Notch signaling is a major signaling pathway that occurs in many tissues at in nearly all stages of development. In Drosophila, Notch and its ligands, Delta and Serrate, physically interact as a part of activation of the signal. Notch activation requires the endocytic adaptor protein Epsin to facilitate the endocytosis of the ligand Delta. Our laboratory has discovered that Epsin activity is regulated by ubiquitination. Liquid facets, the gene coding for the Drosophila protein related to Epsin, was discovered to be an enhancer of the fat facets (faf) mutant eye phenotype. faf codes for a deubiquitinating enzyme. Epsin has been determined to be a key substrate of the activity of Faf in the eye. An F1 screen for dominant suppressors of the faf phenotype was performed to identify the E3 ubiquitin ligase whose substrate is Epsin. The E2 ubiquitin-conjugating enzyme UbcD1 was isolated as a strong suppressor of the faf mutant eye defect. UbcD1 has previously been identified as a strong suppressor of faf. An RNAi approach was used to study UbcD1’s role in Epsin ubiquitination further. The data that I obtained do not elucidate the UbcD1’s role in Epsin ubiquitination, but instead suggest an alternative role that should be considered. / text

Ubiquitination is an extremely important post-translational modification, regulating a wide variety of cellular processes including proteasomal degradation, subcellular targeting, endocytosis and DNA repair. The HECT class of E3 ligases catalyse the final step of ubiquitin conjugation to the substrate; the Nedd4 family make up 9 members of this class in humans, and are implicated in pathologies ranging from congenital ion channel misregulation to cancer, via the TGF-ß signalling pathway. The Nedd4-like proteins contain WW substrate recognition domains, which recognise and bind proline–rich PY motifs. This work focuses on the interaction between Nedd4 and PMEPA1, a membrane protein showing altered expression in prostate cancer and a known Nedd4 substrate. PMEPA1 is recognised as important in several cancers, although its detailed function is not yet known; it is upregulated in prostate cancer and has been postulated to decrease cellular androgen receptor (AR) via a negative feedback loop involving Nedd4 in a ubiquitin-proteasome dependent process. Misregulation of the AR response to testosterone is associated with a more advanced form of prostatic cancer and poor patient prognosis, for which there are currently few treatment options available. PMEPA1 contains two well-documented canonical (PPxY) PY motifs both required for interaction with Nedd4. This work details the identification and characterisation of a third motif with a variant PY (vPY) sequence, QPTY. Our findings indicate that the loss of this vPY motif leads to a significant reduction in Nedd4-dependent ubiquitination in vitro. In addition, the nature of the amino acid residues surrounding the vPY motif also appears to play a role in the functionality of the site. Alongside these experiments, immunodetection of protein levels in HeLa and LNCaP cell lysates was used in conjunction with confocal microscopy to shed light on the interactions between PMEPA1, Nedd4 and AR in vitro and in vivo. A possible non-proteasomal role for ubiquitination in the PMEPA1-AR interaction, as opposed to the simple ubiquitin-proteasome mediated degradation previously proposed, is discussed in the light of this new data. This work has expanded previous knowledge of the specificity of the PY-WW interaction, as well as providing a basis for further investigation, and possibly clinical targeting, of the role of PMEPA1 and AR in prostate cancer.

