Connective Tissue Growth Factor (CTGF) is a matricellular protein that has been shown to mediate cell adhesion, and as a consequence, it regulates cell proliferation, migration, differentiation and gene transcription. Although previous in vivo and in vitro studies supported the anabolic role of CTGF in skeletogenesis, to date mechanisms of this effect remain unknown. So far, no specific receptor has been identified for CTGF, although previous studies have shown that integrins can serve as functional signaling receptors for CTGF. The CTGF-integrin interaction initiates intracellular signaling cascades that ultimately regulate cell cytoskeleton reorganization, gene transcription and cell function. To study the effect of CTGF on osteoblasts, we first conducted adhesion assays using the MC3T3-E1 osteoblastic cell line. We confirmed that osteoblasts adhere to rCTGF in a concentration-dependent manner and we showed this adhesion has characteristics of integrin mediated adhesions. Next, we used an array of blocking antibodies directed against the individual alpha and beta; integrin subunits that are known to be expressed in osteoblasts. Significant decreases in cell adhesion were observed upon treatment with anti-alpha-v or anti-beta1 blocking antibodies. Subsequent coimmunoprecipitation analyses demonstrated that CTGF interacts with alpha-v and beta1 integrins in osteoblasts. Furthermore, we showed that the specificity of this CTGF-integrin interaction occurs in the C-terminal domain (fourth module) of CTGF. The immunefluorescence staining of cells cultured on substrates of rCTGF, fibronectin (positive control) or BSA (negative control) demonstrated that osteoblast adhesion to rCTGF results in actin cytoskeleton reorganization, focal adhesion formation, enhanced cell spreading and Rac activation. These series of events are necessary for proper cell-matrix interaction and integrins' downstream signaling initiation. Next, through alkaline phosphatase (ALP) staining and activity assays, as well as Alizarin red staining, we demonstrated that osteoblast attachment to CTGF matrix enhances cell maturation, bone nodule formation and matrix mineralization. To investigate whether the effect of CTGF on osteoblast differentiation involves activation of specific signaling molecules, we performed Western blot and chromatin immunoprecipitation (ChIP) assays. Osteoblasts cultured on rCTGF expressed higher levels of both total and phosphorylated forms of focal adhesion kinase (FAK) and extracellular signal-regulated kinase (ERK) compared to the cells cultured on BSA. In addition, these osteoblasts showed an increase in runt-related transcription factor 2 (Runx2) binding to the osteocalcin gene promoter compared to the negative control. These experiments confirmed CTGF's effect on enhancing osteoblast differentiation through regulation of important signaling molecules. In another series of experiments, we used primary osteoblasts isolated from CTGF KO mice, their WT littermates, or WT cells infected to overexpress (OE) CTGF to study the effect of different levels of endogenous CTGF on osteoblast cytoskeleton reorganization and motility. Our assays showed enhanced cell adhesion, spreading and Rac expression in CTGF OE osteoblasts, while in CTGF KO osteoblasts, cell adhesion, spreading and Rac expression were significantly decreased. In contrast, CTGF OE osteoblasts that showed high adhesion and spreading, exhibited diminished cell motility and low levels of RhoA expression, while KO cells migrated quickly and expressed high levels of RhoA. Together, these experiments establish CTGF as an adhesion protein for osteoblasts; they demonstrate that the alpha-v beta1 integrin is a functional signaling receptor for CTGF; they confirm that osteoblast differentiation is enhanced when cultured on CTGF matrix through activation of regulatory signaling molecules; and finally, these experiments establish a role for CTGF in the regulation of small RhoGTPases expression, which in turn implies a significant role for CTGF in cell cytoskeleton reorganization and motility. / Cell Biology

No description available.

No description available.

