Overactivity of platelet-derived growth factor receptor (PDGFR) is a frequent event in many types of solid tumors. Therefore, it is of great importance to uncover the mechanisms that regulate PDGF/PDGFR signalling, to develop efficient inhibitors targeting this pathway. The first step of downregulation of PDGFR activity upon ligand binding is internalization; thus we investigated how endocytosis pathways affect PDGFR signaling. We showed that in Ras-transformed fibroblasts, the internalization of PDGFR is shifted from the routine clathrin-dependent endocytosis to macropinocytosis, which results in enhanced PDGFR activity and subsequent downstream signalling, promoting anchorage-independent growth. We were also interested in how intracellular trafficking regulates signalling attenuation of PDGFR. We found that His-domain containing protein tyrosine phosphatase (HD-PTP) positively regulates phosphorylation level of the ubiquitin-ligases c-Cbl and Cbl-b; consistently, silencing of HD-PTP led to a decreased level of PDGFR ubiquitination (paper II). Consequently, internalized PDGFR could not be sorted properly and escaped degradation. This resulted in enhanced activation of phospholipase C γ (PLCγ) and changed kinetics of signal transducer and activator of transcription (STAT) 3 signalling, which further increased colony formation of HD-PTP silenced cells in soft agar, indicating a tumor suppressor role of HD-PTP. Activation of PDGFR leads to stimulation of downstream pathways. We identified Fer kinase as a critical signal transducer downstream of PDGFR in a proteomic screen. We showed that Fer kinase is essential for PDGF-induced STAT3 activation; as a result (paper III), Fer depletion severely blunted the ability of PDGFR signalling to promote anchorage-independent growth in soft agar and delayed tumor initiation in a mouse model. The crosstalk between host and tumor plays a critical role in tumor progression. At present most anti-cancer drugs are targeting tumor cells; we were interested in how targeting tumor host cells affects the efficacy of anti-tumor therapy. We found that selective PDGFRβ inhibition in host cells exerted tumor inhibitory effects on growth and vascularization of tumors with autocrine PDGF signaling, whereas tumors lacking such stimulation show only minor response on tumor growth (paper IV). Meanwhile, we demonstrated that PDGF/PDGFRβ signalling promotes expression of NG2, a marker for pericytes.

No description available.