The Estrogen Receptor alpha (ER alpha) is a multi-domain transcription factor that has been extensively studied due to its known involvement in breast cancer treatment and progression. Subsequent studies have shown coregulators are extensively involved in modulating the transcriptional activation of ER alpha and many of these proteins possess enzymatic functions. Coregulators are divided into two categories, coactivators which enhance transcriptional output and corepressors which decrease transcriptional output. One protein responsible for Angelman syndrome, E6-associated protein (E6-AP) was found to be a coactivator of ER alpha and possessed ubiquitin ligase activity; however, the ubiquitin ligase activity has been shown not to be essential to E6-AP coactivation ability. The current work was undertaken to explore the role of E6-AP in the regulation of ER alpha. E6-AP was found to play a role in a unique ligand-independent degradation pathway. Because the degradation effect was ligand-independent, it was proposed that the degradation signal mediating the event occurred through phosphorylation of E6-AP. In silico analysis of E6-AP indicated several potential phosphorylation sites on the E6-AP protein. Numerous phosphorylation sites of E6-AP were confirmed by western blot and mass spec indicating a possible phosphorylation signal mediating E6-AP/ER alpha interaction. Because it has been shown that the ligase function of E6-AP is not required for its coactivation, we then examined E6-AP coactivation of ERα in the presence of ligand. One well studied gene TFF-1 (pS2) was examined as a model ERα target gene. Estrogen-mediated transcription from TFF-1 was decreased with knockdown of E6-AP in both MCF-7 and T47D cell lines. Furthermore, under E6-AP knockdown conditions, ChIP of p300, a known histone acetyl transferase (HAT), indicated a reduced recruitment to the TFF-1 promoter in both cell lines. Interestingly, the reduced recruitment of p300 had a cell specific effect on phosphorylated RNA polymerase II (pRNA pol II) recruitment indicating cell specific functions of E6-AP. Further investigation also found a gene specific effect for E6-AP on pRNA pol II recruitment. The current work provides a new role for E6-AP as a coactivator of ER alpha in the form of a scaffold allowing creation of fully functional transcription complexes in a gene and cell specific manner.

UEV1A and MMS2 are two human genes whose proteins share greater than ninety percent sequence identity. Both Uev1A and Mms2 are ubiquitin-conjugating enzyme variants (Uev) that lack the active cystine residue required for the ubiquitin conjugation reaction. They work with the ubiquitin-conjugating enzyme Ubc13 to create Lys63-linked polyubiquitin chains which have recently been found to cause cellular signals not involving target protein degradation. Only the Mms2-Ubc13 complex functions in DNA repair in mammalian cells. In contrast, only the Uev1A-Ubc13 complex is involved in TRAF2- and TRAF6-mediated NF-¦ÊB activation by ubiquitinating NEMO/IKKg. UEV1B is a splice variant of UEV1A containing different N-terminal coding sequences and its cellular function is currently unknown. <p>The NF-¦ÊB signaling pathway has been regarded as a primary pro-survival and anti-apoptotic response. UEV1A expression is positively correlated to tumor formation, suggesting that it plays a role in tumorigenesis. Furthermore, experimental overexpression of UEV1A alone is sufficient to activate NF-¦ÊB, which is reversible upon suppression of UEV1A expression by RNA interference. Overexpression of UEV1A also protects cells from stress-induced apoptosis and confers a growth advantage under serum-deprived conditions. These observations collectively support UEV1A as a candidate proto-oncogene.<p> The objective of this research was to obtain monoclonal antibodies (Mabs) capable of recognizing Uev1, but not Mms2. This is a challenging task given very few possible epitopes that may distinguish the two proteins. The four sequences unique to Uev1A are located in the 30 a.a. N terminal region, the single substitution at a.a. 104, the 14 a.a. core region from a.a. 116-129 (7/14 variable) and the C-terminus at a.a. 167-169.<p>Hybridoma cells were produced vis ¨¢ vis fusions of B cells from Uev1A-immunized mice with FO myeloma cells lacking the ability to produce immunoglobulin. The hybridoma cells were screened using enzyme immunoassays (EIAs) for reactivity with Uev1A and Mms2. Ascities fluid was produced for five Mabs named LN1, LN2, LN2A, LN2B and LN3. EIA and Western blotting of Uev1A-deletion constructs revealed that Mab LN1 binds specifically to amino acids (a.a.) 10-30 of the unique Uev1A N-terminal sequence, that Mabs LN2, LN2A and LN2B bind specifically to the unique a.a. 110-130 region of Uev1A, and that Mab LN3 binds specifically to a.a. 30-44 of Uev1A common to Mms2.<p> Competition analysis of unconjugated Mabs versus horse radish peroxidase (HRP)-conjugated Mabs for binding to Uev1A permitted epitope mapping for the five Mabs. The results indicate that Mabs LN1 and LN3 inhibit each other from binding to their distinct sequences which are spatially adjacent. Mabs LN2 and LN2B inhibit each other, but not Mab LN2A. Additionally, Mab LN2A is unable to inhibit Mabs LN2 or LN2B. In summary, Mabs have been aquired to three catergories which were originally desired: one to the unique N-terminus (Mab LN1), three to the unique core sequence (Mabs LN2, LN2B and LN2A), and one to the same region Ubc13 binds to Uev1A and Mms2 (Mab LN3). The potential applications of these Mabs are discussed.

