Les enzymes lipolytiques sont solubles en phase aqueuse mais hydrolysent des substrats insolubles. Leurs activités lipolytiques dépendent donc fortement de l’organisation des substrats lipidiques présents sous forme de structures interfaciales telles que des émulsions, des micelles, des liposomes, ou des bicouches lipidiques. Les propriétés cinétiques et la spécificité de substrat de ces enzymes résultent de l’étape initiale d’adsorption à l’interface lipide-eau et des interactions entre le substrat et le site actif. Dans le cadre de ce travail de thèse, la technique des films monomoléculaires a été utilisée pour étudier en détail les étapes séquentielles d’adsorption, de catalyse et d’inhibition de l’enzyme à l’interface lipide-eau. Dans une première partie, nous avons réalisé la caractérisation physico-chimique de la lipase gastrique de chien (DGL), avec l’étude :  de son adsorption sur un film non substrat de dilauroylphosphatidylcholine ; ‚ de l’hydrolyse interfaciale de la 1,2-dicaprine dans des films mixtes en présence d’Orlistat. Concernant l’étape de catalyse, nous avons étudié l’effet du propeptide sur la spécificité de substrat et l’activité interfaciale de la phospholipase A2 sécrétée de groupe X de souris. Enfin, dans une troisième partie, nous avons comparé les propriétés interfaciales de la lipase YLLIP2 de la levure Yarrowia lipolytica qui serait un bon candidat pour l’enzymothérapie de substitution chez les patients atteints d’insuffisance pancréatique exocrine (IPE), la lipase pancréatique humaine et la DGL. Nos résultats ont confirmé le rôle d’YLLIP2 en tant qu’excellent « substitut » non seulement de la HPL en cas d’IPE, mais aussi de la DGL. / Lipolytic enzymes are water-soluble whereas their substrates are insoluble in water. Their lipolytic activities depend strongly on the organization of the lipid substrates present in interfacial structures such as oil-in-water emulsions, micelles, liposomes, and membrane bilayers. The kinetic properties and substrate specificity of these enzymes result from both their adsorption at the lipid-water interface, and the interactions occurring between the substrate and the active site. In this thesis work, the monomolecular film technique was used to study in details the sequential steps of adsorption, catalysis and inhibition of model enzymes at the lipid-water interface. In a first part, we performed the physico-chemical characterization of the dog gastric lipase (DGL), by studying:  its adsorption onto a dilauroylphosphatidylcholine non-substrate film; ‚ its interfacial hydrolysis of 1,2-dicaprin in mixed films with various amounts of Orlistat. Regarding the catalysis step, we studied the effect of the propeptide on the substrate specificity and interfacial activity of the murine group X secreted phospholipase A2. A model of this enzyme with its propeptide was built from the available 3D structure of the corresponding mature human enzyme. Finally, in the third part, we compared the interfacial kinetic properties of YLLIP2 lipase of the yeast Yarrowia lipolytica which has been identified as a good candidate for enzyme replacement therapy for patients with exocrine pancreatic insufficiency (EPI), human pancreatic lipase and DGL. Our results confirmed the role of YLLIP2 as an excellent "substitute" not only for HPL in case of PEI, but also for the DGL at acidic pH values.

Viral proteases play crucial roles in the life cycle and maturation of many viruses by processing the viral polyprotein after translation and in some cases cleaving host proteins associated with the immune response. The essential role of viral proteases makes them attractive therapeutic targets. In this thesis, I provide an introductory summary of viral proteases, their structure, mechanism, and inhibition, while the breadth of this thesis focuses on the Hepatitis C virus (HCV) NS3/4A and Zika virus (ZIKV) NS2B/NS3 viral proteases. HCV NS3/4A protease inhibitors (PIs) have become a mainstay in combination therapies. However, drug resistance remains a major problem against these PIs. In this thesis, I applied insights from the HCV substrate envelope (SE) model to develop strategies for designing PIs that are less susceptible to resistance. Also, I used the HCV NS3/4A protease as a model system to decipher the molecular mechanism and role of fluorination in HCV PIs potency and drug resistance. The drug design strategies described in this thesis have broad applications in drug design. The ZIKV is an emerging global threat, and currently, with no treatment available. In this thesis, I described the discovery, biochemical and antiviral evaluation of novel noncompetitive quinoxaline-based inhibitors of the ZIKV NS2B/NS3 protease. The inhibitors are proposed to interfere with NS2 binding to NS3, thereby preventing the protease from adopting the closed and active conformation. The inhibitors from this work will serve as lead compounds for further inhibitor development toward the goal of developing antivirals.

