La voie non-mévalonate est fortement présente chez les plantes et les bactéries mais est absente chez les mammifères. C'est pourquoi inhiber la synthèse des isoprénoïdes et identifier un inhibiteur de cette voie enzymatique contribuera grandement à la recherche de nouveaux antibiotiques, antifongiques et herbicides. Les propriétés uniques de la 1-deoxy-D-xylulose 5-phosphate reductoisomérase (DXR), l'enzyme centrale de cette voie enzymatique, en font une cible très intéressante pour la synthèse de nouveaux composés. La Fosmidomycine agit comme un inhibiteur de la DXR et reste aujourd'hui, avec son homologue acétylé FR90098, la référence en termes d'inhibiteur même si de nombreux efforts ont été faits pour la synthèse d'analogues depuis plusieurs années comme expliqué dans le premier chapitre avec la mise en relation de la structure des composés et leur activité. L'analyse de la diffraction des rayons X de la DXR avec la Fosmidomycine où le substrat naturel montre que la fonction phosphonate ou phosphate interagit avec une poche polaire hautement spécifique dans le site actif de l'enzyme permettant peu de modifications. Par comparaison, la fonction acide hydroxamique qui chélate le cation de l'enzyme offre la possibilité de modifications par l'introduction d'autres fonctions complexantes. Dans ce contexte, de nombreuses modifications comme l'introduction de fonctions carbamoylphosphinate, amidoxime, N-hydroxyurée et dérivées d'uraciles comme unités complexantes ont été synthétisées pour trouver des nouvelles familles d'inhibiteurs de la DXR. Toutes ces fonctions possèdent des propriétés de chélation intéressantes. En effet, elles ont déjà conduit à de puissants inhibiteurs de différentes métalloenzymes. / The non-mevalonate pathway is highly present in higher plants, protozoa and bacteria but as no equivalent in mammals. That is why shut down isoprenoid biosynthesis and identify a non-mevalonate pathway inhibitor would greatly contribute to the search for safer antibiotics, antimalarials and for our concern herbicides. The unique properties of the 1-deoxy-D-xylulose 5-phosphate reductoisomerase (DXR), the central enzyme of this pathway, make it a remarkable and attractive target for drug design. Fosmidomycin acts as an inhibitor of DXR and still remains, along with its N-acetyl homologue FR90098, one of the most potent inhibitor ever known even if extensive work on the development of Fosmidomycin analogue derivatives have been developed since the last decade as demonstrated in the first chapter with the development of a structure activity relationship of all the potential inhibitors of this enzyme already reported in the literature. The X-ray diffraction analysis of the co-crystals of DXR and Fosmidomycin or substrate shows that the phosphonic/phosphate group interacts with a highly specific polar pocket in the enzyme site, allowing only few structural modifications. By contrast, the cation chelating subunit represented by the hydroxamic acid function offers fine tuning possibilities for the complexation abilities as well as potential secondary interactions with the NADPH cofactor or directly with the enzyme. In this context, several modifications such as the introduction of carbamoylphosphinate, amidoxime, N-hydroxyurea and uracil complexing subunits have been made in order to find new families of DXR inhibitors. All of these functions show promising chelation capabilities as they already led to potent inhibitors of different metalloenzymes.

Metalloproteins account for at least half of all known proteins. Metal ions often facilitate chemical that are energetically and/or kinetically challenging. Metal ion-dependent proteins are responsible for a myriad of essential biological functions, including respiration, biosynthesis of essential amino acids, nitrogen fixation, oxygen transport, photosynthesis and metabolisms (e.g. glycolysis and citric acid cycle). Not surprisingly, a growing number of disorders (e.g. various cancers, phenylketonuria, Wilson’s disease) are associated with mutations in metalloenzymes. A general introduction of the importance of metals in biology is presented in chapter 1. This thesis is aimed at obtaining a greater understanding of the structure and function of three metalloenzymes, ketol acid reductoisomerase (KARI), purple acid phosphatase (PAP) and metallo β lactamase (MβL). Chapter 2 examines the structure and dynamics of plant KARI. KARI is an enzyme in the branched-chain amino acid (BCAA) biosynthesis pathway. KARI is a binuclear Mg2+ enzyme that catalyses the conversion of 2-acetolactate (AL) into (2R)-2,3-dihydroxy-3-isovalerate or 2-aceto-2-hydroxybutyrate into (2R, 3R)-2,3-dihydroxy-3-methylvalerate in the presence of NADPH. To date, the only reported structures for a plant KARI are those of the spinach enzyme-Mn2+-(phospho) ADP ribose-(2R,3R)-2,3-dihydroxy-3-methylvalerate complex and the spinach KARI-Mg2+-NADPH-N-hydroxy-N-isopropyloxamate complex, where N-hydroxy-N-isopropyloxamate (IpOHA) is a predicted transition-state analog. These studies demonstrate that the enzyme is consisted of two domains, N- domain and C- domain, with the active site at the interface of these domains. In this chapter, the structures of the rice KARI-Mg2+ and rice KARI-Mg2+-NADPH complexes were determined to 1.55 and 2.8 Å resolutions, respectively. Comparisons of all the available plant KARI structures have revealed several major differences. Firstly, the N-domain is rotated up to 15o relative to the