UEV1A and MMS2 are two human genes whose proteins share greater than ninety percent sequence identity. Both Uev1A and Mms2 are ubiquitin-conjugating enzyme variants (Uev) that lack the active cystine residue required for the ubiquitin conjugation reaction. They work with the ubiquitin-conjugating enzyme Ubc13 to create Lys63-linked polyubiquitin chains which have recently been found to cause cellular signals not involving target protein degradation. Only the Mms2-Ubc13 complex functions in DNA repair in mammalian cells. In contrast, only the Uev1A-Ubc13 complex is involved in TRAF2- and TRAF6-mediated NF-¦ÊB activation by ubiquitinating NEMO/IKKg. UEV1B is a splice variant of UEV1A containing different N-terminal coding sequences and its cellular function is currently unknown. <p>The NF-¦ÊB signaling pathway has been regarded as a primary pro-survival and anti-apoptotic response. UEV1A expression is positively correlated to tumor formation, suggesting that it plays a role in tumorigenesis. Furthermore, experimental overexpression of UEV1A alone is sufficient to activate NF-¦ÊB, which is reversible upon suppression of UEV1A expression by RNA interference. Overexpression of UEV1A also protects cells from stress-induced apoptosis and confers a growth advantage under serum-deprived conditions. These observations collectively support UEV1A as a candidate proto-oncogene.<p> The objective of this research was to obtain monoclonal antibodies (Mabs) capable of recognizing Uev1, but not Mms2. This is a challenging task given very few possible epitopes that may distinguish the two proteins. The four sequences unique to Uev1A are located in the 30 a.a. N terminal region, the single substitution at a.a. 104, the 14 a.a. core region from a.a. 116-129 (7/14 variable) and the C-terminus at a.a. 167-169.<p>Hybridoma cells were produced vis ¨¢ vis fusions of B cells from Uev1A-immunized mice with FO myeloma cells lacking the ability to produce immunoglobulin. The hybridoma cells were screened using enzyme immunoassays (EIAs) for reactivity with Uev1A and Mms2. Ascities fluid was produced for five Mabs named LN1, LN2, LN2A, LN2B and LN3. EIA and Western blotting of Uev1A-deletion constructs revealed that Mab LN1 binds specifically to amino acids (a.a.) 10-30 of the unique Uev1A N-terminal sequence, that Mabs LN2, LN2A and LN2B bind specifically to the unique a.a. 110-130 region of Uev1A, and that Mab LN3 binds specifically to a.a. 30-44 of Uev1A common to Mms2.<p> Competition analysis of unconjugated Mabs versus horse radish peroxidase (HRP)-conjugated Mabs for binding to Uev1A permitted epitope mapping for the five Mabs. The results indicate that Mabs LN1 and LN3 inhibit each other from binding to their distinct sequences which are spatially adjacent. Mabs LN2 and LN2B inhibit each other, but not Mab LN2A. Additionally, Mab LN2A is unable to inhibit Mabs LN2 or LN2B. In summary, Mabs have been aquired to three catergories which were originally desired: one to the unique N-terminus (Mab LN1), three to the unique core sequence (Mabs LN2, LN2B and LN2A), and one to the same region Ubc13 binds to Uev1A and Mms2 (Mab LN3). The potential applications of these Mabs are discussed.