Breast cancer, as the most prevalent cancer in women, is responsible for more than 15% of new cancer cases and about 6.9% of all cancer-related death in the US. A major cause of therapeutic failure in breast cancer is the development of resistance to chemotherapy, especially for triple-negative breast cancer (TNBC). Therefore, how to overcome chemoresistance is the major challenge to improve the life expectancy of breast cancer patients. Our studies demonstrate that TNBC cells surviving the chronic treatment of chemotherapeutic drugs show significantly higher expression of the dual serine/threonine and tyrosine protein kinase (DSTYK) than non-treated parental cells. In our in vitro cellular models, DSTYK knockout via the CRISPR/Cas9-mediated technique results in apoptotic cell death of chemoresistant cells upon drug treatment. Moreover, DSTYK knockout promotes chemotherapeutic drug-induced tumor cell death in an orthotopic mouse model. These findings suggest that DSTYK exerts an important and previously unknown role in promoting chemoresistance. Our studies provide fundamental insight into the role of DSTYK in chemoresistance in TNBC cells and lay the foundation for the development of new strategies targeting DSTYK for improving TNBC therapy.

No / Recent studies have revealed extensive cortical demyelination in patients with progressive multiple sclerosis (MS). Demyelination in gray matter lesions is associated with activation of microglia. Macrophages and microglia are known to express myeloperoxidase (MPO) and generate reactive oxygen species during myelin phagocytosis in the white matter. In the present study we examined the extent of microglial activation in the cerebral cortex and the relationship of microglial activation and MPO activity to cortical demyelination. Twenty-one cases of neuropathologically confirmed multiple sclerosis, with 34 cortical lesions, were used to assess microglial activation. HLA-DR immunolabeling of activated microglia was significantly higher in demyelinated MS cortex than control cortex and, within the MS cohort, was significantly greater within cortical lesions than in matched non-demyelinated areas of cortex. In homogenates of MS cortex, cortical demyelination was associated with significantly elevated MPO activity. Immunohistochemistry revealed MPO in CD68-positive microglia within cortical plaques, particularly toward the edge of the plaques, but not in microglia in adjacent non-demyelinated cortex. Cortical demyelination in MS is associated with increased activity of MPO, which is expressed by a CD68-positive subset of activated microglia, suggesting that microglial production of reactive oxygen species is likely to be involved in cortical demyelination.

No / Heterozygous germline defects in a gene encoding a type II receptor for bone morphogenetic proteins (BMPR-II) underlie the majority of inherited cases of the vascular disorder known as pulmonary arterial hypertension (PAH). However, the precise molecular consequences of PAH causing mutations on the function of the receptor complex remain unclear. We employed novel enzymatic and fluorescence activity based techniques to assess the impact of PAH mutations on pre-mRNA splicing, nonsense-mediated decay (NMD) and receptor complex interactions. We demonstrate that nonsense and frameshift mutations trigger NMD, providing further evidence that haplo-insufficiency is a major molecular consequence of disease-related BMPR2 mutations. We identified heterogeneous functional defects in BMPR-II activity, including impaired type I receptor phosphorylation, receptor interactions and altered receptor complex stoichiometry leading to perturbation of downstream signalling pathways. Importantly, these studies demonstrate that the intracellular domain of BMPR-II is both necessary and sufficient for receptor complex interaction. Finally and to address the potential for resolution of stoichiometric balance, we investigated an agent that promotes translational readthrough of a BMPR2 nonsense reporter construct without interfering with the NMD pathway. We propose that stoichiometric imbalance, due to either haplo-insufficiency or loss of optimal receptor-receptor interactions impairs BMPR-II mediated signalling in PAH. Taken together, these studies have identified an important target for early therapeutic intervention in familial PAH.