Les molécules classiques du CMH de classe II sont responsables de la présentation de peptides exogènes par les cellules présentatrices d’antigène aux lymphocytes T CD4+. Cette présentation antigénique est essentielle à l’établissement d’une réponse immunitaire adaptative. Cependant, la reconnaissance d’auto-antigènes ainsi que l’élimination des cellules du Soi sont des problèmes à l’origine de nombreuses maladies auto-immunes. Notamment, le diabète et la sclérose en plaque. D’éventuels traitements de ces maladies pourraient impliquer la manipulation de la présentation antigénique chez les cellules dont la reconnaissance et l’élimination engendrent ces maladies. Il est donc primordial d’approfondir nos connaissances en ce qui concerne les mécanismes de régulation de la présentation antigénique. La présentation antigénique est régulée tant au niveau transcriptionnel que post-traductionnel. Au niveau post-traductionnel, diverses cytokines affectent le processus. Parmi celles-ci, l’IL-10, une cytokine anti-inflammatoire, cause une rétention intracellulaire des molécules du CMH II. Son mécanisme d’action consiste en l’ubiquitination de la queue cytoplasmique de la chaîne bêta des molécules de CMH II. Cette modification protéique est effectuée par MARCH1, une E3 ubiquitine ligase dont l’expression est restreinte aux organes lymphoïdes secondaires. Jusqu’à tout récemment, il y avait très peu de connaissance concernant la structure et les cibles de MARCH1. Considérant son impact majeur sur la présentation antigénique, nous nous sommes intéressé à la structure-fonction de cette molécule afin de mieux caractériser sa régulation ainsi que les diverses conditions nécessaires à son fonctionnement. Dans un premier article, nous avons étudié la régulation de l’expression de MARCH1 au niveau protéique. Nos résultats ont révélé l’autorégulation de la molécule par formation de dimères et son autoubiquitination. Nous avons également démontré l’importance des domaines transmembranaires de MARCH1 dans la formation de dimères et l’interaction avec le CMH II. Dans un second article, nous avons investigué l’importance de la localisation de MARCH1 pour sa fonction. Les résultats obtenus montrent la fonctionnalité des motifs de localisation de la portion C-terminale de MARCH1 ainsi que la présence d’autres éléments de localisation dans la portion N-terminale de la protéine. Les nombreux mutants utilisés pour ce projet nous ont permis d’identifier un motif ‘‘VQNC’’, situé dans la portion cytoplasmique C-terminale de MARCH1, dont la valine est requise au fonctionnement optimal de la molécule. En effet, la mutation de la valine engendre une diminution de la fonction de la molécule et des expériences de BRET ont démontré une modification de l’orientation spatiale des queues cytoplasmiques. De plus, une recherche d’homologie de séquence a révélé la présence de ce même motif dans d’autres ubiquitines ligases, dont Parkin. Parkin est fortement exprimée dans le cerveau et agirait, entre autre, sur la dégradation des agrégats protéiques. La dysfonction de Parkin cause l’accumulation de ces agrégats, nommés corps de Lewy, qui entraînent des déficiences au niveau du fonctionnement neural observé chez les patients atteints de la maladie de Parkinson. La valine comprise dans le motif ‘’VQNC’’ a d’ailleurs été identifiée comme étant mutée au sein d’une famille où cette maladie est génétiquement transmise. Nous croyons que l’importance de ce motif ne se restreint pas à MARCH1, mais serait généralisée à d’autres E3 ligases. Ce projet de recherche a permis de caractériser des mécanismes de régulation de MARCH1 ainsi que de découvrir divers éléments structuraux requis à sa fonction. Nos travaux ont permis de mieux comprendre les mécanismes de contrôle de la présentation antigénique par les molécules de CMH II. / Classical MHC class II molecules are responsible for the presentation of exogenous peptides to CD4+ T cells, which is essential for the establishment of the adaptive immune response. However, problems with recognition of auto-antigens and the subsequent cell elimination are at the root of numerous autoimmune diseases. Manipulation of the antigen presentation pathway in order to eliminate cells that present self-antigens could serve as potential treatments of many autoimmune disorders. It is therefore essential to deepen our knowledge regarding the mechanisms regulating antigen presentation. Antigen presentation is regulated both transcriptionally and post-translationally. Whereas many cytokines affect the latter, IL-10, an anti-inflammatory cytokine, causes the intracellular retention of MHC II molecules. This phenotype is the result of the ubiquitination of MHC II -chain cytoplasmic tail by MARCH1. MARCH1 is an E3 ubiquitin ligase expressed in secondary lymphoid organs. Until recently, little was known about the structure-function and the targets of MARCH1. Considering its major impact on antigen presentation, we were interested to study this E3 ligase in order to reveal how it is regulated and what are the required conditions for its function. In a first report, we have investigated the regulation of MARCH1’s protein expression. Our results revealed its autoregulation via dimer formation and autoubiquitination. In addition, we have demonstrated the involvement of MARCH1’s transmembrane domains for dimerization and MHC II interaction. In a second article, we highlighted the importance of MARCH1 localization for its function. Our results indicated that localization motifs in the C-terminal portion of MARCH1 were functional and revealed the presence of some sorting elements in the N-terminal portion of the molecule. A panel of mutant were used and allowed us to identify a ‘’VQNC’’ motif, located in the C-terminal cytoplasmic portion of MARCH1, in which the valine is central for the molecule’s function. Indeed, point-mutation of the valine led to a decrease in MARCH1 ability to relocate MHC II whereas BRET experiments revealed a modification in the spatial organization of the cytoplasmic tails. Moreover, a blast of sequence homology showed the presence or that same motif in others ubiquitine ligases, one of which is Parkin. Parkin is highly expressed in the brain and seems to be implicated in protein aggregates’ degradation. It was reported that malfunction of Parkin leads to the accumulation of aggregates, called Lewy bodies, responsible for the neural functions deficiencies observed in patients with Parkinson disease. Interestingly, a family for which the sickness was genetically transmitted has a mutated valine in the VQNC motif. We believe that the importance of this motif is not restricted to MARCH1 and could be generalized to others E3 ubiquitin ligases. This project enabled us to characterize the regulation mechanisms of MARCH1. In addition, we discovered various structural elements required for its function. Altogether, our data allows for a better understanding of the mechanisms controlling MHC II molecules antigen presentation.

<p>As a means of further localising the interaction, truncated fragments derived from the C-terminal region of YB-1, were tested for their interaction with the RING finger domain of RBBP6 using three different assays: a directed yeast 2-hybrid assay, co-immunoprecipitation and NMR chemical shift perturbation analysis. Our results suggest that the entire 62 amino acid region at the C-terminal domain of YB-1 may be involved in the interaction with RBBP6. Using chemical shift perturbation analysis, this study provides an indication of where YB-1 binds to the RING finger. This represents the first step towards the design of therapeutics aimed at modulating the interaction between RBBP6 and YB-1 as a means of targeting the oncogenic effects of YB-1. In order to identify E2 enzymes involved in the ubiquitination of YB-1, we examined the efficiencies of selected E2s in an in vitro ubiquitination assay. UbcH5c and UbcH7 were both found to catalyse the ubiquitination of YB-1 in conjuction with RBBP6, whereas Ubc13 was not. Finally, we show using NMR that two single-point mutations of the RING finger-like domain are sufficient to abolish homodimerisation of the domain. These will be used in future studies to investigate the requirement for homodimerisation on the ubiquitination activity of RBBP6.</p>

