Faculty Bibliography

Numerous studies have demonstrated the critical role of angiogenesis for successful osteogenesis during endochondral ossification and fracture repair. Vascular endothelial growth factor (VEGF), a potent endothelial cell-specific cytokine, has been shown to be mitogenic and chemotactic for endothelial cells in vitro and angiogenic in many in vivo models. Based on previous work that (1) VEGF is up-regulated during membranous fracture healing, (2) the fracture site contains a hypoxic gradient, (3) VEGF is up-regulated in a variety of cells in response to hypoxia, and (4) VEGF is expressed by isolated osteoblasts in vitro stimulated by other fracture cytokines, the hypothesis that hypoxia may regulate the expression of VEGF by osteoblasts was formulated. This hypothesis was tested in a series of in vitro studies in which VEGF mRNA and protein expression was assessed after exposure of osteoblast-like cells to hypoxic stimuli. In addition, the effects of a hypoxic microenvironment on osteoblast proliferation and differentiation in vitro was analyzed. These results demonstrate that hypoxia does, indeed, regulate expression of VEGF in osteoblast-like cells in a dose-dependent fashion. In addition, it is demonstrated that hypoxia results in decreased cellular proliferation, decreased expression of proliferating cell nuclear antigen, and increased alkaline phosphatase (a marker of osteoblast differentiation). Taken together, these data suggest that osteoblasts, through the expression of VEGF, may be in part responsible for angiogenesis and the resultant increased blood flow to fractured bone segments. In addition, these data provide evidence that osteoblasts have oxygen-sensing mechanisms and that decreased oxygen tension can regulate gene expression, cellular proliferation, and cellular differentiation

Modulation of biological pathways governing osteogenesis may accelerate osseous regeneration and reduce the incidence of complications associated with fracture healing. Transforming growth factor beta1 (TGF-beta1) is a potent growth factor implicated in the regulation of osteogenesis and fracture repair. The use of recombinant proteins, however, has significant disadvantages and has limited the clinical utility of these molecules. Targeted gene therapy using adenovirus vectors is a technique that may circumvent difficulties associated with growth factor delivery. In this study, we investigate the efficacy of replication-deficient adenoviruses containing the human TGF-beta1 and the bacterial lacZ genes in transfecting osteoblasts in vitro and osseous tissues in vivo. We demonstrate that adenovirus-mediated gene therapy efficiently transfects osteoblasts in vitro with the TGF-beta1 virus causing a marked up-regulation in TGF-beta1 mRNA expression even 7 days after transfection. Increased TGF-beta1 mRNA expression was efficiently translated into protein production and resulted in approximately a 46-fold increase in TGF-beta1 synthesis as compared with control cells (vehicle- or B-galactosidase-transfected). Moreover, virally produced TGF-beta1 was functionally active and regulated the expression of collagen IalphaI (5-fold increase) and the vascular endothelial growth factor (2.5-fold increase). Using an adenovirus vector encoding the Escherichia coli LacZ gene, we demonstrated that adenovirus-mediated gene transfer efficiently transfects osteoblasts and osteocytes in vivo and that transfection can be performed by a simple percutaneous injection. Finally, we show that delivery of the hTGF-beta1 gene to osseous tissues in vivo results in significant changes in the epiphyseal plate primarily as a result of increased thickness of the provisional calcification zone

Distraction osteogenesis is a powerful technique capable of generating viable osseous tissue by the gradual separation of osteotomized bone edges. Although the histologic and ultrastructural changes associated with this process have been extensively delineated, the molecular events governing these changes remain essentially unknown. We have devised a rat model of mandibular distraction osteogenesis that facilitates molecular analysis of this process. Such information has significant clinical implications because it may enable targeted therapeutic manipulations designed to accelerate osseous regeneration. In this study, we have evaluated the expression of transforming growth factor beta-1, a major regulator of osteogenesis during endochondral bone formation and development, and osteocalcin, an abundant noncollagenous extracellular matrix protein implicated in the regulation of mineralization and bone turnover. The right hemimandible of 36 adult male rats was osteotomized, and a customized distraction device was applied. Animals were allowed to recover and, after a 3-day latency period, were distracted at a rate of 0.25 mm twice daily for 6 days followed by a 2- or 4-week consolidation period. Distraction regenerate was harvested after the latency period, days 2, 4, or 6 of distraction, and after 2 or 4 weeks of consolidation and processed for Northern analysis (n = 4 at each time point) and immunohistochemical localization of TGF-beta1 (n = 2 at each time point). Six sham-operated animals (i.e., skin incision without osteotomy) were also killed (immediately postoperatively), and the mandibles were harvested and prepared in a similar fashion. Equal loading and transfer of RNA for Northern analysis was ensured by stripping and probing membranes with a probe against GAPDH (a housekeeping gene). Our results demonstrate that the spatial and temporal patterns of TGF-beta1 mRNA expression and protein production coincide with osteoblast migration, differentiation, and extracellular matrix synthesis. In addition, we demonstrate that TGF-beta1 production may be an important regulator of vasculogenesis during mandibular distraction osteogenesis. Finally, we have shown that osteocalcin gene expression coincides temporally with mineralization during rat mandibular distraction osteogenesis