No / Cyclic adenosine monophosphate (cAMP)-dependent signaling modulates platelet function at sites of vascular injury. Here we show that thrombospondin-1 (TSP-1) prevents cAMP/protein kinase A (PKA) signaling through a CD36-dependent mechanism. Prostaglandin E(1) (PGE(1)) induced a robust inhibition of both platelet aggregation and platelet arrest under physiologic conditions of flow. Exogenous TSP-1 reduced significantly PGE(1)-mediated inhibition of both platelet aggregation and platelet arrest. TSP-1 prevented PGE(1)-stimulated cAMP accrual and phosphorylation of PKA substrates, through a mechanism requiring phosphodiesterase3A. TSP-1 also inhibited VASP phosphorylation stimulated by the nonhydrolyzable cAMP analog, 8-bromo-cAMP, indicating that it may regulate cAMP-mediated activation of PKA. The inhibitory effect of TSP-1 on cAMP signaling could be reproduced with a peptide possessing a CD36 binding sequence of TSP-1, while the effects of TSP-1 were prevented by a CD36 blocking antibody. TSP-1 and the CD36 binding peptide induced phosphorylation of Src kinases, p38 and JNK. Moreover, inhibition of Src kinases blocked TSP-1-mediated regulation of cAMP concentrations and the phosphorylation of VASP, indicating that TSP-1 modulated the cAMP/PKA signaling events through a tyrosine kinase-dependent pathway downstream of CD36. These data reveal a new role for TSP-1 in promoting platelet aggregation through modulation of the cAMP-PKA signaling pathway.

No / OBJECTIVE: Development of treatment resistance and adverse toxicity associated with classical chemotherapeutic agents highlights the need for safer and effective therapeutic approaches. Herein, we examined the effectiveness of a combination treatment regimen of 5-fluorouracil (5-FU) and curcumin in colorectal cancer (CRC) cells. METHODS: Wild type HCT116 cells and HCT116+ch3 cells (complemented with chromosome 3) were treated with curcumin and 5-FU in a time- and dose-dependent manner and evaluated by cell proliferation assays, DAPI staining, transmission electron microscopy, cell cycle analysis and immunoblotting for key signaling proteins. RESULTS: The individual IC50 of curcumin and 5-FU were approximately 20 microM and 5 microM in HCT116 cells and 5 microM and 1 microM in HCT116+ch3 cells, respectively (p<0.05). Pretreatment with curcumin significantly reduced survival in both cells; HCT116+ch3 cells were considerably more sensitive to treatment with curcumin and/or 5-FU than wild-type HCT116 cells. The IC50 values for combination treatment were approximately 5 microM and 1 microM in HCT116 and 5 microM and 0.1 microM in HCT116+ch3, respectively (p<0.05). Curcumin induced apoptosis in both cells by inducing mitochondrial degeneration and cytochrome c release. Cell cycle analysis revealed that the anti-proliferative effect of curcumin and/or 5-FU was preceded by accumulation of CRC cells in the S cell cycle phase and induction of apoptosis. Curcumin potentiated 5-FU-induced expression or cleavage of pro-apoptotic proteins (caspase-8, -9, -3, PARP and Bax), and down-regulated anti-apoptotic (Bcl-xL) and proliferative (cyclin D1) proteins. Although 5-FU activated NF-kappaB/PI-3K/Src pathway in CRC cells, this was down-regulated by curcumin treatment through inhibition of IkappaBalpha kinase activation and IkappaBalpha phosphorylation. CONCLUSIONS: Combining curcumin with conventional chemotherapeutic agents such as 5-FU could provide more effective treatment strategies against chemoresistant colon cancer cells. The mechanisms involved may be mediated via NF-kappaB/PI-3K/Src pathways and NF-kappaB regulated gene products.

No / We identify cytochrome P450 1A1 (CYP1A1) as a target for tumor-selective drug development in bladder cancer and describe the characterization of ICT2700, designed to be metabolized from a prodrug to a potent cytotoxin selectively by CYP1A1. Elevated CYP1A1 expression was shown in human bladder cancer relative to normal human tissues. RT112 bladder cancer cells, endogenously expressing CYP1A1, were selectively chemosensitive to ICT2700, whereas EJ138 bladder cells that do not express CYP1A1 were significantly less responsive. Introduction of CYP1A1 into EJ138 cells resulted in 75-fold increased chemosensitivity to ICT2700 relative to wild-type EJ138. Negligible chemosensitivity was observed in ICT2700 in EJ138 cells expressing CYP1A2 or with exposure of EJ138 cells to CYP1B1- or CYP3A4-generated metabolites of ICT2700. Chemosensitivity to ICT2700 was also negated in EJ138-CYP1A1 cells by the CYP1 inhibitor alpha-naphthoflavone. Furthermore, ICT2700 did not induce expression of the AhR-regulated CYP1 family, indicating that constitutive CYP1A1 expression is sufficient for activation of ICT2700. Consistent with the selective activity by CYP1A1 was a time and concentration-dependent increase in gamma-H2AX protein expression, indicative of DNA damage, associated with the activation of ICT2700 in RT112 but not EJ138 cells. In mice-bearing CYP1A1-positive and negative isogenic tumors, ICT2700 administration resulted in an antitumor response only in the CYP1A1-expressing tumor model. This antitumor response was associated with detection of the CYP1A1-activated metabolite in tumors but not in the liver. Our findings support the further development of ICT2700 as a tumor-selective treatment for human bladder cancers.

