VIH-1, l’agent étiologique du Syndrome de l’Immunodéficience Acquise, est un rétrovirus qui infecte les cellules immunitaires et détourne leur machinerie cellulaire pour se répliquer rapidement. Lors de l’infection, le génome ARN est rétrotranscrit en ADN par la transcriptase inverse virale (RT), puis l’insertion du génome proviral dans l’ADN de la cellule hôte est une étape obligatoire du cycle viral catalysée par l’enzyme virale Intégrase (IN). L’interaction de l’IN avec son cofacteur essentiel, la protéine nucléaire LEDGF/p75, dirige l’intégration à l’intérieur de gènes dans des régions fortement exprimées de la chromatine, ce qui permet la production efficace de nouveaux virions. Les Inhibiteurs Allostériques Intégrase-LEDGF (INLAIs) sont une nouvelle classe de molécules antirétrovirales se liant à l’IN au site de liaison de LEDGF/p75. Conçus pour inhiber compétitivement l’interaction protéine-protéine IN-LEDGF/p75, ils inhibent également les activités enzymatiques de l’Intégrase et augmentent son niveau de multimérisation.Nous avons étudié plusieurs nouvelles séries d’INLAIs de la société Mutabilis, et avons pu démontrer que ces molécules inhibent l’intégration, mais ont aussi un effet antirétroviral plus puissant et indépendant de LEDGF/p75 post-intégration au cours de la maturation des virions, qui conduit à la production de virus non infectieux, ayant une morphologie excentrique caractérisée par un défaut d’encapsidation du génome viral. Lors de l’infection de cellules par ces virus, le cycle viral s’arrête à l’étape de rétrotranscription du génome viral. Nous avons montré que ces virions contiennent pourtant un génome viral stable et fonctionnel, une RT active et l’ARNtLys3 qui sert d’amorce à la rétrotranscription, et ont également conservé leur immunoréactivité pour les lymphocytes B et T. En évaluant l’impact du polymorphisme de l’IN au voisinage du site de liaison, nous avons identifié le variant polymorphe Ala125, pour lequel l’INLAI MUT-A perd concomitamment son effet sur la maturation des virions et sur la multimérisation de l’IN, tandis qu’il inhibe aussi bien l’intégration et l’interaction IN-LEDGF, prouvant que l’effet tardif des INLAIs est associé à l’induction de la multimérisation de l’IN. Nous avons pu associer la multimérisation de l’IN à une déstabilisation du dimère par les INLAIs en analysant les co-structures de MUT-A avec les intégrases polymorphes. Les INLAIs, outre leur intérêt thérapeutique sont de remarquables réactifs qui ont permis de démontrer le rôle essentiel de l’intégrase à trois étapes clés du cycle viral du VIH-1 : la rétrotranscription, l’intégration et la maturation des virions. / HIV-1, the causative agent of AIDS, is a retrovirus that infects immune cells and hijacks their cell machinery to achieve rapid replication. In the course of infection, the RNA genome is reverse transcribed into DNA by the viral Reverse Transcriptase (RT) before the obligatory insertion of the proviral genome into the host cell DNA catalyzed by the viral enzyme Integrase (IN). The interaction of IN with its essential cofactor, the nuclear protein LEDGF/p75, targets integration within gene introns in highly transcribed chromatine regions, which allows efficient production of new virions. IN-LEDGF Allosteric Inhibitors (INLAIs) are a novel class of antiretroviral molecules binding IN at the LEDGF/p75-binding site. Designed to competitively inhibit IN-LEDGF/p75 protein-protein interaction, they are also capable of inhibiting IN enzymatic activities and raising the IN multimerization level.We studied several new INLAI series from the company Mutabilis. We could demonstrate that these molecules inhibit integration, but also have a more potent, LEDGF-independent, antiretroviral effect during virion maturation, resulting in the production of non-infectious virions. Virions produced upon INLAI treatment have an eccentric morphology characterized by an encapsidation defect of the viral genome, and lead to an infection block at reverse transcription. Yet, we showed that these virions package a stable and functional viral genome, an active RT and the tRNALys3 primer for reverse transcription, and also keep their immunoreactivity towards B- and T-cell lymphocytes. When evaluating the influence of polymorphism at the edge of the binding site, we identified the IN Ala125 polymorphic variant which causes the concomitant loss of MUT-A effect on virion maturation and IN multimerization, whereas inhibition of integration and IN-LEDGF interaction are maintained. This proves that INLAIs exert their late stage effect through induction of IN multimerization. We could associate IN multimerization to INLAI-induced dimer destabilization by analyzing MUT-A co-structures with polymorphic integrases. Beside their therapeutic interest INLAIs are highly valuable reagents that allowed to demonstrate the essential role of integrase at three key steps of the HIV-1 replication cycle, reverse transcription, integration and virus maturation.

