Alan J. Wonacott

Structure of UDP-N-acetylglucosamine enolpyruvyl transferase, an enzyme essential for the synthesis of bacterial peptidoglycan, complexed with substrate UDP-N-acetylglucosamine and the drug fosfomycin.

BACKGROUND UDP-N-acetylglucosamine enolpyruvyl transferase (MurA), catalyses the first committed step of bacterial cell wall biosynthesis and is a target for the antibiotic fosfomycin. The only other known enolpyruvyl transferase is 5-enolpyruvylshikimate-3-phosphate (EPSP) synthase, an enzyme involved in the shikimic acid pathway and the target for the herbicide glyphosate. Inhibitors of enolpyruvyl transferases are of biotechnological interest as MurA and EPSP synthase are found exclusively in plants and microbes. RESULTS The crystal structure of Escherichia coli MurA complexed with UDP-N-acetylglucosamine (UDP-GlcNAc) and fosfomycin has been determined at 1.8 A resolution. The structure consists of two domains with the active site located between them. The domains have a very similar secondary structure, and the overall protein architecture is similar to that of EPSP synthase. The fosfomycin molecule is covalently bound to the cysteine residue Cys115, whereas UDP-GlcNAc makes several hydrogen-bonding interactions with residues from both domains. CONCLUSIONS The present structure reveals the mode of binding of the natural substrate UDP-GlcNAc and of the drug fosfomycin, and provides information on the residues involved in catalysis. These results should aid the design of inhibitors which would interfere with enzyme-catalyzed reactions in the early stage of the bacterial cell wall biosynthesis. Furthermore, the crystal structure of MurA provides a model for predicting active-site residues in EPSP synthase that may be involved in catalysis and substrate binding.

Low resolution study of crystalline l-lactate dehydrogenase

The electron density distribution of dogfish muscle lactate dehydrogenase, based on 2000 independent terms extending to 5 A resolution, shows the shape of the tetrameric molecule and of the individual subunits. Each of the five heavy-atom derivatives used in the calculation substitutes at one or more of three sites A, B and C. The heavy-atom compounds (sodium p-hydroxymercuribenzoate, dimercury acetate, Baker dimercurial, platinum ethylene diamine dichloride, and sodium aurichloride) all react with a sulphydryl group at A to some extent. Sodium aurichloride substitutes at site B with some loss of isomorphism, while the platinum compound occupies site C. Anomalous dispersion measurements have been used to determine the absolute configuration of the molecule and to improve phasing. A second electron density map includes 2000 extra terms which have been phased using only two of these compounds, extending the resolution to approximately 4 A. It is possible to trace only some parts of the polypeptide chain of the subunits in either map since the molecule does not appear to have any distinctive secondary structure. Adenosine, a competitive inhibitor, has been found to bind close to the “nonessential” thiol at site A, at the surface between adjoining subunits. The subunit is divided into two parts by a narrower “neck”; the “essential” thiol group is at site B in this region. On diffusion of the coenzyme, nicotinamide adenine dinucleotide, into grown crystals, the crystal symmetry is lowered; the modification is probably caused by a change in quaternary structure. A similar change is caused by certain parts of the coenzyme.

The structure of the nicotinamide-adenine dinucleotide coenzyme when bound to lactate dehydrogenase.

The electron density distribution of M4 dogfish lactate dehydrogenase at 5 A resolution has been re-calculated using improved phases derived in part from a new heavy atom derivative. Difference maps have been calculated for an adenosine and a nearly isomorphous NAD derivative, permitting the positioning of the coenzyme on the molecule and identification of its adenosine moiety. A skeletal model has been fitted to the electron density which clearly establishes the “open” structure of the bound coenzyme. The coenzyme is bound at the base of a deep cleft with the nicotinamide end buried particularly deeply within the subunit. The reactive (A-type) hydrogen comes within 13.5 A of the essential thiol group, but is separated from it by substantial protein electron density. No two of the four coenzyme binding sites approach closer than 21 A.

Structure and Mechanism of Lactate Dehydrogenase

The enzyme lactate dehydrogenase (LDH) catalyses the conversion of lactate to pyruvate in the presence of the coenzyme nicotinamide adenine dinucleotide (NAD). It is composed of four subunits each of molecular weight 35,000 (Appella and Markert, 1961; Cahn et al, 1962). The subunits are single polypeptide chains of either H (heart) or M (muscle) variety (Wieland and Pfleiderer, 1957; Markert and Moller, 1959). Hybridization of these chains gives rise to five possible isoenzymes per species. This paper concerns itself with the properties of the M4 isoenzyme of dogfish (Squalus acanthius). Rossmann et al (1967) have previously reported that the apo-enzyme crystallizes in space group F422 with a = 146.9, c = 155.2A, and one polypeptide chain per asymmetric unit. Adams et al (1969) showed that the molecular center coincided with the intersection of a defined set of mutually perpendicular two-fold axes in the crystal lattice. They also reported a low resolution (5.0A) structure in which the boundaries of the molecule and subunits could be traced. In addition some properties of the binary coenzyme complex were discussed.