Thèse diffusée initialement dans le cadre d'un projet pilote des Presses de l'Université de Montréal/Centre d'édition numérique UdeM (1997-2008) avec l'autorisation de l'auteur.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

The ubiquitination process is vital to maintain the protein homeostasis in the cell. With high specificity it regulates degradation of proteins by tagging them with a small protein called ubiquitin. Four proteins are involved to perform the process and in this thesis one of these proteins is studied. This protein is called Ro52 and belongs to the TRIM protein family. It posses E3 ligase activity because of a N-terminal RING-domain and therefore it is responsible for the last step in the ubiquitination process. The structure of Ro52 is not totally solved and the function of the protein’s four domains is not fully understood. In this thesis three constructs of two domains from Ro52 (RING and B-box) is investigated by circular dichroism (CD), nuclear magnetic resonance (NMR) spectroscopy and auto-ubiquitination assay by Western blot. The goal was to gain deeper insight in structural and functional properties of these domains. In the end only two constructs were investigated because of time limitations. It was shown by NMR that one construct has similar structure as the wild type but lower stability, possibly due to shorter N-terminal region. Comparison of the results from CD measurements showed that the constructs were well structured but did not reveal any significant differences in secondary structure between the constructs. Functional analysis by Western blot encountered unexpected problems and no results were obtained. The current thesis provides a basis for further investigation of variant constructs jointly expressing the RING-B-box domains, and shows that even small changes may alter structure and stability in ways that might affect functional properties.

The muscarinic acetylcholine type 2 receptor (M2 mAChR), a member of the GPCR superfamily, is found throughout the parasympathetic nervous system where it controls pulmonary, urinary, and cardiac function, and neurotransmission. The molecular mechanisms that regulate M2 mAChR availability at the cell surface are an important component in controlling these physiological events. Since beta-arrestin proteins are known to regulate the activity of other GPCRs, we sought to identify their role in regulating M2 mAChR activity, a topic that remains contentious in the field. To achieve this goal we utilized mouse embryonic fibroblasts (MEFs) derived from beta-arrestin knockout mice lacking one or both isoforms (MEF KO1, KO2, or KO1/2 cells) in addition to exogenous expression of beta-arrestin mutants. This study demonstrates that agonist-induced internalization of M2 mAChR is beta-arrestin- and clathrin-dependent, and that the receptor stably co-localizes with beta-arrestin in early endosomal vesicles suggesting it behaves as a class B receptor. Next, we sought to identify beta-arrestin s function in regulating the post-endocytic trafficking (down-regulation) of the M2 mAChR. MEF KO1/2 cells were unable to down-regulate M2 mAChRs whereas MEF KO1 or KO2 cells retained the ability to do so. In MEFwt cells, both M2 mAChR and beta-arrestin exhibited basal ubiquitination that increased following agonist stimulation. Receptor degradation appeared to be regulated by the ubiquitination status of beta-arrestin 2, since expression of a chimeric â-arrestin 2 form fused to ubiquitin increased both constitutive and agonist-promoted down-regulation, whereas expression of a beta-arrestin 2 mutant lacking putative ubiquitination sites, beta-arrestin 2K18R, K107R, K108R, K207R, K296R, significantly blocked degradation while internalization and stable association remained intact. Upon further analysis, the beta-arrestin 2K18R, K107R, K108R, K207R, K296R mutant blocked delivery of M2 mAChR to the late endosome/lysosome, presumably where degradation occurs. Inhibition of proteasome-dependent recycling of ubiquitin blocked receptor down-regulation without affecting internalization or the ubiquitination state of the M2 mAChR while ubiquitination of beta-arrestin 2 diminished significantly. These results support a role for ubiquitinated beta-arrestin in mediating M2 mAChR sorting and degradation in the lysosome. Collectively, these studies give us new insight on the function of beta-arrestin in regulating the activity of the M2 mAChR.

