Faculty Bibliography

Mandibular fractures, resulting from either trauma or reconstructive surgery, can be challenging craniofacial problems. The morbidity of failed fracture healing is significant and may require bone grafting. Donor site morbidity and finite amounts of autogenous bone are major drawbacks of autogenous bone grafting. Similarly, the use of allografts and xenografts may be associated with an increased risk of rejection, infection, and nonunion. To circumvent the limitations of bone grafting, research efforts have focused on formulating a suitable bone substitute. The purpose of our study was to evaluate the efficacy of type I collagen implants in repairing critical sized mandibular defects in rats. Twelve male Sprague-Dawley rats (200-300g) were divided equally into control and experimental groups. Full thickness, round, four millimeter in diameter defects were created in the ramus of the right mandible of all rats using an electrical burr at low speed. The defects were irrigated of all bone chips, and either filled with a precisely fitted disk of allogenic collagen type I gel (experimental animals) or left empty (control animals). Animals were killed 6 weeks after surgery and healing of the bone defects was assessed in a blinded fashion using radiologic and histologic analysis. Radiologic analysis of the control group revealed a clear circular right mandibular defect in all animals, whereas the collagen disk implant group revealed an indistinct to nonexistent right mandibular defect in all animals. Densitometric analysis revealed a significant difference between these groups (* P = 0.01). Similarly, gross analysis of control mandibles revealed a 4mm round, soft-tissue filled defect, while implanted defects demonstrated gross bone spanning the defect. Finally, histologic analysis of all control mandibles revealed clearly demarcated bony edges at the defect border with connective tissue spanning the defect. In contrast, histological analysis of all implanted mandibles revealed indistinct bony edges at the defect border with a thin layer of osteoblasts and viable bone spanning the defects. We have demonstrated the ability of type I collagen to promote healing of a membranous bony defect that would not otherwise heal at 6 weeks. The suitability of type I collagen as a carrier matrix provides ample opportunity for tissue-engineered approaches to further facilitate bony defect healing. Promoting bone formation through tissue engineering matrices offers great promise for skeletal healing and reconstruction.

Vascular disruption secondary to fracture creates a hypoxic gradient of injury wherein the oxygen tension at the center of the wound is very low. In vivo this hypoxic microenvironment stimulates the expression of a variety of cytokines from inflammatory cells, fibroblasts, endothelial cells, and osteoblasts. In order to begin to dissect this complex system, we have examined the effects of hypoxia on isolated osteoblast gene expression in vitro. Understanding gene expression in this system may facilitate the development of targeted therapeutic modalities designed to accelerate fracture repair and reduce complications. Using an established model of in vitro hypoxia, we have analyzed the expression of genes involved in bone matrix production and turnover. Subconfluent neonatal rat calvarial osteoblasts were exposed to hypoxia (pO(2) = 35-40 mm Hg) and total cellular RNA was collected at 0, 3, 6, 24, and 48 h. Northern analysis was used to analyze the expression patterns of (1) transforming growth factors (TGFs)-beta1, -beta2, and -beta3 and their type I receptor; (2) collagens I and III; and (3) tissue inhibitor of metalloproteinase-1. We have demonstrated a marked elevation of TGF-beta1 gene expression within 3 h of hypoxia. Although neither TGF-beta2 nor TGF-beta3 expression was affected by hypoxia, the TGF-beta type I receptor was substantially upregulated within 6 h. In addition, extracellular matrix scaffolding molecules (collagens I and III) were markedly, but differentially, upregulated. Finally, we have demonstrated that the expression of an inhibitor of extracellular matrix turnover, the tissue inhibitor of metalloproteinase-1, was strikingly decreased in response to hypoxia. These results imply that hypoxia can affect osseous healing by altering the expression of cytokines, bone-specific extracellular matrix molecules, and their regulators

Angiogenesis, the formation of new blood vessels, is crucial to the process of fracture healing. Vascular disruption after osseous injury results in an acidic, hypoxic wound environment. We have previously shown that osteoblasts can produce vascular endothelial growth factor (VEGF) in response to a variety of stimuli. In this study we examined pH and lactate concentration, two components of the putative fracture extracellular microenvironment, and determined their relative contribution to regulation of rat calvarial osteoblast VEGF production under both normoxic and hypoxic conditions. Our results demonstrate that pH and lactate concentration do independently affect osteoblast VEGF mRNA and protein production. Acidic pH (7.0) significantly decreased VEGF production, under normoxic and hypoxic conditions (P < 0.05), compared with neutral pH (7.4). This decrease was primarily transcriptionally regulated, because the rate of VEGF mRNA degradation was unchanged at pH 7.0 vs. 7.4. Similarly, an elevated lactate concentration (22 mM) also depressed osteoblast elaboration of VEGF at both neutral and acidic pH (P < 0.001). Furthermore, the effects of increasing acidity and elevated lactate appeared to be additive

