Glycosaminoglycans (GAGs) such as heparin, heparan sulfate (HS), chondroitin sulfate (CS) and dermatan sulfate (DS) are sulfated polysaccharides that exist on animal cell surfaces and in the extracellular matrix. GAGs are important in providing structural and hydrating support and interaction points for proteins of varied functions, for example growth factors and homeostasis regulatory proteins. Hepatocyte Growth Factor (HGF) is a protein growth factor that regulates cell growth, survival, proliferation, chemotaxis, cell morphology, tissue regeneration and angiogenesis. It is involved in embryogenesis, wound healing and many cancers. In this project, the interactions between the GAG binding N and NK -domains of HGF (HGF-N and HGF-NK) and different types of GAGs are characterised with biophysical techniques. GAG oligosaccharides were produced by enzymatic digestion and purified by preparative gel filtration and ion exchange chromatography. Different constructs of HGF were cloned from human cDNA, expressed with the Pichia pastoris expression system, purified to homogeneity and characterised by mass spectrometry and nuclear magnetic resonance (NMR). The dissociation constants between the different HGF protein constructs, different heparin oligosaccharide lengths and the drug Fondaparinux were shown by isothermal calorimetry (ITC) to vary between 0.35 and 9.26 μM. It was found that the entropy contribution was favourable for short oligosaccharides and disfavourable for long oligosaccharides and that the enthalpy contribution was less important for shorter oligosaccharides than for longer oligosaccharides. NMR titrations of CS, DS, heparin, Fondaparinux and sulfated maltose into 15N labelled protein samples showed that all ligands bind to the same HGF-N binding site, but different binding modes exists. The binding site consists of three regions, with the α2-helix and L2 loops playing key roles (residues 70-84). All GAGs also utilise the N-terminal residues 32-42, whereas long heparin oligosaccharides can also utilise a binding region formed mainly by the β2-strand (residues 59-64, 66, 95, 96). The GAG binding mode changes if HGF-N has an N-terminal truncation and the β2- strand residues become more important, emphasising the role of the N-terminal residues in the HGF-GAG interaction. Spin-labelled fully sulfated heparin-derived hexasaccharide was used to determine its binding direction on the HGF-N surface. Affinity chromatography confirmed the importance of the N-terminal residues and that HGF binds to all investigated GAGs. The oligomeric states of HGF-N and HGF-NK were investigated by AUC, gel filtration and ITC. The results suggest that the proteins oligomerise like beads on a string for long oligosaccharides. An HGF-N self-associating dimer with a slow on/off rate was characterised by affinity chromatography, gel filtration and NMR.

Bone formation and remodeling require continuous generation of osteoprogenitors from bone marrow stromal cells (MSC), which are regulated by local growth factors and hormones with putative roles in mesenchymal proliferation and differentiation. Hepatocyte growth factor (HGF) and its receptor c-Met are widely expressed in MSC and are thought to play a key role in the interactions between cells. 1,25-dihydroxyvitamin D (1,25OHD) is the most active metabolite of vitamin D. 1,25OHD binds to its nuclear/membrane vitamin D receptor (VDR) and generates appropriate biological responses. The purpose of this study was to investigate the regulation of proliferation and differentiation by HGF in human bone marrow-derived stromal cells (hMSC). We examined the impact of HGF on hMSC cell-cycle regulation and the combination effects of HGF and 1,25OHD on hMSC osteogenic differentiation to enhance our knowledge of hMSC regulation. hMSC isolated from bone marrow were plated and grown in DMEM supplemented with 3% FBS incubated at 37C with 5% CO2 in air. HGF treatment of hMSCs reduced the rate of cell proliferation and this result was not due to apoptosis or cell senescence. Real-time RT-PCR and Western blot analysis showed increased gene and protein expression of the cell-cycle inhibitors p53, p21, and p27 after HGF treatment. These results appear to be specific because HGF did not significantly alter the gene expression level of other cell-cycle mediators such as RB, cyclin D1, CDK2, CDK4, or CDK6. Transfection of siRNA specific for cMet, the HGF receptor, eliminated the HGF anti-proliferation effect. cMet siRNA also eliminated the increase in p53, p21, and p27, further supporting a role for these cell-cycle inhibitors in HGF¡¯s regulation of hMSC. These results suggest that treatment of hMSC with HGF slows cell proliferation by increasing the expression of p53, p21, and p27. The reduced rate of cell proliferation did not appear to be due to cell differentiation, because treatment of hMSC with HGF alone did not induce cell differentiation. However, HGF in combination with a known osteogenic differentiation activator, 1,25OHD, significantly increased cell maturation/differentiation compared to 1,25D alone, as indicated by an increase in osteocalcin mRNA (a marker for osteogenic differentiation). Whereas HGF had no effect on 1,25OHD synthesis per se, HGF did induce 1,26OHD receptor (VDR) gene expression. HGF up-regulated the expression of the p63 gene, a member of the p53 family. Knocking down the p63 gene reduced the HGF effect on VDR expression and eliminated the HGF-induced up-regulation of the osteogenic differentiation markers osteopontin (OPN) and bone sialoprotein (BSP). Moreover, the ChIP assay shows that p63 was able to bind to the VDR promoter, possibly explaining the mechanism of p63-mediated VDR up-regulation. These results indicate that HGF can also induce hMSC osteogenic differentiation when combined with 1,25OHD by up-regulating 1,25OHD receptor VDR expression.

No description available.

