One of the current goals in cancer research is to discover and validate novel molecular targets that may be useful for diagnostic and therapeutic purposes in fighting this disease. The PRL phosphatases (PRL-1, PRL-2, and PRL-3) are low molecular weight protein tyrosine phosphatases with unknown biological function(s) that have gained attention from cancer researchers in the past couple of years, mainly due to reports that these phosphatases may play important roles in tumor progression and metastasis. Motivated by the particular urgent need for molecular targets in pancreatic cancer this work was undertaken to determine what role PRL proteins played in pancreatic cancer biology and to determine if targeting PRLs would be effective in treating this disease. In this dissertation, it was found that both PRL-1 and PRL-2, but not PRL-3 are upregulated in pancreatic adenocarcinomas, suggesting that some cancer cells are dependent upon their activity for continued proliferation and survival. To validate this hypothesis, siRNAs were used in cell-based assays to evaluate the biological consequences of PRL-1 and/or PRL-2 inhibition. It was found that perturbations in PRL phosphatase signaling result in reduced proliferation, migration and especially the ability to grow in soft agar. Oligonucleotide microarray analysis revealed that many Erk and/or Akt dependent stress and growth factor inducible genes were differentially regulated between pancreatic cancer cells treated with PRL-targeting siRNA and their non-targeting siRNA treated counterparts. Subsequently, PRL knockdown was found to alter serum induced as well as amino acid deprivation induced Akt and Erk phosphorylation in multiple pancreatic cancer cell lines, suggesting that PRLs function upstream of these key pathways. Interestingly, we show that PRL proteins in cell free assays exhibit higher activity on doubly phosphorylated phosphatidylinositol substrates than tyrosine-phosphorylated peptides, suggesting that the biological substrate(s) might include non-protein molecules. These data support the hypothesis that PRL-1 and PRL-2 might play important biological roles in pancreatic cancer cells and further studies should be undertaken to determine the usefulness of these phosphatases as potential molecular biomarkers and targets.

Les trois enzymes de la famille PRL (PRL-1,2,3), représentent un groupe de protéines tyrosine phosphatases intrigantes, impliquées dans un très grand nombre de maladies. Elles ont gagné beaucoup d'attention ces dernières années dans le contexte tumoral puisqu’elles sont très fréquemment associées à la formation de métastases, la prolifération cellulaire, l'invasion et la migration cellulaire. Bien que les propriétés oncogéniques des PRLs ne fassent plus de doute, ces phosphatases sont très peu caractérisées dans un contexte physiologique. Ainsi, aucun substrat biologique n’a été clairement identifié à ce jour et très peu de fonctions biologiques leurs sont associées, malgré un très haut degré de conservation inter-espèce suggérant un rôle critique de ces protéines dans les fonctions cellulaires. La caractérisation de la souris déficiente pour PRL-2 indique que cette phosphatase est probablement impliquée tout au long du développement. La phosphorylation des protéines est un phénomène important pour la régulation de l'angiogenèse en jouant des fonctions critiques et réversibles sur les voies de signalisation cellulaire. Cela nous a incités, dans le cadre de cette thèse, à étudier le rôle de PRL-2 dans la morphogenèse vasculaire en utilisant à la fois des modèles in vitro et un modèle de souris déficientes pour PRL-2. Dans le modèle murin de l'angiogenèse rétinienne, nous avons trouvé que la délétion de PRL-2 entraîne un retard dans la formation du plexus vasculaire avec une réduction de la vascularisation. De plus, on observe une angiogenèse et une ramification excessive au niveau du front vasculaire chez les souriceaux. En effet, le front de croissance des rétines de souris PRL-2 KO présente une densité vasculaire plus élevée associée à un bourgeonnement endothélial actif et multidirectionnel. Ces données présentent PRL-2 comme un composant potentiel du réseau de signalisation complexe qui orchestre la néo-angiogenèse. Nous avons ensuite examiné les conséquences de la perte de PRL-2 sur le comportement des cellules endothéliales in vitro. Nous avons observé des modifications de migration et d’invasion dans différents modèles. En particulier, l’absence de PRL-2 accroît le bourgeonnement de capillaires dans des tests d'angiogenèse in vitro, ce qui est en accord avec les données in vivo. En utilisant plusieurs siRNA, nous avons montré que la voie de signalisation de PRL-2 est liée à la signalisation VEGF/VEGFR. En effet, la stimulation des cellules endothéliales par le VEGF dépend de la présence ou de l'absence de PRL-2. En outre, une autre voie altérée par la régulation négative de PRL-2, est celle de Notch ainsi que celle du facteur de transcription Hey2. Ceci est cohérent avec les données in vivo où nous avons démontré un effet important sur la différenciation artérioveineuse. Ces données introduisent PRL-2 