The well-described detrimental effects of ionizing radiation on the regeneration of bone within a fracture site include decreased osteocyte number, suppressed osteoblast activity, and diminished vascularity. However, the biologic mechanisms underlying osteoradionecrosis and the impaired fracture healing of irradiated bone remain undefined. Ionizing radiation may decrease successful osseous repair by altering cytokine expression profiles resulting from or leading to a change in the osteoblastic differentiation state. These changes may, in turn, cause alterations in osteoblast proliferation and extracellular matrix formation. The purpose of this study was to investigate the effects of ionizing radiation on the proliferation, maturation, and cytokine production of MC3T3-E1 osteoblast-like cells in vitro. Specifically, the authors examined the effects of varying doses of ionizing radiation (0, 40, 400, and 800 cGy) on the expression of transforming growth factor-beta1 (TGF-beta1), vascular endothelial growth factor (VEGF), and alkaline phosphatase. In addition, the authors studied the effects of ionizing radiation on MC3T3-E1 cellular proliferation and the ability of conditioned media obtained from control and irradiated cells to regulate the proliferation of bovine aortic endothelial cells. Finally, the authors evaluated the effects of adenovirus-mediated TGF-beta1 gene therapy in an effort to 'rescue' irradiated osteoblasts. The exposure of osteoblast-like cells to ionizing radiation resulted in dose-dependent decreases in cellular proliferation and promoted cellular differentiation (i.e., increased alkaline phosphatase production). Additionally, ionizing radiation caused dose-dependent decreases in total TGF-beta1 and VEGF protein production. Decreases in total TGF-beta1 production were due to a decrease in TGF-beta1 production per cell. In contrast, decreased total VEGF production was secondary to decreases in cellular proliferation, because the cellular production of VEGF by irradiated osteoblasts was moderately increased when VEGF production was corrected for cell number. Additionally, in contrast to control cells (i.e., nonirradiated), conditioned media obtained from irradiated osteoblasts failed to stimulate the proliferation of bovine aortic endothelial cells. Finally, transfection of control and irradiated cells with a replication-deficient TGF-beta1 adenovirus before irradiation resulted in an increase in cellular production of TGF-beta1 protein and VEGF. Interestingly, this intervention did not alter the effects of irradiation on cellular proliferation, which implies that alterations in TGF-beta1 expression do not underlie the deficiencies noted in cellular proliferation. The authors hypothesize that ionizing radiation-induced alterations in the cytokine profiles and differentiation states of osteoblasts may provide insights into the cellular mechanisms underlying osteoradionecrosis and impaired fracture healing

Although it is one of the most commonly occurring craniofacial congenital disabilities, craniosynostosis (the premature fusion of cranial sutures) is nearly impossible to prevent because the molecular mechanisms that regulate the process of cranial suture fusion remain largely unknown. Recent studies have implicated the dura mater in determining the fate of the overlying cranial suture; however, the molecular biology within the suture itself has not been sufficiently investigated. In the murine model of cranial suture fusion, the posterior frontal suture is programmed to begin fusing by postnatal day 12 in rats (day 25 in mice), reliably completing bony union by postnatal day 22 (day 45 in mice). In contrast, the sagittal suture remains patent throughout the life of the animal. Using this model, this study sought to examine for the first time what differences in gene expression--if any--exist between the two sutures with opposite fates. For each series of experiments, 35 to 40 posterior frontal and sagittal suture complexes were isolated from 6-day-old Sprague-Dawley rat pups. Suture-derived cell cultures were established, and ribonuicleic acid was derived from snap-frozen, isolated suture tissue. Results demonstrated that molecular differences between the posterior frontal and sagittal suture complexes were readily identified in vivo, although these distinctions were lost once the cells comprising the suture complex were cultured in vitro. Hypothetically, this change in gene expression resulted from the loss of the influence of the underlying dura mater. Significant differences in the expression of genes encoding extracellular matrix proteins existed in vivo between the posterior frontal and sagittal sutures. However, the production of the critical, regulatory cytokine transforming growth factor beta-1 was equal between the two suture complexes, lending further support to the hypothesis that dura mater regulates the fate of the overlying cranial suture

