Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Biological Sciences

Content Description

1 online resource (xiv, 202 pages) : illustrations (some color)

Dissertation/Thesis Chair

Melinda Larsen

Committee Members

Paolo Forni, Gabriella Fuchs, Morgan Sammons, Bijan Dey


BMP, Cell idenity, FGF, scRNA Sequencing, Stroma, Submandibular Gland, Epithelial cells, Salivary glands, Submandibular gland, Myofibroblasts, Morphogenesis

Subject Categories

Cell Biology | Developmental Biology | Molecular Biology


Organogenesis is the process organs go through where cellular communications coordinate all a developing organ needs. What organs need are more cells, in the right place, doing the right job. In the salivary gland, we know that stromal cells are important for organogenesis and that they coordinate the epithelium’s form and functions. However, specific stromal contributions have focused on epithelial quantity and placement. There is less information about how the stroma directs the epithelium towards certain functions. Here we used organoids as a model for understanding what stromal signaling directs epithelial cell fate. We found that stromal cell state is just as important as their signaling array. FGF2 signaling promotes an in-vivo like state in stromal cells that supports proacinar epithelium. One signaling factor from this stromal state that directs the epithelium are BMPs. Together, BMP and FGF2 signaling can replace the stroma and allows proacinar marker, AQP5, expression in the epithelium. However, the stroma also has a negative proacinar signal that happens when FGF2 is absent. Without FGF2, the stroma changes into myofibroblasts and appear inhibitory via TGFβ expression. This stromal toggling is an important biological aspect that needs controlling as stromal cells become clinically applied. We used these stromal cell states to validate artificial scaffolds meant for inducing salivary gland regeneration. We found that our artificial scaffolds alone could prevent the myofibroblast switch. The scaffold’s myofibroblast inhibition characteristics comparable with endogenous ECM. These data demonstrate that the scaffolds combined with properly maintained stroma could become a regenerative cell therapy.