Protein Crystallography

Prof D W Rice - Harrison Chair in Structural Biology

My laboratory is concerned with the use of X-ray crystallography to determine the structure/function relationship in proteins. Such studies provide a powerful route towards understanding basic biological mechanisms, contribute towards the rational design of new drugs and underpin the use of enzymes for industrial and biomedical applications.

A major programme in the laboratory concerns progress towards the design of novel herbicides and a recent highlight includes the structure determination of A. thaliana imidazoleglycerol-phosphate dehydratase, an enzyme of histidine biosynthesis and a target for the experimental family of triazole phosphonate herbicides (1). The structure is composed of twenty-four identical subunits arranged in 432 symmetry (Fig 1a) and shows how the formation of a novel dimanganese cluster is crucial to the assembly of the active 24mer from an inactive trimeric precursor and to the formation of the active site of the enzyme. Molecular modelling suggests that the substrate is bound to the manganese cluster as an imidazolate moiety (Fig 1b) which subsequently collapses to yield a diazafulvene intermediate. The mode of imidazolate recognition exploits pseudo-symmetry at the active site arising from a combination of the assembly of the particle and pseudo-symmetry present in each subunit as a result of gene duplication. This provides an intriguing example of the role of evolution in the design of Nature’s catalysts and has opened up the opportunity of analysing the mode of inhibitor binding as a contribution towards a programme of rational herbicide design.

IGPD particle and close-up of active site

Figure 1(a).[left] Diagram showing a surface representation of the IGPD particle as viewed down the three-fold axis. Twenty-one of the twenty-four subunits are shown with residues of one trimer omitted to show the large internal cavity. Each of the trimers is coloured differently with manganese ions coloured purple. (b) [right] A close-up view of the IGPD active site to show how the reaction intermediate (shown in a stick format) interacts with the 2 manganese ions in the active site pocket (shown as purple spheres).

A further theme in the laboratory is a systematic programme designed to determine the structures of essential proteins in bacteria as the targets for the development of novel antibiotics. As part of this project we have recently determined the structure of B. subtilis glutamate racemase (RacE) (2, Fig 2a). This enzyme is responsible for the conversion of L-glutamate to D-glutamate (D-Glu) to provide this essential building block of the peptidoglycan layer in bacterial cell walls. The structure of a complex of B. subtilis RacE with D-Glu has revealed that the glutamate is buried in a deep pocket, whose formation at the interface of the enzyme’s two domains involves a large-scale conformational rearrangement (Fig 2b). Analysis of the structure reveals that these domains are related by pseudo-2-fold symmetry, which superimposes the two catalytic cysteine residues, which are located at equivalent positions on either side of the -carbon of the substrate from which a proton is transferred. Whilst these two domains show only limited sequence identity (14%), the structural similarity suggests that the racemase activity of RacE arose as a result of an ancient gene duplication event. The structure of the complex we have determined is dramatically different from that seen in studies elsewhere on the enzyme from A. pyrophilus and provides new insights into the RacE mechanism and an explanation for the potency of a family of RacE inhibitors which have been developed as experimental antibiotics (Fig 2c). Furthermore, modelling studies on inhibitor binding have highlighted residues on the enzyme surface that might be important in the acquisition of drug resistance thus contributing towards our development of a new family of antibiotics.

Figure 2(a).[left] A diagram of the final 2Fo-Fc map for B. subtilis glutamate racemase calculated at 1.75 A and showing the location of the enzyme bound D-glutamate. (b)[centre] A schematic representation of the superposed structures of the B. subtilis RacE/glutamate complex (green), and the A. pyrophilus MurI/glutamine complex (blue). The elements of secondary structure in the B. subtilis enzyme are identified. The position of the bound glutamate in B. subtilis RacE is shown in CPK format and the domain closure that accompanies substrate binding can be seen. (c)[right] A schematic diagram indicating the modelled positions of the 4S and 4R isomers of 4-substituted D-glutamic acid inhibitors of RacE indicating the alternative positions for the two hydrophobic benzene substituents. In the 4S isomer (lower) the benzene ring can be seen to lie in a pocket bounded by a cluster of largely hydrophobic side chains. In contrast in the 4R isomer the ring appears to form adverse contacts with the side chains of Cys40 and Glu153 providing an explanation for the lack of antibacterial activity of this isomer.

Further extensions to our programme of antibiotic research includes the structure determination of enzymes involved in the biosynthesis of fatty acids (3, 4, 5, Fig 3a), in quorum sensing (6, Fig 3b), in the non-mevalonate-dependent pathway of isoprenoid biosynthesis (Fig 3c) and of a series of bacterial GTPases which control unknown but essential functions within the cell (7, Fig 3d). These enzymes form targets for the discovery of novel anti-infectives against both bacteria and important parasites including malaria and toxoplasma.

