Mark P. Roach

Rational molecular design of a catalytic site: engineering of catalytic functions to the myoglobin active site framework

Abstract Proteins that contain the heme prosthetic group are responsible for many different types of catalytic activity. Understanding the mechanisms through which a particular type of catalytic activity is favored over the others remains a significant challenge. Recently, the most common strategy for structure–function studies for a particular enzyme has involved substitution of amino acid residues by site-directed mutagenesis followed by investigations of the effect of the substitution on the catalytic activity of that system. This work describes a significant departure from this common strategy. Instead, we seek to convert a non-enzymatic hemoprotein into one that is capable of catalytic activity. In so doing, we expect to gain an understanding of the general structural requirements for particular enzymatic functions. Comparison of X-ray crystal structures of myoglobin and peroxidases reveals differences in arrangement of amino acid residues in the heme pockets. On the basis of these structural differences and the reaction mechanism of peroxidases, we have rationally designed several myoglobin mutants in order to convert myoglobin into a peroxidase-like enzyme. We have discovered that the location of the distal histidine in the active site provides a critical balance between the formation and subsequent decay of the oxo-ferryl porphyrin radical cation (compound I), a catalytic species for one- and two-electron oxidation and oxygen transfer reactions. The mutants prepared in this work have been altered in such a manner that they have permitted compound I to be observed in myoglobin for the first time. This allows us to investigate mechanistic details under single turnover conditions by use of double mixing stopped-flow spectroscopy. Furthermore, some of the mutants we have constructed might be useful as good catalysts for asymmetric oxidations. In this short review, we describe our attempts to elucidate structure–function relationships on the activation of the oxygen–oxygen bond of peroxides by hemoproteins.

Proximal ligand control of heme iron coordination structure and reactivity with hydrogen peroxide: investigations of the myoglobin cavity mutant H93G with unnatural oxygen donor proximal ligands.

The role of the proximal heme iron ligand in activation of hydrogen peroxide and control of spin state and coordination number in heme proteins is not yet well understood. Although there are several examples of amino acid sidechains with oxygen atoms which can act as potential heme iron ligands, the occurrence of protein-derived oxygen donor ligation in natural protein systems is quite rare. The sperm whale myoglobin cavity mutant H93G Mb (D. Barrick, Biochemistry 33 (1994) 6546) has its proximal histidine ligand replaced by glycine, a mutation which leaves an open cavity capable of accommodation of a variety of unnatural potential proximal ligands. This provides a convenient system for studying ligand-protein interactions. Molecular modeling of the proximal cavity in the active site of H93G Mb indicates that the cavity is of sufficient size to accommodate benzoate and phenolate in conformations that allow their oxygen atoms to come within binding distance of the heme iron. In addition, benzoate may occupy the cavity in an orientation which allows one carboxylate oxygen atom to ligate to the heme iron while the other carboxylate oxygen is within hydrogen bonding distance of serine 92. The ferric phenolate and benzoate complexes have been prepared and characterized by UV-visible and MCD spectroscopies. The benzoate adduct shows characteristics of a six-coordinate high-spin complex. To our knowledge, this is the first known example of a six-coordinate high-spin heme complex with an anionic oxygen donor proximal ligand. The benzoate ligand is displaced at alkaline pH and upon reaction with hydrogen peroxide. The phenolate adduct of H93G Mb is a five-coordinate high-spin complex whose UV-visible and MCD spectra are distinct from those of the histidine 93 to tyrosine (H93Y Mb) mutant of sperm whale myoglobin. The phenolate adduct is stable at alkaline pH and exhibits a reduced reactivity with hydrogen peroxide relative to that of both native ferric myoglobin, and the exogenous ligand-free derivative of ferric H93G Mb. These observations indicate that the identity of the proximal oxygen donor ligand has an important influence on both the heme iron coordination number and the reactivity of the complex with hydrogen peroxide.

Crystallization and initial spectroscopic characterization of the heme-containing dehaloperoxidase from the marine polychaete Amphitrite ornata.

The heme-containing dehaloperoxidase from Amphitrite ornata was crystallized from an unbuffered solution containing 30% PEG 8000 and 200 mM ammonium sulfate by the hanging-drop vapor-diffusion method. Dark-red bipyramidal crystals are orthorhombic in space group P2(1)2(1)2(1) with unit-cell dimensions a = 68.5, b = 68.4 and c = 61.1 A. The asymmetric unit contains two subunits related by a non-crystallographic twofold axis. The crystals scatter beyond 2 A resolution. The native data have been collected and one single-site mercury derivative has been found. SIRAS phasing was used to determine the positions of the heme Fe atoms and structure determination is in progress. A preliminary spectroscopic investigation indicates that the heme is protoporphyrin IX and its coordination sphere resembles that of a typical heme peroxidase, i.e. histidine ligated. Detailed spectroscopic and electrochemical studies are now under way.

Thiolate Adducts of the Myoglobin Cavity Mutant H93G as Models for Cytochrome P-450

A recent development in the field of heme proteins has been the engineering of cavity mutants in which the axial coordinating residue is replaced by a smaller, noncoordinating residue, leaving a cavity that can then be filled by exogenous ligands. In this manner, potential models for the cysteinate-ligated cytochrome P-450 monooxygenases have been prepared using the H93G cavity mutant of sperm whale myoglobin in which the coordinating histidine has been replaced by glycine. Magnetic circular dichroism (MCD) spectroscopy has been used for structural characterization of several ferric and ferrous thiolate-H93G adducts. Ferric mixed-ligand complexes can be prepared with neutral sixth ligands. The model breaks down in two cases: (i) when anionic ligands are added to the ferric-thiolate complex and (ii) on reduction of the ferric-thiolate complex. Results are discussed in the context of the H93C mutants of myoglobin and the stabilizing influences on the cytochrome P-450 heme-cysteinate complex.