The Role Of CTGF In BMP9-induced Osteogenesis And In Tumorigenesis

The overall objective of this research is to study the effects of connective tissue growth factor (CTGF, also named as CCN2) on BMP9- induced osteogenesis and on osteosarcomas.The CCN proteins contain six members, namely CCN1 to CCN6, which are small secreted cysteine-rich proteins. The CCN proteins are modular proteins, containing up to four functional domains. Many of the CCN members are induced by growth factors, cytokines, or cellular stress. The CCNs show a wide and highly variable expression pattern in adult and in embryonic tissues. The CCN proteins can integrate and modulate the signals of integrins, BMPs, VEGF, Wnts, and Notch. The involvement of integrins in mediating CCN signaling may provide diverse context-dependent responses in distinct cell types.In this project, the most important CCN family members, CCN1, CCN2, CCN3 and CCN4, were evaluated in BMP9-induced osteogenesis and osteosarcomas. Of the two aspects, CCN2 (CTGF) showed significant effects both in the inhibition of the BMP9-induced bone formation and in the proliferation and migration of osteosarcoma cell line 143B. Thus, CTGF was chosen for more extensive study. To simplify the depiction, this study was divided into two parts.Part 1: The effect of CTGF on BMP9-induced osteogenesis.In Chapter 1, we isolated MEFs from mouse embryos, 12.5 to 13.5 days postcoitum. We then established immortalized MEFs (iMEFs) through the immortalization of these MEFs by integration of SV40 T gene mediated by retrovirus vector. The establishment of iMEFs greatly facilitated this study both on BMP9-induced osteogenesis and on osteosarcoma cell lines.In Chapter 2, we evaluated the effects of four CCN proteins, CCN1, 2, 3 and 4, on the BMP9-induced bone formation. Subconfluent iMEFs were co-infected with AdBMP9 and AdR-CCN1, Ad-CCN2, Ad-CCN3, Ad-CCN4, or, AdGFP adenoviruses and osteogeinc differentiation early marker alkaline phosphatase (ALP) activity was measured by quantitative colorimetric analysis at day 7. Results showed that CCN2 (CTGF) had the strongest inhibitory effect on BMP9-induced bone formation. These results promoted us to focus on CTGF for further study.Since CTGF is a mosaic protein that contains four modules, to further elucidate which module or modules contribute to this inhibitory function, terminal and internal deletions of CTGF were made in Chapter 3. Using molecular biology techniques, we did a series of cloning and finally generated six adenoviruses which could express CTGF full length, CTGFCD1 (no module IV), CTGFCD2 (no module III and IV), CTGFID1 (no module II), CTGFID2 (no module III), CTGFID3 (no module II and III). The presence of CTGF full length or CTGF deletions in these recombinated adenoviruses was confirmed by PCR, and the expression of these CTGF constructs was confirmed by Western blot with anti-Flag antibody, due to the fusion 3×Flag tag in these constructs. These CTGF full length and deletion mutant adenoviruses provided useful tools for dissecting the functional domains of CTGF in bone formation.In Chapter 4, we evaluated the effects of CTGF full length and its deletion mutants on the BMP9-induced bone formation in vitro. Subconfluent iMEFs were co-infected with AdBMP9 and AdR-CTGF, CTGFCD1, CTGFCD2, CTGFID1, CTGFID2, CTGFID3, or AdRFP adenoviruses and osteogeinc differentiation early marker alkaline phosphatase (ALP) activity was measured by ALP (histochemical) staining at day 7. Results showed that CTGF full length, CTGFCD1 and CTGFCD2 could inhibit BMP9-induced bone formation, while CTGFID1, CTGFID2, and CTGFID3 had a much more significant inhibitory effect.In Chapter 5, we evaluated the effects of CTGF full length and its deletion mutants on the BMP9-induced bone formation in vivo. BMP9+RFP or BMP9+CTGF mutants expressing iMEFs were implanted subcutaneously in nude mice. Ectopic osseous masses were retrieved at 6 weeks. Micro-CT, H & E staining, and Masson's Trichrome staining were employed to inspect the general and detailed information of the retrieved bone masses. Results showed that compared with BMP9+RFP control, BMP+CTGFCD2 and BMP9+CTGFCD1 yielded much larger bone masses, while the bone masses of BMP+CTGFID1/2/3 are much smaller. BMP9+CTGF full length didn't change the bone mass volume. Histological studies showed that CTGF deletions affect the BMP9-induced bone formation by the promotion or inhibition of iMEFs proliferation. However, CTGF and its deletions didn't affect the BMP9-induced iMEFs differentiation. These results suggest that CTGF module IV inhibited the BMP9-induced bone formation, while module I may promote this effect.Part 2: The effect of CTGF on osteosarcoma. In Chapter 1, using crystal violet staining and wound healing assay, we studied the effects of CCN1, 2, 3, and 4 on osteosarcoma cell line 143B and MG63. Results showed that CCN1 and CCN2 promoted the proliferation of 143B cells and promoted the migration of MG63 cells, while CCN3 and CCN4 had no effect or only had slightly inhibitory effect on the proliferation and migration of these osteosarcoma cell lines. Considering we had focused on CCN2 (CTGF) for detailed study in Part One, we also chose CCN2 (CTGF) for further study in Part Two.In Chapter 2, we generated two osteosarcoma cell lines, 143B-Luc and MG63-Luc by retrovirus-mediated firefly luciferase gene integration. The constitutively express of luciferase of the two cell lines were confirmed by luciferase assay in vitro. The luciferase activities of 143B-Luc and MG63-Luc cells were more than 1,000,000, while the values of the controls, 143B and MG63 cells were lower than 100. The establishment of 143B-Luc and MG63-Luc provided a useful tool for our in vivo study of CTGF.In Chapter 3, we tested the effect of CTGF on 143B-Luc cells in vivo by direct CTGF adenovirus infection. 143B-Luc cells were infected with Ad-CTGF and Ad-GFP for 18 h, and injected subcutaneously into the back of nude mice. Injected mice were tracked by Xenogen imaging, and tumors were harvested at day 32. Results showed that the volumes of tumors of CTGF group were much larger than those of the GFP group, suggested that CTGF could promote the growth of osteosarcoma cells.In chapter 4, we tested the effect of CTGF on MG63-Luc cells in vivo, using iMEFs as CTGF gene delivery vehicle. iMEFs were infected with either Ad-CTGF or Ad-GFP for 18 h. Infected iMEFs were mixed with MG63-Luc (1:1 ratio), to a total number of 2×106, and injected subcutaneously into the back of nude mice. MG63-Luc cells, without mix with iMEFs, were injected as control. Injected mice were tracked by Xenogen imaging, and tumors were harvested at day 25. Results showed that the volumes of tumors of iMEFs+CTGF group were much larger than those of the iMEFs+GFP group. There is no significant difference between the iMEFs+GFP group and the MG63-Luc only group. It suggested CTGF promoted osteosarcoma cell growth, and iMEFs was a good gene delivery vector. In summary, our studies demonstrated that Module IV could exert an inhibitory effect on the BMP9-induced osteogenesis in vivo, while Module I could promote the bone formation. Our study also demonstrated that CTGF could promote the proliferation of osteosarcoma cell line both in vitro and in vivo. In addition, iMEFs expressing CTGF dramatically increased tumor volume in vivo, indicated that CTGF function was closely related to the context or microenviroment.Ultimately, our results could offer the reference for the clinical use of BMP9 combined with CTGF deletions in bone formation, and the use of CTGF as a therapeutic target for osteosarcomas. Our results also enhanced the theory of stem cell differentiation v.s, proliferation.