Connective-Tissue Growth Factor (CTGF/CCN2) Induces Astrogenesis and Fibronectin Expression of Embryonic Neural Cells In Vitro

Abstract

Connective-tissue growth factor (CTGF) is a modular secreted protein implicated in multiple cellular events such as chondrogenesis, skeletogenesis, angiogenesis and wound healing. CTGF contains four different structural modules. This modular organization is characteristic of members of the CCN family. The acronym was derived from the first three members discovered, cysteine-rich 61 (CYR61), CTGF and nephroblastoma overexpressed (NOV). CTGF is implicated as a mediator of important cell processes such as adhesion, migration, proliferation and differentiation. Extensive data have shown that CTGF interacts particularly with the TGFβ, WNT and MAPK signaling pathways. The capacity of CTGF to interact with different growth factors lends it an important role during early and late development, especially in the anterior region of the embryo. ctgf knockout mice have several cranio-facial defects, and the skeletal system is also greatly affected due to an impairment of the vascular-system development during chondrogenesis. This study, for the first time, indicated that CTGF is a potent inductor of gliogenesis during development. Our results showed that in vitro addition of recombinant CTGF protein to an embryonic mouse neural precursor cell culture increased the number of GFAP- and GFAP/Nestin-positive cells. Surprisingly, CTGF also increased the number of Sox2-positive cells. Moreover, this induction seemed not to involve cell proliferation. In addition, exogenous CTGF activated p44/42 but not p38 or JNK MAPK signaling, and increased the expression and deposition of the fibronectin extracellular matrix protein. Finally, CTGF was also able to induce GFAP as well as Nestin expression in a human malignant glioma stem cell line, suggesting a possible role in the differentiation process of gliomas. These results implicate ctgf as a key gene for astrogenesis during development, and suggest that its mechanism may involve activation of p44/42 MAPK signaling. Additionally, CTGF-induced differentiation of glioblastoma stem cells into a less-tumorigenic state could increase the chances of successful intervention, since differentiated cells are more vulnerable to cancer treatments.

Fig 1. Exogenous CTGF protein increases the number of GFAP-positive cells of a neural progenitor culture.

Double immunostaining showing GFAP (A and B green) and Nestin (C and D red) expression of untreated and CTGF-treated astrocytes. E and F show merged pictures of GFAP and Nestin immunostaining together with nuclei-DAPI staining. Scale bars 50 μm. G shows the percentage of cells that were unstained or stained with GFAP and/or Nestin. CTGF was able to increase the number of GFAP-positive cells and, more dramatically, the number of Nestin/GFAP-positive cells.

Fig 2. Exogenous CTGF protein increases the number of Sox2-positive cells of a neural progenitor culture.

Immunostaining showing Sox2 (A and D) expression of untreated and CTGF-treated cells. C and F show merged pictures of Sox2 immunostaining together with nuclei-DAPI staining. Scale bars 10 μm. G shows the percentage of cells that were positive for Sox2.

Fig 3. Recombinant CTGF increases expression and deposition of fibronectin.

Immunostaining and immunoblotting for fibronectin and laminin. (A-D) Untreated or CTGF-treated progenitor neural cells immunostained for fibronectin (A, B) and laminin (C, D). (E) Immunoblotting for fibronectin and laminin of untreated (lane 1) or CTGF-treated neural progenitor cells (lane 2). CTGF incubation of neural progenitor cells for 120 h increased the expression of fibronectin threefold (compare untreated versus CTGF bars of graph in E). Scale bar 50 μm.

Fig 4. CTGF increases activation of p44/42 MAPK (ERK1/2).

Immunoblot for phosphorylated p44/42 MAPK (ERK 1 and 2) of untreated or CTGF-treated neural progenitor cells. CTGF increased the expression of phosphorylated p44/42 MAPK almost sevenfold (compare lanes 1 and 2 and the graph of the densitometry). As a loading control we used total p44/42 MAPK and tubulin.

Fig 5. CTGF does not increase p38 MAPK or p54/46 SAPK/JNK MAPK pathways.

Immunoblot for phosphorylated p38 MAPK and phosphorylated p46/p54 MAPK of untreated or CTGF-treated neural progenitor cells. CTGF did not change the levels of expression of these proteins (compare lanes 1 and 2). As a loading control we used total p46/p54 MAPK and tubulin.

Fig 6. Exogenous CTGF protein increases GFAP and reduces Sox2 expression in human cancer stem cells.

Immunoblotting showing GFAP, β-tubulin III and Sox2 expression in untreated and 1 nM and 5 nM CTGF-treated cells. CTGF incubation of human glioma stem cells for 120 h increased GFAP and reduced Sox2 expression in a dose-dependent manner. Alpha-tubulin was used as a loading control. **P < 0.001 and *P < 0.005.

Fig 7. Exogenous CTGF protein increases the number of GFAP-, CD133- and Nestin-positive cells of human cancer stem cells.

Immunostaining showing GFAP (A and D), CD133 (H and K) and Nestin (O and R) expression of untreated and CTGF-treated astrocytes. C and F, J and M and Q and T show merged pictures of GFAP, CD133 and Nestin immunostaining together with nuclei-DAPI staining respectively. Scale bars 25 μm. G, N and U show the fluorescence intensity of cells that were stained with GFAP, CD133 and Nestin respectively.