Estradiol regulates daily rhythms underlying diet-induced obesity in female mice

Abstract

The circadian system is a critical regulator of metabolism and obesity in males, but its role in regulating obesity in females is poorly understood. Because there are sex differences in the development of obesity and susceptibility to obesity-related disorders, we sought to determine the role of estrogens in regulating the circadian mechanisms underlying diet-induced obesity. When fed high-fat diet, C57BL/6J male mice gain weight, whereas females are resistant to diet-induced obesity. Here, we demonstrate that estradiol regulates circadian rhythms in females to confer resistance to diet-induced obesity. We found that ovariectomized females with undetectable circulating estrogens became obese and had disrupted daily rhythms of eating behavior and locomotor activity when fed a high-fat diet. The phase of the liver molecular circadian rhythm was also altered by high-fat diet feeding in ovariectomized mice. Estradiol replacement in ovariectomized females a fed high-fat diet rescued these behavioral and tissue rhythms. Additionally, restoring the daily rhythm of eating behavior in ovariectomized females with time-restricted feeding inhibited diet-induced obesity and insulin resistance. Together, these data suggest that the circadian system is a target for treating obesity and its comorbidities in women after menopause, when circulating levels of estrogens are too low to protect their circadian rhythms.

Keywords: circadian; estrogen; female; ovariectomy; time-restricted feeding.

Fig. 1.

Estradiol replacement inhibits diet-induced obesity in ovariectomized (OVX) mice. Female C57BL/6J mice were OVX and implanted with tubing containing either vehicle (red) or 17β-estradiol (E2; blue). A: body weights of mice fed low-fat diet (6–8 wk old) for 2 wk and then 45% high-fat diet for 1 wk (8–9 wk old; arrow indicates that high-fat diet was initiated after body weight was collected at 8 wk; n = 26–27/group). B: body composition was measured at 9 wk of age (n = 9–11/group). C and D: energy intake (kcal/wk; C) and total activity (D) were measured during 1 wk of high-fat diet feeding (8–9 wk of age; n = 26–27/group). E: fasting blood glucose was measured at 9 wk of age (n = 8–10/group). Data are means ± SE. *P < 0.01.

Fig. 2.

Circulating estradiol inhibits disruption of the daily rhythm of eating behavior by high-fat diet feeding in female mice. Females were ovariectomized and treated with vehicle (A, C, D, G, and H; OVX + vehicle) or 17β-estradiol (OVX + E2; B, E, F, I, and J). Representative actograms of eating behavior in OVX mice treated with vehicle (A) or 17β-estradiol (E2; B) and fed low-fat diet (LFD) for 1 wk and then high-fat diet (HFD) for 1 wk (open bar is lights on and black bar is lights off; red asterisks in A and B show when the diet was switched). Representative circular histograms (C–F) and mean (± SE) amplitudes (G and I) and phases (H and J) of eating behavior rhythms on the last day of LFD feeding (indicated by pink arrow in A and light blue arrow in B) and last day of HFD feeding (indicated by red arrow in A and dark blue arrow in B). Circular histogram scale: inner circle, 5 eating events; outer circle, 10 eating events; 0–12: lights on; 12–0: lights off. The lengths and directions of the vectors represent the amplitudes and phases of the eating behavior rhythms, respectively; n = 17/group, except for H (n = 14/group; 3 OVX + vehicle mice had arrhythmic eating behavior during HFD feeding and, therefore, had no phases). *P < 0.05 vs. LFD.

Fig. 3.

Circulating estradiol enhances the amplitude of the daily locomotor activity rhythm during high-fat diet feeding. Females were ovariectomized and treated with vehicle (OVX + vehicle; A, C, E, and F) or 17β-estradiol (OVX + E2; B, D, G, and H). Representative actograms of locomotor activity (6-min bins; scaled max count: 7), measured by passive infrared sensors, of OVX females treated with vehicle (A) or E2 (B). Mice were fed low-fat diet (LFD) for 1 wk and high-fat diet (HFD) for 1 wk (red asterisk indicates when the diet was switched). Mean (± SE) activity profiles (C and D), amplitudes (E and G), and mesors (F and H) of locomotor activity rhythms during LFD feeding (days 1–6) and HFD feeding (days 8–13); n = 17 mice/group. *P < 0.001 vs. LFD.

Fig. 4.

Circulating estradiol prevents the phase advance of the liver circadian clock during high-fat feeding in female mice. Intact (gray) and ovariectomized (OVX) PER2::LUC females treated with vehicle (red) or 17β-estradiol (E2; blue) were fed high-fat diet for 1 wk. Tissues were cultured, and bioluminescence was measured. The mean phases (± SE) of the peaks of the bioluminescence rhythms were plotted relative to last lights on. Light and dark are indicated by the open and black bars, respectively. Number of rhythmic samples/number of samples cultured is shown for each tissue. *P < 0.05. SCN, suprachiasmatic nucleus; WAT, white adipose tissue.

Fig. 5.

Time-restricted feeding inhibits diet-induced obesity and glucose intolerance in ovariectomized mice. Ovariectomized mice were fed high-fat diet either ad libitum (red) or only during the 12-h dark phase (time-restricted feeding; black) beginning at 8 wk of age (indicated by arrow in A). A and B: mean (± SE) body weights were measured weekly (A), and adiposity was measured at 14 wk of age (B). C and D: energy intake (C) and activity counts (D) were summated from 8 through 14 wk of age. A–D: n = 16 ad libitum; n = 19 time-restricted feeding. E and F: mean blood glucose (E) and area under the curve (arbitrary units; F) are shown for glucose tolerance tests performed at Zeitgeber time (ZT) 6 (mice were fasted ZT0–6) at 15 wk of age (n = 11/group). G and H: fasting blood glucose (G) and homeostatic model assessment of insulin resistance (HOMA-IR; H) (n = 6–7/group) were measured at 16 wk old. I: the mean phases (± SE; n = 4–6/group) of the peaks of the bioluminescence rhythms of tissues cultured from ovariectomized females fed high-fat diet ad-libitum (red) or time-restricted (black) were plotted relative to last lights on. The phases of tissues explanted from intact females (gray) fed high-fat diet for 1 wk are shown for reference. Light and dark are indicated by the white and black bars, respectively. *P < 0.05.

Fig. 6.

Proposed model of circadian regulation of diet-induced obesity in female mice. A: in females with physiological levels of circulating estradiol, daily rhythms are not disrupted by high-fat diet feeding, and the mice are resistant to obesity and metabolic dysfunction. B: when circulating estradiol is very low, consumption of high-fat diet disrupts daily rhythms, and the mice develop severe obesity and insulin resistance. C: when the eating rhythm is restored by time-restricted feeding (food available only during the 12-h active phase) in females with low circulating estradiol, the mice have improved insulin resistance and only moderate obesity.