The etiology of craniosynostosis is unknown. The elucidation of the biological pathways responsible for this disorder has been hampered by an inability to evaluate cranial sutures before, during, and after cranial suture fusion. The programmed fusion of the rat posterofrontal (PF) suture postnatally provides an excellent model to study the molecular events that occur during cranial suture fusion. Previous experiments have implicated transforming growth factor beta (TGF-beta) growth factors in the regulation of PF suture fusion. The purpose of these experiments was to localize the expression of high-affinity receptors for these growth factors during cranial suture fusion. Four rats were sacrificed on postnatal days 8, 12, 17, and 40 (N = 16). The PF and sagittal sutures were harvested and prepared for immunohistochemical localization of TGF-beta receptor 1 and receptor 2 (Tbeta-RI, Tbeta-RII) protein. Results indicate that immunostaining for Tbeta-RI and Tbeta-RII is markedly increased in the dura mater and osteoblasts of the sutural margin of the PF suture during active suture fusion (on postnatal days 12, 17, and 40) compared with the osteoblasts and dura mater underlying the patent sagittal suture. These results, in combination with the authors' previous findings as well as studies supporting a role for TGF-beta molecules in the regulation of osteogenesis, implicate TGF-beta signaling in the regulation of suture fusion. The possible mechanisms of ligand-receptor interaction are discussed

The role of transforming growth factor beta (TGF-beta) in the regulation of cranial suture fusion has been studied by various qualitative techniques such as in situ hybridization and immunohistochemistry. Although the relative expression of TGF-beta isoforms has been assessed in these studies, increased expression of TGF-beta has not been demonstrated in a quantitative fashion. Therefore, the purpose of this study was to quantify TGF-beta production by fusing (posterofrontal [PF]) and nonfusing (sagittal) mouse sutures using two different quantitative TGF-beta assays. The PF and sagittal sutures of 25-day-old mice were harvested and cultured separately in vitro. Culture media conditioned for 48 hours were collected after 3, 6, 9, 12, 15, 18, 21, 24, 27, and 30 days of culture, and total TGF-beta production was assessed using a TGF-beta bioassay. For a quantitative TGF-beta1 immunoassay, media conditioned for 48 hours were collected after 3, 5, 7, 9, 14, 22, and 28 days of culture. The TGF-beta bioassay revealed large amounts of total TGF-beta activity in both PF and sagittal sutures during the first week of culture, with decreasing amounts thereafter. Absolute TGF-beta activity in conditioned media collected from PF sutures at several early time points was higher than those obtained from sagittal sutures; however, these differences were not statistically significant. The results of the TGF-beta1 immunoassay (enzyme-linked immunosorbent assay) were similar to the bioassay in that the highest TGF-beta1 levels were noted during the first week of culture period and decreased thereafter. Analysis of variance of these samples, however, revealed significantly more TGF-beta1 protein in samples collected from the PF suture compared with the sagittal suture on days 3 and 5 of culture (p < 0.05). TGF-beta1 levels in the conditioned media obtained from PF sutures remained elevated compared with the sagittal suture on days 7 and 9; however, these differences were not statistically significant. Increased production of TGF-beta in the conditioned media of fusing PF sutures is the first such quantitative demonstration of growth factor upregulation during suture fusion and supports the hypothesis that TGF-beta expression may be important in cranial suture fusion

A significant body of literature supports a role for the dura mater underlying cranial sutures in the regulation of sutural fate. These studies have implicated regional differentiation of the dura mater based on association with fusing and patent rat cranial sutures. The purpose of these experiments was to isolate and characterize dural cells associated with fusing (posterior frontal) and patent (sagittal) rat cranial sutures. Six-day-old rats were killed, and the dura mater underlying the posterior frontal and sagittal sutures was harvested. Dural cells were briefly trypsinized and allowed to reach confluence. Two litters (10 animals per litter) were used for each set of experiments. Cells were harvested after the first and fifth passages for analysis of vimentin and desmoplakin expression (characteristic of human meningeal cells), cellular proliferation, density at confluence (a measure of cellular contact inhibition), and alkaline phosphatase production. In addition, bone nodule formation and collagen I production were analyzed in first passage cells. The results indicate that suture-derived dural cells can be established and that these cells coexpress vimentin and desmoplakin. In addition, it is demonstrated that first-passage sagittal suture-derived dural cells proliferate significantly faster and have decreased cellular contact inhibition than posterior frontal suture-derived cells (p < 0.01). Finally, it is shown that suture-derived dural cells have osteoblast-like properties, including alkaline phosphatase production, collagen I expression, and bone nodule formation in vitro. The possible mechanisms by which regional differentiation of suture-derived dural cells occur are discussed

