Breast cancer is unanimously considered a highly heterogeneous disease due to its diverse molecular features. Breast tumor kinase (BRK), also known as protein tyrosine kinase 6 (PTK6), is a non-receptor tyrosine kinase that is highly expressed in over 80% of breast carcinomas. The role and mechanism of action of enzymatically activated BRK in breast pathology are unclear. The objectives of this project were to reveal the effect of BRK activation on cell migration, proliferation and tumorigenesis. We also aimed to determine the mechanism of action of BRK in the promotion of cell proliferation. We used BRK-negative cells (MCF10A, MDA-MB-231 and HEK293) to generate three sets of stable cell lines that stably expressed GFP alone, GFP-BRK-WT or GFP-BRK-Y447F (constitutively active) by retroviral infections. We also stably knocked down BRK from BRK-positive cells BT20 and SKBR3 by RNA interference using shRNAs against BRK. Western blotting, immunoprecipitation and qPCR studies were conducted to evaluate protein expression, protein-protein interaction and mRNA expression, respectively. Both sets of cell lines were used to determine the effect of BRK on cell proliferation (automated cell counter), cell migration (transwell and wound healing assay), transformation (colony formation assay) and tumor formation (mouse Xenograft assay). To investigate the mechanism of action of BRK, we validated downstream of tyrosine kinases 1 (Dok1), a tumor suppressor, as a BRK substrate. Deletion or site-directed mutagenesis was performed to map BRK-targeted tyrosines in Dok1 protein. Results obtained from this research project showed that stable expression of the constitutively active mutant of BRK (BRK-Y447F) in MDA-MB-231 cells led to a significant increase in the cell proliferation, migration rate and promoted colony formation and drastically enhanced tumor formation in athymic nude mice in comparison to control cells. Additionally, depletion of BRK abrogated the migration of BT20 and SKBR3 cells. Furthermore, we showed that BRK interacts with and phosphorylates Dok1, inducing Dok1 downregulation via a ubiquitin-proteasome-mediated mechanism. Together, our results show that the activation of BRK is essential for mammary gland tumorigenesis and suggest that targeting of Dok1 for degradation is a novel mechanism of action of BRK in the promotion of cell proliferation, migration and tumor formation.

SRMS (Src-Related tyrosine kinases lacking C-terminal Regulatory tyrosine and N terminal Myristoylation Sites) belongs to a family of non-receptor tyrosine kinases, which also includes breast tumor kinase (BRK). SRMS was first identified in 1994 in a screen for the genes that regulate the growth and differentiation of neuroepithelial cells. This 54 kDa protein spanning 488 amino acids, consists of the prototypical Src homology 3 (SH3), Src homology 2 (SH2) and a tyrosine kinase domain. While BRK has been documented for its expression in over 60 % of breast carcinomas, information on SRMS on similar grounds remains absent from the literature. Furthermore, unlike BRK, knowledge of how SRMS regulates its enzymatic activity as well as the identification of its substrates remains unknown. The work in this thesis demonstrates that SRMS is potentially expressed in the majority of breast carcinomas. To understand the biochemical and cellular functions of SRMS, a series of mutants comprising point mutations as well as the deletion of the N-terminal region and the functional, SH3 and SH2 domains, were generated and assessed for enzymatic activity in cells. This study demonstrates for the first time that the wild type protein is apparently constitutively active and that its N-terminal region regulates its enzymatic activity. As well, three critical amino acid residues in the protein namely, lysine 258 (ATP binding site), tyrosine 380 (auto-phosphorylation site) and tryptophan 223 (intramolecular interaction) have been characterized. All three residues have been determined to be essential for the enzymatic activity of SRMS. Finally, the adapter protein Dok1 has been characterized as a novel substrate of SRMS. The results from the present study underscore the potential significance of the catalytically active non-receptor tyrosine kinase, SRMS that should serve as a foundation upon which further research may ensue in the context of breast tumorigenesis.

