Abstract

SummaryCircadian (approximately daily) rhythms are a pervasive property of mammalian cells, tissues, and behaviour, ensuring physiological and metabolic adaptation to solar time. Models of daily cellular timekeeping revolve around transcriptional feedback repression, whereby CLOCK and BMAL1 activate the expression of clock proteins PERIOD (PER) and CRYPTOCHROME (CRY), which in turn repress CLOCK/BMAL1 activity. CRY proteins are thus considered essential negative regulators of the oscillation; a function supported by behavioural arrhythmicity of CRY-deficient mice when kept under constant conditions. Challenging this interpretation, however, we find evidence for persistent circadian rhythms in mouse behaviour and cellular PER2 levels when CRY is absent. CRY-less oscillations are variable in their expression and have a shorter period than wild type controls. Importantly, we find classic circadian hallmarks such as temperature compensation and determination of period by casein kinase 1{delta}/{varepsilon} activity to be maintained. In the absence of CRY-mediated transcriptional feedback repression and rhythmic Per2 transcription, PER2 protein rhythms are sustained for several cycles, accompanied by circadian variation in protein stability. We suggest that, whereas circadian transcriptional feedback imparts robustness and functionality onto biological clocks, the core timekeeping mechanism is post-translational. Our findings suggest that PER proteins normally act as signalling hubs that transduce timing information to the nucleus, imparting daily rhythms upon the activity of transcriptional effectors. Highlights[tpltrtarr] PER/CRY-mediated negative feedback is dispensable for mammalian circadian timekeeping [tpltrtarr]Circadian variation in PER2 levels persists in the absence of rhythmic Per2 transcription [tpltrtarr]CK1 and GSK3 are plausible mechanistic components of a cytoscillator mechanism [tpltrtarr]CRY-mediated feedback repression imparts robustness to biological timekeeping In briefCircadian turnover of mammalian clock protein PERIOD2 persists in the absence of canonical transcriptional feedback repression and rhythmic clock gene activity, demanding a re-evaluation of cellular clock function and evolution.