Herein is described a novel, high-efficiency, non-viral, protocol for delivery of genes into permeabilized primary perichondrial cells forming a first step towards ex vivo gene therapy for the repair of full-thickness articular cartilage defects. Using this protocol, a plasmid carrying a marker beta-galactosidase (beta-gal) gene, driven by a strong mammalian promoter, was transfected into primary perichondrial cells with greater than 70% efficiency. This was achieved by utilizing a three step method: 1) Primary cells were permeabilized using a mild detergent (lysolecithin), 2) The beta-gal plasmid was allowed to associate with a polycation (poly-L-lysine) core covalently linked to a receptor ligand (transferrin) forming the DNA/transferrin-poly-L-lysine complex (DTPL complex), and 3) Cationic liposomes were introduced to the DTPL complex to form the DTPLL complex. This mixture was used to transfect primary perichondrium cells. More than 70% of the primary perichondrial cells were found to be positive for beta-gal activity. For in vivo assessment, D,D-L,L-polylactic acid (PLA) scaffolds (3 mm x 3.7 mm) seeded with the transfected cells were implanted into experimentally created osteochondral defects in rabbit knees. The transformed cells continued to express beta-gal in vive for the entire test period of 7 days, as detertnined by the beta-gal assay. We have previously demonstrated that adding exogenous TCF-beta 1 can enhance the chondrocytic phenotype of perichondrial cells (1,2). These studies were initiated in order to assess the usefulness of transfected perichondrium cells as vehicles for the localized delivery of TGF-beta 1 into the repair site. To this end, we have developed a TGF-beta 1 expression vector and show that cells transfected with this construct overexpress TGF-beta 1 specific mRNA. This system can now be used for the delivery of TGF-beta 1 into the articular repair site to enhance the chondrocytic phenotype of implanted cells.