Construction of ex vivo and in vivo PRCs. Phase shifts after the addition of DEX (closed circles with a black line) or vehicle (open circles with a gray line) into the culture media were plotted against the phase of their addition (A). Data are expressed as mean ± SEM (n = 4–5 for each time point). The ex vivo PRC for DEX was constructed by calculating the difference in the phase shift caused by DEX from that of the respective mean value obtained from the control group (B). The phase-dependent effect of DEX on the nasal circadian clock in vivo were examined by injecting DEX (closed circles with a black line) or PBS (open circles with a gray line) at 4 different circadian phases and measuring the first peak phases of circadian PER2 rhythm in cultured nasal mucosa (C). Data are expressed as mean ± SEM (n = 5 for each time point). The in vivo PRC for DEX was constructed by calculating the difference in first circadian peak phases between the DEX and control groups (D). In both the ex vivo and in vivo experiments, phase shifts were phase dependent, with statistical significance in the DEX group but not in the control group (two-way ANOVA with a post hoc Bonferroni test; *, P < .05 and **, P < .01). Dose dependency of phase shifts induced by DEX was examined ex vivo at the maximum phase-delaying phase (CT12) and phase-advancing phase (CT18) (E). Abscissa shows final concentration of DEX. The data representing 0M (vehicle control) and 10−7M are the also shown in A. Data are expressed as mean ± SEM (n = 4–6 for each time point). Statistically significant differences were detected in phase shifts at CT12 and CT18 (one-way ANOVA with a post hoc Bonferroni test; **, P < .01 vs vehicle; †, P < .05 10−7M vs 10−8M).
