Tracking expression of virally mediated BMP-2 in gene therapy for bone repair.

Ex vivo gene therapy using stem cells transduced with viral vectors is a useful method for delivering a therapeutic protein to augment bone repair in animal models. However, the duration of cell-mediated protein production and the fate of the transduced cells are unknown. We constructed an adenoviral vector encoding Myc epitope tagged bone morphogenetic protein (BMP)-2 gene (AdBMP-2). Rat bone marrow cells transduced with this vector produced biologically active BMP-2 protein, which was confirmed by Western blot analysis and alkaline phosphatase assay. Implantation of bone marrow cells infected ex vivo with AdBMP-2 caused orthotopic bone formation in mouse hindlimbs and bony union of critical-sized mouse radial defects. Immunohistochemical analysis revealed that rBMCs expressed Myc epitope-tagged BMP-2 protein for 14 days in vivo and became incorporated in the endochondral fracture callus. This novel adenovirus encoding for epitope-tagged BMP-2 can be used for immunohistochemical tracking of transduced cells in ex vivo gene therapy for bone repair.

BMP signaling is required for septation of the outflow tract of the mammalian heart.

Bone morphogenetic proteins (BMPs) constitute a family of approximately 20 growth factors involved in a tremendous variety of embryonic inductive processes. BMPs elicit dose-dependent effects on patterning during gastrulation and gradients of BMP activity are thought to be established through regulation of the relative concentrations of BMP receptors, ligands and antagonists. We tested whether later developmental events also are sensitive to reduced levels of BMP signaling. We engineered a knockout mouse that expresses a BMP type II receptor that lacks half of the ligand-binding domain. This altered receptor is expressed at levels comparable with the wild-type allele, but has reduced signaling capability. Unlike Bmpr2-null mice, mice homozygous for this hypomorphic receptor undergo normal gastrulation, providing genetic evidence of the dose-dependent effects of BMPs during mammalian development. Mutants, however, die at midgestation with cardiovascular and skeletal defects, demonstrating that the development of these tissues requires wild-type levels of BMP signaling. The most striking defects occur in the outflow tract of the heart, with absence of septation of the conotruncus below the valve level and interrupted aortic arch, a phenotype known in humans as persistent truncus arteriosus (type A4). In addition, semilunar valves do not form in mutants, while the atrioventricular valves appear unaffected. Abnormal septation of the heart and valve anomalies are the most frequent forms of congenital cardiac defects in humans; however, most mouse models display broad defects throughout cardiac tissues. The more restricted spectrum of cardiac anomalies in Bmpr2(deltaE2) mutants makes this strain a key murine model to understand the embryonic defects of persistent truncus arteriosus and impaired semilunar valve formation in humans.

Further characterization of BC3H1 myogenic cells reveals lack of p53 activity and underexpression of several p53 regulated and extracellular matrix-associated gene products.

To catalog factors that may contribute to the completion of myogenesis, we have been looking for molecular differences between BC3H1 and C2C12 cells. Cells of the BC3H1 tumor line, though myogenic, are nonfusing, and withdraw from the cell cycle only reversibly, whereas cells of the C2C12 line fuse, differentiate terminally, and express several muscle-specific gene products that BC3H1 cells do not. Relative to C2C12 cells, BC3H1 cells underaccumulated cyclin-dependent kinase inhibitor p21 and underaccumulated transcripts for p21, GADD45, CDO, decorin, osteopontin, H19, fibronectin, and thrombospondin-1 (tsp-1). Levels of accumulation of H19, tsp-1, and larger isoforms of fibronectin messenger ribonucleic acid (mRNA) were found to increase in response to expression of myogenic regulatory factors as shown by their accumulation in differentiated myogenically converted 10T1/2 cells but not in 10T1/2 fibroblasts. BC3H1s accumulated a temperature-insensitive, geldanamycin-sensitive, misfolded form of p53 incapable of transactivating a p53 responsive reporter, consistent with underexpression of p21, GADD45, and tsp-1. BC3H1 and C2C12 cells were similar with respect to upregulation of p27 protein, downregulation of mitogen-activated protein kinase phosphatase-1 (MKP-1) protein, upregulation of retinoblastoma (Rb) mRNA, and nuclear localization of hypophosphorylated Rb. Cells of both lines expressed the muscle-specific 1b isoform of MEF2D. Although nonfusing in the short term, after more than 18 d in differentiation medium, some cultures of BC3H1 cells formed viable multinucleated cells in which the nuclei did not reinitiate synthesis of DNA in response to serum. Our findings suggest participation of tsp-1 and specific isoforms of fibronectin in myogenesis and suggest additional avenues of research in myogenesis and oncogenesis.