**5. The CCN family**

The CCN family makes up a group of six highly conserved, secreted, extracellular matrix‐ associated proteins that regulate diverse cellular functions, including skeletal development, wound healing, fibrosis, and cancer. Originally named after the three identified members cysteine‐rich 61 (Cyr61, CCN1), connective tissue growth factor (CTGF, CCN2), and nephro‐ blastoma overexpressed (Nov, CCN3)—this family also includes the Wnt‐induced secreted proteins 1–3 (i.e., WISP1/CCN4, WISP2/CCN5, and WISP3/CCN6). Members of the CCN family share a unique and conserved modular structure and interact with and orchestrate cellular responses to extracellular factors via direct binding to cell surface receptors, including integrins, Notch1, neurotrophic tyrosine kinase receptor type 1 (TrkA), low‐density lipopro‐ tein receptor‐related proteins (LRPs), and HSPGs. CCN proteins can also mediate biological functions by interacting with growth factors such as tumor growth factor beta (TGF‐β), vascular endothelial growth factor, and bone morphogenetic proteins (BMPs) and by associ‐ ating with other ECM proteins including fibronectin and fibulin 1C. Through these interac‐ tions, CCN proteins serve both distinct and overlapping biological roles. Consequently, deregulation of their expression or activities contributes to the pathobiology of several diseases, many of which may arise when inflammation or tissue injury becomes chronic, including vascular diseases, fibrosis, arthritis, and cancer.

#### **5.1. Structure and function of CCN family members**

CCN proteins are cysteine‐rich and share a modular structure (Modules I–IV), with an N‐ terminal secretory peptide followed by four conserved domains with sequence homologies to insulin‐like growth factor‐binding proteins (IGFBPs), von Willebrand factor type C repeat (VWC), thrombospondin type I repeat (TSP1), and a carboxyl‐terminal domain (CT) that contains a cysteine‐knot motif [212]. The order of these modules has been strictly conserved during evolution, suggesting that it is critically important for these proteins. Each module is involved in protein binding and contains conserved hydrophobic, polar, and cysteine residues. Module I shares 32% sequence homology with the N‐terminal cysteine‐rich regions of the IGF‐ binding proteins and contains a GCGC‐CXXC motif that is involved in IGF binding. Module II includes a stretch of 70 amino acids with sequence identity to von Willebrand factor as well as various thrombospondins, collagens, and mucins [212]. This domain has been shown to mediate protein oligomerization [213]. Module III is a TSP1 repeat that contains the WSXCSXXCG motif, which is thought to be implicated in the binding of sulfated glycoconju‐ gates and to be important for cell attachment [212,214]. The last module, Module IV, occurs at the carboxy‐terminus of various extracellular proteins and is the least conserved of the four domains at the nucleotide sequence level. It consists of several cysteine residues that adopt a cysteine‐knot motif. This motif comprises a complex structure of two‐stranded β‐sheets that lie face to face and are linked by three interlocking disulfide bridges [215] and occurs in TGF‐ β, PDGF, and nerve growth factor (NGF). It is critical for several of the biological functions of CCN proteins and is thought to mediate dimerization and binding to cell surface receptors. CCN5 is the only family member that lacks the CT domain [216]. A variable, central hinge region that is susceptible to proteolytic processing by MMPs and other proteases links the amino‐terminus and carboxy‐terminus of these proteins, yielding two halves that bind distinct cell surface receptors [217]. It is not clear whether the individual properties of each of the four modules govern the biological properties of the CCN protein or if it is the combination of the modules and other sequences within the protein that do so. However, all of the modules are highly interactive with a number of other molecules, which include cell surface receptors, ECM components, growth factors, and structural proteins [218].

#### **5.2. Cell surface receptors mediating CCN functions in cartilage and bone**

**5. The CCN family**

The CCN family makes up a group of six highly conserved, secreted, extracellular matrix‐ associated proteins that regulate diverse cellular functions, including skeletal development, wound healing, fibrosis, and cancer. Originally named after the three identified members cysteine‐rich 61 (Cyr61, CCN1), connective tissue growth factor (CTGF, CCN2), and nephro‐ blastoma overexpressed (Nov, CCN3)—this family also includes the Wnt‐induced secreted proteins 1–3 (i.e., WISP1/CCN4, WISP2/CCN5, and WISP3/CCN6). Members of the CCN family share a unique and conserved modular structure and interact with and orchestrate cellular responses to extracellular factors via direct binding to cell surface receptors, including integrins, Notch1, neurotrophic tyrosine kinase receptor type 1 (TrkA), low‐density lipopro‐ tein receptor‐related proteins (LRPs), and HSPGs. CCN proteins can also mediate biological functions by interacting with growth factors such as tumor growth factor beta (TGF‐β), vascular endothelial growth factor, and bone morphogenetic proteins (BMPs) and by associ‐ ating with other ECM proteins including fibronectin and fibulin 1C. Through these interac‐ tions, CCN proteins serve both distinct and overlapping biological roles. Consequently, deregulation of their expression or activities contributes to the pathobiology of several diseases, many of which may arise when inflammation or tissue injury becomes chronic,