C-domain, expanding the active site by up to 4 Å. Secondly, an α-helix in the C-domain that includes residues V510-T519 and forms part of the active site moves by ~ 3.9 Å upon binding of NADPH. Thirdly, the 15 C-terminal amino acid residues in the rice KARI-Mg2+ complex are disordered. In the rice KARI-Mg2+-NADPH complex and spinach KARI structures, many of the 15 residues bind to NADPH and the N-domain and cover the active site. Fourthly, the location of the metal ions within the active site can vary by up to 2.7 Å. The new structures have thus, led to the proposal of an induced-fit mechanism. In this proposed induced-fit mechanism, (i) substrate enters the active site, (ii) active site is closed during catalysis, and (iii) the opening of active site facilitates product release. PAP is also a binuclear metalloenzyme and is capable of utilizing a heterovalent active site to hydrolyse a broad range of phosphomonoester substrates. Chapter 3 examines the catalytic mechanism of PAP based on several new crystal structures. The red kidney bean PAP structure in complex in sulphate was determined to 2.4 Å. This sulphate-bound structure provides insight into the pre-catalytic phase of its reaction cycle. This stucture demonstrates the significance of an extensive hydrogen-bonding network in the second coordination in initial substrate binding and orientation prior to hydrolysis. Most importantly, the two metal ions, Fe3+ and Zn2+, are five-coordinate in this structure, with only one nucleophilic μ-hydroxide present in the metal-bridging position. In combination with kinetic, crystallographic and spectroscopic data, all PAP structures form the proposal of a comprehensive eight-step model for the catalytic mechanism of purple acid phosphatases in general. To date, no reliable method for producing recombinant PAP at levels suitable for structural biology have been reported. Natural sources are the only way so far to obtaining PAP in a large quantity. Attempts to produce active and recombinant PAP from Mycobacterium marinum using bacterial are found in chapter 4. In brief, in combination with Nus fusion tag, Rosetta (DE3) strain and lower temperature (e.g. 25oC), expression of soluble and mycobacterial PAP becomes possible. However, this soluble protein is non-functional and thus, switching into other expression system (e.g.algal sytem) is the only approach to obtain soluble and functional protein. In algal expression system, human PAP was attempted. Preliminary results indicate that some PAP activity was observed when expressed in algal system. Chapter 5 focuses on the investigation of metallo β lactamase (MβL) from Klebsiella pneumoniae (Kp-MβL). This enzyme requires one or two Zn2+ ions for catalysis. Kinetic properties of Kp-MβL for the hydrolysis of various β-lactam substrates (e.g. benzyl-penicillin, cefoxitin, imipenem and meropenem) were investigated and the role of the metal ions in catalysis was also examined. Kinetic data demonstrate that Klebsiella pneumoniae MβL can degrade a broad spectrum of β-lactam antibiotics, with a high preference for cephems and carbapenems. Kinetic data from pH dependence studies has revealed that catalysis of benzyl-penicillin and meropenem is preferred at acidic pH. The kcat vs pH profile demonstrates that catalysis is enhanced by protonation, thus it is likely that the relevant group is responsible for the donation of a proton to the product or leaving group. In this case, a doubly Lewis activated, bridging hydroxide molecule has been speculated. A single protonation event (pKa ~7) is also observed in kcat/Km vs pH profile. Since benzyl-penicillin does not have an acidic moiety in this pH range, this event is likely to be associated with the free enzyme. His 79 and 139 have been speculated to enhance substrate binding. In contrast, catalysis of both cefoxitin and imipenem is favoured at alkaline pH, leading to the proposal that a terminally bound water is likely to form a nucleophile. A bell-shaped pH profile for kcat/Km is observed for cefoxitin and imipenem substrates. pKa of ~ 9-9.5 is likely to be associated with Lys161, which enhances substrate binding. In Chapter 6, a novel MβL from Serratia proteamaculans (Spr-MβL) is investigated. This chapter includes expression, purification and preliminary characterization of this MβL using steady-state kinetics. Expression of this enzyme in Rosetta (DE3) plysS E. coli strain yields only a small amount of soluble enzyme (1 mg/ 6 L culture). To improve the amount of soluble protein, Spr-MβL was subjected to several rounds of in vitro evolution. About two-fold gain in solubility was achieved by this method along with a five-fold increase in β-lactamase activity. Further rounds of directed evolution are now planned. The kinetic behaviour for Spr-MβL-catalysed the hydrolysis of three β-lactam substrates, penicillin, cefoxitin and imipenem were also studied. Kinetic data suggest that a water molecule bridging the two Zn2+ ions is the likely nucleophile in the reaction with penicillin while the reaction-initiating nucleophile is likely to be a terminally bound hydroxide in the reaction with cephalothin and imipenem (Chapter 6). In summary, this project has led to a better understanding of the structures of KARI and PAP prior to catalysis. This project has also aided in the understanding of catalytic mechanism of MβLs and the role the metal ions play. The knowledge gained will facilitate the development of new chemotherapeutics and herbicides.

