Timing of Food Intake Drives the Circadian Rhythm of Blood Pressure

Abstract

Timing of food intake has become a critical factor in determining overall cardiometabolic health. We hypothesized that timing of food intake entrains circadian rhythms of blood pressure (BP) and renal excretion in mice. Male C57BL/6J mice were fed ad libitum or reverse feeding (RF) where food was available at all times of day or only available during the 12-h lights-on period, respectively. Mice eating ad libitum had a significantly higher mean arterial pressure (MAP) during lights-off compared to lights-on (113 ± 2 mmHg vs 100 ± 2 mmHg, respectively; P < 0.0001); however, RF for 6 days inverted the diurnal rhythm of MAP (99 ± 3 vs 110 ± 3 mmHg, respectively; P < 0.0001). In contrast to MAP, diurnal rhythms of urine volume and sodium excretion remained intact after RF. Male Bmal1 knockout mice (Bmal1KO) underwent the same feeding protocol. As previously reported, Bmal1KO mice did not exhibit a diurnal MAP rhythm during ad libitum feeding (95 ± 1 mmHg vs 92 ± 3 mmHg, lights-off vs lights-on; P > 0.05); however, RF induced a diurnal rhythm of MAP (79 ± 3 mmHg vs 95 ± 2 mmHg, lights-off vs lights-on phase; P < 0.01). Transgenic PERIOD2::LUCIFERASE knock-in mice were used to assess the rhythm of the clock protein PERIOD2 in ex vivo tissue cultures. The timing of the PER2::LUC rhythm in the renal cortex and suprachiasmatic nucleus was not affected by RF; however, RF induced significant phase shifts in the liver, renal inner medulla, and adrenal gland. In conclusion, the timing of food intake controls BP rhythms in mice independent of Bmal1, urine volume, or sodium excretion.

Keywords: Bmal1; blood pressure; circadian rhythms; sodium excretion; time-restricted feeding.

Graphical Abstract

Figure 1.

Food and water intake (average of the last 2 days) during ad libitum feeding and RF in WT mice. Data are presented as 12-h average (A and C) and 24-h average (B and D). Repeated measures two-way ANOVA was used in (A), (C), and indicated significant Feeding X Time interactions (P < 0.05; bars indicate significant post hoc comparisons). Student’s t-tests for unpaired data were used in (B), (D).

Figure 2.

MAP and HR of the last 2 days during ad libitum feeding and RF in WT mice. Data are presented as 48-h trace with shaded areas indicating lights off (A and C) as well as 12-h average of the last 2 days (B and D). Repeated measures two-way ANOVA was used in comparing the 12-h average data and indicated significant Feeding × Time interactions (P < 0.05; bars indicate significant post hoc comparisons).

Figure 3.

Urine volume (A), urinary sodium excretion (UNaV; B), urinary potassium excretion (UKV; C), urine excretion of ET-1 (D), and aldosterone (E) during ad libitum feeding and RF in WT mice. Data are presented as 12-h average of the last 2 days. Repeated measures two-way ANOVA indicated significant Feeding × Time interactions (P < 0.05; bars indicate significant post hoc comparisons).

Figure 4.

Food and water intake (average of the last 2 days) during ad libitum feeding and RF in Bmal1KO mice. Data are presented as 12-h average (A and C) and 24-h average (B and D). Repeated measures two-way ANOVA was used in (A) and (C), indicating significant Feeding × Time interactions (P < 0.05; bars indicate significant post hoc comparisons). Student’s t-test for unpaired data was used in (B), (D).

Figure 5.

MAP and HR of the last 2 days during ad libitum feeding and RF in Bmal1KO mice. Data are presented as 48-h trace (A and C) as well as 12-h average of the last 2 days (B and D). Repeated measures two-way ANOVA comparing the 12-h average data indicated significant Feeding × Time interactions (P < 0.05; bars indicated significant post hoc comparisons).

Figure 6.

Urine volume (A), urinary sodium excretion (UNaV; B), urinary potassium excretion (UKV; C), urine excretion of ET-1 (D), and aldosterone (E) during ad libitum feeding and RF in Bmal1KO mice. Data are presented as 12-h average of the last 2 days. Repeated measures two-way ANOVA was used.

Figure 7.

GFR of C57BL6/J mice at Day 6 of either ad libitum feeding or RF. GFR was measured at ZT4 and ZT16. Mixed effects two-way ANOVA reveals significant effect of Time, but no effect of Feeding or Time × Feeding interaction.

Figure 8.

Plasma hormone levels of C57BL6/J mice at Day 6 of either ad libitum feeding or RF. Plasma levels of corticosterone (A), insulin (B), IGF-1 (C), and leptin (D) were measured from samples collected at six time points through a 24-h cycle. Repeated measures two-way ANOVA results are indicated on the right of each panel.

Figure 9.

Lumicycle data of tissue cultures from Per2^Luc+/− knock-in mice. Representative traces of bioluminescence in the SCN and renal cortex (A) and renal inner medulla (B) were shown. Acrophase (time of peak) of the SCN, renal cortex, inner medulla, adrenal gland, and liver were shown (C). Student’s t-test indicated significant phase differences between ad libitum and RF for inner medulla, adrenal gland, and liver (n = 3–4/group).

Figure 10.

MAP of C57BL6/J mice at Day 5 and Day 6 of either ad libitum feeding, RF (07:00 am–07:00 pm feed), or offset feeding (01:00 pm–01:00 am feed) under constant dark conditions. Data were presented as 48-h trace (A) as well as 12-h averages of the feeding and fasting periods for RF group (B) and offset feeding group (C). Student’s t-test indicated significant fasting-feeding differences in the 12-h average MAP data for either feeding time. (D) Rayleigh plots for phase of peak HR, MAP, and activity from C57BL6/J mice during the last 4 days of ad libitum feeding, RF, or offset feeding. Circles represent the phase for individual mice and arrows represent the average phase for each feeding condition. Timing of food availability for each group is represented by the blue (ad libitum), red (RF), and brown (OF) bands inside the circle.

Figure 11.

Schematic summary of the impact of meal timing on BP and urine production. When mice are fed ad libitum, both BP and urine volume follow diurnal rhythms that are high during the lights-off phase and low during the lights-on phase. When mice were reverse fed with food available only during the inactive period, the diurnal rhythm of BP is inverted; however, the diurnal rhythm of urine production remains intact.