Transient exposure to transforming growth factor beta 3 under serum-free conditions enhances the biomechanical and biochemical maturation of tissue-engineered cartilage.

A goal of cartilage tissue engineering is the production of cell-laden constructs possessing sufficient mechanical and biochemical features to enable native tissue function. This study details a systematic characterization of a serum-free (SF) culture methodology employing transient growth factor supplementation to promote robust maturation of tissue-engineered cartilage. Bovine chondrocyte agarose hydrogel constructs were cultured under free-swelling conditions in serum-containing or SF medium supplemented continuously or transiently with varying doses of transforming growth factor beta 3 (TGF-beta3). Constructs were harvested weekly or bi-weekly and assessed for mechanical and biochemical properties. Transient exposure (2 weeks) to low concentrations (2.5-5 ng/mL) of TGF-beta3 in chemically defined medium facilitated robust and highly reproducible construct maturation. Constructs receiving transient TGF-beta3 exposure achieved native tissue levels of compressive modulus (0.8 MPa) and proteoglycan content (6-7% of wet weight) after less than 2 months of in vitro culture. This maturation response was far superior to that observed after continuous growth factor supplementation or transient TGF-beta3 treatment in the presence of serum. These findings represent a significant advance in developing an ex vivo culture methodology to promote production of clinically relevant and mechanically competent tissue-engineered cartilage constructs for implantation to repair damaged articular surfaces.

Effects of Runx2 genetic engineering and in vitro maturation of tissue-engineered constructs on the repair of critical size bone defects.

Genetic and tissue engineering strategies are being pursued to address the clinical limitations of current bone grafting materials. Based on our previous work demonstrating that overexpression of the Runx2 osteoblastic transcription factor and in vitro construct maturation synergistically enhanced in vivo mineralization in an ectopic site (Byers et al., Tissue Eng 2004;10:1757-1766), we examined the effects of these two parameters on the repair of critical size bone defects. Primary rat bone marrow stromal cells transduced with Runx2 or control (no Runx2 insert) retroviral vector were seeded onto 3D fused deposition-modeled polycaprolactone scaffolds. Runx2-modified cells produced biologically-equivalent mineralized matrices at nearly 2-fold higher rates than control cells. Constructs cultured in vitro for 1 day (immature) or 21 days (mineralized) were subsequently implanted into critical size calvaria defects in syngeneic rats, and bone healing was analyzed by micro-CT and histomorphometry at 28 days. Runx2-modified and control constructs precultured for 1 day healed to a greater extent than defects receiving no implant. Cell-free scaffolds yielded equivalent levels of bone formation as constructs precultured for 1 day. Interestingly, defects treated with control cell-seeded constructs precultured for 21 days exhibited low bone formation compared to other construct treatments, and repair was comparable to empty defects. In contrast, Runx2-modified constructs precultured for 21 days contained twice as much bone as control constructs precultured for 21 days and equivalent levels of new bone as cell-free and 1 day precultured constructs. These results demonstrate interplay between Runx2 genetically-modified cells and in vitro construct maturation in bone healing responses.

The effect of applied compressive loading on tissue-engineered cartilage constructs cultured with TGF-beta3.

In this study, we report that the sequential application of physiologic deformational loading after culturing with the growth factor TGF-beta3 (for 2-3 weeks) yields significantly stiffer chondrocyte-seeded agarose constructs than cultures in which deformational loading was applied during the initial 2-3 week TGF-beta3 exposure period. Using this culture protocol, engineered constructs were found to reach Young's modulus and GAG levels similar to that of native (parent) articular cartilage after only 42 days of culture. The present study extends the work on the mechanical preconditioning of engineered cartilage constructs to include transient supplementation with TGF-beta3 in a clinically-relevant, chemically-defined, serum-free media formulation.

Runx2/Cbfa1-genetically engineered skeletal myoblasts mineralize collagen scaffolds in vitro.

