Type IIB procollagen n-propeptide inhibits angigenesis in cartilage

Abstract

Purpose: A fragment of type II procollagen N-propeptide, PIIBNP, contains the classic integrin binding sequence RGDRGD. This sequence is conserved across species, indicating potential biological function. We have previously shown that recombinant PIIBNP induces death of tumor cells that express integrin αvβ3 and αvβ5. Data from HUVEC cell tube formation assay, mouse aortic ring assay and mouse corneal assay showed that recombinant PIIBNP inhibits angiogenesis both in vitro and in vivo (1). PIIBNP is synthesized by chondrocytes liberated to matrix in cartilage. The goal of this study was to determine the role that PIIBNP plays in cartilage in keeping cartilage avascular and intact. Methods: Detection of PIIBNP from cartilage. Cartilage was isolated from limbs of P1 mice. Cartilage was smashed by crusher and PIIBNP was extracted with SDS loading buffer, resolved by SDS-PEGE, and detected by western blotting. Mouse chondrocyte isolation. Chondrocytes were obtained by enzymatic dissociation of ventral regions of rib cages from newborn mice. Chondrocytes were incubated with or without the presence of MMP inhibitor for 48 hours. The medium from chondrocytes culture were used as the conditioned medium. Cartilage-aorta co-culture: Bovine and mice bones were used as control in the cartilage-aorta co-culture experiments. Mouse cartilage or bone was placed into inserts of a 24-well plate and cocultured with aorta in 3D matrigel for 8 days. The microvessel outgrowth was visualized under a light microscope (NIKON ECLIPS E800) and images were digitally photographed using a Q capture Retiga 2000R camera, and quantified with an Image J software. Results: We have been able to detect PIIBNP in cartilage (Fig. 1A) and chondrocytes conditioned medium (Fig. 1B). In order to reasonably hypothesize an effect of PIIBNP in cartilage, we sought to determine weather cartilage inhibits angiogenesis in mouse aortic ring assays. The results from cartilage-aorta coculture experiments showed that microvessel outgrowth was seen only in control and in the presence of bone, but not in the presence of cartilage (Fig. 2A). Since we and other have detected PIIBNP in cartilage, it is possible that the PIIBNP in cartilage inhibited the microvessel outgrowth. To further confirm this, we performed aorta angiogenesis assay in the presence of chondrocyte conditioned medium and found that the chondrocyte conditioned medium inhibited 80 percent of the microvessel outgrowth as compared with the control (Fig B). MMT inhibitor GM6001 inhibits the production of PIIBNP and the cartilage treated with GM6001 did not inhibit microvessel outgrowth in mouse aortic ring assay. These data suggested that natural PIIBNP in cartilage inhibits angiogenesis. To further investigate the mechanism behind the antiangiogenesis of PIIBNP, we treated HUVEC cells in culture with recombinant proteins. Treatments with PIIBNP, but not GST or mPIIBNP, induced disappearance of VE-cadherin at cell-cell junctions (Fig 3A), suggesting that the engagement of PIIBNP on cell surface elicited a signaling events involved in the inhibition of angiogenesis by PIIBNP. This hypothesis is supported by the fact that the PIIBNP, but not GST or mPIIBNP, induced activation of the Src kinase (Fig 3B), consistent with previous report that integrin αvβ3 signaling induces Src activation and VE-cadherin internalization. Conclusions: We previously showed that recombinant PIIBNP induced tumor cell death via binding to integrin αvβ3 and αvβ5, and suppresses tumor growth in nude mice. The data presented here suggested that the natural occurring PIIBNP in cartilage most likely inhibits angiogenesis. We have detected the existence of PIIBNP from P1 chondrocyte conditioned medium which also inhibits angiogenesis. Since PIIBNP reaches the highest level during cartilage formation and it kills tumor cell, it is an excellent candidate for the molecular mechanism by which cartilage remains avascular and intact.