In vitro characterization of human salivary gland stem/progenitor cells (SGSCs) in comparison to two immortalized lines (HSG and HSY)? ? Previously culture conditions were established for clonal expansion of cells in low calcium medium, which lead to the expression of mesenchymal markers (epithelial-mesenchymal transformation, EMT), and re-epithelialization in normal calcium medium, leading to partial regain of ductal and acinar markers (mesenchymal-epithelial transformation, MET). Cells expanded in low calcium medium and replated in Matrigel formed 3D structures that expressed some but not all ductal markers, and some, but not all acinar markers. ? ? This year, primary cells grown under low calcium conditions were compared to two immortalized cell lines (HSG and HSY), which are the only lines available for the study of salivary differentiation, in order to gain some insight into ways that we can modify our conditions to get more complete re-establishment of the ductal and acinar morphology and markers. Unlike SGSCs grown in low calcium medium, HSG and HSY do not express mesenchymal markers. All three cells types were found to express the ductal marker, cytokeratin 19, and acinar markers, aquaporin 5, chromogranin B and amylase. When plated into 50% Matrigel, all three cell types were found to form 3D aggregates, and aquaporin 5 was dramatically upregulated. Based on a recent publication suggesting that acinar structures are better supported by type I collagen, similar experiments were performed using rat tail collagen gels, however, this culture conditions did not support the formation of 3D structures of any of the cell types. ? ? The cell surface markers that are expressed by the cells grown in low calcium and high calcium medium are being examined. Based on recent studies that pericytes that express CD146 may represent a source of local progenitor cells, we examined salivary glands by immunohistochemistry, and clonal lines by FACS for CD146 expression. As expected CD146 cells were observed surrounding blood vessels in the intact gland. In newly established clonal lines and in freshly isolated cells, CD146 is highly expressed. Current studies are aimed at determining if pre-sorting for CD146 gives rise to clonal populations of cells that can more efficiently undergo EMT and then MET in vitro. ? ? In vivo transplantation? ? A method for transplanting ex vivo expanded cells subcutaneously into immunocompromised mice in conjunction with a mixture of Matrigel with polyglycolic acid polymer (PGA) was established, and it was found that mini structures with ductile elements (cytokeratin 19 positive) surrounded by clusters of cells with acinar markers (aquaporin and amylase) were formed. In order to determine if these structures could be functional, cells were placed with the Matrigel/PGA scaffold into damaged salivary glands (surgical wound). Again, the human SGSCs produced small gland-like structures and RT-PCR for human markers showed expression of acinar and ductal elements. However, human salivary proteins, histatin or chromogranin B, were not detected in the mouse saliva, indicating that a functional connection was not made between the remaining mouse salivary gland and the human transplant. Interestingly, while HSG and HSY transplants did express ductal and acinar markers, their organization was quite different from that of the SGSCs. They formed carcinoma-like masses of cells. Cell transplantation was also performed with type I collagen gels, but none of the three cells types grew within these constructs. ? ? Based on these results, it is apparent that better ways are needed to encourage MET to form functional ductal and acinar elements in the appropriate architecture.
