Role of CCN2/CTGF/Hcs24 in Bone Growth

Satoshi Kubota; Masaharu Takigawa* takigawa@md.okayama-u.ac.jp    Department of Biochemistry and Molecular Dentistry, Okayama University, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan
* Corresponding author. E‐mail address:

I Introduction

In mammals, the development of most bones in the skeleton is initiated by the preparation of the cartilage prototype and is accomplished through endochondral ossification or its related process. Endochondral ossification occurs mainly by growth plate chondrocytes in a precisely organized architecture with four‐dimensional (i.e., temporal and spatial) polarity of cytodifferentiation, in collaboration with vascular endothelial cells and osteoblasts, which invade through the cartilage canal. This complex biological process is under the control of a vast number of systemic hormones and local growth factors. Analogous to a full orchestra that needs a conductor to play polyphonic music with multiple instruments, certain key molecules have to modulate the functioning status of these signaling molecules and integrate the extracellular information in a harmonized manner to maintain the proper biological architecture of the growth plate and to promote the endochondral ossification toward bone growth. We suggest that CCN2, which is a classical member of the CCN protein family and was formerly called the connective tissue growth factor (CTGF)/hypertrophic chondrocyte‐specific gene product 24 (Hcs24)/ecogenin, is one such conductor/modulator of bone growth.

The functional characteristics of CCN2 as a conductor/modulator are represented by the fact that CCN2 is capable of promoting both the growth and differentiation of most mesenchymal cells involved in endochondral ossification. This apparently contradictory functionality may explain how CCN2 acts as a central driver of cartilage/bone growth and regeneration. In this article, we first briefly introduce the novel family of modulator proteins, namely the CCN family. Thereafter, we summarize advances in CCN2 research, mainly from the viewpoint of the molecular and cellular biology of bone growth and regeneration.

II The CCN Family

A General Structure and Terminology

The acronym “CCN” does not represent any structural or functional aspect of this gene family, but is a simple assemblage of the three classical members recognized earlier. This “family name” was first given in 1993 by Bork from the first letters of Cyr61 (cysteine‐rich protein 61), CTGF (connective tissue growth factor), and NOV (nephroblastoma‐overexpressed gene). Thereafter, three additional members were newly classified in this family (Brigstock, 1999; Lau and Lam, 1999; Perbal, 2004; Perbal and Takigawa, 2005; Takigawa et al., 2003). During this emerging period, a number of different names were independently assigned to each member by different research groups, thereby resulting in significant confusion in terminology (Baxter et al., 1998; Grotendorst et al., 2000; Moussad and Brigstock, 2000). To resolve this problem, a unified nomenclature was proposed in 2003, in which each member was renamed numerically under the family name, CCN (Brigstock et al., 2003). The relevance between the unified and original name of each member is summarized in Fig. 1A. Mammals possess six members in their genome so far, and no evidence for another member has yet been described.

Although the genomic organization of the members demonstrates remarkable variety, all of the gene products retain a highly characteristic structure. Except for CCN6, these proteins consist of four conserved protein modules connected in tandem in order, following an N‐terminal signal peptide for secretion. It should be noted that these four modules—insulin‐like growth factor binding protein (IGFBP), von Willebrand factor type C repeat (VWC), thrombospondin type 1 repeat (TSP1), and C‐terminal cysteine‐knot (CT) modules—contain conserved cysteine residues and are highly interactive with a variety of other molecules, thus providing a structural basis for the multiple functionality of CCN member proteins (Perbal and Takigawa, 2005). Interestingly, each module is encoded by the corresponding single exon, and the boundaries of exons in mRNA strictly correspond to the module boundary in protein (Fig. 1B). Therefore, the CCN members are supposed to have developed by exon shuffling along the course of evolution (Patthy, 1987). In this context, the conserved cysteine residues involved in all of the modules are supposed to determine the proper tertiary structure of each module by intramodular interactions rather than by intermodular/intermolecular interactions (Brigstock, 1999). In the mRNAs, the coding regions are flanked by 5′‐ and 3′‐untranslated regions (UTR) that contain critical regulatory elements of gene expression, which will be described later (Kubota and Takigawa, 2002).

B Members of the CCN Family

1 CCN1/Cyr61/Cef10

The first member of the CCN family was identified as one of the immediate‐early genes in mouse fibroblasts in 1989. Because of the 38 cysteine residues in the gene product, which is currently recognized to be conserved among all of the CCN family proteins, this gene was designated cysteine‐rich 61 (cyr61) (O'Brien et al., 1990; Simmons et al., 1989). Mammalian ccn1 genes are quite compact with relatively short introns, as observed in ccn2. In terms of protein structure, CCN1/Cyr61 is uniquely characterized by spacer/hinge amino acid stretches located between the VWC and TSP modules (Bork, 1993). However, the functional significance of this intermodular peptide is still unclear.

The CCN1 protein is known to mediate integrin‐associated cell adhesion events while promoting wound healing, angiogenesis, and chondrogenesis, as also observed in the case of CCN2 (Chen et al., 2001a,b; Jedsadayanmata et al., 1999; Kireeva et al., 1997; Schober et al., 2002). In addition, when we stimulated chondrocytic cells with different types of cytokines, ccn1 and ccn2 displayed a similar response to all the stimuli examined, thus indicating that they may be engaged in similar missions in the metabolism of skeletal tissues (Moritani et al., 2005). Nevertheless, a difference in their biological roles was also clearly represented by the phenotypic difference between knockout (KO) mice of ccn1 and ccn2. Most of ccn1 KO mice were reported to be embryonic lethal, due to a failure of embryonic vascular system development in placenta (Mo et...