Isoform-specific C-terminal phosphorylation drives autoinhibition of Casein Kinase 1

Abstract

Casein kinase 1 {delta} (CK1{delta}) controls essential biological processes including circadian rhythms and Wnt signaling, but how its activity is regulated is not well understood. CK1{delta} is inhibited by autophosphorylation of its intrinsically disordered C-terminal tail. Two CK1 splice variants, {delta}1 and {delta}2, are known to have very different effects on circadian rhythms. These variants differ only in the last 16 residues of the tail, referred to as the extreme C-termini (XCT), but with marked changes in potential phosphorylation sites. Here we test if the XCT of these variants have different effects in autoinhibition of the kinase. Using NMR and HDX-MS, we show that the {delta}1 XCT is preferentially phosphorylated by the kinase and the {delta}1 tail makes more extensive interactions across the kinase domain. Mutation of {delta}1-specific XCT phosphorylation sites increases kinase activity both in vitro and in cells and leads to changes in circadian period, similar to what is reported in vivo. Mechanistically, loss of the phosphorylation sites in XCT disrupts tail interaction with the kinase domain. {delta}1 autoinhibition relies on conserved anion binding sites around the CK1 active site, demonstrating a common mode of product inhibition of CK1{delta}. These findings demonstrate how a phosphorylation cycle controls the activity of this essential kinase. SignificanceSubtle control of kinase activity is critical to physiologic modulation of multiple physiological processes including circadian rhythms. CK1{delta} and the closely related CK1{varepsilon} regulate circadian rhythms by phosphorylation of PER2, but how kinase activity itself is controlled is not clear. Building on the prior observation that two splice isoforms of CK1{delta} regulate the clock differently, we show that the difference maps to three phosphorylation sites in the variably spliced region (XCT) that cause feedback inhibition of the kinase domain. More broadly, the data suggest a general model where CK1 activity on diverse substrates can be controlled by signaling pathways that alter tail phosphorylation. These inhibitory phosphorylation sites could also be targets for new therapeutic interventions.