en tant que composant novateur du réseau de signalisation complexe qui orchestre l'angiogenèse développementale et, éventuellement, pathologique. / The three Phosphatase of Regenerative Liver (PRL-1, -2, -3) represent an intriguing group of protein tyrosine phosphatases that has been implicated in a number of diseases. They have gained much attention in the context of cancer. Indeed, they have been constantly associated with metastasis, cell proliferation, cell invasion and migration. To date, however, little is known about their physiological function and no biological substrates have been clearly identified. All three PRLs are highly conserved among mammals, underscoring the idea that they might have important roles in cellular functions. Characterization of the PRL-2 knockout mouse indicates that this phosphatase is likely involved throughout development. Protein phosphorylation is implicated in angiogenesis by playing critical and reversible functions in cell signaling pathways. This prompted us, in the frame of this thesis, to investigate the role of PRL-2 in vascular morphogenesis using both in vitro models and genetic loss-of-function mouse models. In the retinal angiogenesis mouse model, we found that PRL-2 deletion leads to delay in the formation of the retinal vascular plexus with a reduction of the advancing vasculature across the vitreal surface. Furthermore, excessive angiogenesis and branching at the leading edge in 6 day old pups is observed. Indeed, the growing front of PRL-2 KO mouse retinas showed a higher vascular density due to active, multidirectional hypersprouting of the vasculature. This data introduces PRL-2 as a potential component of the complex signaling network that orchestrates neo-angiogenesis. Based on these findings, we have examined whether the absence of PRL-2 can modify the behavior of endothelial cells in vitro. We showed an altered migration in various assays. In particular, sprouting in in vitro angiogenesis assays is altered, which is in agreement with the in vivo data. By using several siRNA, we showed that the signaling pathway of PRL-2 is dependent to VEGF/VEGFR signaling. Indeed, the stimulation of endothelial cells by VEGF is dependant of the presence or absence of PRL-2. Furthermore, other targets altered by PRL-2 downregulation, are Notch and the Hey2 transcription factor, which is consistent with in vivo data as we showed a strikingly effect on arteriovenous differentiation. Taken together, these data introduce PRL-2 as a novel component of the complex signaling network that orchestrates developmental and, possibly, pathological angiogenesis.

Indiana University-Purdue University Indianapolis (IUPUI) / Phosphatases of regenerating liver, consisting of PRL-1, PRL-2 and PRL-3, belong to a novel protein tyrosine phosphatases subfamily, whose overexpression promotes cell proliferation, migration and invasion and contributes to tumorigenesis and metastasis. However, although great efforts have been made to uncover the biological function of PRLs, limited knowledge is available on the underlying mechanism of PRLs’ actions, therapeutic value by targeting PRLs, as well as the physiological function of PRLs in vivo. To answer these questions, we first screened a phage display library and identified p115 RhoGAP as a novel PRL-1 binding partner. Mechanistically, we demonstrated that PRL-1 activates RhoA and ERK1/2 by decreasing the association between active RhoA with GAP domain of p115 RhoGAP, and displacing MEKK1 from the SH3 domain of p115 RhoGAP, respectively, leading to enhanced cell proliferation and migration. Secondly, structure-based virtual screening was employed to discover small molecule inhibitors blocking PRL-1 trimer formation which has been suggested to play an important role for PRL-1 mediated oncogenesis. We identified Cmpd-43 as a novel PRL-1 trimer disruptor. Structural study demonstrated the binding mode of PRL-1 with the trimer disruptor. Most importantly, cellular data revealed that Cmpd-43 inhibited PRL-1 induced cell proliferation and migration in breast cancer cell line MDA-MB-231 and lung cancer cell line H1299. Finally, in order to investigate the physiological function of PRLs, we generated mouse knockout models for Prl-1, Prl-2 and Prl-3. Although mice deficient for Prl-1 and Prl-3 were normally developed, Prl-2-null mice displayed growth retardation, impaired male reproductive ability and insufficient hematopoiesis. To further investigate the in vivo function of Prl-1, we generated Prl-1-/-/Prl-2+/- and Prl-1+/-/Prl-2-/- mice. Similar to Prl-2 deficient male mice, Prl-1-/-/Prl-2+/- males also have impaired spermatogenesis and reproductivity. More strikingly, Prl-1+/-/Prl-2-/- mice are completely infertile, suggesting that, in addition to PRL-2, PRL-1 also plays an important role in maintaining normal testis function. In summary, these studies demonstrated for the first time that PRL-1 activates ERK1/2 and RhoA through the novel interaction with p115 RhoGAP, targeting PRL-1 trimer interface is a novel anti-cancer therapeutic treatment and both PRL-1 and PRL-2 contribute to spermatogenesis and male mice reproductivity.