No / PURPOSE: Cytochrome P450 2W1 (CYP2W1) is a monooxygenase detected in 30% of colon cancers, whereas its expression in nontransformed adult tissues is absent, rendering it a tumor-specific drug target for development of novel colon cancer chemotherapy. Previously, we have identified duocarmycin synthetic derivatives as CYP2W1 substrates. In this study, we investigated whether two of these compounds, ICT2705 and ICT2706, could be activated by CYP2W1 into potent antitumor agents. EXPERIMENTAL DESIGN: The cytotoxic activity of ICT2705 and ICT2706 in vitro was tested in colon cancer cell lines expressing CYP2W1, and in vivo studies with ICT2706 were conducted on severe combined immunodeficient mice bearing CYP2W1-positive colon cancer xenografts. RESULTS: Cells expressing CYP2W1 suffer rapid loss of viability following treatment with ICT2705 and ICT2706, whereas the CYP2W1-positive human colon cancer xenografts display arrested growth in the mice treated with ICT2706. The specific cytotoxic metabolite generated by CYP2W1 metabolism of ICT2706 was identified in vitro. The cytotoxic events were accompanied by an accumulation of phosphorylated H2A.X histone, indicating DNA damage as a mechanism for cancer cell toxicity. This cytotoxic effect is most likely propagated by a bystander killing mechanism shown in colon cancer cells. Pharmacokinetic analysis of ICT2706 in mice identified higher concentration of the compound in tumor than in plasma, indicating preferential accumulation of drug in the target tissue. CONCLUSION: Our findings suggest a novel approach for treatment of colon cancer that uses a locoregional activation of systemically inactive prodrug by the tumor-specific activator enzyme CYP2W1.

