Resetting the circadian clock of Alzheimer's mice via GLP-1 injection combined with time-restricted feeding

Abstract

Circadian rhythm disturbances are the most common symptoms during the early onset of AD. Circadian rhythm disorders aggravate the deposition of amyloid plaques in the brains of AD patients. Therefore, improving the circadian rhythm of AD patients might slow down the pathological development of neurodegeneration. Circadian regulation is driven by a master clock in suprachiasmatic nuclei (SCN) and peripheral clock located in peripheral organs. The rhythmic feeding-fasting cycle has been proved to dominant cue to entrain peripheral clocks. We hypothesized that dietary intervention to a certain period of time during the dark phase might entrain the clock and reset the disrupted daily rhythms of AD mice. In this study, exogenous glucagon-like peptide-1 (GLP-1) treatment, time-restricted feeding (TRF), and the combination were used to examine the effect of overall circadian rhythm and neurodegenerative pathogenesis of transgenic AD mice. It was confirmed that GLP-1 administration together with time-restricted feeding improves circadian rhythm of 5 × FAD mice including the physiological rhythm of the activity-rest cycle, feeding-fasting cycle, core body temperature, and hormone secretion. Furthermore, GLP-1 and TRF treatments improved the diurnal metabolic homeostasis, spatial cognition, and learning of 5 × FAD mice. The aberrant expression of clock genes, including Baml1, Clock, and Dbp, was improved in the hypothalamus, and pathological changes in neurodegeneration and neuroinflammation were also observed in AD mice with dual treatment.

Keywords: Alzheimer’s disease; amyloid-β; circadian rhythm; glucagon-like peptide-1; time-restricted feeding.

FIGURE 1

Alzheimer’s disease mice exhibit circadian rhythm disturbances. (A) Representative locomotor activity records of 5 × FAD mice (AD) and wild-type mice (WT), respectively. Locomotor activity was defined as the moving distance per unit time (2 min). Each horizontal line represents 24 h. Periods of darkness are indicated by grey backgrounds. The black and white bars on the button indicate 12 h-dark and 12 h-light periods, respectively. (B) Percentage of the locomotor activity in the dark and light/total activity (24 h). n = 7 per group, ****p < 0.0001, using two-way ANOVA followed by Sidak t test. (C) Representative meal duration records of two groups. (D) Percentage of the meal duration in the dark and light/total meal duration (24 h). n = 7 per group, ****p < 0.0001, using Two-Way ANOVA followed by Sidak t test. (E) Representative photographs of PER2:LUC mice in vivo monitoring from each time point at 6 h intervals. (F,G) Raw photon count data of individual bioluminescence rhythms from E. n = 3 per group, n = 3 per group, *p < 0.015, **p < 0.01 using Two-Way ANOVA test. (H) GLP-1 secretion levels in both groups. (I) The protocol design of our study.

FIGURE 2

Effect of GLP-1 and TRF on circadian rhythm disorder in mice models of Alzheimer’s disease. (A) Representative locomotor activity records of each group. The green line represents the time of intraperitoneal injection of GLP-1 to this group of mice. The blue area delineates 4-h of feeding time. (B) Activity during the light phase (m). n = 7 per group, *p < 0.05 vs. AD group using One-Way ANOVA followed by Dunnett’s test. (C) Activity during the dark phase (m). (D) Ratio of the activity in the dark and light/total activity in each group. n = 7 per group, ****p < 0.0001 vs. AD mice using two-way ANOVA followed by Dunnett’s test. (E) Representative meal duration records of each group. (F) Meal duration during the light phase (s). n ≥ 4 per group, *p < 0.05, ***p < 0.0001 vs. AD group using One-Way ANOVA followed by Dunnett’s test. (G) Meal duration during the dark phase (s). n ≥ 4 per group. (H) Ratio of meal duration in the dark and light/total meal duration (24 h) in each group. n ≥ 4 per group, *p < 0.05, ***p < 0.0001 vs. AD mice using Two-Way ANOVA followed by Dunnett’s test. (I) Body temperature was measured every 3 h for 24 h continuously.

FIGURE 3

Influence of GLP-1and TRF on hormone secretion in mice. (A–D) Secretion levels of GLP-1, melatonin, cortisol, and orexin A in the five groups of mice detected by ELISA, respectively. n = 6 each group, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. AD group, using two-way ANOVA followed by Dunnett’s test. n = 6 per test and per group.

FIGURE 4

Effect of GLP-1and TRF on glucose metabolism. (A–D) Individual responses in blood glucose to identical intraperitoneal glucose tolerance test (IPGTT) conducted at four time points throughout the day (n = 6 per time point and per group). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared to the AD group using two-way ANOVA followed by Dunnett’s test. (E) Corresponding ΔAUC for blood glucose. n = 6, *p < 0.05, ***p < 0.001 vs. AD group using One-Way ANOVA followed by Dunnett’s test.

FIGURE 5

Changes in learning and memory capacity under the influence of GLP-1 and TRF. Cognition evaluation of the mice by the MWM. (A) Escape latency onto a hidden platform during the training trials of the Morris water maze test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared to the AD group, using One-Way ANOVA followed by Dunnett’s test or Dunn’s test. (B) Average swimming speed during the training trials. (C) Representative swimming tracks of the different groups of mice in the probe trial. (D) Numbers of crossing the platform during the probe trial. *p < 0.05 vs. AD group, using The Kruskal-Wallis test. (E) Percentage of time spent in the target quadrant in probe trial. *p < 0.05, ***p < 0.001 compared to the AD group, using one-way ANOVA followed by Dunnett’s test. n = 8–11 animals per group.

FIGURE 6

Novel object recognition task reveals improvement of GLP-1 and TRF on long-term memory and attention deficits in AD mice. (A) Recognition index was defined as the time to explore a familiar or new object/total time to explore both objects. n = 6, Two-tailed t-test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (B) Discrimination index is calculated as the difference between the time spent exploring a new object and a familiar object/the total time spent exploring both objects. n = 6, One-Way ANOVA, *p < 0.05 vs. AD group.

FIGURE 7

Impact of GLP-1 and TRF on circadian rhythm-related gene expression. (A–C) Relative expression level of circadian rhythm-related genes in the hypothalamus, liver and hippocampus tissues at four time points throughout the day, respectively. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. AD group, using two-way ANOVA followed by Dunnett’s test. n ≥ 5 per time point and per group.

FIGURE 8

Role of GLP-1 and TRF on induced decrease in Aβ, GFAP, and IBA1 expression. (A–C) Representative hippocampus and cortex images from three groups stained for Aβ (MOAB-2), GFAP, and (IBA1), including a 4× field and a 20× field. Scale bars are 200 and 50 μm respectively. (D–F) The quantitative analysis of the positive areas of the three proteins in the hippocampal and cortical regions, respectively. Hippocampus: Averages of 4 fields per slice, n = 4 animals per group were quantified. Cortex: Averages of 5 fields per slice were quantified. n = 4. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Aβ: The Kruskal-Wallis test. GFAP: The Kruskal-Wallis test was used for quantitative analysis of hippocampal regions and Brown-Forsythe and Welch ANOVA tests for cortical regions. IBA1 of hippocampus: One-Way ANOVA followed by Dunnett’s test. IBA1 of cortex: The Kruskal-Wallis test.