Circadian Rhythms: Phosphorylating the CLOCK

The molecular circadian clock is a cell-autonomous system organized to create a conserved transcriptional-based negative feedback loop. This feedback loop is initiated when a transcription factor complex (positive arm) induces the expression of negative elements which, after a time-delay, feed back to inhibit the positive arm. The kinetic properties of circadian systems are set through post-translational modifications (PTMs) acting on nearly every component¹. Phosphorylation is generally regarded as the PTM primarily responsible for setting the 24-hour period length through altering the stability or accessibility of the negative (repressor) complex. While a number of studies have focused on comprehensive dissection of these negative element PTMs^2–4, the molecular rhythms community has just begun to dig into the regulation and function of PTMs in the positive arm. In this issue of Cell Cycle Spengler et al.⁵ fill in some of the story by investigating the role of phosphorylation on the transcription factor CLOCK in the mammalian circadian system and present data on its possible regulation by the GSK-3β kinase.

CLOCK forms a heterodimeric complex with another transcription factor, BMAL1, and together they regulate expression of circadianly regulated genes. Both trigger each other’s phosphorylation but little is known about the function of these PTMs. The authors have taken a bioinformatic approach to identify potential phosphorylation sites on CLOCK, narrowing down the possibilities within the 885-residue protein to focus on a region encompassing 36 conserved residues, 425–461, containing predicted kinase targets. CLOCK proteins harboring serine/threonine to alanine mutations in this region result in decreased phosphorylation and an increase in protein stability, thus defining a discreet region on CLOCK important for regulated turnover, termed a “phospho-degron”. The authors show that these phopho-deficient mutants still interact with BMAL1 and that BMAL1-mediated degradation of CLOCK requires the phospho-degron. Circadian physiology is also impacted as the mutant CLOCK proteins have attenuated transcriptional activity and an apparent phase delay in synchronized cells.

Previously a mass-spectrometry based study had shown that S427, within the phospho-degron, is phosphorylated in vivo⁶; however, that study did not establish function to the site. Spengler et al. further focus on two residues, including the in vivo phosphorylated S427, and S431. Mutation of S431A almost completely blocks all BMAL1-dependent CLOCK phosphorylation, while S427 is one of two predicted GSK-3β target sites within the phospho-degron. These data led the authors to hypothesize a conserved role for GSK-3β in regulating CLOCK specifically at S427, with S431 acting as its priming site. Along these lines, they present compelling evidence that GSK-3β can phosphorylate CLOCK in vitro and manipulation of upstream activators/inhibitors of GSK-3β predictably alters CLOCK phosphorylation and stability. Unfortunately, S427A and S431A mutations were not assessed individually for CLOCK stability, and were not included in the mutant strains when the authors looked at BMAL1-dependent CLOCK degradation or the circadian phase delay. If the proposed model is that GSK-3β mediates these clock effects specifically through S427 it would be interesting to look at S427A and S431A using the same conditions in order to determine if they impact CLOCK functions.

While this study defines a new region important for proper phosphorylation and function of CLOCK it also raises a few interesting questions. For instance, why would reduced phosphorylation and increased stability lead to an attenuation of transcriptional activity? One possibility is that phosphorylation is structurally required for CLOCK/BMAL1 promoter occupancy or, as the authors make the case, that there is an intermediate, unstable, phosphorylated form more transcriptionally active: such a model has recently been suggested based on data from fungal circadian oscillators⁷ whose molecular architecture is quite similar to the mammalian clock studied here. Additionally, given that small molecule inhibitors of GSK-3β specifically decrease period in cycling mammalian cells⁸, it is interesting that in this study the proposed role of GSK-3β is in promoting degradation of CLOCK. The authors suggest that reduction of CLOCK phosphorylation results in the observed phase delay, through longer protein half-life. This is perhaps opposite to what one might expect when inactivation of the responsible kinase leads to period decreases. However, here the authors are looking at one specific target, CLOCK, which may not be the only circadian protein regulated by GSK-3β. While further work will be needed to understand how GSK-3β activity affects other kinetic parameters of the circadian clock, Spengler et al have taken an important first step in the analysis.