In vivo disruption of PKA activity through targeted mutations of the RIα regulatory subunit

Abstract

The cAMP-dependent protein kinase (PKA) is expressed in all animal cells and plays a key regulatory role in many diverse physiological processes. The PKA holoenzyme is a heterotetramer that exists as a dimer of regulatory (R) subunits that serves to sequester two catalytic (C) subunits in the absence of cAMP. The R-subunit isoforms have the same general domain organization but are not functionally redundant and differ in their biochemical characteristics, expression levels and subcellular localization. The isolation of mutant R-subunits with useful experimental properties has allowed us to interrogate PKA-dependent signaling in vivo in a manner that cannot be accomplished using traditional knockout strategies. Genetic disruption of any of the R-subunits results in a compensatory increase in the RIα subunit, which occurs through reduced turnover of RIα protein when bound to C-subunit in the holoenzyme. This switch to the RIα holoenzyme can confound interpretation of experimental results as basal PKA activity levels are significantly elevated due to increased cAMP sensitivity of the RI-containing holoenzyme. Furthermore, any requirements for subcellular targeting of PKA must be closely scrutinized in light of the ability of a number of A-Kinase Anchoring Proteins (AKAPs) to bind RI, although for most AKAPs this occurs with greatly reduced affinity when compared to RII subunits. The specific goals of this thesis were to utilize two genetically modified mouse lines harboring mutations in distinct domains of the RIα subunit to: a) assess the specific role of PKA in fluid homeostasis by disrupting kinase activity through expression of a dominant-negative mutation (RIαB) and, b) examine the role of compartmentalized RIα signaling <italic>in vivo</italic> through the introduction of alanine substitutions (RIα<super>ALA</<super>) in the dimerization/docking (D/D) domain, an N-terminal hydrophobic sequence that is required for the vast majority of PKA-AKAP interactions and subcellular targeting of the holoenzyme. A point mutation (G324D) in the C-terminal cAMP-binding site renders RIαB (B site mutant) insensitive to physiological levels of cAMP while Cre-LoxP technology allowed us to target RIαB to distinct cell populations. Although PKA-dependent phosphorylation is required for apical trafficking and exocytosis of the renal water channel, AQP2, the precise role of PKA in short- and long-term regulation of AQP2 remains unclear. I found that RIαB expression in renal principal cells is sufficient to cause nephrogenic diabetes insipidus while dramatically decreasing total AQP2 protein levels. The effect of RIαB was independent of changes in AQP2 mRNA and persisted with both acute (DDAVP treatment) and chronic (24-hour dehydration) activation of the V2R, suggesting a previously undetermined role for PKA in regulating protein stability and whole cell abundance of AQP2. To assess the role of RIα-AKAP interactions <italic>in vivo</italic>, we introduced alanine mutations (V22A, I27A) into the D/D that have been shown to disrupt binding to dual-specific AKAPs <italic>in vitro</italic>. RIα<super>ALA</super> abolished binding to the RI-selective AKAP, SKIP; surprisingly, these mutations also resulted in dramatic increases in RIα protein in conjunction with apparent decreases in RIα mRNA. In mouse embryonic fibroblasts (MEFs), the protein stability of WT RIα was increased by proteasome inhibition whereas the elevated RIα<super>ALA</super> was unaffected suggesting that AKAP interactions may be a required part of the proteasome degradation pathway for RIα. RIα<super>ALA</super> expression in MEFs also attenuated CREB phosphorylation and induction of the early response gene, c-Fos in response to elevated cAMP. Mice expressing the RIα<super>ALA</super> mutation had reduced locomotor activity and c-Fos induction was reduced in the striatum in response to the dopamine type 2 receptor antagonist, haloperidol. This work has identified roles for PKA in regulating fluid homeostasis as well as a possible role for the D/D domain in the stability of RIα and its subsequent impact on cAMP-dependent transcriptional regulation.