The ability of immature animals and newborns to orchestrate successful calvarial reossification is well described. This capacity is markedly attenuated in mature animals and in humans greater than 2 years of age. Previous studies have implicated the dura mater as critical to successful calvarial reossification. The authors have previously reported that immature, but not mature, dural tissues are capable of elaborating a high expression of osteogenic growth factors and extracellular matrix molecules. These findings led to the hypothesis that a differential expression of osteogenic growth factors and extracellular matrix molecules by immature and mature dural tissues may be responsible for the clinically observed phenotypes (i.e., immature animals reossify calvarial defects; mature animals do not). This study continues to explore the hypothesis through an analysis of transforming growth factor (TGF)-beta3, collagen type III, and alkaline phosphatase mRNA expression. Northern blot analysis of total RNA isolated from freshly harvested immature (n = 60) and mature (n = 10) dural tissues demonstrated a greater than three-fold, 18-fold, and nine-fold increase in TGF-beta3, collagen type III, and alkaline phosphatase mRNA expression, respectively, in immature dural tissues as compared with mature dural tissues. Additionally, dural cell cultures derived from immature (n = 60) and mature dura mater (n = 10) were stained for alkaline phosphatase activity to identify the presence of osteoblast-like cells. Alkaline phosphatase staining of immature dural cells revealed a significant increase in the number of alkaline phosphatase-positive cells as compared with mature dural tissues (p < 0.001). In addition to providing osteogenic humoral factors (i.e., growth factors and extracellular matrix molecules), this finding suggests that immature, but not mature, dura mater may provide cellular elements (i.e., osteoblasts) that augment successful calvarial reossification. These studies support the hypothesis that elaboration of osteogenic growth factors (i.e., TGF-beta33) and extracellular matrix molecules (i.e., collagen type III and alkaline phosphatase) by immature, but not mature, dural tissues may be critical for successful calvarial reossification. In addition, these studies suggest for the first time that immature dural tissues may provide cellular elements (i.e., osteoblasts) to augment this process

Normal bone growth and repair is dependent on angiogenesis. Fibroblast growth factor-2 (FGF-2), vascular endothelial growth factor (VEGF), and transforming growth factor-beta (TGFbeta) have all been implicated in the related processes of angiogenesis, growth, development, and repair. The purpose of this study was to investigate the relationships between FGF-2 and both VEGF and TGFbeta in nonimmortalized and clonal osteoblastic cells. Northern blot analysis revealed 6-fold peak increases in VEGF mRNA at 6 h in fetal rat calvarial cells and MC3T3-E1 osteoblastic cells after stimulation with FGF-2. Actinomycin D inhibited these increases in VEGF mRNA, whereas cycloheximide did not. The stability ofVEGF mRNA was not increased after FGF-2 treatment. Furthermore, FGF-2 induced dose-dependent increases in VEGF protein levels (P < 0.01). Although in MC3T3-E1 cells, TGFbeta1 stimulates a 6-fold peak increase in VEGF mRNA after 3 h of stimulation, we found that both TGFbeta2 and TGFbeta3 yielded 2- to 3-fold peak increases in VEGF mRNA levels noted after 6 h of stimulation. Similarly, both TGFbeta2 and TGFbeta3 dose dependently increased VEGF protein production. To determine whether FGF-2-induced increases in VEGF mRNA may have occurred independently of TGFbeta, we disrupted TGFbeta signal transduction (using adenovirus encoding a truncated form of TGFbeta receptor II), which attenuated TGFbeta1 induction of VEGF mRNA, but did not impede FGF-2 induction ofVEGF mRNA. In summary, FGF-2-induced VEGF expression by osteoblastic cells is a dose-dependent event that may be independent of concomitant FGF-2-induced modulation of TGFbeta activity.