Genetic engineering of progenitor and stem cells is an attractive approach to address cell sourcing limitations associated with tissue engineering applications. Bone tissue engineering represents a promising strategy to repair large bone defects, but has been limited in part by the availability of a sustained, mineralizing cell source. This study examined the in vitro mineralization potential of primary skeletal myoblasts genetically engineered to overexpress Runx2/Cbfa1, an osteoblastic transcriptional regulator essential to bone formation. These cells were viable at the periphery of 3D fibrous collagen scaffolds for 6 weeks of static culture. Exogenous Runx2 expression induced osteogenic differentiation and repressed myogenesis in these constructs relative to controls. Runx2-modified cells deposited significant amounts of mineralized matrix and hydroxyapatite, as determined by microcomputed tomography, histological analysis, and Fourier transform infrared spectroscopy, whereas scaffolds seeded with control cells exhibited no mineralized regions. Although mineralization by Runx2-engineered cells was confined to the periphery of the construct, colocalizing with cell viability, it was sufficient to increase the compressive modulus of constructs 30-fold relative to controls. This work demonstrates that Runx2 overexpression in skeletal myoblasts may address current obstacles of bone tissue engineering by providing a potent cell source for in vitro mineralization and construct maturation. Additionally, the use of genetic engineering methods to express downstream control factors and transcriptional regulators, in contrast to soluble signaling molecules, represents a robust strategy to enhance cellular activities for tissue engineering applications.

Runx2/Cbfa1 stimulates transdifferentiation of primary skeletal myoblasts into a mineralizing osteoblastic phenotype.

Runx2, a transcriptional activator downstream of bone morphogenetic protein (BMP) signaling, is essential to osteoblastic differentiation and bone formation and maintenance. BMPs activate complex signaling networks, utilizing numerous signaling molecules and transcription factors to induce expression of osteoblastic markers in mesenchymal cell types. However, the role of Runx2 in this process, particularly in an environment independent of the other regulatory elements modulated by BMPs, remains poorly understood. In the present study, we used retroviral gene delivery to examine the effects of sustained Runx2 expression in primary myoblasts. Runx2 inhibited myogenesis, as demonstrated by suppression of MyoD and myogenin mRNA levels and reduced myotube formation. Additionally, Runx2-stimulated osteogenesis including osteoblastic gene expression, alkaline phosphatase activity, and biological mineral deposition. Notably, these osteogenic markers were induced to significantly greater levels than those observed in BMP-2-treated controls. These results demonstrate that direct exogenous expression of the Runx2 transcription factor, only one of numerous downstream targets of BMP signaling, is sufficient to induce transdifferentiation of myogenic cells into a mineralizing osteogenic lineage. This work underscores the potency of Runx2 as a regulator of osteogenesis and cell differentiation and provides new insights into the plasticity of committed mesenchymal cells.

Exogenous Runx2 expression enhances in vitro osteoblastic differentiation and mineralization in primary bone marrow stromal cells.

Bone marrow stromal cells represent a promising cell source for cell-based therapeutic and bone tissue-engineering applications, but are restricted by a low frequency in healthy marrow, an age-related decrease in osteogenic capacity, and a propensity for dedifferentiation during in vitro expansion. To address these limitations, retroviral gene delivery was used to examine the effects of sustained and elevated expression of the Runx2 osteoblastic transcription factor on osteoblastic gene and protein expression and mineralization in primary rat bone marrow stromal cells. Runx2 overexpression upregulated several osteoblast-specific genes, including collagen type I and osteocalcin, and enhanced alkaline phosphatase activity and biological mineral deposition. Forced Runx2 expression in combination with dexamethasone increased matrix mineralization compared with exogenous Runx2 expression or dexamethasone treatment alone, whereas dexamethasone-free control cultures displayed minimal mineralization. These additive effects suggest complementary interactions between Runx2 and dexamethasone-responsive regulatory factors. Finally, Runx2 overexpression in stromal cell cultures undergoing considerable in vitro expansion resulted in higher matrix mineralization capacity compared with controls, which completely lost the ability to produce mineralized matrix even in the presence of dexamethasone. These findings provide a novel strategy for cell-based therapeutic applications requiring significant numbers of osteogenic cells to synthesize mineralized constructs for the treatment of large bone defects.

Synergy between genetic and tissue engineering: Runx2 overexpression and in vitro construct development enhance in vivo mineralization.

