Relative Functions of Gαs and its Extra Large Variant XLαs in the Endocrine System

In addition to Gαs, GNAS encodes several other transcripts that show parent-of-origin specific expression. Among those, XLαs is a variant of Gαs derived from a distinct promoter and an upstream first exon that splices onto exons 2–13 of Gαs. Thus, XLαs is identical to Gαs over a long stretch of C-terminal amino acids. XLαs is derived exclusively from the paternal allele. In contrast, Gαs is expressed predominantly from the maternal allele in some tissues, including the renal proximal tubule, thyroid, and pituitary. XLαs knockout mice and some children with paternal GNAS deletions show severe early postnatal defects involving suckling defects, hypoglycemia, reduced adiposity, and impaired endocrine response to hypoglycemia. Crossed to an out-bred strain, XLαs knockout mice have a ˜20% survival rate and display a hypermetabolic, lean phenotype. Disease-causing GNAS mutations that are on the paternal allele also affect XLαs in most cases, and therefore, changes in XLαs activity are predicted to contribute to the pathogenesis of various diseases. Particularly, inactivating GNAS mutations causing progressive osseous heteroplasia, a disorder of severe ectopic bone formation, are almost exclusively present on the paternal GNAS allele. Despite these findings showing that XLαs is a critical protein playing roles in early postnatal adaptation to feeding, glucose and energy metabolism, and likely, mesenchymal stem cell fate, the cellular roles of XLαs remain almost completely unknown. We and others have shown that it can stimulate cAMP formation when overexpressed in cultured cells. On the other hand, data from mouse models indicate that XLαs and Gαs differ from each other importantly regarding their cellular roles, and in fact, some data indicate that XLαs actions may oppose Gαs actions. To gain novel insights into the cellular actions of XLαs and clarify these conflicting data, we performed a series of studies using cell culture and transgenic mice in which XLαs expression is targeted to the renal proximal tubule in vivo. Furthermore, we generated a mouse model of pseudohypoparathyroidism type-Ib, in which XLαs is biallelically expressed in addition to other molecular defects observed in patients with this disorder. Our investigations show that the transgenic overexpression of XLαs in the proximal tubule enhances basal cAMP signaling pathways and Gαs-mediated responses in this tissue, indicating that XLαs can mimic Gαs in vivo. On the other hand, the hypocalcemia observed in our pseudohypoparathyroidism type-Ib mouse model was rescued when these mice were crossed to XLαs knockout mice, indicating that increased XLαs expression due to loss of imprinting contributes to the hypocalcemia caused by Gαs deficiency and parathyroid hormone resistance. Moreover, our cell culture studies indicate that XLαs trafficking is significantly different from Gαs trafficking upon activation. These findings indicate that XLαs can mediate the signaling of some hormones, such as parathyroid hormone, differently from Gαs, and that in certain cases, it antagonizes the signaling mediated by the latter. It will be important to determine the cellular actions of XLαs relative to Gαs to improve our understanding of cAMP signaling pathways and the pathogenesis of diseases caused by GNAS mutations.