Le suppresseur de tumeur DOK1 (downstream of tyrosine kinases1) est une protéine régulatrice de voies de signalisation impliquées dans des processus cellulaires tel que la prolifération, la migration et l'apoptose. Le rôle suppresseur de tumeur de DOK1 a été démontré dans des modèles animaux. Les souris knock-out pour DOK1 présentent une forte susceptibilité de développer des leucémies, des tumeurs malignes hématologiques, des adénocarcinomes pulmonaires, ainsi que des sarcomes histiocytaires agressifs. En outre, nous avons rapporté précédemment que le gène DOK1 peut être muté et son expression réprimée dans différentes tumeurs malignes humaines, telles que les lignées cellulaires de lymphome de Burkitt (BL) et la leucémie lymphoïde chronique (LLC). Cependant, les mécanismes de dérégulation de DOK1 restent inconnus, notamment dans les processus de carcinogenèse induite ou non par des oncovirus. Dans ce projet de thèse, nous avons d'abord caractérisé le promoteur de DOK1 et le rôle du facteur de transcription E2F1 comme le principal régulateur de l'expression de DOK1. Nous avons démontré pour la première fois la contribution de DOK1 dans la réponse cellulaire au stress par son rôle suppresseur de prolifération cellulaire et promoteur d'apoptose. Nous avons trouvé que l'expression du gène DOK1 est réprimée dans une variété de cancers humains, y compris le cancer de la tête et du cou, les lymphomes de Burkitt et les cancers du poumon. Cette répression est due à l'hyperméthylation aberrante de DOK1. Nous avons donc étudié les événements épigénétiques, qui sont souvent altérés dans les cancers, et leurs implications dans la répression de DOK1 dans les lignes cellulaire cancéreuses de la tête et du cou. Nous nous sommes par la suite intéressés aux mécanismes de dérégulation de DOK1 par le virus d'Epstein Barr dans le cadre de sa propriété oncogénique dans les lymphocytes B humains ainsi que dans les lignes cancéreuses du lymphome de Burkitt. Nos résultats apportent de nouvelles informations sur les mécanismes de régulation de l'expression de DOK1 dans la carcinogenèse induite ou non par des oncovirus, ce qui pourrait le définir comme un biomarqueur potentiel de cancer et comme une cible intéressante pour des thérapies épigénétiques / The newly identified tumor suppressor DOK1 (downstream of tyrosine kinases1) inhibits cell proliferation, negatively regulates MAP kinase activity, opposes leukemogenesis, and promotes cell spreading, motility, and apoptosis. DOK1 also plays a role in the regulation of immune cell activation, including B cells. The tumor suppressor role of DOK1 was demonstrated in animal models. DOK1 knockout mice show a high susceptibility to develop leukemia, hematological malignancies as well as lung adenocarcinomas and aggressive histiocytic sarcoma. In addition, we previously reported that the DOK1 gene can be mutated and its expression is down-regulated in human malignancies such as Burkitt’s lymphoma cell lines (BL) and chronic lymphocytic leukemia (CLL). However, very little is known about the mechanisms underlying DOK1 gene regulation and silencing in viral- and non viral-related tumorigenesis. In the present project, we first characterized the DOK1 promoter. We have shown the role of E2F1 transcription factor as the major regulator of DOK1 expression and how DOK1 plays a role in DNA stress response though opposing cell proliferation and promoting apoptosis. We demonstrated that DOK1 gene expression is repressed in a variety of human cancers, including head and neck, Burkitt’s lymphoma and lung cancers, as a result of aberrant hypermethylation. We investigated the link between the epigenetic events and DOK1 silencing in non viral head and neck cancer cell lines, and by Epstein Barr virus in relation to its oncogenic activity in human B cells and neoplasia such as Burkitt’s lymphoma. These data provide novel insights into the regulation of DOK1 in viral and non viral-related carcinogenesis, and could define it as a potential cancer biomarker and an attractive target for epigeneticbased therapy