Tissue engineering has emerged as a promising strategy to generate bone-grafting substrates. These approaches, however, are limited by an insufficient supply of committed osteoprogenitor cells and dedifferentiation of osteogenic cells during in vitro culture. To address these limitations, we engineered bone marrow stromal cells to constitutively express the osteoblastic transcription factor Runx2/Cbfa1, using retroviral gene delivery. These Runx2-modified cells were integrated into three-dimensional polymeric scaffolds to create tissue-engineered constructs. Compared with control stromal cells, Runx2 overexpression significantly upregulated osteoblastic differentiation and mineralization in vitro and in vivo in an ectopic, nonosseous subcutaneous site. More importantly, in vitro construct development to create a mineralized template before implantation dramatically enhanced subsequent in vivo mineralized tissue formation, providing a novel templating tissue-engineering strategy to improve in vivo mineralization. Finally, Runx2 overexpression and in vitro construct development synergistically enhanced in vivo mineralization compared with in vitro construct development or genetic engineering alone. This work provides a novel integrated genetic and tissue-engineering strategy to create mineralized templates for generating robust bone-grafting material.

Cell-type-dependent up-regulation of in vitro mineralization after overexpression of the osteoblast-specific transcription factor Runx2/Cbfal.

Functional expression of the transcriptional activator Runx2/Cbfal is essential for osteoblastic differentiation and bone formation and maintenance. Forced expression of Runx2 in nonosteoblastic cells induces expression of osteoblast-specific genes, but the effects of Runx2 overexpression on in vitro matrix mineralization have not been determined. To examine whether exogenous Runx2 expression is sufficient to direct in vitro mineralization, we investigated sustained expression of Runx2 in nonosteoblastic and osteoblast-like cell lines using retroviral gene delivery. As expected, forced expression of Runx2 induced several osteoblast-specific genes in NIH3T3 and C3H10T1/2 fibroblasts and up-regulated expression in MC3T3-E1 immature osteoblast-like cells. However, Runx2 expression enhanced matrix mineralization in a cell-type-dependent manner. NIH3T3 and IMR-90 fibroblasts overexpressing Runx2 did not produce a mineralized matrix, indicating that forced expression of Runx2 in these nonosteogenic cell lines is not sufficient to direct in vitro mineralization. Consistent with the pluripotent nature of the cell line, a fraction (25%) of Runx2-expressing C3H10T1/2 fibroblast cultures produced mineralized nodules in a viral supernatant-dependent manner. Notably, bone sialoprotein (BSP) gene expression was detected at significantly higher levels in mineralizing Runx2-infected C3H10T1/2 cells compared with Runx2-expressing cultures which did not mineralize. Treatment of Runx2-infected C3H10T1/2 cultures with dexamethasone enhanced osteoblastic phenotype expression, inducing low levels of mineralization independent of viral supernatant. Finally, Runx2 overexpression in immature osteoblast-like MC3T3-E1 cells resulted in acceleration and robust up-regulation of matrix mineralization compared with controls. These results suggest that, although functional Runx2 is essential to multiple osteoblast-specific activities, in vitro matrix mineralization requires additional tissue-specific cofactors, which supplement Runx2 activity.

Enhanced expression of the osteoblastic phenotype on substrates that modulate fibronectin conformation and integrin receptor binding.

Integrins represent the primary mechanism of cell-extracellular matrix interactions and control cell morphology, proliferation, and differentiation. We have previously shown that substrate-dependent modulation of adsorbed fibronectin (Fn) conformation alters alpha5beta1 integrin binding to Fn and directs C2C12 myoblast proliferation and differentiation (Mol. Biol. Cell 10 (1999) 785). The model substrates used in these experiments were bacteriological (untreated) polystyrene (B), tissue culture polystyrene (T), and type-I collagen-coated T (C). In the present study, we examined MC3T3-EI osteoblast-like cell differentiation on Fn-coated B, T, and C substrates. Immunofluorescence staining revealed substrate-dependent differences in integrin alpha5beta1 binding and clustering into focal adhesions (C > T > B), consistent with our previous integrin binding analysis. Alkaline phosphatase activity and matrix mineralization showed substrate-dependent differences (C > T > B, p < 0.05). Similar trends were observed for alkaline phosphatase, osteocalcin, and bone sialoprotein gene expression. Blocking experiments with antibodies directed against Fn completely inhibited matrix mineralization on Fn-coated C, indicating that Fn is critical to expression of the osteoblastic phenotype on this extracellular matrix component. These substrate-dependent differences in osteoblast differentiation correlated with differences in alpha5beta1 binding, suggesting that these differences arise from substrate modulation of integrin-matrix interactions. Substrate-dependent modulation of cell function may provide a versatile mechanism to control cell responses in numerous biomedical applications.