Cytochromes P450 are the major xenobiotics converting enzymes. They are classified as mixed function monooxygenases (MFO). Isoform 1A1 is a extrahepatic form found mainly in the lung and other tissues. It is strongly induced by polycyclic aromatic hydrocarbons and their derivatives via the Ah receptor. As a marker reaction for this enzyme can be used hydroxylation of Sudan I, which has previously been widely used as a azo dye in industry, but since 1980s it is banned for coloring food and cosmetics for its negative influence on the organism. NADPH:cytochrome P450 reductase is the major electron donor for cytochrome P450 catalyzed monooxygenation reactions. Another electron carrier for cytochrome P450 catalyzed reactions is cytochrome b5. It was shown that cytochrome b5 can stimulate, inhibit or have no effect on P450 catalyzed reactions. This thesis aims to evaluate the influence of the ration between NADPH:cytochrome P450 reductase and cytochrome b5 on cytochrome P450 1A1 catalyzed Sudan I hydroxylation. The main goal is to characterize the influence of electron donor and electron transfer ratios on hydroxylation of Sudan I, and to determine the kinetic parameters KM and VMAX for selected protein ratios. Partial aims of the thesis were to characterize the recombinant proteins used in this study...

UHRF1 est une protéine nucléaire responsable du maintien et de la régulation de l'épigénome des cellules. Elle favorise la prolifération cellulaire et est surexprimée dans la plupart des cancers. TIP60, l'un des partenaires le plus important d’UHRF1, est impliqué dans le remodelage de la chromatine et la régulation transcriptionnelle grâce à son activité acétyltransférase. Ensemble, les deux protéines régulent la stabilité et l'activité d'autres protéines telles que la DNMT1 et la p53. Le but de cette étude était d'explorer le mécanisme d'interaction entre UHRF1 et TIP60 en visualisant cette interaction dans les cellules. La microscopie par imagerie à temps de vie de fluorescence et d'autres techniques de biologie moléculaire ont été utilisées. Les résultats ont montré que UHRF1 interagit directement avec le domaine MYST de TIP60 et cette interaction se produit dans la phase S du cycle cellulaire. Les deux protéines ont également montré une réponse similaire aux dommages à l'ADN, ce qui prédit une cohérence dans leur fonction dans le mécanisme de réparation de l'ADN. La surexpression de TIP60 a également induit la baisse du niveau d’UHRF1 et de DNMT1 ainsi qu’une induction d'apoptose dans les cellules ce qui suggère un rôle de TIP60 dans la régulation des fonctions oncogéniques d’UHRF1. / UHRF1 is a nuclear protein maintaining and regulating the epigenome of cells. Its promotes proliferation and is found upregulated in most of cancers. TIP60 is one of the important interacting partner of UHRF1 and is involved in chromatin remodeling and transcriptional regulation through its acetyltransferase activity. Together they regulate the stability and activity of other proteins such as DNMT1 and p53. The aim of this thesis was to explore the mechanism of interaction between UHRF1 and TIP60 by visualizing this interaction in cells. Fluorescent lifetime imaging microscopy and other molecular biology techniques were employed for this purpose. Results of this study showed that UHRF1 interacts directly to the MYST domain of TIP60 and this interaction prevails in the S-phase of cell cycle. Both proteins also showed a similar response to DNA damage predicting a coherence in their function in DNA repair mechanism. Overexpression of TIP60 also downregulated UHRF1 and DNMT1 and induced apoptosis in cells suggesting a role of TIP60 in regulation of oncogenic functions of UHRF1.

