Tissue engineering has emerged as a promising alternative to conventional orthopaedic grafting therapies. The general paradigm for this approach, in which phenotype-specific cells and/or bioactive growth factors are integrated into polymeric matrices, has been successfully applied in recent years toward the development of bone, ligament, and cartilage tissues in vitro and in vivo. Despite these advances, an optimal cell source for skeletal tissue repair and regeneration has not been identified. Furthermore, the lack of robust, functional orthpaedic tissue interfaces, such as the boneligament enthesis, severely limits the integration and biological performance of engineered tissue substitutes. This works aims to address these limitations by spatially controlling the commitment of primary dermal fibroblasts toward an osteoblastic (bone cell) lineage within three-dimensional polymeric matrices. The overall objective of this project was to investigate transcription factor-based gene therapy strategies for the differentiation of fibroblasts into a mineralizing cell source for orthopaedic tissue engineering applications. Our central hypothesis was that fibroblasts genetically engineered to express Runx2 via conventional and biomaterial-mediated ex vivo gene transfer approaches will differentiate into a mineralizing osteoblastic phenotype.

As a first step toward testing this hypothesis, we investigated retroviral gene delivery of the osteogenic transcription factor Runx2 as a mineralization induction strategy in primary dermal fibroblasts. We found that a combination of constitutive Runx2 overexpression and supplementation with the steroid hormone dexamethasone (DEX) synergistically induced osteogenic differentiation, including bone sialoprotein gene expression, alkaline phosphatase activity, and biological mineral deposition in primary dermal fibroblast monolayer cultures. This unexpected result suggested that Runx2-engineered fibroblasts have the capacity to create mineralized templates for bone repair and may be a potential cell source for bone tissue engineering applications. Furthermore, the complete absence of native osteoblastic phenotype in hormone-only treated cultures suggested that these cells could be utilized as a robust experimental model to study the Runx2-dependent mechanisms of DEX-induced osteogenesis.

Further characterization of Runx2-engineered fibroblasts involved investigation of the cellular and molecular pathway(s) driving the induction of osteogenesis in this non-osteoblastic cellular phenotype. More specifically, we used these cells as a model system to study the effect of DEX on Runx2 serine phosphorylation and the functional role of this phosphorylation state during osteoblastic differentiation. We demonstrated that DEX decreased Runx2 phosphoserine levels, particularly on Serine¹²⁵, in parallel with the upregulation of MAPK phosphatase-1 (MKP-1). Mutation of Ser¹²⁵ to glutamic acid, mimicking constitutive phosphorylation, inhibited Runx2-induced osteogenic differentiation, which was not rescued by DEX treatment. Conversely, mutation of Serine¹²⁵ to glycine, mimicking constitutive dephosphorylation, markedly increased osteogenic differentiation, which was enhanced by but did not require additional DEX supplementation. The DEX-induced decrease in Runx2 phosphorylation correlated with upregulation of MKP-1 through a glucocorticoid-receptor-dependent mechanism. Furthermore, inhibition of MKP-1 abrogated the effect of DEX on Runx2 phosphoserine levels. Collectively, these results demonstrated that DEX induces osteogenesis, at least in part, by modulating the phosphorylation state of a negative regulatory serine residue (Ser¹²⁵) on Runx2 via MKP-1. This work identifies a previously unreported mechanism for glucocorticoid-induced osteogenic differentiation and provides insights into the role of Runx2 phosphorylation during skeletal development.