Certains cancers, comme celui du sein et de la prostate forment préférentiellement des métastases au niveau des os et sont dits ostéotropiques. Les SIBLINGs sont des protéines non-collagéniques de la matrice osseuse notamment impliquées dans la régulation de la minéralisation de cette matrice. Nos travaux antérieurs et présents montrent que certaines protéines de la famille des SIBLINGs sont exprimées de manière ectopique au niveau des cellules de cancers considérés comme ostéotropiques. Notre travail a tout dabord permis de mettre en évidence pour la première fois lexpression des SIBLINGs : (a) BSP dans les lésions (pré)néoplasiques du col de lutérus ; (b) DMP1 dans les cancers pulmonaires humains et (c) DSPP dans les cancers prostatiques et sa corrélation avec lagressivité des tumeurs. Dautre part, nous avons contribué significativement à la compréhension des mécanismes responsables de lactivation et de la régulation de la transcription du gène de la BSP dans les cellules cancéreuses et ostéoblastiques. Dans une première étude, nous avons réalisé des constructions comprenant différentes délétions de la région promotrice humaine du gène de la BSP que nous avons ensuite transfectées de manière transitoire dans les cellules cancéreuses mammaires (MDA-MB-231 et MCF-7) et dostéosarcome humain (Saos-2). La comparaison des profils de transfection obtenus pour ces trois lignées nous a permis de déterminer une région du promoteur importante pour la régulation transcriptionnelle de la BSP et suffisante pour expliquer lactivité du promoteur dans les cellules cancéreuses mammaires. Suite à la recherche des sites cis potentiellement actifs, nous avons identifié un élément CRE (cAMP Responsive Element) comme étant principalement responsable de lexpression de la BSP dans les cellules cancéreuses mammaire tandis quun élément FRE (Fibroblast growth factor Response Element) est par contre un acteur incontournable de cette même expression dans les cellules de la lignée osseuse. Des expériences de retard sur gel nous ont ensuite permis didentifier les facteurs CREB-1, Jun-D et Fra-2 comme étant associés à lélément CRE aussi bien dans les lignées cancéreuses mammaires que dans les Saos-2. Nous avons montré que dans nos modèles cellulaires : (a) CREB-1 agit de manière constitutive et positive sur la régulation du gène de la BSP et que (b) Jun D et Fra-2, sont des facteurs de transcription importants pour la régulation transcriptionnelle du gène de la BSP dans les cellules cancéreuses mammaires. Dans une seconde étude basée sur lobservation que lexpression endogène de BSP est fortement augmentée, tant au niveau protéique que de lARNm, dans les cellules Saos-2 confluentes par rapport aux cellules préconfluentes, nous avons voulu déterminer le rôle potentiel du facteur de transcription Runx2 dans la régulation transcriptionnelle du gène de la BSP au niveau de ce modèle de différenciation ostéoblastique. Le facteur de transcription Runx2 est en effet connu comme étant essentiel au développement ostéoblastique, à la différenciation et à la formation osseuse. Ce facteur régule positivement ou négativement lexpression des gènes ostéoblastiques en interagissant avec divers co-facteurs transcriptionnels. Nous avons démontré quun site de liaison pour le facteur Runx2 présent au niveau du promoteur de la BSP lie ce facteur dans les cellules préconfluentes et à un degré significativement moindre dans les cellules post-confluentes suggérant que ce facteur jouerait un rôle répresseur de lexpression du gène de la BSP. Etant donné quil a été démontré que la déacétylase dhistones HDAC3 est capable dinteragir avec Runx2 pour réprimer le promoteur dautre gène osseux, nous avons vérifié lhypothèse selon laquelle HDAC3 et Runx2 seraient présents au niveau du promoteur de la BSP pour réprimer lexpression de la BSP dans les cellules préconfluentes. Des expériences dimmunoprécipitation de la chromatine nous ont permis de mettre en évidence la présence concomitante de Runx2 et de HDAC3 au niveau du promoteur de la BSP dans les cellules Saos-2 préconfluentes et non dans les cellules post-confluentes. Finalement, la suppression de Runx2, dHDAC3 et des deux simultanément par interférence à lARN dans les cellules Saos-2 préconfluentes induit une augmentation de lexpression de la BSP dans ces cellules. Ces travaux ont donc permis de démontrer pour la première fois que la levée de la répression exercée par laction conjointe de Runx2 et de HDAC3 au promoteur de la BSP est nécessaire pour permettre lexpression de cette protéine au cours de la différenciation ostéoblastique. Lensemble de nos résultats nous a permis de dresser de nouveaux modèles de lactivation et de la régulation du gène de la BSP dans les cellules ostéoblastiques au cours de leur différenciation mais aussi dans les cellules cancéreuses.