Abstract Tumor susceptibility gene TSG101 plays an important role in cellular functions including intracellular protein sorting, vesicular trafficking, and transcription regulation. Our previous results from yeast two-hybrid screening show that TSG101 interacts with a novel transcriptional repressor protein, KLIP1. In this study, we demonstrated in vivo interaction between TSG101 and KLIP1 in nucleus of 293 cells using co-immunoprecipitation assay and confocal imaging. In addition, we found KLIP1 could be modified in a modality of either poly- and mono-ubiquitination when exogenously expressed in 293 cells in conjunction with either wild type His-tagged ubiquitin or a mutant His-tagged ubiquitin (K0-Ub) which has no capability of forming polyubiquitin chain. Furthermore, we found that TSG101 could increase 60 kDa-KLIP1, but decrease 71 kDa-KLIP1 levels of monoubiquitinated KLIP1 protein species in a dose dependent manner. These results indicate that TSG101 might regulate KLIP1 protein function through affecting its monoubiquitin modification status. Further investigation using wildtype pHA-KLIP1 and mutant pHA-KLIP1-M6 containing mutation in its 6 lysine residues for possible ubiquitin modification revealed that wildtype HA-KLIP1, but not HA-KLIP1-M6, could inhibit transcription activity of thymidine kinase (TK) promoter. In conclusion, our results support that TSG101 interacts and acts as a transcriptional co-repressor of KLIP1 by keeping it in 60 kDa-monoubiquitinated status in the nucleus, where KLIP1 functions as a transcription repressor for TK promoter. Further experiment using mutant HA-KLIP1 expression plasmid containing single mutation in the 6 lysine sites should reveal the exact location of ubiquitin-modified lysine site for monoubiquitinated species of KLIP1 protein.

Immunoglobulin-containing and proline-rich receptor-1 (IGPR-1), is a newly identified cell adhesion molecule and is expressed in various cell types including, epithelial and endothelial cell origin. IGPR-1 regulates cell-cell adhesion and promotes angiogenesis, activated by shear stress and mediates endothelial cells response to shear stress. Moreover, IGPR-1 expression is upregulated in colon cancer and supports colon tumor growth. IGPR-1 contains a single extracellular immunoglobulin domain, a transmembrane domain and followed by a cytoplasmic proline-rich C-terminus. We demonstrate that ubiquitin E3 protein ligase neural precursor cell expressed developmentally down regulated 4 (NEDD4) binds to and ubiquitinates IGPR-1. Furthermore, among the four WW domains, the C-terminus WW domain#4 selectively mediates the binding of NEDD4 with IGPR-1. We used in vitro ubiquitination assay and identified UbcH6, as an E2 conjugating enzyme required for NEDD4-mediated ubiquitination of IGPR-1. Taken together, our data identifies NEDD4/Ubc6H ubiquitination system as a major pathway involved in the ubiquitination of IGPR-1. / 2021-06-08T00:00:00Z

Le virus de la mosaïque jaune du navet (TYMV) est un excellent modèle pour la réplication des virus à ARN simple brin de polarité positive. Ce petit virus de plante code l’ensemble desprotéines nécessaires à sa réplication sous forme d’un précurseur polyprotéine (206K). Du NauC-terminus, celui-ci porte une activité méthyltransférase, protéase à cystéine (PRO),hélicase et ARN-polymérase ARN dépendante (POL). Comme tous les virus à ARN(+)connus, son complexe de réplication est étroitement lié à des membranes cellulaires. Dans lecas du TYMV, c’est l’enveloppe des chloroplastes qui est impliquée. Un des acteurs clés de laréplication est la polymérase qui permet la synthèse de nouveaux génomes. La régulation de son activité implique dans un premier temps son clivage de la 206K par PRO. Ainsi libérée,elle est recrutée aux chloroplastes par une interaction directe avec le domaine PRO du produit de clivage 140K, afin de former le complexe de réplication. Un second clivage par PRO contenu dans 140K, à la jonction PRO-hélicase, permet de poursuivre le cycle de réplication.Récemment, il a également été montré que l’activité POL était régulée par la voie ubiquitine-protéasome durant le cycle viral. Ubiquitinilée par la cellule hôte, elle est adressée au protéasome où elle sera dégradée. Cependant, PRO, grâce à sa seconde fonction ubiquitine hydrolase, est capable de la protéger de cette dégradation. Afin de caractériser d’un point de vue structural ce mécanisme de régulation de la réplication, nous avons cristallisé, à l’aide d’un contaminant, PRO et avons résolu sa structure. L’empilement cristallin est