The etiology of keloid formation is diverse. They are characterized grossly as thick scar tissue that extends beyond the boundaries of the original wound. Histologically, keloids are composed of excessive collagen with an abnormally large number of partially or totally occluded microvessels. This occlusion of keloid microvessels has been hypothesized to contribute to a hypoxic microenvironment within these pathological scars. Vascular endothelial growth factor (VEGF), a potent endothelial cell mitogen, and proangiogenic cytokine have been implicated in normal and pathological wound healing. The purpose of this study was to evaluate the amount of VEGF protein production by fibroblast cell lines derived from keloids and normal human dermal skin in hypoxic compared with normoxic culture conditions. By enzyme-linked immunosorbent protein assay, VEGF was increased in both keloid and normal human dermal fibroblasts in hypoxia over normoxic controls. There was not, however, a significant difference between upregulation of VEGF protein when comparing the keloid and normal fibroblast groups. As the result of the data, alternative hypotheses for hypoxia-induced keloid formation were explored: (1) downstream modulation or signal transduction of VEGF, (2) VEGF production from cells other than fibroblasts, (3) the importance of matrix accumulation stimulated by hypoxia, or (4) increased migration of cells (other than fibroblasts) specific to keloid biology. These hypotheses may help explain the possible role of hypoxia in the pathogenesis of keloid formation. Future studies involving in situ hybridization or immunohistochemical analysis may offer greater insight into the mechanisms underlying keloid formation. Ultimately, our therapeutic goal is the utilization of biomolecular approaches for the suppression of keloid formation

To date, many efforts to engineer cartilage have focused on matrix construction with the goal of producing a durable construct as cartilage replaces the resorbing matrix. However, the importance of matrix construction is at least matched by the challenge of efficient chondrocyte extraction, culture expansion, and prevention of dedifferentiation. This challenge is underscored by the large number of chondrocytes needed for a clinically significant construct such as an ear. Because human rib provides a large, readily available source of hyaline cartilage, the authors evaluated human rib chondrocyte extraction and found that maximum viable cell yield occurred after a 6-hour digestion. They also evaluated human microtic auricular remnant chondrocyte extraction and identified fibroblast contamination as a shortcoming of this potential source of chondrocytes. Initially, rib chondrocytes proliferated in vitro with a doubling time of approximately 1 week. As the cells were passaged, proliferation decreased such that the cells stopped proliferating and adopted a large, spindle-shaped morphology by passage 6. Interestingly, no increase in proliferation was noted when rib chondrocytes were stimulated with transforming growth factor beta 1, bone morphogenetic protein 2, and basic fibroblast growth factor. The major obstacles to the use of autologous rib chondrocytes in matrix construction are the low cell yield from a small piece of rib and the limited proliferation that these cells will undergo in vitro. Further investigation of culture systems and mitogenic cytokines may help resolve these limitations

Poorly healing mandibular osteotomies can be a difficult problem in reconstructive surgery. Many therapies have been attempted to augment the healing of mandibular fractures, defects, or osteotomies, but these methods have substantial drawbacks or have been ineffective. The difficulty in treating poorly healing bony defects has led to the exploration of gene therapy as a possible approach to supplement or accelerate mandibular fracture healing. To understand at what point the introduction of a suitable gene candidate might be of benefit in mandibular healing, it is imperative to examine the temporal expression of bone growth factors in a model of membranous bone healing. Insulinlike growth factors (IGFs) I and II are two such bone growth factor candidates because of their known potent in vitro as well as in vivo effects on bone formation. In this study the authors demonstrate the temporal pattern of IGF I and IGF II gene expression during mandibular osteotomy healing using a rat model. Their data reveal that IGF I and IGF II were elevated 7 days after a mandibular osteotomy that was held in external fixation. The upregulation of IGF I and IGF II during mandibular bone healing underscores the importance of these growth factors in bone repair. Gene therapy utilizing recombinant viral constructs containing IGFs I and II may be of benefit during mandibular bone healing in an effort to augment clinical scenarios of poor or retarded bony repair

Angiogenesis is essential to both normal and pathological bone physiology. Vascular endothelial growth factor (VEGF) has been implicated in angiogenesis, whereas transforming growth factor-beta1 (TGF-beta1) modulates bone differentiation, matrix formation, and cytokine expression. The purpose of this study was to investigate the relationship between TGF-beta1 and VEGF expression in osteoblasts and osteoblast-like cells. Northern blot analysis revealed an early peak of VEGF mRNA (6-fold at 3 h) in fetal rat calvarial cells and MC3T3-E1 osteoblast-like cells after stimulation with TGF-beta1 (2.5 ng/ml). The stability of VEGF mRNA in MC3T3-E1 cells was not increased after TGF-beta1 treatment. Actinomycin D inhibited the TGF-beta1-induced peak in VEGF mRNA, whereas cycloheximide did not. Blockade of TGF-beta1 signal transduction via a dominant-negative receptor II adenovirus significantly decreased TGF-beta1 induction of VEGF mRNA. Additionally, TGF-beta1 induced a dose-dependent increase in VEGF protein expression by MC3T3-E1 cells (P < 0.01). Dexamethasone similarly inhibited VEGF protein expression. Both TGF-beta1 mRNA and VEGF mRNA were concurrently present in rat membranous bone, and both followed similar patterns of expression during rat mandibular fracture healing (mRNA and protein). In summary, TGF-beta1-induced VEGF expression by osteoblasts and osteoblast-like cells is a dose-dependent event that may be intimately related to bone development and fracture healing