Carbon catabolite repression (CCR) is a mechanism by which micro-organisms preferentially utilize more easily metabolizable carbon sources in comparison to less easily metabolizable carbon sources. It prevents the organisms from unnecessary expenditure of energy and enables them to exploit the nutrients in appropriate manner. It represents a complex system of gene regulation. The main aim of this study was to study the intracellular localization of proteins involved in CCR including CreA, CreB, CreC and CreD in A. nidulans in repressing and derepressing conditions. The major regulatory protein involved in CCR in A. nidulans is CreA. It is a DNA-binding repressor, but very little is known about the molecular events that allow CreA function to result in appropriate regulation in response to carbon source. To determine the amount and localization of CreA in different carbon sources, strains were made over-expressing GFP and HA tagged CreA. Western analysis showed that high levels of full length CreA can be present in cells that show normal responses to carbon catabolite repression, whether they are grown in repressing or derepressing media. Hence the amount of CreA is similar in both the conditions and thus degradation of CreA is not a key step in carbon catabolite repression. Fluorescence microscopy studies have shown that CreA is in the nucleus under repressing and derepressing carbon conditions and this is not affected by the absence of CreB or CreD, the other important proteins in A. nidulans. Thus mere localization of CreA in nucleus is not sufficient to cause carbon catabolite repression and there is some modification process involved for CreA to act as a repressor protein in CCR. CreB is a deubiquitinating protein and CreC is a protein containing five WD 40 repeats, a putative nuclear localization signal (NLS) and a proline rich region and both the proteins are present in the cell in a complex. CreB was localized using strains that over-expresses GFP tagged CreB and fluorescence microscopy. CreB is present mainly in the cytoplasm in both repressing and derepressing conditions. Moreover, intracellular localization of CreB is unaffected by the presence or absence of CreD. However, the amount of CreB was higher in a creD+ background as compared to a creD34 mutant background, implying that the presence of CreD affects the amount of CreB in the cell. CreC was localized by using strain that over-expresses YFP tagged CreC and it is also present mainly in the cytoplasm. CreD contains arrestin domains and PY motifs and is highly similar to the Rod1p and Rog3p from S. cerevisiae. CreD is proposed to be involved in ubiquitination process in CCR in A. nidulans. Localization studies have shown that CreD is present throughout the cell in a punctate pattern with more in the cytoplasm than in the nucleus. CreB and CreD co-localize in some regions of the cell whereas in other regions either CreB or CreD is present. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1346526 / Thesis (Ph.D.) - University of Adelaide, School of Molecular and Biomedical Science, 2008

Carbon catabolite repression (CCR) is a mechanism by which micro-organisms preferentially utilize more easily metabolizable carbon sources in comparison to less easily metabolizable carbon sources. It prevents the organisms from unnecessary expenditure of energy and enables them to exploit the nutrients in appropriate manner. It represents a complex system of gene regulation. The main aim of this study was to study the intracellular localization of proteins involved in CCR including CreA, CreB, CreC and CreD in A. nidulans in repressing and derepressing conditions. The major regulatory protein involved in CCR in A. nidulans is CreA. It is a DNA-binding repressor, but very little is known about the molecular events that allow CreA function to result in appropriate regulation in response to carbon source. To determine the amount and localization of CreA in different carbon sources, strains were made over-expressing GFP and HA tagged CreA. Western analysis showed that high levels of full length CreA can be present in cells that show normal responses to carbon catabolite repression, whether they are grown in repressing or derepressing media. Hence the amount of CreA is similar in both the conditions and thus degradation of CreA is not a key step in carbon catabolite repression. Fluorescence microscopy studies have shown that CreA is in the nucleus under repressing and derepressing carbon conditions and this is not affected by the absence of CreB or CreD, the other important proteins in A. nidulans. Thus mere localization of CreA in nucleus is not sufficient to cause carbon catabolite repression and there is some modification process involved for CreA to act as a repressor protein in CCR. CreB is a deubiquitinating protein and CreC is a protein containing five WD 40 repeats, a putative nuclear localization signal (NLS) and a proline rich region and both the proteins are present in the cell in a complex. CreB was localized using strains that over-expresses GFP tagged CreB and fluorescence microscopy. CreB is present mainly in the cytoplasm in both repressing and derepressing conditions. Moreover, intracellular localization of CreB is unaffected by the presence or absence of CreD. However, the amount of CreB was higher in a creD+ background as compared to a creD34 mutant background, implying that the presence of CreD affects the amount of CreB in the cell. CreC was localized by using strain that over-expresses YFP tagged CreC and it is also present mainly in the cytoplasm. CreD contains arrestin domains and PY motifs and is highly similar to the Rod1p and Rog3p from S. cerevisiae. CreD is proposed to be involved in ubiquitination process in CCR in A. nidulans. Localization studies have shown that CreD is present throughout the cell in a punctate pattern with more in the cytoplasm than in the nucleus. CreB and CreD co-localize in some regions of the cell whereas in other regions either CreB or CreD is present. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1346526 / Thesis (Ph.D.) - University of Adelaide, School of Molecular and Biomedical Science, 2008