tel que le Cterminus d’un domaine PRO est inséré dans la crevasse catalytique du domaine PRO suivant,nous fournissant ainsi des informations structurales sur son activité endopeptidase. Dans un second temps, afin d’avoir des informations sur sa seconde activité, nous avons réalisé un complexe stable entre PRO et une molécule d’ubiquitine afin de le cristalliser et résoudre sa structure. Enfin, nous avons initié l’étude cristallographique de la polymérase. / Turnip Yellow Mosaic Virus is an excellent model for positive-stranded RNA virus replication. It’s a small plant virus whose replication machinery is encoded in the viralgenome as a single polyprotein (206K). From N- to C-terminus, this 206K harbors amethyltransferase, a cysteine proteinase (PRO), a helicase and an RNA-dependent RNApolymerase (POL). As in all RNA(+) viruses known, the replication complex is bound to cellmembranes. For TYMV, it ,is the chloroplast envelope that is implicated. A key component inthe replication process is the polymerase, that allows the synthesis of new genomes. The regulation of its activity involves initially its cleavage by PRO from the 206K. Once liberated,the polymerase is recruited to the chloroplasts through a direct interaction with the PROdomain, contained in 140K, in order to form the replication complex. A second cleavage of140K by PRO at the PRO-helicase junction allows to continue the replication cycle. Recently,it has also been shown that the POL activity was regulated by the ubiquitin-proteasomesystem during the viral multiplication cycle. Ubiquitinilated by the host cell, POL is addressed to the proteasome where it is degraded. However, PRO, due to its second function as an ubiquitin hydrolase, is able to protect POL from its degradation. In order to characterizethis mechanism of replication regulation with a structural point of view, we crystallized,assisted by contaminant, PRO and resolved its structure. In the crystal packing, the C-terminalof a PRO domain is inserted into the catalytic cleft of the next PRO domain, thus providing us structural informations on its endopeptidase activity. In a second step, in order to obtain information about its second activity, we made a stable complex between PRO and amolecule of ubiquitin in order to cristallise it. Finally, we initiated the crystallographic studyof POL.Keywords:

Les hétérotopies nodulaires périventriculaires (HNP) sont des malformations du cortex cérébral caractérisées par la formation d’amas de neurones dans des parties inappropriées du cerveau. Elles peuvent être responsables d’une multitude de troubles tels qu’une hypotonie, un déficit intellectuel ou des épilepsies. Notre équipe a montré que le gène NEDD4L codant pour une E3 ubiquitine ligase était responsable d’HNP associées à des syndactilies. Des études cellulaires et une approche in utero nous ont permis de montrer une instabilité de la protéine mutante et des problèmes lors de la neurogénèse, le positionnement neuronal et la translocation terminale. Des études plus approfondies mettent en lumière le rôle critique de NEDD4L dans différentes voies de signalisation. En effet, alors que l’excès de protéine NEDD4L WT conduit à une dérégulation de DAB1 et de la voie mTORC1, l’instabilité des protéines mutantes conduit à une dérégulation des voies mTOC1 et AKT. L’ensemble de ces données permet de mieux comprendre le rôle critique de NEDD4L dans la régulation des voies mTOR et sa contribution dans le développement cortical. / Neurodevelopmental disorders with periventricular nodular heterotopia (PNH) are etiologically heterogeneous, and their genetic causes remain in many cases unknown. Here we show that missense mutations in the HECT domain of the E3 ubiquitin ligase NEDD4L lead to PNH associated with toes syndactyly, cleft palate and neurodevelopmental delay. Cellular and expression data showed a sensitivity of PNH-associated mutants to proteasome degradation. Moreover, in utero electroporation approach showed that PNH-related mutants and excess of wild type (WT) NEDD4L affect neurogenesis, neuronal positioning and terminal translocation. Further investigations, including rapamycin based experiments, revealed differential deregulation of pathways involved. Excess of WT NEDD4L leads to a disruption of Dab1 and mTORC1 pathways, while PNH-related mutations are associated with a deregulation of mTORC1 and AKT activities. Altogether, these data provide insights to better understand the critical role of NEDD4L in the regulation of mTOR pathways and their contributions in cortical development.