Ces travaux de thèse s'articulent autour de deux projets : les études structure-fonction de l'aminopeptidase A, d'une part, et celles du récepteur de l'apéline, d'autre part. I/ L'aminopeptidase A (APA, EC 3.4.11.7) est une aminopeptidase monozinc membranaire qui, dans le cerveau, produit l'angiotensine (Ang) III à partir de l'Ang II. L'Ang III est l'un des principaux peptides effecteurs du système rénine-angiotensine cérébral qui exerce un effet stimulateur tonique sur le contrôle central de la pression artérielle chez le rat hypertendu. Ainsi le blocage de l'APA par un inhibiteur spécifique et sélectif, l'EC33 ou sa prodrogue, le RB150, normalise la pression artérielle dans deux modèles expérimentaux d'hypertension artérielle (HTA). L'APA constitue une cible thérapeutique potentielle pour le traitement de l'HTA qui justifie le développement de nouveaux inhibiteurs de cette enzyme plus puissants et plus sélectifs que l'EC33 et avec un profil pharmacodynamique et pharmacocinétique amélioré par rapport au RB150. Pour cela, nous avons construit un modèle tridimensionnel (3D) de l'APA sur la base de la structure cristallographique de l'APA humaine récemment publiée. Nous avons ensuite validé ce modèle par des études structure-fonction par modélisation moléculaire et mutagénèse dirigée en démontrant l'implication, d'un résidu du sous-site S1 dans la spécificité de substrat acide de l'APA et de deux résidus formant le sous-site S2' interagissant avec le résidu P2' acide d'inhibiteurs tripeptidiques précédemment développés dans le laboratoire.II/ L'apéline est le ligand naturel du récepteur orphelin humain APJ (ApélineR), un récepteur à sept domaines transmembranaires couplé aux protéines G. L'apéline et son récepteur sont impliqués dans le maintien de l'équilibre hydrique et des fonctions cardiovasculaires. L'ApélineR constitue une cible thérapeutique potentielle dans le traitement de l'insuffisance cardiaque et des rétentions hydriques. Etant donné que la demi-vie de l'apéline dans la circulation sanguine est de l'ordre de la minute, l'objectif est de développer des analogues de l'apéline métaboliquement stables. Pour développer de tels composés, nous avons entrepris de comprendre comment l'apéline se lie à son récepteur et comment elle l'active. Dans ce but, nous avons construit un modèle 3D de l'ApélineR basé sur la structure cristallographique du récepteur aux chimiokines, CXCR4. Nous avons validé ce modèle par des études structure-fonction par modélisation moléculaire et mutagénèse dirigée. Nous avons identifié à la surface du récepteur, les résidus acides des boucles extracellulaires qui interagissent avec les résidus basiques de l'apéline. Nous avons ensuite développé des analogues de l'apéline-17 (K17F) métaboliquement stables par deux stratégies différentes. Premièrement, nous avons substitué chacun des résidus de l'apéline par son énantiomère de la série D ou par un acide aminé synthétique. Deuxièmement, nous avons ajouté une chaîne fluoroalkyle à l'extrémité N-terminale de l'apéline. Ces deux stratégies ont permis d'obtenir plusieurs composés dont les plus actifs sont le P92 et le LIT01-196 qui conservent des propriétés pharmacologiques identiques à celles de K17F et qui présentent une demi-vie plasmatique largement supérieure à celle du peptide endogène. Ces deux analogues se sont révélés particulièrement actifs in vivo avec une capacité à diminuer la pression artérielle et à réduire la sécrétion de vasopressine dans le sang conduisant à une augmentation de la diurèse aqueuse. Les modèles 3D validés de l'APA et de l'ApélineR seront utilisés pour des campagnes de criblage in silico de chimiothèques virtuelles afin de découvrir de nouveaux inhibiteurs de l'APA et des agonistes de l'ApélineR qui pourraient conduire à terme à de nouveaux candidats-médicaments. Ces composés pourraient être utiles pour le traitement de l'HTA et de l'insuffisance cardiaque. / The doctoral work was divided in two parts, one on the structure-function studies of aminopeptidase A, and the second one, on those of the apelin receptor. I/ Aminopeptidase A (APA) is a membrane bound monozinc aminopeptidase which generates, in the brain, angiotensin (Ang) III from Ang II. Ang III is one of the main effector peptides of the brain renin-angiotensin system, which exerts a tonic stimulatory action on the control of blood pressure in hypertensive rats. Thus, the blockade of brain APA by a specific and selective inhibitor, EC33 or its prodrug, RB150, normalizes blood pressure in two animal models of arterial hypertension (HTA). APA constitutes a potential therapeutic target for the treatment of HTA that justifies the development of more potent and selective APA inhibitors than EC33, with enhanced pharmacodynamic and pharmacokinetic profiles when compared to RB150. With this aim, we built a three dimensional (3D) model of APA based on the recently published crystal structure of human APA. We validated this model by structure-function studies combining molecular modeling and site-directed mutagenesis demonstrating the crucial role of one residue in the S1 subsite responsible for substrate specificity of APA for N-terminal acidic amino-acid residues and two other residues constituting the S2' subsite of APA involved in the binding of the P2' acidic residue of tripeptidic inhibitors, previously developed in the laboratory. II/ Apelin is the endogenous ligand of the human orphan receptor named APJ (ApelinR), a G protein-coupled receptor. Apelin and ApelinR are involved in the control of body fluid homeostasis and cardiovascular functions. ApelinR constitutes a potential therapeutic target for the treatment of heart failure and water retentions. Given that apelin half-life in the blood circulation is in the minute range, we aimed to develop potent metabolically stable apelin analogs.. In this context, it is necessary to understand how apelin binds to ApelinR and how it is activated. To do so, we build a 3D model of ApelinR based on the crystal structure of the chemokine receptor, CXCR4. We validated this model by structure-function studies by molecular modeling and site-directed mutagenesis. We showed that apelin interacts with the receptor through interactions between the basic residues of the peptide and the acidic residues of the ApelinR, located in the extracellular loops. ,We then developed metabolically stable apelin-17 (K17F) analogs following two different strategies. First, we substituted each residue of K17F by its D-isomer or a synthetic amino-acid. Secondly, we added a fluoroalkyl chain at the N-terminal part of K17F. These two strategies allowed to significantly improve plasma half-life of the modified peptides for several hours without modifying their pharmacological properties as compared to K17F. Two apelin metabolically stable analogs, P92 and LIT01-196, were found to have significantly higher in vivo activity than K17F with a strong capacity to decrease blood pressure and to inhibit vasopressin release in the blood stream inducing an increased aqueous diuresis. These new validated 3D models will be now used to perform in silico screening of virtual chemical libraries to discover new APA inhibitors and ApelinR agonists that could ultimately lead to new drug candidates. These compounds could be useful for the treatment of HTA and heart failure.