The evolutionary conserved insulin and nutrient signaling network regulates growth andmetabolism. Nutrients are directly utilized for growth or stored, mostly as triglycerides. InDrosophila, activation of insulin/nutrient signaling in the fat body (the fly equivalent of liverand adipose tissue), causes an increase in fat stores composed of several small-size lipiddroplets (LDs). Conversely, fasting produces an increase in LD size and a decrease in fatcontents. The TOR kinase and its substrate S6 kinase (S6K) play a central role in this response,and particularly in Drosophila, they have been shown to orchestrate cell-autonomous andhormone-controlled growth. However, despite extensive research studies on different modelorganisms (mouse, fly, worm) to decipher the molecular and physiological functions of S6K,nothing is known about how its degradation is regulated.Taking advantage of the inducible RNA interfering (RNAi) library from NIG (Japan), we haveperformed three genetic screens to identify novel regulators of steroidogenesis, lipidmetabolism and dS6K-dependent growth. First, RNAi lines were screened in the ring gland; anorgan that controls the progression of the developmental steps by producing the steroidhormone ecdysone. Out of 7,000 genes screened, 620 positive candidates were identified toproduce developmental arrest and/or overgrowth phenotypes. Then, we challenged 4,000 genesby RNAi screening able to recapitulate the larger sized LD phenotype as obtained uponstarvation, leading to the identification of 24 potential candidates. Finally, the RNAi lines werescreened for their ability to enhance a growth phenotype dependent of the Drosophila S6K(dS6K). Out of 7,000 genes screened, 45 genes were identified as potential negative regulatorsof dS6K. These genes were further used to design a novel protein-protein interaction networkcentered on dS6K through the available data from yeast-2-hybrid (Y2H) assay. The most potentinteractors were then analyzed by treatment of cultured S2 cells with the corresponding doublestrand RNA (dRNA). Western blotting thus, allowed us to discriminate between the geneproducts that regulate dS6K levels versus those that regulate its phosphorylation, as a hallmarkfor its kinase activity. Interestingly, archipelago (ago), which encodes a component of an SCFubiquitinligase known to regulate the degradation of dMyc, Cyclin E and Notch, was identifiedas a negative regulator of dS6K-dependent growth. Based on the Y2H available data showingthat Ago and dS6K interact each other and the presence of a putative Ago-interaction motif indS6K, we hypothesized that Ago causes an ubiquitin-mediated degradation of dS6K. Ourmolecular data showed that loss of ago caused an elevated level of dS6K, which confirms arole of Ago in controlling dS6K degradation. Altogether our findings emphasize the importanceof the saturating screening strategies in Drosophila to identify novel regulators of metabolicand signaling pathways.

Lfc is a guanine nucleotide exchange factor (GEF) for RhoA and is negatively regulated by its association with the microtubule array. Tctex-1, a light chain subunit of the dynein motor complex, was identified as an Lfc-interacting protein in a yeast two-hybrid screen. In mouse embryonic fibroblast (MEF) cells, over-expression of Tctex-1 represses Lfc-induced actin stress fiber and focal adhesion complex formation. Here, we present biochemical evidence obtained from a real-time, nuclear magnetic resonance (NMR)-based assay indicating that the microtubule exerts its inhibitory effect on Lfc through a mechanism that is dependent on the presence of Tctex-1. We also present NMR structure data showing that Lfc and the dynein intermediate chain (DIC) bind to different surfaces of Tctex-1. The biochemical and structural data together support a model in which Lfc is recruited to the microtubules through the dynein cargo adaptor function of Tctex-1, resulting in inhibition of Lfc function.

