes in this regard. Initial studies (see publication #138) show that a trinuclear complex we have synthesized is very active in oxidative cleavage of DNA plasmids. Further studies are in progress, aimed at a better mechanistic understanding, discovering possible hydrolytic processes, and extending the investigations to peptides or proteins.
Paramagnetic NMR
Thorough characterization of complexes generated in our group inlucdes characterization of NMR-spectroscopy. Interestingly, a number of dicopper(II) complexes we have studies show very sharp, but paramagnetically shifted proton NMR signals. Ligand-based NMR signals for copper(II) complexes is unusual because of unfavorable relaxation times, but the binuclear nature of the complexes is seen to be responsible (see publication #140). Further investigation will seek to broaden the scope of complex type exhibiting this kind of NMR spectroscopic behavior, and the application to characterization paramagnetic multinuclear copper complexes involved in O2-binding or oxidative chemistry. We have also studied (publication #146) and are continuing to study in some detail the highly interesting NMR-spectroscopic behavior of antiferromagnetically coupled Fe(III)-X-Cu(II) (S = 2) systems, relevant to our heme-copper model program (see above).