CCN proteins are cysteine‐rich and share a modular structure (Modules I–IV), with an N‐ terminal secretory peptide followed by four conserved domains with sequence homologies to insulin‐like growth factor‐binding proteins (IGFBPs), von Willebrand factor type C repeat (VWC), thrombospondin type I repeat (TSP1), and a carboxyl‐terminal domain (CT) that contains a cysteine‐knot motif [212]. The order of these modules has been strictly conserved during evolution, suggesting that it is critically important for these proteins. Each module is involved in protein binding and contains conserved hydrophobic, polar, and cysteine residues. Module I shares 32% sequence homology with the N‐terminal cysteine‐rich regions of the IGF‐ binding proteins and contains a GCGC‐CXXC motif that is involved in IGF binding. Module II includes a stretch of 70 amino acids with sequence identity to von Willebrand factor as well as various thrombospondins, collagens, and mucins [212]. This domain has been shown to mediate protein oligomerization [213]. Module III is a TSP1 repeat that contains the WSXCSXXCG motif, which is thought to be implicated in the binding of sulfated glycoconju‐ gates and to be important for cell attachment [212,214]. The last module, Module IV, occurs at the carboxy‐terminus of various extracellular proteins and is the least conserved of the four domains at the nucleotide sequence level. It consists of several cysteine residues that adopt a cysteine‐knot motif. This motif comprises a complex structure of two‐stranded β‐sheets that lie face to face and are linked by three interlocking disulfide bridges [215] and occurs in TGF‐ β, PDGF, and nerve growth factor (NGF). It is critical for several of the biological functions of CCN proteins and is thought to mediate dimerization and binding to cell surface receptors. CCN5 is the only family member that lacks the CT domain [216]. A variable, central hinge

including vascular diseases, fibrosis, arthritis, and cancer.

**5.1. Structure and function of CCN family members**

176 Composition and Function of the Extracellular Matrix in the Human Body

Despite the structural similarities that CCN proteins have to other protein domains as described above, their interactions are unique because of their ability to bind extracellular factors via their modular domains. CCN proteins have been shown to interact specifically with cell surface receptors such as HSPGs, integrins, and LRPs, accounting for their ability to regulate numerous cellular functions. CCN2, the most studied of the CCN family members, shares common functionality with CCN1 with respect to interaction with integrins and HSPGs, causing comparable biological effects. LRP1 is another common receptor shared by CCN2 and CCN1; however, the target cell and biological consequence differ between the two [219,220].

Direct binding of CCN proteins to integrins present on cellular surface drives many of their effects on cartilage and bone. Integrins comprise a large family of cell‐cell and cell‐matrix receptors that signal both from the ECM to the cytoplasm and from the cell to the matrix (inside‐ out and outside‐in) [221]. Integrins are αβ heterodimers and can be present in a number of configurations. The different combinations of α and β receptors define what ECM molecules a cell interacts with. These receptors regulate many cellular functions such as proliferation, differentiation, motility, and developmental processes among others [221]. A decrease in the interaction of β integrins with matrix molecules is observed during the pathogenesis of osteoarthritis, and thus the disruption of integrin‐matrix interaction causes cellular dysfunc‐ tion in cartilage [222]. Integrins that have been identified in cartilage include α5β1, αvβ3, αvβ5, α6β1, α1β1, α2β1, α10β1, and α3β1. As is the case with matrix molecules, integrins are expressed differentially in specific regions of cartilage and during development and pathogenesis.

Other matrix receptors are present in cartilage. One example is NG2, a transmembrane proteoglycan in cartilage with the matricellular protein type VI collagen as its ligand [72–74]. NG2 interaction with type VI collagen may be an important interaction in determining the progression of sarcomas [223]. Annexin V, or anchorin II, binds to type II collagen and is mainly expressed in the superficial zone of articular cartilage [224]. CD44 binds to hyaluronic acid; blocking of this receptor causes a loss of the matrix of cartilage [225].

As a matricellular protein, CCN2 also binds the fibronectin receptor (α5β1) and aggrecan, which are major components of the ECM [226]. The interaction of α5β1 with fibronectin in cartilage causes loss of the differentiated state [227]. Blocking of this integrin causes a loss of the differentiation of pre‐hypertrophic chondrocytes [228]. A knockout mouse that has loss of β<sup>1</sup> specifically in chondrocytes exhibits a disease similar to chondrodysplasia [222]. β? is the most expressed integrin β subunit in osteoarthritic chondrocytes [229].

CCN2 is induced by signaling from retinoids in cartilage in the growth plate [230]. CCN2 causes differentiation of the chondrocytes and is important for matrix deposition and signal‐ ing from cellular receptors [226,231]. Retinoids are important signaling molecules involved in the cartilage becoming hypertrophic during endochondral ossification.