Lfc is a guanine nucleotide exchange factor (GEF) for RhoA and is negatively regulated by its association with the microtubule array. Tctex-1, a light chain subunit of the dynein motor complex, was identified as an Lfc-interacting protein in a yeast two-hybrid screen. In mouse embryonic fibroblast (MEF) cells, over-expression of Tctex-1 represses Lfc-induced actin stress fiber and focal adhesion complex formation. Here, we present biochemical evidence obtained from a real-time, nuclear magnetic resonance (NMR)-based assay indicating that the microtubule exerts its inhibitory effect on Lfc through a mechanism that is dependent on the presence of Tctex-1. We also present NMR structure data showing that Lfc and the dynein intermediate chain (DIC) bind to different surfaces of Tctex-1. The biochemical and structural data together support a model in which Lfc is recruited to the microtubules through the dynein cargo adaptor function of Tctex-1, resulting in inhibition of Lfc function.

Computational microRNA (miRNA) target identification in animal genomes is a challenging problem due to the imperfect pairing of the miRNA with the target site. Techniques based on sequence alone are prone to produce many false positive interactions. Therefore, integrative techniques have been developed to utilize additional genomic, structural features, and evolu- tionary conservation information for reducing the high false positive rate. We propose that the context of a putative miRNA target in a protein-protein interaction (PPI) network can be used as an additional filter in a computational miRNA target pr ediction algorithm. We compute several graph theoretic measures on human PPI network as indicators of network context. We assess the performance of individual and combined contextual measures in increasing the precision of a popular miRNA target prediction tool, TargetScan, using low throughput and high throughput datasets of experimentally verified human miRNA targets. We used clas- sification algorithms for that assessment. Since there exists only miRNA targets as training samples, this problem becomes a One Class Classification (OCC) problem. We devised a novel OCC method, DiVo, based on simple distance metrics and voting. Comparative analysis with the state of the art methods show that, DiVo attains better classification performance. Our eventual results indicate that topological properties of target gene products in PPI networks are valuable sources of information for filtering out false positive miRNA target genes. We show that, for targets of a number of miRNAs, netwo rk context correlates better with being a target compared to a sequence based score provided by the prediction tool.

Computational microRNA (miRNA) target identification in animal genomes is a challenging problem due to the imperfect pairing of the miRNA with the target site. Techniques based on sequence alone are prone to produce many false positive interactions. Therefore, integrative techniques have been developed to utilize additional genomic, structural features, and evolu- tionary conservation information for reducing the high false positive rate. We propose that the context of a putative miRNA target in a protein-protein interaction (PPI) network can be used as an additional filter in a computational miRNA target prediction algorithm. We compute several graph theoretic measures on human PPI network as indicators of network context. We assess the performance of individual and combined contextual measures in increasing the precision of a popular miRNA target prediction tool, TargetScan, using low throughput and high throughput datasets of experimentally verified human miRNA targets. We used clas- sification algorithms for that assessment. Since there exists only miRNA targets as training samples, this problem becomes a One Class Classification (OCC) problem. We devised a novel OCC method, DiVo, based on simple distance metrics and voting. Comparative analysis with the state of the art methods show that, DiVo attains better classification performance. Our eventual results indicate that topological properties of target gene products in PPI networks are valuable sources of information for filtering out false positive miRNA target genes. We show that, for targets of a number of miRNAs, network context correlates better with being a target compared to a sequence based score provided by the prediction tool.