Tissue engineering and ex vivo gene therapy can be used synergically as tool to regenerate bone, which overcome the problems of currently available bone replacements. Recently, a new source of mesenchymal stromal cells (MSCs) has been found in the umbilical cord; human umbilical cord perivascular cells (HUCPVCs) provide an alternative to bone marrow derived MSCs and due to their easy harvest, fast expansion, and non-immunogeneic and immunomodulatory phenotype we hypothesized that HUCPVCs are a putative candidate cell source for osteogenic ex vivo gene therapy. This work proposes the generation of cocktails of genetically modified HUCPVCs and their cryopreservation as an “off the shelf” therapeutic. This approach involves the engineering of osteogenic cell populations, by genetically modifying HUCPVCs using recombinant adenoviruses to deliver four fundamental genes for bone formation: bone morphogenetic protein 2 (BMP-2), runt-related transcription factor 2 (Runx2), Osterix (OSX/SP7) transcription factor and vascular endothelial growth factor (VEGF). Our results show that HUCPVCs can be efficiently modified by adenoviruses and can be cryopreserved without affecting the production efficiency and bioactivity of proteins of interest produced by the cells. Moreover, overexpression of BMP2, Runx2 and SP7 enhances ALP activity levels in HUCPVCs and upregulates ALP, OPN, COL1A1 and OCN gene expression; data that provides the first evidence of the effects of combinational expression of BMP2, Runx2 and SP7. Furthermore, we report for the first time the genetic modification of human BMSCs to express SP7 and Runx2, which enhances their ALP activity and matrix mineralization capacity.

Tissue engineering and ex vivo gene therapy can be used synergically as tool to regenerate bone, which overcome the problems of currently available bone replacements. Recently, a new source of mesenchymal stromal cells (MSCs) has been found in the umbilical cord; human umbilical cord perivascular cells (HUCPVCs) provide an alternative to bone marrow derived MSCs and due to their easy harvest, fast expansion, and non-immunogeneic and immunomodulatory phenotype we hypothesized that HUCPVCs are a putative candidate cell source for osteogenic ex vivo gene therapy. This work proposes the generation of cocktails of genetically modified HUCPVCs and their cryopreservation as an “off the shelf” therapeutic. This approach involves the engineering of osteogenic cell populations, by genetically modifying HUCPVCs using recombinant adenoviruses to deliver four fundamental genes for bone formation: bone morphogenetic protein 2 (BMP-2), runt-related transcription factor 2 (Runx2), Osterix (OSX/SP7) transcription factor and vascular endothelial growth factor (VEGF). Our results show that HUCPVCs can be efficiently modified by adenoviruses and can be cryopreserved without affecting the production efficiency and bioactivity of proteins of interest produced by the cells. Moreover, overexpression of BMP2, Runx2 and SP7 enhances ALP activity levels in HUCPVCs and upregulates ALP, OPN, COL1A1 and OCN gene expression; data that provides the first evidence of the effects of combinational expression of BMP2, Runx2 and SP7. Furthermore, we report for the first time the genetic modification of human BMSCs to express SP7 and Runx2, which enhances their ALP activity and matrix mineralization capacity.