Mycobacterium tuberculosis, the causative pathogen of tuberculosis, currently infects one-third of the world’s population, resulting in two million deaths annually. This clearly shows that tuberculosis is one of the most serious diseases of our times. The often unpleasant side effects from the current drugs, combined with the difficulty of ensuring patient compliance, and the emergence of drug-resistant and multidrug-resistant strains, makes the need for new and better drugs urgent. In this thesis, all the steps, from cloning, purification, crystallization, to activity determination, and structure determination are presented for two different M. tuberculosis enzymes. The structures, which were modeled from X-ray crystallographic data, provide the framework for structure-based drug design. Here, new potential inhibitors can be tailor-made based on the specific interactions in the enzyme’s active site. The bacteria have two different peptidyl-prolyl cis-trans isomerases that catalyze the isomerization of peptide bonds preceding proline residues, a process of high importance for correct folding. Here we present the structure of peptidyl-prolyl cis-trans isomerase A, an enzyme present inside the bacteria, and distinguish it from the B form of the enzyme, which is membrane bound, placing its active site outside the bacteria. The enzyme 1-deoxy-D-xylulose-5-phosphate reductoisomerase catalyzes the second step within the non-mevalonate pathway, which leads to the production of isopentenyl diphosphate. This compound is the precursor of various isoprenoids, vital to all living organisms. In humans, isopentenyl diphosphate is produced via a different pathway, indicating that all the enzymes within the non-mevalonate pathway may be suitable drug targets in M. tuberculosis. Several structures of both wild type and mutant 1-deoxy-D-xylulose-5-phosphate reductoisomerase in complex with different substrates, and also with the known inhibitor fosmidomycin, provide valuable information not only to the field of drug design, but also, in this case, into the catalysis.

La synthèse enzymatique de terpénoides chez les mammifères provient de la voie mevalonique. Récemment une voie différente a été découverte et s'est révélée être prépondérante pour de nombreux organismes comme les plantes et bactéries. L'identification d'un inhibiteur de cette cascade enzymatique permettrait le développement d'une nouvelle famille d'herbicide. Les caractéristiques de la 1-déoxy-D-xylulose 5-phosphate réductoisomérase (DXR) font de cette enzyme très spécifique une cible pour la synthèse de nouveaux composés. La Fosmidomycine ainsi que son analogue acétylé le plus proche, FR-900098 restent des références pour l'inhibition de la DXR. Dans ce contexte, l'ensemble des molécules décrites dans la littérature en tant qu'inhibiteurs a été classé en fonction des modifications apportées sur le substrat naturel ou la Fosmidomycine. A partir de l'ensemble de ces informations, cinq familles ont été synthétisées pour trouver un nouveau motif complexant. Pour deux d'entre elles, le squelette de base contient un acide phosphonique et un acide phosphinique sur lequel a été introduit la diversité moléculaire grâce aux réactions de Pudovik et de couplage pallado-catalysé. Les autres motifs complexant originaux sont constitués d'une fonction carbonyle et d'un hétérocycle en α ou β. Après optimisation de la synthèse des précurseurs, la diversité a été introduite à l'aide, par exemple, d'une réaction de trois composantes permettant la préparation d'hétérocycle. Enfin, deux modifications ont été faites sur le bras espaceur : l'introduction d'atomes de fluor pour modifier les propriétés physicochimiques ou d'un atome d'azote, point d'attache de nouveaux groupements. / The non-mevalonate pathway is widely found in higher plants and in many eubacteria, including pathogenic ones, but not in mammals. Identifying a non-mevalonate pathway inhibitor would greatly contribute to the search for new herbicides. The unique properties of 1-Deoxy-D-xylulose 5-phosphate reductoisomerase make it remarkable and rational target for drug design. The phosphonohydroxamic acid Fosmidomycin, which acts through inhibition of DXR, is a natural compound produced in the fermentation of Streptomyces and still remains, with its N-acetyl homologue FR900098, one of the most active compounds. First of all, the enzyme and all the potential inhibitors tested in literature were classified in order to understand the global quest for therapeutically useful compounds. In this context, we designed and synthesized five different families of Fosmidomycin analogues containing a new chelating unit. Two targets molecules families bearing a phosphinophonic acid as common core were imagined. Divergent approach allowed the introduction of the chemical diversity thank to powerful pallado-catalyzed coupling reaction. The other families containing carbonyl group and heterocycle in α‐ and β‐position were regarded as highly potent complexing units. Chemical diversity was introduced mainly at the end of the synthesis. For one of them convergent ring formation using three-components reaction was developed. Finally two modifications of the Fosmidomycin linker were performed by the introduction of fluorine atoms on the parent structure as well as the replacement of a carbon by a nitrogen atom in order to create a new point of modifications.

