There exists a dire need for improved therapeutics to achieve predictable and effective bone regeneration. Non-viral gene therapy is a safe method that can efficiently transfect target cells, therefore is a promising approach to overcoming the drawbacks of protein delivery of growth factors. The goal of this study was to employ cost-effective biomaterials to deliver genetic materials (DNA or RNA) in a controlled manner in order to address the high cost issues, safety concerns, and lower transfection efficiencies that exist with protein and gene therapeutic approaches. To achieve our goal, we set several aims: 1) To assess the bone regeneration capacity of polyethylenimine (PEI)-chemically modified ribonucleic acid (cmRNA) (encoding bone morphogenetic protein-2 (BMP-2)) activated matrices, compared to PEI-plasmid DNA (BMP-2)-activated matrices. 2) To explore the osteogenic potential of cmRNA-encoding BMP-9, in comparison to cmRNA-encoding BMP-2. 3) To use collagen membranes as integral components of a guided bone regeneration protocol and to enhance the bioactivity of collagen membranes by incorporating plasmid DNA (pDNA) or cmRNA encoding bone morphogenetic protein-9 (BMP-9). 4) To test whether the delivery of pDNA encoding BMP-2 (pBMP-2) and fibroblast growth factor-2 (pFGF-2) together can synergistically promote bone repair in a leporine model of diabetes mellitus, a condition that is known to be detrimental to union. 5) To investigated whether there is a synergistic effect on bone regeneration following delivery of pBMP-2 and pFGF-2, insulin and/or vitamin D. These investigations together provided new insights regarding the appropriate treatment methods for patients with fractures. Here we develop and test a non-viral gene delivery system for bone regeneration in challenging animal models utilizing a scaffold carrying PEI-nucleic acid complexes. We utilized three kinds of pDNA encoding either BMP-2, BMP-9 or FGF-2 as well as two kinds of cmRNA encoding either BMP-2 or BMP-9 formulated into PEI complexes. The fabricated nanoplexes were assessed for their size, charge, in vitro cytotoxicity, and capacity to transfect human bone marrow stromal cells (BMSCs). The in vivo functional potency of different nanoplexes embedded in scaffolds was evaluated using a calvarial bone defect model in rats, diaphyseal long bone radial defects in a diabetic rabbit model and intramuscular implantation in a diabetic rat. The results indicate that our non-viral gene delivery system induced migration and differentiation of resident cells to enhance bone regeneration. Together these findings suggest that scaffolds loaded with non-viral vectors harboring cmRNA or pDNA encoding osteogenic proteins may be a powerful tool for stimulating bone regeneration with significant potential for clinical translation.