NMP kinases from M.tuberculosis(H. Munier-Lehmann collaboration with G. Labesse & S. Pochet)
Two of the five NMP kinases from M.tuberculosis were studied this year in priority: the TMP kinase and UMP kinase. A "drug design" approach was used for TMP kinase by molecular modelling, based on the three-dimentional structure of E.coli and yeast TMP kinase (Lavie et al., 1997 and 1998), as resolved by X-ray crystallography. Three positions in dTMP were retained to design analogues. The position 2 of the heterocycle was primarily selected. Ten compounds have been synthesized and tested for their activity as either substrates or inhibitors of the purified recombinant enzyme. One of these analogues is a competitive inhibitor at concentrations around 10-6M. The corresponding nucleosides of dTMP analogues exhibited also a high affinity for the TMP kinase of M.tuberculosis. This observation allows the design of new specific inhibitors, modified in position 5' of the sugar, to increase the capability of these compounds to cross the biological membranes.
UMP kinase is another interesting target for specific antibacterial agents. In fact, the bacterial enzyme, which is highly specific for UMP, is a hexamer regulated by UTP and GTP, while its eukaryote homologue is a monomer, able to phosphorylate both UMP and CMP. The M.tuberculosis enzyme has been overproduced in E.coli as a fusion protein. A tag-sequence consisting of six histidines in alternance with six asparagines was placed at its N-terminus. The extra-sequence has no effect on the stability and catalytic activity of the bacterial enzyme and allows its purification in large amounts in a single step. UMP kinase from M. tuberculosis is stable and soluble until 14 mg/ml, contrary to E. coli enzyme which is practically insoluble over 0,1 mg/ml. Recently, crystals in the presence of UTP were obtained in collaboration with B. Gomes Guimaraes and P. Alzari (Unit of Structural Biochemistry). The conditions of crystallization are currently being optimised. A "heavy" UTP derivative has been synthesized to allow the structure resolution in the near future.
CMP kinase from E. coli(L. Assairi, H. Munier-Lehmann, A. Ofiteru, A.-M. Gilles collaboration with T. Bertrand, P. Briozzo & J. Gallay)
n order to better understand the mechanism of recognition of CMP and dCMP by the bacterial CMP kinases, the structure of E. coli enzyme in complex with the two substrates or with ß-arabinosyl CMP and dideoxy CMP, has been resolved by X-ray crystallography. The study of these structures demonstrates the contribution of 2'OH and 3'OH from the pentose to the ligand-induced conformational changes of CMP kinase. Substitution by site-directed mutagenesis of Arg181 and Asp 185, which interacts directly with 3'OH of pentose, significantly decreased the phosphorylation of both CMP and dCMP. Substitution of Ser101 which forms an hydrogen bond between Asp185 and CMP or dCMP, affected moderately the phosphorylation of the CMP but decreased dramatically the phosphorylation of dCMP. Ser101 is the only residue strictly conserved among 40 aminoacids of a domain specific to bacterial CMP kinases, called INSERT. Time-resolved fluorescence study of the conformation and dynamics of ATP site showed that anthraniloyl-2'-dATP, once bound to the protein, is protected from aqueous environment, due to the movement of LID domain which covers the ATP-binding site.
UMP kinase from other bacterial species(L. Assairi, T. Borza, C. Gagyi, A.-M. Gilles collaboration with N. Bucurenci & M. Straut)
Cloning and overproduction of UMP kinase from different bacterial species (H. influenzae and N. meningitidis, for Gram negative bacteria ; B. subtilis and S. pneumoniae, for Gram positive bacteria) have been undertaken. All of these proteins are soluble in usual buffers but unstable in the absence of UTP. The biochemical analysis of recombinant proteins showed important differences between the enzymes belonging to the two families of microorganisms: the UMP kinase from Gram positive bacteria, contrary to Gram negative bacteria, exhibit an almost absolute requirement for GTP as activator.
NAD kinase from Neisseria meningitidis(L. Assairi)
NAD kinase is a nucleotide kinase producing NADP from ATP. Due to the role played by NADP in numerous metabolic reactions, this enzyme is essential in both prokaryotes and eucaryotes. Our purposes have been the followings: structural study, analysis of ATP and NAD recognition sites, study of catalytic mechanism and search of specific inhibitors. We have cloned by PCR the NAD kinase gene of N. meningitidis and overproduced the protein by using several expression vectors. A protein carrying a His-tag at its C-terminus, being the most soluble and active, has been selected. The characterization of NAD kinase by usual physico-chemical and kinetic methods is currently being performed.
5. Identification of proteins with unknown function Gagyi, T. Borza, L. Assairi, A.-M. Gilles collaboration with M. Straut & N. Bucurenci)
We have undertaken the study of a protein from E. coli with unknown structure and function, previously isolated by pseudo-affinity chromatography and identified by 2-D gel electrophoresis and mass spectrometry. This monomeric protein of 18.2 kD (YajQ) has been uniformly labelled by 15N and/or 13C. The secondary structure has just been resolved by homo- and heteronuclear NMR spectroscopy. The NMR data allows a detailed description of the molecule, indicating a structure able to bind single stranded-RNA species. A search for the natural ligand of this protein is currently being performed. In addition, we have carried out a study of a cluster consisting of four genes present only in Salmonella species and coding respectively a permease (DeoP), a deoxyribokinase (dRK), a repressor (DeoQ) and a protein of unknown function (OrfX). The DeoQ repressor regulates exclusively this cluster of genes. It is interesting to mention that DeoR, a repressor of deoCABD operon and homologous to DeoQ is able to repress the newly described operon. The DeoQ and OrfX proteins have been overproduced and characterized. Recently, crystals of OrfX have been obtained.
6. Two-dimensional gel electrophoresisLaurent-Winter)
The global analysis of bacterial proteomes has been pursued in collaboration with several units of the Pasteur Institute, particularly in pathogenic microorganisms,. Fifteen proteins have been identified in Vibrio cholerae which expression is regulated by VicH an analog of H-NS protein (Unité de Régulation de l'Expression Génétique). The analysis of mutant of Staphylococcus aureus, which is modified in the two compounds regulation system has allowed to select ten proteins which could be potential targets of mutated regulatory gene (Unité de Biochimie Microbienne). Ten proteins have been identified in M. tuberculosis as potential targets for a transcriptional activator belonging to AraC, which activity is essential for the virulence (Unité de Génétique Mycobactérienne). Finally, in encapsulated yeast, Cryptococcus neoformans, 2-D gel electrophoresis associated with immune blot analysis techniques have permitted the identification of a new protein of 40 kDa, involved in the humoral response of animals surviving to the infection with Cryptococcus (Unité de Mycologie).
7. Mass spectrometryA. Namane, J.-C. Rousselle)
During the last year, we have continued to perform analyses by mass spectrometry with the main purpose to characterize and identify bacterial proteins. A sensitivity ranging between 100 fentomoles and 1 picomole is reached in the identification of proteins separated by 2D-gel elecrophoresis. More than 800 analyses have been performed. For the "Programme Transversal de Recherche", entitled "Glycosylation d'épitopes immunodominants de M. tuberculosis" (G. Marchal), 16 new exported and potentially glycosylated proteins, among which, six are unknown. In collaboration with the Unité d'Interactions Moléculaire (A. Jacquier), we identified in yeast by mass spectrometry protein complexes belonging to mitochondrial ribosomes. Among these, sixteen proteins, which have not been previously described, belong to the small subunit of yeast mitochondrial ribosome.