The recent emergence of a number of new viral diseases as well as the re-emergence of tuberculosis (TB), indicate an urgent need for new drugs against viral and bacterial infections. Coronavirus nsp1 has been shown to induce suppression of host gene expression and interfere with host immune response. However, the mechanism behind this is currently unknown. Here we present the first nsp1 structure from an alphacoronavirus, Transmissible gastroenteritis virus (TGEV) nsp1. Contrary to previous speculation, the TGEV nsp1 structure clearly shows that alpha- and betacoronavirus nsp1s have a common evolutionary origin. However, differences in conservation, shape and surface electrostatics indicate that the mechanism for nsp1-induced suppression of host mRNA translation is likely to be different in the alpha- and betacoronavirus genera. The Modoc virus is a neuroinvasive rodent virus with similar pathology as flavivirus encephalitis in humans. The flaviviral methyltransferase catalyses the two methylations required to complete 5´ mRNA capping, essential for mRNA stability and translation. The structure of the Modoc NS5 methyltransferase domain was determined in complex with its cofactor S-adenosyl-L-methionine. The observed methyltransferase conservation between Modoc and other flaviviral branches, indicates that it may be possible to identify drugs that target a range of flaviviruses and supports the use of Modoc virus as a model for general flaviviral studies. 1-deoxy-D-xylulose 5-phosphate reductoisomerase (DXR) is part of the methylerythritol phosphate (MEP) pathway that produces essential precursors for isoprenoid biosynthesis. This pathway is used by a number of pathogens, including Mycobacterium tuberculosis and Plasmodium falciparum, but it is not present in humans. Using a structure-based approach, we designed a number of MtDXR inhibitors, including a novel fosmidomycin-analogue that exhibited improved activity against P.falciparum in an in vitro blood cell growth assay. The approach also allowed the first design of an inhibitor that bridge both DXR substrate and co-factor binding sites, providing a stepping-stone for further optimization.

Since the discovery of Mycobacterium tuberculosis (Mtb) as the bacterial agent causing tuberculosis, the permanent eradication of this disease has proven challenging. Although a number of drugs exist for the treatment of tuberculosis, 1.7 million people still die every year from this infection. The current treatment regimen involves lengthy combination therapy with four different drugs in an effort to combat the development of resistance. However, multidrug-resistant and extensively drug-resistant strains are emerging in all parts of the world. Therefore, new drugs effective in the treatment of tuberculosis are much-needed. The work presented in this thesis was focused on the early stages of drug discovery by applying different hit identification and hit expansion strategies in the exploration of two new potential drug targets, glutamine synthetase (GS) and 1-deoxy-D-xylulose-5-phosphate reductoisomerase (DXR). A literature survey was first carried out to identify new Mtb GS inhibitors from compounds known to inhibit GS in other species. Three compounds, structurally unrelated to the typical amino acid derivatives of previously known GS inhibitors, were then discovered by virtual screening and found to be Mtb GS inhibitors, exhibiting activities in the millimolar range. Imidazo[1,2-a]pyridine analogues were also investigated as Mtb GS inhibitors. The chemical functionality, size requirements and position of the substituents in the imidazo[1,2-a]pyridine hit were investigated, and a chemical library was designed based on a focused hierarchical design of experiments approach. The X-ray structure of one of the inhibitors in complex with Mtb GS provided additional insight into the structure–activity relationships of this class of compounds. Finally, new α-arylated fosmidomycin analogues were synthesized as inhibitors of Mtb DXR, exhibiting IC50 values down to 0.8 µM. This work shows that a wide variety of aryl groups are tolerated by the enzyme. Cinnamaldehydes are important synthetic intermediates in the synthesis of fosmidomycin analogues. These were prepared by an oxidative Heck reaction from acrolein and various arylboronic acids. Electron-rich, electron-poor, heterocyclic and sterically hindered boronic acids could be employed, furnishing cinnamaldehydes in 43–92% yield.