No description available.

Le cancer de la prostate (PCa) est le deuxième cancer le plus courant chez les hommes au niveau mondial. Le suppresseur de la signalisation des cytokines 1 (SOCS1) est considéré comme un suppresseur de tumeur en raison de la fréquente répression épigénétique de ce gène dans de nombreux cancers. Il a été reporté que SOCS1 inhibait l’activation de STAT3 induite par l’IL-6, ainsi que les cyclines et les kinases dépendantes des cyclines dans les cellules malignes de la prostate. D’autre part, il a été montré que SOCS1 n’était pas essentiel lors du contrôle de la signalisation de l’IL-6 dans les hépatocytes dépourvus de cette protéine, cependant elle est essentielle pour atténuer la signalisation du facteur de croissance des hépatocytes (HGF) via son récepteur MET. MET est un récepteur de tyrosine kinases qui est surexprimé dans le PCa agressif et métastatique. Notre hypothèse de recherche propose que la répression de SOCS1 par méthylation du promoteur et la dérégulation de l’expression de MET et de sa signalisation, sont des mécanismes pathogéniques liés au développement et à la progression du PCa. Nous avons généré des lignées de cellules PC3 et DU145 stables exprimant SOCS1. Les cellules ont été stimulées avec HGF et l’activation des voies de signalisation a été évaluée par immunobuvardage. Des essais in vitro de migration, de prolifération et d’invasion ont été effectués en présence de HGF. Des gènes de transition épithélio-mésenchymateuse ont été évalués par PCR quantitatif en présence ou non du facteur de croissance. Les cellules du PCa transfectées ou pas avec SOCS1 ont été inoculées dans des souris NOD SCID gamma de façon sous-cutanée ou orthoptique afin d’évaluer respectivement la croissance tumorale et la formation de métastases. Les tumeurs reséquées ont été analysées histologiquement et biochimiquement. Nos résultats montrent que SOCS1 atténue l'activation de MET induite par HGF et la phosphorylation d’ERK dans les cellules PC3, ainsi que la phosphorylation d’ERK et d’AKT dans les cellules DU145. SOCS1 inhibe également la prolifération cellulaire induite par HGF, ainsi que la migration et l’invasion in vitro. De plus, SOCS1 réduit l’expression des gènes de transition épithélio-mésenchymateuse impliqués dans la dégradation des composants de la matrice extracellulaire dans les cellules DU145 mais pas dans les cellules PC3. La surexpression de SOCS1 a stimulé l’augmentation de déposition de collagène, in vivo. Les tumeurs formées par les cellules exprimant SOCS1 étaient de taille significativement plus petites avec une réduction de la prolifération comparé aux tumeurs provenant des cellules contrôles. En outre, SOCS1 a inhibé la formation de métastases à distance dans un modèle orthotopique. En conclusion, nous suggérons que SOCS1 est un suppresseur de tumeur indispensable de la prostate, et qu’au moins une partie de sa fonction a lieu via la régulation négative de la signalisation du récepteur MET. / Abstract : Prostate cancer (PCa) is the second most common cancer among men worldwide. Suppressor of cytokine signaling 1 (SOCS1) is considered a tumor suppressor due to frequent epigenetic repression of the SOCS1 gene in several human malignancies. Inactivation of SOCS1 also occurs in PCa by gene methylation and micro-RNA-mediated repression. SOCS1 has been reported to inhibit IL-6-induced STAT3 activation and down-regulates cyclins and cyclin-dependent kinases in PCa cells. It has been shown that SOCS1 is not required to control IL-6 signaling in SOCS1-deficient hepatocytes, but is essential to attenuate hepatocyte growth factor (HGF) signaling via its receptor MET. This protein is a receptor tyrosine kinase (RTK), overexpressed in aggressive and metastatic PCa. Thus we hypothesized that the repression of SOCS1 via promoter methylation and deregulated MET expression and signaling are inter-related pathogenic mechanisms in PCa development and progression. We generated stable SOCS1-expressing PCa cell lines (PC3 and DU145) using lentiviral transduction followed by clonal selection via limiting dilution. Cells were stimulated with HGF and downstream signaling events were assessed by Western blot. Proliferation, migration and invasion assays were also conducted in the presence of HGF in vitro. Epithelial mesenchymal transition genes were evaluated by qPCR in the presence or absence of the growth factor. The PCa cells transfected with SOCS1 and non-transfected controls were inoculated into NOD SCID gamma mice as xenografts or as orthotopic tumors to assess tumor growth and metastasis formation, respectively. Resected tumors were further analyzed histologically and biochemically. Our results showed that SOCS1 attenuates HGF-induced MET activation and ERK phosphorylation in PC3 and DU145 PCa cell lines. SOCS1 inhibited HGF induced cell proliferation, migration and invasion in vitro. Additionally, SOCS1 decreased epithelial mesenchymal transition genes involved in the degradation of extracellular matrix components in DU145 cells but not in PC3. In vivo, SOCS1 overexpression leads to an increase of collagen deposition. Tumors formed by SOCS1 expressing cells were significantly smaller in size with reduced cell proliferation compared to tumors arising from control cells. Furthermore, SOCS1 inhibited distant metastasis formation in the orthotopic model. Overall our results suggest that SOCS1 has a tumor suppressor role in PCa evolution and part of this function is mediated by the negative regulation of MET receptor signalling and down-regulation of genes supporting migration and invasion processes such as matrix metalloproteinases.