Poorly healing mandibular fractures and osteotomies can be troublesome complications of craniomaxillofacial trauma and reconstructive surgery. Gene therapy may offer ways of enhancing bone formation by altering the expression of desired growth factors and extracellular matrix molecules. The elucidation of suitable candidate genes for therapeutic intervention necessitates investigation of the endogenously expressed patterns of growth factors during normal (i.e., successful) fracture repair. Transforming growth factor beta1 (TGF-beta1), its receptor (Tbeta-RII), and the extracellular matrix proteins osteocalcin and type I collagen are thought to be important in long-bone (endochondral) formation, fracture healing, and osteoblast proliferation. However, the spatial and temporal expression patterns of these molecules during membranous bone repair remain unknown. In this study, 24 adult rats underwent mandibular osteotomy with rigid external fixation. In addition, four identically treated rats that underwent sham operation (i.e., no osteotomy) were used as controls. Four experimental animals were then killed at each time point (3, 5, 7, 9, 23, and 37 days after the procedure) to examine gene expression of TGF-beta1 and Tbeta-RII, osteocalcin, and type I collagen. Northern blot analysis was used to compare gene expression of these molecules in experimental animals with that in control animals (i.e., nonosteotomized; n = 4). In addition, TGF-beta1 and T-RII proteins were immunolocalized in an additional group of nine animals killed on postoperative days 3, 7, and 37. The results of Northern blot analysis demonstrated a moderate increase (1.7 times) in TGF-beta1 expression 7 days postoperatively; TGF-beta1 expression returned thereafter to near baseline levels. Tbeta-RII mRNA expression was downregulated shortly after osteotomy but then increased, reaching a peak of 1.8 times the baseline level on postoperative day 9. Osteocalcin mRNA expression was dramatically downregulated shortly after osteotomy and remained low during the early phases of fracture repair. Osteocalcin expression trended slowly upward as healing continued, reaching peak expression by day 37 (1.7 times the control level). In contrast, collagen type IalphaI mRNA expression was acutely downregulated shortly after osteotomy, peaked on postoperative days 5, and then decreased at later time points. Histologic samples from animals killed 3 days after osteotomy demonstrated TGF-beta1 protein localized to inflammatory cells and extracellular matrix within the fracture gap, periosteum, and peripheral soft tissues. On postoperative day 7, TGF-beta1 staining was predominantly localized to the osteotomized bone edges, periosteum, surrounding soft tissues, and residual inflammatory cells. By postoperative day 37, complete bony healing was observed, and TGF-beta1 staining was localized to the newly formed bone matrix and areas of remodeling. On postoperative day 3, Tbeta-RII immunostaining localized to inflammatory cells within the fracture gap, periosteal cells, and surrounding soft tissues. By day 7, Tbeta-RII staining localized to osteoblasts of the fracture gap but was most intense within osteoblasts and mesenchymal cells of the osteotomized bone edges. On postoperative day 37, Tbeta-RII protein was seen in osteocytes, osteoblasts, and the newly formed periosteum in the remodeling bone. These observations agree with those of previous in vivo studies of endochondral bone formation, growth, and healing. In addition, these results implicate TGF-beta1 biological activity in the regulation of osteoblast migration, differentiation, and proliferation during mandibular fracture repair. Furthermore, comparison of these data with gene expression during mandibular distraction osteogenesis may provide useful insights into the treatment of poorly healing fractures because distraction osteogenesis has been shown to be effective in the management of these difficult clinical cases

Vascular disruption secondary to fracture leads to a hypoxic zone of injury where the oxygen tension at the center of the wound is quite low. In this dynamic microenvironment, a number of growth factors are elaborated to stimulate the synthetic processes of fracture repair. Previously the authors have shown the hypoxia-induced increase of vascular endothelial growth factor expression in osteoblasts. The purpose of these experiments was to examine osteoblast expression of insulinlike growth factors (IGF) I and II--cytokines believed to play a role in increased collagen synthesis, chemotaxis, and proliferation of osteoblasts in response to hypoxia. Primary cell cultures of osteoblasts isolated from neonatal rat calvaria were subjected to hypoxia (PO2 = 35 mmHg) for 0, 3, 6, 24, and 48 hours. Northern blot analysis of ribonucleic acid (RNA) from resulting cultures demonstrated a more than 60% increase in IGF-II messenger RNA (mRNA) expression after 3 hours of hypoxia. IGF-II mRNA expression continued to increase through later time points to 200% and 260% of baseline at 24 and 48 hours respectively. In contrast, IGF-I demonstrated no significant change in mRNA expression compared with baseline control (normoxia) cultures. In these experiments the authors have demonstrated a hypoxia-induced increase in IGF-II but not IGF-I in primary osteoblasts. The differential expression of these two growth factors may underscore important differences in the behavior of osteoblasts in the hypoxic fracture microenvironment. Taken together, these data add additional support to the theory that hypoxia induces gene-specific changes in expression of molecules important to extracellular matrix formation for successful bone healing

Gene therapy has moved from the promise of laboratory investigation to the reality of clinical practice in just the last decade. Various methods for delivery of genes to host cells have been developed and utilized both in vitro and in vivo. From the perspective of the plastic surgeon, gene therapy holds the promise to augment healing in clinical situations that remain difficult to treat, such as chronic wounds, osteoradionecrosis, or possibly to expedite current clinical practices, such as distraction osteogenesis. The authors chose to investigate the potential for gene therapy in osseous tissues using a replication-deficient adenovirus vector to deliver the marker transgene beta-galactosidase. An adenovirus vector is ideal for use in situations in which transgene expression is desired for only a relatively short period of time, such as wound and fracture healing. Utilizing a rat mandibular osteotomy model, they demonstrated that, using an adenoviral vector, foreign genes can be delivered in a simple fashion and can be expressed in a reliable manner within and around the osteotomy site for at least 10 days. Furthermore, there was no evidence of transfection of distant tissues associated with local application of the adenovirus vector. With this information, clinicians may now attempt to deliver osteogenic and angiogenic genes in a site-specific fashion to improve and expedite osseous healing