Thyroid Tumorigenesis is typically a well understood process, with well delineated oncogenic factors. Follicular and papillary thyroid cancers are typically survivable, with 5-year survival rates being >95% for Stage I-III of both cancer types. Anaplastic thyroid cancer, in contrast, lacks this prognosis, and is the most lethal of all endocrine-related cancers. The median survival time after a diagnosis is generally between 6-8 months, with a 5-year survival rate of <10%. Current treatment for anaplastic thyroid cancers routinely meet roadblocks, as resistance is quickly developed. Even non-discriminatory kinase inactivators, such as sorafenib, which are generally considered a drug of last resort, are unable to effect survival rates. As such, there is a clear need for further investigation of the causes of anaplastic thyroid cancer mechanisms. Previous work in the Carr lab revealed a novel regulatory pathway of an oncogene that is associated with several other endocrine-related cancers, as well as other non-endocrine-related cancers. Specifically, the Runt-related transcription factor 2 (Runx2) was found to be suppressed via direct binding of the thyroid hormone receptor beta 1 isoform (TRß1) to its proximal promotor. Runx2 was previously shown to be associated with increasing malignancy, with Runx2 occurring at low-levels in indolent cell lines, whilst occurring at high-levels in more malignant cell lines. TRß1, conversely, exhibited the opposite relationship. Endogenous levels of TRß1 were found to be high in indolent cell lines and were depleted in malignant cell lines. These findings were further confirmed via tissue microarrays. Restoration of TRß1 in malignant cell lines diminished Runx2 mRNA and protein levels, which was corroborated by evidence from electrophoretic mobility-shift assays, and chromatin immunoprecipitations that TRß1 was able to directly bind Runx2 promotor 1. Current studies have investigated the nuclear protein profile that associates with TRß1 to alter Runx2 transcription. Through EMSA-to-Mass Spectrometry methodologies, as well as novel DNA pulldown techniques, binding partners have been elucidated. Findings have also been confirmed via classical immunoprecipitations. Specifically, our findings show that TRß1 complexes with the brahma-related gene 1 (BRG1) protein, the nuclear co-repressor (NCOR), and BRG1-associated protein 60 (BAF60). BRG1 functions by preferentially recruiting histone deacetylases (HDAC), with BRG1 and the HDAC’s acting to alter chromatin, and thus transcription. Future studies aim at examining whether other proteins complex with TRß1 to alter Runx2 transcription, and whether these complexes are altered in aggressive cell lines.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22646号 / 医博第4629号 / 新制||医||1044(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 篠原 隆司, 教授 松田 秀一, 教授 安達 泰治 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

No description available.

Transcriptional regulation of CD8+ T cell differentiation during acute and chronic viral infections is an intricate web made up of many of transcription factors. While several transcription factors have been elucidated in this process, there are still many more that remain elusive. In this work, we look into the role of two transcription factors, IRF4 and Runx2, and their role in CD8+ T cell terminal effector cells and memory precursor cells during acute LCMV-Armstrong infection. We found that IRF4 expression was regulated by TCR signal strength during infection, and that IRF4 expression levels directly correlated with the magnitude of the effector cell response. IRF4 was also shown to regulate T-bet and Eomes, two transcription factors critical for CD8+ T cell differentiation into effector and memory cells. From these results, we were interested in the potential role of IRF4 during chronic LCMV-clone 13 infection, where ratios of T-bet and Eomes are critical for viral clearance. We found that haplodeficiency of IRF4 in the T cell compartment lead to an increase in the ratio of Eomes to T-bet in T cells, which in turn affected the proportion of Eomeshi versus T-bethi cells and resulted in a loss in ability to clear viral infection. Irf4+/-Eomes+/- compound heterozygous mice were generated to test if decreasing Eomes expression would rescue the Irf4+/- phenotype. Irf4+/-Eomes+/- mice were phenotypically similar to WT mice in terms of Eomes to T-bet ratios, and were able to clear viral infection, demonstrating a critical role of IRF4 in regulating T-bet and Eomes during chronic viral infection. Next we looked into the role of Runx2 during acute LCMV-Armstrong infection and found that Runx2-deficient pathogen-specific CD8+ T cells had a defect in the total number of memory precursor cells compared to WT controls. We further showed that Runx2 was inversely correlated with TCR signal strength, and that Runx2 expression was repressed by IRF4. From these work, we have introduced two more transcription factors that are critical for CD8+ T cells differentiation during acute and chronic viral infection. Given the sheer number of transcription factors known to regulate these processes, having a full understanding of the transcriptional network will allow us to find the best targets for therapeutic intervention for treatments ranging from vaccine development and autoimmunity to cancer immunotherapy and treatment of chronic viral infections.