Figure 3(a) [top, left] Diagram illustrating the position of mutations in a bacterial enoyl reductase that leads to resistance to the antibiotic triclosan. (b) [top, right] Electron density map around the zinc ion that lies at the active site of LuxS, a protein involved in quorum sensing. (c) [lower, left] The structure of the trimer of 4-diphosphocytidyl-2C-methyl-D-erythritol synthase an enzyme involved in isoprenoid biosynthesis. (d) [lower, right] The structure of the GDP complex of B. subtilis YsxC, an essential bacterial GTPase

A further theme in the laboratory is the use of crystallography to probe those aspects of protein structure that are responsible for stability, substrate specificity and catalysis. Studies in this area include the analysis of enzymes from hyperthermophiles, psychrophiles and halophiles. These programmes address the molecular basis of substrate specificity (7, 8) and question our understanding of how proteins are able to withstand high temperatures (9, 10, 11) and to catalyse chemistry at low temperatures and in high salt (Fig 4).

Figure 4 Diagram showing the strongly negative electrostatic potential on the surface of the glucose dehydrogenase from the extreme halophile Haloferax mediterranei which arises from an almost uniform coverage of the surface by glutamate or aspartate residues.

Selected Publications

[1] Catalysis by imidazolegycerol-phosphate dehydratase: evolution of a metalloenzyme. S.E. Glynn, P.J. Baker, S.E. Sedelnikova, C.L. Davies, T.C. Eadsforth, C.W. Levy, H.F Rodgers, G.M. Blackburn, T.R. Hawkes, R. Viner and D.W. Rice. Structure (2005) in press

[2] Substrate induced conformation changes in Bacillus subtlisis glutamate racemase and their implications for drug discovery. S. Ruzheinikov, M.A. Taal, S.E. Sedelnikova, P.J. Baker and D.W. Rice. Structure (2005) in press

[3] Molecular basis of triclosan activity. C.W. Levy, A. Roujeinikova, S. Sedelnikova, P.J. Baker, A.R. Stuitje, A.R. Slabas, D.W. Rice and J.B. Rafferty. Nature (1999) 398, 383-384

[4] Triclosan inhibits the growth of Plasmodium falciparum and Toxoplasma gondii by inhibition of Apicomplexan Fab I. R McLeod, S P Muench, J B Rafferty, D E Kyle, E J Mui, M J Kirisitis, D G Mack, C W Roberts, B U Samuel, R E Lyons, M Dorris, W K Milhous and D W Rice. International Journal of Parasitology, (2001) 31, 109-113.

[5] Delivery of antimicrobials into parasites. B.U. Samuel, B. Hearn, D. Mack, P. Wender, J. Rothbard, M.J. Kirisits, E. Mui, S Wernimont, C.W. Roberts, S.P. Muench, D.W. Rice, S.T Prigge and R. McLeod. PNAS (2003), 100, 14281-14286.

[6] The 1.2 A structure of a novel quorum sensing protein, Bacillus subtilis LuxS. S.N. Ruzheinikov, S.K. Das, S.E. Sedelnikova, A. Hartley, S.J. Foster, M.J. Horsburgh, A.G. Cox, C.W. McCleod, A. Mekhalfia, G.M. Blackburn, D.W. Rice, P.J. Baker. J Mol Biol (2001) 313, 111-122.

[7] Analysis of the open and closed conformations of the GTP-binding protein YsxC from Bacillus subtilis. S.N. Ruzheinikov, S.K. Das, S.E. Sedelnikova, P.J. Baker, P.J. Artymiuk, J. García-Lara, S.J. Foster and D.W. Rice. J. Mol. Biol. (2004) 265-278.

[8] The crystal structure of Thermotoga maritima maltosyltransferase and its implications for the molecular basis of the novel transfer specificity. A Roujeinikova, C Raasch, J Burke, P J Baker, W Liebl and D W Rice. J Mol Biol (2001), 312, 119-131.

[9] Crystal structure of T. maritima 4--glucanotransferase and its acarbose complex: implications for substrate specificity and catalysis. A. Roujeinikova, C. Raasch., S.E. Sedelnikova, W. Liebl and D.W. Rice. J Mol Biol (2002) 321, 149-162

[10] Substrate specificity and mechanism from the structure of Pyrococcus furiosus galactokinase. A Hartley, SE Glynn, V Barynin, PJ Baker, SE Sedelnikova, C Verhees, D de Geus , J van der Oost, DJ Timson, RJ Reece, DW Rice. J. Mol. Biol (2004) 337, 387-398.

[11] Crystal Structure of Pyrococcus furiosus Phosphoglucose Isomerase: Implications for Substrate Binding and Catalysis. J M. Berrisford, J Akerboom, A P. Turnbull, D de Geus, S E. Sedelnikova, I Staton, CW. McLeod, C H. Verhees, J van der Oost, D W. Rice, and P J. Baker. J. Biol. Chem., (2003), 278: 33290 - 33297.

[12] The structures of inhibitor complexes of Pyrococcus furiosus phosphoglucose isomerase provide insights into substrate binding and catalysis. Berrisford JM; Akerboom J; Brouns S; Sedelnikova SE; Turnbull AP; van der Oost J; Salmon L; Hardre R; Murray IA; Blackburn GM; Rice DW; Baker PJ. J. Mol.Biol. (2004) 343, 649-657.