La voie de signalisation TGF-β joue un rôle biphasique durant la cancérogenèse. Mon laboratoire a identifié une nouvelle protéine inhibitrice de la voie TGF-β, WWP1. WWP1 est une E3 ubiquitine ligase qui induit la polyubiquitination et la dégradation du récepteur de type I au TGF-β. De plus, le gène WWP1 est amplifié dans une large proportion de cancers mammaires et prostatiques, suggérant que WWP1 pourrait jouer un rôle clé dans les processus de cancérogenèse liés au TGF-β. Mon projet de thèse était donc de caractériser la régulation de l’activité catalytique de WWP1 ainsi que son mécanisme d’action dans la cellule. Mes résultats montrent qu’à l’état basal, WWP1 est monoubiquitinée, son activité de polyubiquitination étant réduite par l’effet inhibiteur qu’exercent les domaines C2 et/ou WW sur son domaine HECT. En présence de substrats, la protéine WWP1 « s’ouvre » et peut alors induire la polyubiquitination et la dégradation de ses substrats. De plus, nous avons observé qu’un mutant de WWP1, détecté dans un cancer de la prostate, est incapable de s’autoréguler selon ce modèle. Il présente une plus forte activité ligase envers lui-même et ses substrats, ce qui entraîne une atténuation de la réponse cytostatique du TGF-β pouvant conférer une activité oncogénique à WWP1. De plus, nous avons identifié STARD13 comme un nouveau partenaire de WWP1. STARD13 est une protéine à activité RhoGAP, considérée comme un suppresseur de tumeur. Nous avons montré que STARD13 permet l’association de WWP1 avec la GTPase RhoA, entraînant ainsi la polyubiquitination et la dégradation de RhoA. De façon intéressante, le complexe WWP1/STARD13 est impliqué dans le remodelage de l’architecture du cytosquelette en dégradant préférentiellement la forme activée de RhoA. Ces résultats ont permis d’identifier un nouveau rôle de WWP1 qui pourrait jouer un rôle essentiel durant la migration des cellules cancéreuses lors du processus métastatique. La caractérisation de nouveaux mécanismes de régulation et d’action de WWP1 devrait permettre à terme d’identifier si WWP1 est un marqueur diagnostique dans le cancer et/ou une nouvelle cible thérapeutique pour le développement de médicaments anticancéreux. / The TGF-β pathway plays a biphasic role during cancerogenesis. My laboratory identified a new protein, WWP1, as a negative regulator of TGF-β signaling. WWP1 is an E3 ubiquitin ligase that triggers polyubiquitination and degradation of TGF-β type I receptor. A genomic amplification of WWP1 is found in a large portion of mammary and prostatic tumors, suggesting a key role for WWP1 during carcinogenesis related to TGF-β. My thesis project was to determine the regulation of the catalytic activity of WWP1 and a new molecular mechanism of action of WWP1 whose deregulation can be implicated in cancerogenesis. My results indicate that at steady states, WWP1 is monoubiquitinated, its polyubiquitination activity being silenced due to the inhibitory effects of C2 or/and WW domains on its Hect domain. In presence of substrates, WWP1 is « opened » and induces polyubiquitination and degradation of its substrates. Moreover, a WWP1 mutation found in prostate cancer disrupts this regulatory mechanism. It possesses an increased ligase activity towards itself and its substrates, which leads to the attenuation of TGF-β cytostatic signaling, a consequence that could conceivably confer tumorigenic properties to WWP1. We also identified STARD13 as a novel WWP1 interacting partner. STARD13 has a RhoGAP activity, and is considered as a tumor suppressor. We have shown that STARD13 mediates the association of WWP1 with the GTPase RhoA, ultimately leading to RhoA polyubiquitination and degradation. Interestingly, the WWP1/STARD13 complex is involved in the actin cytoskeleton rearrangement by preferentially targeting the active form of RhoA for degradation. These results reveal a previously unrecognized role for WWP1, which could play a key role in the migration of cancer cells during metastasis. Characterization of new regulation and action mechanisms for WWP1 should allow identifying whether WWP1 is a diagnosis biomarker in cancer and/or a new therapeutic target for the development of anticancer drugs.