Dietary Fat Rewires the Circadian Clock to Track the Seasons

The circadian clock evolved to keep biological processes synchronized with the 24-hour light-dark cycle, but it must also adapt across the year as seasonal changes shift the relative lengths of day and night. How the mammalian clock accomplishes this seasonal recalibration — and whether diet plays a role — has been poorly understood. This study from Levine, Ptáček, Fu and colleagues at UCSF, published in Science, provides a mechanistic answer: dietary fat modifies phosphorylation of the clock repressor protein PER2 at serine 662 (S662), and this single molecular event is both necessary and sufficient to control the rate at which mice entrain to seasonal photoperiods.

The researchers first established that high-fat diet (HFD, 45% calories from fat) and caloric restriction (CR, 40% below control) have opposite effects on behavioral entrainment. Wild-type mice on regular chow advanced daily locomotor activity onset at roughly 0.25 hours/day when shifted from equinox (12:12 light-dark) to a winter cycle (4:20 LD). Mice on HFD advanced at only half that rate, accumulating a ~2-hour phase delay by day 30. Conversely, CR mice advanced at ~0.5 hours/day — twice the control rate — accumulating a ~4-hour phase advance over 20 days. For summer adaptation (shift to 20:4 LD), effects were reversed: HFD accelerated phase-delay entrainment while CR slowed it, with CR mice achieving only 3.65 hours of delay versus 5.6 hours in controls within the first two days.

Genetic manipulation of PER2 dosage confirmed the clock's central role. Mice with five extra copies of human PER2 (PER2-TgWT) failed to shift activity patterns within 60 days under a winter cycle, while Per2 knockout mice entrained in just 13 days at 0.48 hours/day. The phospho-null mutation PER2-S662G (serine to glycine, preventing phosphorylation) advanced activity at ~1.25 hours/day, fully entraining to winter cycles in as few as 4 days, while the phospho-mimetic PER2-S662D (serine to aspartate) failed to advance within 60 days. For summer cycles, PER2-S662D mice entrained rapidly to ZT19.5 within 2 days, while PER2-S662G mice could only shift to ZT14.5 over the entire experiment — an ~8-hour difference in behavioral phase after 30 days.

Western blotting of immunoprecipitated PER2 from hypothalamic lysates of PER2-TgWT mice confirmed that just one week of HFD significantly increased PER2-S662 phosphorylation and nuclear PER2 protein, consistent with enhanced CK1δ activity. Critically, when PER2-S662G mice — who cannot be phosphorylated at this site — were given HFD, the diet had no significant effect on their entrainment to either seasonal photoperiod, demonstrating that phosphorylation of S662 is the required mediator of HFD's circadian effects. Fasting had the opposite molecular effect: 16 hours of fasting decreased hypothalamic PER2-S662 phosphorylation and nuclear PER2 abundance, and PER2-S662D mice failed to suppress late-dark-phase activity during fasting as wild-type mice did.

RNA-sequencing of hypothalami at ZT16 revealed that fasting altered 929 genes in wild-type mice but only 469 in PER2-S662D mice (299 shared), indicating that the phospho-mimetic mutation broadly blunts the hypothalamic transcriptional response to fasting. Pathway analysis identified differential regulation of polyunsaturated fatty acid (PUFA) metabolism and oxylipin biosynthesis. To directly test whether dietary PUFAs mediated the effect, the authors fed mice a partially hydrogenated version of the HFD (reducing PUFA content while maintaining fat calories). Partially hydrogenated HFD increased hypothalamic PER2-S662 phosphorylation and significantly accelerated entrainment to a summer photoperiod in wild-type mice but not in PER2-S662G mice, confirming that dietary PUFAs modulate circadian phase-shifting specifically through this phosphorylation site. These findings create a coherent model in which seasonal shifts in dietary fat composition — paralleling natural changes in food availability across the year — tune the circadian clock via PER2-S662 phosphorylation to keep behavioral rhythms aligned with seasonal light cycles.