Runx2-expressing fibroblasts were then evaluated within the context of threedimensional polymeric matrices for their potential as a mineralizing cell source for bone tissue engineering applications. Genetically modified fibroblasts were cultured in vitro on three commercially available scaffolds with highly divergent properties, including:​ fused deposition-modeled polycaprolactone (PCL), gas-foamed polylactide-co-glycolide (PLGA), and fibrous collagen disks. We demonstrated that the mineralization capacity of Runx2-engineered fibroblasts is scaffold-dependent, with collagen foams exhibiting tenfold higher mineral volume compared to PCL and PLGA scaffolds. Constructs were differentially colonized by genetically modified fibroblasts, but the scaffold-directed changes in DNA content did not correlate with trends in mineral deposition. Sustained expression of Runx2 upregulated osteoblastic gene expression relative to unmodified control cells and the magnitude of this expression was modulated by scaffold properties. Histological analyses revealed that matrix mineralization co-localized with cellular distribution, which was confined to the periphery of fibrous collagen and PLGA sponges and around the circumference of PCL microfilaments. Fourier transform infrared analysis verified that mineral deposits within Runx2-engineered scaffolds displayed the chemical signature characteristic of carbonate-containing, poorly crystalline hydroxyapatite, whereas control constructs did not contain biologically-equivalent mineral. Importantly, Runx2-transduced fibroblasts formed mineralized templates in vivo after implantation in a subcutaneous, heterotopic site, whereas minimal mineralization was evident in control constructs. Immunohistochemical analysis revealed that Runx2- expressing cells co-localized with mineral deposits in vivo, suggesting that mineral was primarily produced by transplanted donor cells. Taken together, these results establish Runx2-genetic engineering as a strategy for the conversion of a non-osteogenic cellular phenotype into a mineralizing osteoblastic cell source for bone repair.

Finally, we explored the feasibility of spatially regulating Runx2 expression in fibroblasts to engineer heterogeneous bone-soft tissue interfaces. Toward this end, we first demonstrated that biomaterial-mediated retroviral gene transfer is a feasible strategy for the genetic modification and differentiation of fibroblasts into a mineralizing osteoblastic phenotype. Viral uptake from these constructs was found to be highly dependent on the non-covalent adsorption of retroviral vectors to positively-charged poly-L-lysine prior to cell seeding. This observation was leveraged to create a graded distribution of Runx2 retrovirus within tissue engineered constructs. These 3-D retroviral gradients resulted in spatially regulated genetic modification of fibroblasts and, consequently, zonal organization of osteoblastic and fibroblastic cellular phenotypes in vitro. Moreover, implantation of heterogeneous constructs into a subcutaneous, ectopic site resulted in Runx2-induced spatial patterning of mineral deposition and non-mineralized fibroblastic extracellular matrix in vivo. Notably, discrete mineralized nodules co-localizing with transduced cell colonies were distributed throughout the interior of virus-coated constructs, suggesting that a biomaterial-mediated gene transfer approach may circumvent mass transport issues caused by the localization of a dense mineralized shell around the scaffold periphery. Collectively, these results indicate that heterogeneous bone-ligament-mimetic tissue interfaces can be developed by a simple, one step seeding of autologous fibroblasts into polymeric scaffolds containing a graded distribution of the Runx2 retroviral vector. The concept of controlling expression of tissue-specific transcription factors to create spatial gradients of differential cell function within 3-D matrices may be applicable to the development of interfacial zones for a large number of tissue engineering applications.

In summary, this research has established transcription factor-based gene therapy strategies for the conversion of a non-osteoblastic cellular phenotype into a mineralizing cell source for orthopaedic tissue engineering applications. This work is significant because it leverages these genetically engineered fibroblasts to simultaneously (1) elucidate previously unreported molecular pathways involved in bone formation and (2) develop mineralized templates for orthopaedic (bone, ligament) tissue repair. This work is innovative because it utilizes novel biomaterial-mediated gene transfer technologies to engineer bone-soft tissue interfacial zones. Overall, these results are significant toward our ultimate goal of regenerating complex, higher order tissue structures which mimic the cellular and microstructural characteristics of native tissue. Cellular therapies based on primary dermal fibroblasts would be particularly beneficial for patients with a compromised ability to recruit progenitors to the sight of injury as result of traumatic injury, radiation treatment, or osteodegenerative disease.