Function Of Skeletal Matrix Genes

National Institute Of Dental & Craniofacial Research

Investigators

Linked publications & trials

Abstract

Bgn potentially controls bone healing through periosteal cell activation Bone healing involves a complex sequence of physiological events resulting in the formation of new bone in the fracture area. An important tissue in the healing process is the periosteum, a thin membrane surrounding the bone that rapidly expands after fracture and is thought to be the source of progenitor cells needed for bone formation and repair. The periosteum-derived cells (PDCs), which have stem cell character, can differentiate into multiple cell types including bone, cartilage and fat. While Bgn appears important in bone and cartilage formation during fracture healing, its exact role in bone repair is still unclear. Histological analysis of the tissues around fracture sites showed that the fracture-induced thickened periosteum contains high levels of Bgn, suggesting it has a function there. The goal of this project is to first develop novel methods in which to isolate, expand and differentiate periosteal cells and second, to determine if Bgn regulates PDC function during bone repair. While our first interest was to examine Periosteum-derived cells (PDCs) from the long bones (lbPDCs) we wondered if there could be differences between lbPDCs and PDCs from craniofacial bones or, specifically, from the calvaria (cPDCs). In this investigation, we isolated lbPDCs and cPDCs of adult mice and grew them in a novel 3D RAFT system to mimic the native environment of the cells. When cells were subjected to osteogenic differentiation in vitro, the differentiated lbPDCs, but not cPDCs, formed cartilage structures that were incorporated into the matrix. RNAseq analysis showed that differentiated lbPDC, but not differentiated cPDCs, made high levels of collagen type II, aggrecan and collagen type X, further confirming our notion that lbPDCs undergo chondrogenic differentiation under osteogenic conditions. Considering that long bones are formed via endochondral ossification while calvaria is not, our data suggest that PDCs retain a memory of their tissue origin a feature that could be capitalized for future applications in their use in skeletal tissue regeneration. The inflammatory response is necessary at the early stages of bone healing and for the removal of necrotic tissues. Inflammation also stimulates the activation of skeletal stem cells (SSC)s needed for bone regeneration and repair. Biglycan (Bgn), a matrix proteoglycan, is highly expressed in the periosteum, particularly after fracture, and yet its exact role in the inflammation state and in subsequent regulation of bone repair is still unclear. In this study, we established a new fracture technique and examined the role of Bgn in fracture healing using WT and Bgn deficient (KO) mice. Analysis of systemic inflammatory cytokine secretion 24 hours post fracture showed reduction in IL-6 and monocyte chemoattractant protein 1 (MCP1) secretion in Bgn KO mice. RNA analysis and histological examination 3 days post fracture further demonstrated low expression of IL-6 mRNA as well as a significant reduction in macrophage infiltration around the fracture site in Bgn KO compared to WT mice. In addition, after fracture, the periosteum in the Bgn KO did not expand to the same extent as in WT. Callus formation around the fracture site was measured 7, 14, 21 and 28 days post-fracture by CT analysis and showed that fractured bones from Bgn KO mice formed a smaller callus compared to WT mice. Quantification of the callus 14 days post-fracture revealed that the callus volume/total volume and the cross-sectional area was significantly lower in the Bgn KO compared to WT mice. By day 28, most of the callus formed by Bgn KO fractured bones was resorbed and the bone had remodeled into mature lamellar bone. Second harmonic generation microscopy was next used to visualize and quantify the collagen fibrillar structure of the bones. High collagen intensity with minor spacing between the collagen fibers was observed in the WT femurs, whereas Bgn KO femurs showed significantly less collagen intensity with irregular structure and spacing between the collagen fibers. In summary, Bgn deficiency did not prevent callus formation but reduced the inflammation needed for the bone healing process and periosteum expansion, leading to impaired callus and bone formation. Overall our results highlight the importance of the extracellular matrix component Bgn in bone healing and in regulating bone integrity. ColVIa2 regulates bone mass by inhibiting osteoclast function Collagen type VI monomers are composed of three -chains (1, 2, 3) that align in a triple helix that assembles into larger fibrils found in many musculoskeletal tissues. Collagen type VI is widely known for its role in muscular disorders, however its function in bone is still not well understood. Using immunohistochemistry, we examined the expression of collagen type VI during fracture healing and found that it was enhanced in the callus of healing bone. We further examined all three collagen type VI chains by qPCR and found Col62 mRNA was highly upregulated at day 14 post-fracture, suggesting it could have a role in bone remodeling. To determine the role of collagen type VI 2 in bone function we analyzed bone mineral density of femora and whole mouse body X-ray by DEXA and found it was significantly reduced in mice deficient in collagen type VI 2 (Col62 KO). Further micro CT analysis in both femora and vertebrae showed significant decreases in bone volume/tissue volume, trabecular bone number in the Col62 KO compared to wild-type (WT) mice. To try and understand the cellular basis for the decreased bone mass, we performed double labeling assays by calcein injection, and found no difference in trabecular bone formation between WT and Col62 KO mice judged by the mineral appositional rate, bone formation rate, and mineralizing perimeter. This experiment indicated that collagen type VI does not affect bone formation in vivo. To gain insight into mechanistic basis for the low bone mass RNA was extracted from WT and Col62 KO femurs and in Col62 KO RNAseq was performed. Our analysis showed that numerous mRNAs were differentially expressed in proteins related to osteoclast morphology, differentiation, development and formation. Based on these findings we next examined femora sections for the abundance of TRAP positive osteoclasts we discovered that the mutant mice exhibited 2.4 times more osteoclasts compared to WT mice, indicating the primary effect of collagen type VI 2 is on osteclastogenesis. Further analysis of the RNA seq data showed that one of the top regulator effect networks and top upstream regulators affected in the Col62 KO bones were related to TNF signaling. Since TNF is well known to promote osteoclastogenesis, we hypothesized that TNF may be involved in collagen type VI 2 regulation of osteclastogenesis. To test this, solid phase binding assays were performed and showed specific binding of type VI collagen to TNF. When we treated bone marrow stromal cells (BMSCs) from WT and Col62 KO mice with rmTNF protein, we found that Col62 KO cells expressed higher levels of TNF mRNA compared to WT cells suggesting BMSCs from Col62 KO mice are highly sensitive to TNF signaling. Taken together, our data indicate that collagen type VI 2 chain deficiency causes trabecular bone loss by enhancing osteoclast differentiation through enhanced TNF signaling.