We study the roles of multiple factors in the development, maintenance, and expansion of the various pancreatic cell populations, as well as in the function of these different cell types:
1. Hepatic nuclear factor 6 (Hnf6) is a transcription factor that is expressed in all pancreatic cell types early during embryonic development but gets "turned off" in the islet cell lineage prior to hormone expression. Maintenance of Hnf6 expression in islet cells results in islet malformations, poor insulin secretion, and diabetes in mice. To determine which genes are turned "on" or "off" inappropriately in islets over-expressing Hnf6, and what the function of these misregulated genes is in normal islet development and function, we isolated islets from wild-type and Hnf6 over-expressing mice and compared the genes expressed in these two cell populations using microarray analysis. Genes encoding molecules that are involved in glucose-stimulated insulin secretion, cell-cell communication, and cell adhesion were found to be altered. The expression of these genes during normal islet development is currently being analyzed, as is their role in islet formation and function using mutational or transgenic over-expression analysis.
We have also generated mice in which Hnf6 is deleted throughout the entire pancreas. These mice exhibit several abnormalities, including defects in duct differentiation, pancreatitis, and precancerous lesions. Additionally, we generated mice in which Hnf6 was specifically inactivated within endocrine progenitor cells. Loss of Hnf6 prior to differentiation lead to an increased likelihood that putative endocrine progenitors would instead become acinar cells or ducts. Thus, a threshold of Hnf6 must be reached to fully commit an endocrine progenitor to the endocrine lineage
2. Connective tissue growth factor (CTGF) is a secreted factor involved in wound healing, cell migration, and fibrotic disease. We found that CTGF expression was down-regulated in neonatal islets that over-express Hnf6 (see above). In the pancreas, CTGF is expressed in insulin-producing cells, ductal epithelium, and endothelial cells. We have used mice that lack CTGF expression globally to examine its role during islet development. CTGF+/- and CTGF-/- animals show increased numbers of glucagon-producing cells and altered islet architecture similar to that seen in the Hnf6 over-expressing animals. These data suggest that CTGF functions non-cell autonomously to regulate the proportion of different islet endocrine cell types. In addition, CTGF mutant animals show a dramatic decrease in embryonic beta cell proliferation.
Because CTGF-/- animals die at birth, we are currently generating mice with a pancreas-specific or islet-specific deletion of the CTGF gene in order to study the function of CTGF in the postnatal and adult pancreas. We are also investigating the role that CTGF may play in the signaling pathways that regulate the development of the early pancreatic bud, using an ex vivo culture method.
3. Forkhead box M1 (FoxM1) is a transcription factor that regulates cell cycle genes and is essential for regeneration of the liver following partial hepatectomy. We found that mice lacking FoxM1 function throughout their pancreas exhibit reduced beta cell proliferation after birth. However, FoxM1 is dispensable for embryonic beta cell proliferaion. Using a model in which a portion of the pancreas is surgically removed (partial pancreatectomy) and then allowed to regenerate, we have also determined that FoxM1 is not required for beta cell neogenesis, but is required for subsequent islet growth. Studies are currently underway to investigate FoxM1's potential role in regulating beta cell differentiation and function.
During periods of increased insulin demand, such as obesity and pregnancy, it is known that the beta cell mass expands in a compensatory fashion. We found that female mice lacking FoxM1 throughout their pancreas exhibit an increased susceptibility to developing gestational diabetes. Additionally, when placed on a high-fat/high-carbohydrate diet, which induces obesity and insulin resistance, the pancreas-wide FoxM1 knockout mice are more susceptible to developing glucose intolerance and diabetes than are wild-type mice. We will next investigate whether the loss of FoxM1 solely impairs the beta cell's ability to proliferate, or whether impaired beta cell function also contributes to this phenotype.
We are also interested in identifying the signaling pathways that may be regulating FoxM1's activity within the beta cells. It is known that many growth factors (e.g. placental lactogen, insulin-like growth factor-1, and hepatocyte growth factor) can stimulate beta cell proliferation in vivo as well as in vitro. We will investigate whether FoxM1 is downstream of these growth factor signaling pathways using an in vitro mouse islet culture system as well as in vivo transgenic mouse models over-expressing each of these growth factors on the pancreas-wide FoxM1 knockout background.