The physiology and behaviour of mammals are subject to daily oscillations driven by an endogenous circadian clock. The circadian timing system is composed of a central pacemaker in the brain and subsidiary oscillators in most peripheral tissues. Although light–dark cycles are the predominant cue for the brain's pacemaker, cyclic feeding behaviour has a strong effect on clocks that operate in many other tissues. Reporting in Cell, two studies now suggest a strong molecular candidate for the long-sought link between metabolism and components of the internal clock that drive circadian rhythms.

The rhythm-generating mechanism is thought to rely on a feedback loop that involves the transcription factors BMAL1 and CLOCK, which activate the expression of Period (Per) and Cryptochrome (Cry) genes. When PER and CRY proteins accumulate to a critical level, they form complexes with BMAL1–CLOCK heterodimers and thereby repress transcription of their own genes.

...SIRT1 transduces signals originating from cellular metabolites to the circadian clock.

Paolo Sassone-Corsi and colleagues have previously shown that CLOCK is an enzyme with histone acetyltransferase (HAT) activity. CLOCK also acetylates non-histone targets, such as its own dimerization partner BMAL1; this modification is essential for circadian function. If CLOCK is a HAT, the authors reasoned, then there must be a histone deacetylase (HDAC) that acts in the opposite manner. The sirtuin SIRT1 — best known for its potential anti-ageing properties — was the obvious candidate because of its dependence on NAD+ (a factor that has been used as a read-out of metabolic state) and the fact that it preferentially deacetylates the same histone (histone H3) that CLOCK acetylates.

Sassone-Corsi and colleagues now show that the HDAC activity of SIRT1 is controlled in a circadian manner, correlating with the rhythmic acetylation of histone H3 and the clock component BMAL1 by CLOCK. Genetic ablation of the Sirt1 gene or pharmacological inhibition of SIRT1 activity leads to disturbances in the circadian cycle and in the acetylation of histone H3 and BMAL1. Finally, using liver-specific SIRT1-mutant mice, Sassone-Corsi and colleagues showed that SIRT1 contributes to circadian control in vivo — increased BMAL1 acetylation and altered circadian expression of Cry and Per2 genes was observed in the livers of the mutant mice. In a second paper, the group of Ueli Schibler report on another facet of the SIRT1-circadian clock connection. In agreement with Sassone-Corsi's team, these authors showed that SIRT1 associates with the CLOCK–BMAL1 complex but Schibler's group reported that SIRT1 promotes the deacetylation and degradation of PER2, an essential component for both core clock function and the synchronization of circadian oscillators.

It's been a dogma for years that the circadian clock is regulated by transcription feedback loops. These studies now highlight the importance of another loop — an enzymatic one — and indicate that SIRT1 transduces signals originating from cellular metabolites to the circadian clock.