Signal transduction is a crucial event that enables cells to sense and respond to cues from their immediate environment. Guanylyl cyclase C (GC-C) is a member of the family of receptor guanylyl cyclases. GC-C is a single transmembrane protein that responds to its ligands by the production of the second messenger cGMP. The guanylin family of peptides, (including the bacterially produced heat-stable enterotoxin ST) is the ligand for GC-C, elevates intracellular cGMP levels and activates downstream pathways. GC-C regulates the cystic fibrosis transmembrane conductance regulator (CFTR) by inducing phosphorylation by protein kinase G, resulting in chloride ion and fluid efflux. GC-C also regulates cell cycle progression through cGMP-gated Ca2+ channels. These functions are seen in the intestinal epithelium, the primary site for GC-C expression. GC-C as a molecule has been studied in detail, but its functioning in the context of other signaling pathways remains unknown. The aim of the present investigation was to understand the regulation of signal transduction by GC-C and its cross-talk with other signaling pathways operating in the cell. Molecular events that commonly connect components in a signaling pathway are protein phosphorylation and protein-protein interaction. These two aspects are explored in this thesis. The possibility of tyrosine phosphorylation of GC-C has been explored earlier in our laboratory. In vitro studies indicated that the residue Tyr820 was a site for phosphorylation by the Src family of non-receptor tyrosine kinases and those studies also suggested that phosphorylated Tyr820 could bind to the SH2 domain of Src. We generated a nonphosphorylatable mutant of GC-C, GC-CY820F, and a phosphomimetic mutant GC-CY820E to study the effect of phosphorylation of Tyr820, on the functioning of GC-C. A stable cell line of HEK293:GC-CY820F cells was generated and compared with HEK293:GC-CWT. Dose response to ST in the two cell lines showed that cGMP accumulation by GC-CY820F was greater than that of GC-CWT, although the EC50 remained unchanged. The phosphomimetic GC-CY820E mutant receptor was non-responsive to ST. Further in HEK293 cells, phosphorylation of GC-CWT by constitutively active v-Src resulted in decreased ST stimulation and this effect of v-Src was reduced with GC-CY820F. Inhibition of ST stimulation brought about by v-Src required catalytically active Src, as the kinase inactive v-SrcK295R did not inhibit ST stimulation. These results were corroborated by in vitro studies by using the recombinant catalytic domain of GC-C expressed in insect cells and by phosphorylation using a purified kinase, Hck. Observations suggested that phosphorylation of Tyr820 in the catalytic domain of GC-C compromises the guanylyl cyclase activity of GC-C. T84 and Caco-2 colon carcinoma cells endogenously express GC-C. The effect of tyrosine phosphorylation of GC-C was studied by using HgCl2, a known activator of Src kinases, and by the inhibition of protein tyrosine phosphatases using pervanadate, an irreversible inhibitor. Both these ways of achieving increased tyrosine phosphorylation resulted in decreased ST-stimulated cGMP production by GC-C, as suggested from v-Src transfection studies. This decrease was reversed by using a Src kinase specific inhibitor PP2, confirming the role of Src kinases in the inhibition of GC-C activity. Interestingly, in Caco-2 cells that differentiate in culture, the effect of pervanadate on the inhibition of ST-stimulated GC-C activation was dependent on the differentiation stage. Crypt-like cells showed higher inhibition with pervanadate. As they matured into villus-like cells, the effect of pervanadate on GC-C activation was gradually lost. This effect also correlated with a decrease in the expression of Lck, suggesting that in the context of the intestine there could be differential regulation of tyrosine phosphorylation of GC-C along the crypt-villus axis. Intestinal ligated loop assays in rats demonstrated that ST-induced fluid accumulation in the intestine was abrogated on pervanadate treatment. Reduction in this fluid accumulation by pervanadate was not observed with 8-Br-cGMP, a cell permeable analogue of cGMP. This indicated that tyrosine phosphorylation of proteins is important for ST-induced fluid accumulation, and perhaps pervanadate modulates this by phosphorylation of GC-C, thereby causing a reduction in fluid accumulation. Earlier in vitro studies on Src-SH2 binding from the laboratory had suggested the possibility of activation