Persistence of Cyclic AMP Signaling Determines Gs‐couple GPCR Linkage to ERK Activation

Abstract

The Gs‐coupled GPCR PAC1 mediates ERK activation in NS‐1 neuroendocrine cells exclusively through the novel cAMP effector NCS‐Rapgef2 (see Emery et al., Sci. Signal. 6, ra51, 2013; Emery et al., J. Biol. Chem. 289: 10125, 2014). Activation of ERK through this pathway is common to several other Gs‐GPCRs tested, as would be predicted from their ability to elevate cAMP levels. However not all Gs‐coupled GPCRs activate NCS‐Rapgef2 in NS‐1 cells. We have stably expressed a number of Gs‐coupled GPCRs in NS‐1s, at a level (via single‐copy gene transduction and expression from a moderately strong promoter) consistent with expression levels in cells in which their expression is endogenous, as evidenced by expression value (EV) of mRNA for these transcripts upon bulk‐cell transcriptomic analysis of transduced cell lines. Activation of either the beta‐1 or beta‐2 adrenoceptors with isoproterenol causes prolonged cAMP elevation, activation of Epac2 (leading to growth arrest), and activation of PKA (leading to CREB phosphorylation). However, only beta‐1 receptor activation in NS‐1 cells leads to ERK‐dependent neuritogenesis in these cells. The differential effects of beta‐1 and beta‐2 receptor activation in NS‐1 cells inheres in the C‐terminal half of the respective receptor proteins, as deduced from chimeric beta‐1/beta‐2 and beta‐2/beta‐1 receptor‐driven transformation of cell phenotype upon exposure to isoproterenol. To compare cAMP elevation dynamics of the two receptor types, HEK293 cells were transduced with a biosensor (split luciferase PKAR), allowing continuous real‐time cAMP measurement (Emery et al., Peptides 79: 39, 2016). In beta‐2 receptor‐expressing cells, the maximal effect of isoproterenol on cAMP was observed after 10 minutes of treatment and decreased rapidly to about 40% of maximal levels in the 40 min thereafter. In contrast, isoproterenol‐dependent beta‐1 receptor activation caused cAMP activation persisting at maximal levels for at least 40 minutes following agonist addition. The mode of ERK phosphorylation observed following beta‐2 activation (NCS‐Rapgef2‐dependent) results in gene transcription, including up‐regulation of Egr‐1, required for neuritogenesis (Ravni et al., Mol. Pharmacol. 73: 1688, 2008), while ERK activation elicited by beta‐2 receptor activation (NCS‐Rapgef2‐independent) does not have this signaling property. We have observed similar signaling properties, with respect to NCS‐Rapgef2 and ERK activation, for the D1 dopamine receptor, in mice, in vivo, where adenylate cyclase activation occurs through Golf rather than Gs (Jiang et al., eNeuro, ENEURP.0248‐17.2017). We are currently testing the working hypothesis that the Gs‐>NCS‐Rapgef2‐>ERK signaling pathway is exclusive for cAMP‐dependent ERK activation in neuronal and endocrine cells, to which Rapgef2 expression appears to be confined in adult mammals, and that this signaling pathway imparts signaling bias to a subset of Gs‐coupled GPCRs, based on discrimination of cAMP elevation dynamics following ligand engagement with receptor.Support or Funding InformationNIMH Intramural support to L.E.E. via MH002386.