Transforming growth factor-beta1 modulates insulin-like growth factor binding protein-4 expression and proteolysis in cultured periosteal explants.

OBJECTIVE: Periosteum is involved in bone growth and fracture healing and has been used as a cell source and tissue graft for tissue engineering and orthopedic reconstruction including joint resurfacing. Periosteum can be induced by transforming growth factor beta (TGF-beta) or insulin-like growth factor-I (IGF-I) alone or in combination to form cartilage. However, little is known about the interaction between IGF and TGF-beta signaling during periosteal chondrogenesis. The purpose of this study was to determine the effect of TGF-beta1 on IGF binding protein-4 (IGFBP-4) and the IGFBP-4 protease pregnancy-associated plasma protein-A (PAPP-A) expression in cultured periosteal explants. DESIGN: Periosteal explants from rabbits were cultured with or without TGF-beta1. IGFBP-4 and PAPP-A mRNA levels were determined by real-time quantitative PCR. Conditioned medium was analyzed for IGFBP-4 and PAPP-A protein levels and IGFBP-4 protease activity. RESULTS: TGF-beta1-treated explants contained lower IGFBP-4 mRNA levels throughout the culture period with a maximum reduction of 70% on day 5 of culture. Lower levels of IGFBP-4 protein were also detected in the conditioned medium from TGF-beta1-treated explants. PAPP-A mRNA levels were increased 1.6-fold, PAPP-A protein levels were increased threefold, and IGFBP-4 protease activity was increased 8.5-fold between 7 and 10days of culture (the onset of cartilage formation in this model) in conditioned medium from TGF-beta1-treated explants. CONCLUSIONS: This study demonstrates that TGF-beta1 modulates the expression of IGFBP-4 and PAPP-A in cultured periosteal explants.

Serum-free media for periosteal chondrogenesis in vitro.

Organ culture studies involving whole explants of periosteum have been useful for studying chondrogenesis, but to date the standard culture model for these explants has required the addition of fetal bovine serum to the media. Numerous investigators have succeeded in culturing chondrocytes and embryonic cells in serum-free conditions but there have been no studies focused on achieving a defined, serum-free media for culturing periosteal explants. The purpose of the present investigation was to determine if whole periosteal explants can be grown and produce cartilage in serum-free conditions, and to define the minimum media supplements that would be conducive to chondrogenesis. 321 periosteal explants were obtained from the medial proximal tibiae of 31 two month-old NZ white rabbits and cultured using a published agarose suspension organ culture model and DMEM for six weeks. The explants were cultured with and without fetal bovine serum or bovine serum albumin and exposed to transforming growth factor beta alone, a combination of growth factors we call ChondroMix (10 ng/ml transforming growth factor beta, 50 ng/ml basic fibroblast growth factor, and 5 microg/ml growth hormone), and/or ITS+ (2.08 microg/ml each of insulin, transferrin, and selenious acid, plus 1.78 microg/ml linoleic acid and 0.42 mg/ml BSA). Maximal chondrogenic stimulation in this study was observed with the combination of ChondroMix and ITS+. However, the minimal requirement to match or exceed the level of chondrogenic stimulation seen in the standard model (TGF-1 in 10% FBS) was achieved simply by the addition of 2.0 microg/ml insulin in 0.1% BSA-containing medium (p < 0.05). Therefore, based on our results, it would be reasonable to assume that insulin is the component in ITS+ responsible for the observed increase in total cartilage growth. Lower concentrations of insulin were not effective, suggesting that the observed effect of insulin requires activation of the IGF-1 receptor.

Induction of CD-RAP mRNA during periosteal chondrogenesis.

Induction of chondrogenesis and maintenance of the chondrocyte phenotype are critical events for autologous periosteal transplantation, which is a viable approach for cartilage repair. Cartilage-derived retinoic acid-sensitive protein (CD-RAP) is a recently discovered protein that is mainly produced in cartilage. During development, CD-RAP expression starts at the beginning of chondrogenesis and continues throughout cartilage maturation. In order to investigate the involvement of CD-RAP during periosteal chondrogenesis we have determined the nucleotide sequence of the rabbit CD-RAP mRNA and utilized this information to evaluate the temporal and spatial expression pattern of CD-RAP at the mRNA level during chondrogenesis. When the periosteal explants were cultured under chondrogenic conditions, the expression of CD-RAP was induced, as shown by a 40-fold increase in CD-RAP mRNA between days 7 and 10. The temporal expression pattern of CD-RAP closely mimicked that of collagen type IIB mRNA. Also, the CD-RAP mRNA was localized to the matrix forming chondrocytes in the cambium layer of the periosteum by in situ hybridization as indicated by colocalization with collagen type II mRNA and positive safranin O staining. These data suggest a regulatory role of CD-RAP in periosteal chondrogenesis, which is potentially important for both cartilage repair and fracture healing via callus formation.

Temporal expression patterns of BMP receptors and collagen II (B) during periosteal chondrogenesis.

Articular cartilage has a limited ability to repair itself. Periosteal grafts have chondrogenic potential and are used clinically to repair defects in articular cartilage. An organ culture model system for in vitro rabbit periosteal chondrogenesis has been established to study the molecular events of periosteal chondrogenesis in vitro. In this model, bone morphogenetic protein-2 (BMP2) mRNA expression was found to be upregulated in the first 12 h. BMPs usually transduce their signals through a receptor complex that includes type II and either type IA or type IB BMP receptors. Receptors IA and IB play distinct roles during limb development. We have examined the temporal expression patterns for the mRNAs of these receptors using our experimental model. The mRNA expression patterns of these three BMP receptors differed from one another in periosteal explants during chondrogenesis. When these explants were cultured under chondrogenic conditions (agarose suspension with TGF-beta1 added to the media for the first 2 days), the expression of BMPRII mRNA and that of BMPRIA mRNA varied only slightly and persisted over a long time. In contrast, the expression of BMPRIB mRNAwas upregulated within 12 h, peaked at day 5, and fell to a level that was barely detected beyond day 21. Moreover, the expression of BMPRIB mRNA preceded that of collagen type IIB mRNAs, a marker for matrix-depositing chondrocytes. These data support a role for coordinate expression of BMP2 and its receptors early during periosteal chondrogenesis.