Figure 5.

Circadian phase shifts induced by transient treatment with caffeine do not depend on the SCN. Mice maintained under 12:12 h light–dark conditions were injected with vehicle (VEH) or 20 mg·kg⁻¹ caffeine (CAF) at ZT 5 for three consecutive days, then killed at ZT 7 to assess oscillatory PER2::LUC bioluminescence in the liver or SCN. (A, D) Representative detrended data from the SCN (A) or liver (D) of vehicle (VEH)- or caffeine-treated animals. Caffeine caused a phase advance in the liver but not the SCN. (B, E) Average peak phase of PER2::LUC bioluminescence in each tissue indicated in (A) and (D). ##P < 0.01 versus VEH (Student's t-test). (C, F) Period of rhythmic PER2::LUC expression in each tissue indicated in (A) and (D). The distance between peaks 1 and 2 represents period length. All values are expressed as mean ± SEM (n = 4 per group).