This project focuses on identifying and characterizing novel molecules and new mechanisms underlying craniofacial development and their relevance to tissue engineering, with particular focus on salivary and neural crest development. We are addressing the following major questions: 1. How do embryonic salivary glands and other tissues generate their large epithelial surface areas using the process of branching morphogenesis? Specifically, how is cleft formation to delineate buds mediated and regulated? How can we facilitate bioengineering for organ replacement, particularly of salivary glands, by understanding branching morphogenesis and by developing reconstitution approaches? 2. What are the relative roles of the regulation of extracellular matrix, signaling, and selective gene expression versus cell movements in branching morphogenesis and in other major tissue rearrangements such as cranial neural crest development? We are applying a variety of approaches to begin to answer these complex questions, including: laser microdissection;gene expression profiling;RNA interference;whole-embryo, organ, and cell culture;confocal immunofluorescence and brightfield time-lapse microscopy;and a variety of functional inhibition and reconstitution approaches. Replacement of salivary gland function destroyed by radiation therapy for oral cancer or by Sjogren's Syndrome will require restoring enough secretory epithelium to produce adequate volumes of salivary fluid to alleviate xerostomia (salivary hypofunction). This general biological problem of producing large secretory epithelial surface areas in compact organs is solved during embryonic development by the process of branching morphogenesis. During development, a single embryonic bud first develops clefts and buds. It then undergoes repetitive branching to provide the large surface areas needed for effective secretory output. Regardless of whether eventual clinical replacement will involve salivary regeneration or a bioartificial salivary gland, a major challenge is how to create numerous branched epithelial structures. We have been applying a variety of approaches to identify novel mechanisms underlying this process, with a particular focus on extracellular matrix-cell interactions and dynamic movements of both cells and extracellular matrix that drive branching. Our previously published studies established that local, transient fibronectin gene expression is crucial for successful branching of multiple organs including salivary gland and lung, and that salivary branching morphogenesis involves a high level of relatively random cell motility combined with inward progression of fibronectin at forming clefts. However, it was not clear how fibronectin could function to form clefts because it is a non-motor matrix protein. Hypothesizing that another gene was involved, we used laser microdissection/T7-SAGE to identify a gene of unknown function expressed differentially in cells adjacent to clefts but not in buds. Although we have termed this molecule cleftin for convenience, its existing database name is BTB domain-containing gene 7 (Btbd7). We have found that Btbd7/cleftin is developmentally regulated, with striking localization of its mRNA in epithelial cells at the base of extending clefts. Gene silencing experiments using multiple different siRNAs show that Btbd7 is needed for branching morphogenesis of both salivary gland and lung. We have identified the following novel regulatory and morphogenetic pathway: Fibronectin induces Btbd7, which in turn induces Snail2 (Slug) and decreases protein levels of E-cadherin, resulting in cell separation and scattering that promotes inter-epithelial cell gap formation and cleft progression. Specifically, incubation with purified fibronectin induced Btbd7 expression within 20 minutes in salivary epithelia. Btbd7 induced the epithelial-mesenchymal transition mediator Snail2 (Slug) in a tetracycline-regulated expression system using MDCK cells, with minimal or no induction of Snail1 and Twist. Fibronectin also induced Snail2 in salivary epithelia within 1 hour. Because siRNA for Btbd7 blocked Snail2 induction, but siRNA for Snail2 blocked only Snail2 induction with no effect on Btbd7 induction, Btbd7 gene expression is upstream of Snail2. Btbd7 over-expression also decreased protein levels of E-cadherin and loosened MDCK cell-cell associations in 2D and 3D in vitro cultures to induce cell scattering. These findings have been clarified by a close examination of epithelial cell morphology and dynamics during salivary gland branching morphogenesis. Cells at the periphery of salivary gland buds are known to be organized into a semi-columnar epithelium. However, we found that the cells located at the base of developing clefts in contact with fibronectin changed their morphology to become more disorganized and irregularly shaped, suggesting a more motile phenotype. In fact, treatment of salivary epithelial rudiments with purified fibronectin closely mimicked this phenotypic change, and these morphological changes were blocked by Btbd7 siRNA. Direct visualization of cleft formation in living glands using labeled high-molecular weight dextran and GFP-expressing transgenic mice revealed that clefts advance in a discontinuous manner by the formation of transient gaps between cells near the base of forming clefts. This focal loss of cell-cell adhesion between the epithelial cells allows local inward progression of clefts. Based on all of these findings, we propose that fibronectin induces Btbd7 to promote cell shape change, separation, and cleft propagation in branching morphogenesis. The question of whether Btbd7 is a molecule that is not only required for branching morphogenesis but might actively induce this process was addressed by microinjecting a Btbd7-GFP construct in an adenoviral shuttle vector. We observed a modest but statistically significant stimulation of branching morphogenesis by over-expression of Btbd7 after adenoviral microinjection. An inter-laboratory collaborative project in the NIDCR Division of Intramural Research termed the Salivary Gland Molecular Anatomy Project has generated comprehensive gene expression data for developing mouse salivary glands. Spatial and temporal gene expression data from microarrays are now available online at the website: http://sgmap.nidcr.nih.gov/. This collaborative project involved temporal expression data from the research group of Matthew Hoffman, bioinformatics and website establishment by Zheng Wei in the DIR Developmental Mechanisms Unit, and contributions of microanatomical gene expression data from 14 microdissected epithelial sites/ages using 84 microarrays by Kurt Musselmann and co-workers in our group. Craniofacial bones, teeth, and surrounding connective tissues are derived primarily from specific cell populations in cranial neural folds. During embryonic development, the neural folds are formed at the boundary of epidermal ectoderm and neural ectoderm, and they produce the cranial neural crest and olfactory epithelium. Although well-known regulatory factors such as FGF, WNT, and BMP are involved in neural crest and olfactory epithelium formation, little is known about the roles of extracellular matrix proteins in these important in vivo processes. We hypothesized that one or more extracellular matrix proteins play a regulatory role in these developmental processes. Using cDNA microarray analysis, we compared forming cranial neural folds and ventral neural ectoderm to identify two dozen genes expressed 10-fold or higher in newly forming chick cranial neural crest compared to neural tube. We are focusing on strongly differentially expressed extracellular molecules that may help regulate growth factor regulators of neural crest development.
Showing the most recent 10 out of 21 publications
