The anterior pituitary gland comprises five types of endocrine cells plus non-endocrine cells including folliculostellate cells, endothelial cells, and capillary mural cells (pericytes). In addition to being controlled by the hypothalamic–pituitary–target organ axis, the functions of these cells are likely regulated by local cell and extracellular matrix (ECM) interactions. However, these complex interactions are not fully understood. We investigated folliculostellate cell-mediated cell-to-cell interaction. Using S100β-GFP transgenic rats, which express GFP in folliculostellate cells, we designed a three-dimensional cell culture to examine the effects of folliculostellate cells. Interestingly, removal of folliculostellate cells reduced collagen synthesis (Col1a1 and Col3a1). Because pericytes are important collagen-producing cells in the gland, we stained for desmin (a pericyte marker). Removal of folliculostellate cells resulted in fewer desmin-positive pericytes and less desmin mRNA. We then attempted to identify the factor mediating folliculostellate cell–pericyte interaction. RT-PCR and in situ hybridization revealed that the important profibrotic factor transforming growth factor beta-2 (TGFβ2) was specifically expressed in folliculostellate cells and that TGFβ receptor II was expressed in pericytes, endothelial cells, and parenchymal cells. Immunocytochemistry showed that TGFβ2 induced SMAD2 nuclear translocation in pericytes. TGFβ2 increased collagen synthesis in a dose-dependent manner. This action was completely blocked by TGFβ receptor I inhibitor (SB431542). Diminished collagen synthesis in folliculostellate cell-deficient cell aggregates was partially recovered by TGFβ2. TGFβ2-mediated folliculostellate cell–pericyte interaction appears to be essential for collagen synthesis in rat anterior pituitary. This finding sheds new light on local cell–ECM interactions in the gland.

The steroid and xenobiotic receptor (SXR) and its murine ortholog pregnane X receptor (PXR) are nuclear receptors that are expressed mainly in the liver and intestine where they function as xenobiotic sensors. In addition to its role as a xenobiotic sensor, previous studies in our laboratories and elsewhere have identified a role for SXR/PXR as a mediator of bone homeostasis. Here, we report that systemic deletion of PXR results in marked osteopenia with mechanical fragility in female mice as young as 4 months old. Bone mineral density (BMD) of PXR knockout (PXRKO) mice was significantly decreased compared with the BMD of wild-type (WT) mice. Micro-computed tomography analysis of femoral trabecular bones revealed that the three-dimensional bone volume fraction of PXRKO mice was markedly reduced compared with that of WT mice. Histomorphometrical analysis of the trabecular bones in the proximal tibia showed a remarkable reduction in bone mass in PXRKO mice. As for bone turnover of the trabecular bones, bone formation is reduced, whereas bone resorption is enhanced in PXRKO mice. Histomorphometrical analysis of femoral cortical bones revealed a larger cortical area in WT mice than that in PXRKO mice. WT mice had a thicker cortical width than PXRKO mice. Three-point bending test revealed that these morphological phenotypes actually caused mechanical fragility. Lastly, serum levels of phosphate, calcium, and alkaline phosphatase were unchanged in PXRKO mice compared with WT. Consistent with our previous results, we conclude that SXR/PXR promotes bone formation and suppresses bone resorption thus cementing a role for SXR/PXR as a key regulator of bone homeostasis.