of Src family kinases by GC-C. The activation status of Src kinases was monitored by using phosphorylation-state specific antibody, pSFK416. ST stimulation in T84 cells increased Tyr416 phosphorylation of Src kinases in a time dependent manner, indicating that Src kinases are activated downstream of GC-C. This activation of Src kinases was also seen with the endogenous ligand of GC-C, uroguanylin. Interestingly, 8-Br-cGMP a cell permeable analogue of cGMP that is known to mimic other cellular effects of GC-C, namely Cl-secretion and cell cycle progression, did not activate Src kinases, suggesting that the mechanism of Src kinase activation by GC-C could be independent of cGMP. Binding affinities of Src, Lck, Fyn and Yes SH2 domains to Tyr820 phosphorylated GCC peptide were in the nM range, indicating a high affinity of interaction. In vitro GST-SH2 pull down experiments suggested that phosphorylation of Tyr820 in full length GC-C allows interaction of GC-C to the SH2 domain of Src. These studies suggest a dual cross-talk between Src kinases and GC-C; Src phosphorylation inhibits GC-C signaling and stimulation of GC-C by its ligands activates Src kinases. Interaction of proteins containing SH2 and SH3 domains are commonly found in signaling molecules. In accordance with the observation that there are three PXXP motifs in GCC, many SH3 domains could interact with GC-C. GC-C appears to show a preference to bind the SH3 domains of Fyn, Hck, Abl tyrosine kinases, Grb2 and Crk adaptor proteins, the α-subunit of P85 PI3 kinase, PLC-γ and cortactin to various extents. The SH3 domains of spectrin and Nck did not show any detectable interaction with GC-C. In SH3 pull-down assays, the N-terminal SH3 domain of Crk, CrkSH3 (N), bound GC-C maximally, suggesting that Crk is a good candidate for interaction with GC-C. By overlay analysis, the region of GC-C that binds CrkSH3 (N) was narrowed down to the catalytic domain of GC-C containing a ‘PGLP’ motif. Mutations were generated in GC-C at this site to generate GC-CP916Q and GC-CW918R. These mutations compromised the binding of full length receptor to CrkSH3 (N). In cells, CrkII and GC-C co-transfection inhibited the ST stimulation of GC-C. A CrkII mutant, that has compromised binding through its SH3 domain, did not inhibit the activity of GC-C. CrkII from T84 cells co-immunoprecipitated with GC-C and interestingly, the phosphorylated form of CrkII did not, indicating that GC-C - Crk interaction could be regulated by the phosphorylation of Crk. In summary, this study places GC-C, in the context of tyrosine kinase signaling pathway and interaction with the adaptor protein Crk. These studies suggest that GC-C signal transduction can be altered by cross-talk with other signaling events in the cell. Reversible phosphorylation of tyrosine residues inhibits the activity of GC-C, and this is mediated by Src family kinases. Src kinases themselves are activated on stimulation of GC-C by its ligands, possibly because of SH2 domain interaction with GC-C. Association of Crk by its SH3 domain regulates GC-C functioning primarily by inhibiting ST-stimulated cGMP production. This opens up the possibility of GC-C signaling through a multimeric complex involving other binding partners of Crk, and these cross-talks involving GC-C with the two proto-oncogenes, Src and Crk, might have far reaching consequences in the regulation of cellular functions.

3-Deoxy-D-arabino-heptulosonate 7-phosphate (DAHP) synthase (DHS) catalyzes the first step of the shikimate pathway - a pathway involved in Tyrosine (Tyr), Tryptophan (Trp) and Phenylalanine (Phe) biosynthesis - by condensation of phosphoenolpyruvate and erythrose-4-phosphate to DAHP. Our lab previously demonstrated that Arabidopsis thaliana shikimate pathway flux is regulated by Tyr and Trp. This project suggests that A. thaliana DHS1 overexpressor lines have increased Trp accumulation with Tyr treatment, and that an A. thaliana DHS2 overexpressor line treated with Tyr has unchanged Trp accumulation, indicating that AtDHS2 is Tyr-sensitive. Confocal microscopy of all 3 AtDHS isoforms fused to yellow fluorescent protein demonstrates chloroplast localization. Bimolecular fluorescence complementation indicates that protein-protein interactions occur in the cytoplasm, and not in the chloroplast, for AtDHS1 and AtDHS2 with the metabolic regulator At14-3-3ω. These findings suggest that protein-protein interactions could regulate accumulation of AtDHS2 in the chloroplast, and are perhaps modulated by Tyr.