Difference Fourier analysis together with the pattern of sequence conservation has led to the identification of both the glyoxylate and metal binding sites and implicates the C-terminal end of the TIM barrel as the active site, which is consistent with studies of other enzymes with this fold. Two disordered regions of the polypeptide chain lie close to the active site, one of which includes a critical cysteine residue suggesting that conformational rearrangements are essential for catalysis. Structural similarities between isocitrate lyase and both PEP mutase and enzymes belonging to the enolase superfamily suggest possible relationships in aspects of the mechanism.

The NADP-dependent fl-keto acyl carrier protein reductase (BKR) from E. coli has been crystallizcd by the hanging-drop method of vapour diffusion using poly(ethylene glycol) of average molecular weight 1450. The crystals belong to the hexagonal space group P6~22 or P6s22 with unit-cell dimensions a = b = 67.8, c --355.8 ,~. Calculated values for Vm and consideration of the packing suggest that the asymmetric unit contains a dimer. BKR catalyses the first reductive step in the elongation cycle of fatty-acid biosynthesis. It shares extensive sequence homology with the enzyme which catalyzes the second reductive step in the cycle, enoyl acyl carrier protein reductase (ENR), and thus provides an opportunity to study the evolution of enzyme function in a metabolic pathway. The structure determination will permit the analysis of the molecular basis of its catalytic mechanism and substrate specificity.

Isocitrate lyase (ICL) from the filamentous fungus Aspergillus nidulans catalyzes the first committed step of the carbon-conserving glyoxylate bypass. This enzyme has been crystallized by the hanging-drop method of vapour diffusion using polyethylene glycol 2000 as the precipitant. Diffraction patterns show that the crystals diffract to beyond 2.5 A and are probably in space group P4(2)2(1)2 with unit-cell dimensions of a = b = 91.9 and c = 152.7 A, with one molecule in the asymmetric unit. The elucidation of the structure of this enzyme to high resolution will advance the understanding of how the metabolic branch point between the tricarboxylic acid cycle and the glyoxylate bypass is controlled by the affinity of ICL for its substrate isocitrate and contribute to a programme of rational drug design.

The structure of a novel opine dehydrogenase fromArt/1robacter has been solved by isomorphous replacement and provides insights into the mechanism and substrate specificity of the enzyme superfamily to which it belongs. Crown gall opines are the products of the NAD(P)H-dependent reductive condensation between an aketo acid and the a-or co-NH" group of an arrjno acid in a reaction catalysed by a family of enzyntes, generically refened to as the opine dehydrogenases (reviewed in Thompson. J. and Donl:ersloot, J.A. Annu. Rev. Biochem. 1992 61 517-557). The enzymes catalysing this chemistry are encoded on large plasm.ids resident in virulent strains of Agrobacterium. These tumour inducing plasmids are required for crown £all induction and tumouro£enesis involves the excision of a segm~nt of the plasmid DNA ~n which the opine dehydrogenase gene is located. Following integration of the DNA into the plant genome, the plant cell machinery is hijacked to divert resources to the synthesis of opines which permit growth of the tumour. Sequence studies have established that opine dehydrogenases belong to an enzyme superfamily with differential specificity for the keto acid and amino acid partners. Recently, the gene for a novel opine dehydrogenase fi·om Arthrobacter has been sequenced and shown to have 30% sequence identity with the octapine dehydrogenase from Agrobacteriwn twnefaciens. This enzyme is a homodimer of subunit Mr 70.000 and has been overexpressed in E. coli and crystallised. X-ray analysis shows tl1at the crystals which diffract to beyond 1;8 A bel~ng to t11e orJhorhombic space group P2,212 with a= 1 04A, b = 79 A and c = 45A witl1 a single monomer in tl1e asymmeuic unit. The stmcture will be described and insights into the catalytic mechanism and the differential substrate specificity will be discussed. Short chain dehydrogenase/reductases (SDR) constitute a new class of dinucleotide-linked oxidoreductases tl1at use a wide vmietv of substrates. We have detennined the crystal structures of vmiou~ forms of two members of the SDR fanuly: the bacterial 3a,20~ hydroxysteroid dehydrogenase and the humm1 type 1 estrogenic 17~ hydroxysteroid dehydrogenase. Structures of five other SDR's have been detennined elsewhere. These are : rat liver dihydropte1idine reductase, oilseed rape and bacterial enoyl acyl canier protein reductase, mouse lung carbonyl reductase and bacterial 7a-hydroxysteroid dehydrogenase. Despite a low intra-family an1ino acid sequence identity (15-35%), these enzymes are characterized not only by suikingly similm· submut te11iary stil!ctures iliat include a dinucleotide-binding fold, but also an analogous quatemary association and a strictly conserved Tyr-Lys pair at tl1e catalytic end of the active site. The conserved catalytic residues form a Tyr-Lys-Ser uiad in many of the SDR's, owing to a semi-conserved Ser residue. A mechar1ism in wluch ilie Tyr-Lys pair acts as a general acid m1d an elecu·ophile to t11e substrate cmbonyl has been proposed.All seven enzymes of tl1e SDR family f01m homo-dimeiic...

No abstract