Hepatic metabolite responses to 4-day complete fasting and subsequent refeeding in rats

Abstract

Background: Fasting has been widely used to improve various metabolic diseases in humans. Adaptive fasting is necessary for metabolic adaptation during prolonged fasting, which could overcome the great advantages of short-term fasting. The liver is the main organ responsible for energy metabolism and metabolic homeostasis. To date, we lack literature that describes the physiologically relevant adaptations of the liver during prolonged fasting and refeeding. For that reason, this study aims to evaluate the response of the liver of Sprague-Dawley (SD) rats to prolonged fasting and refeeding.

Methods: Sixty-six male SD rats were divided into the fasting groups, which were fasted for 0, 4, 8, 12, 24, 48, 72, or 96 h, and the refeeding groups, which were refed for 1, 3, or 6 days after 96 h of fasting. Serum glucose, TG, FFA, β-hydroxybutyrate, insulin, glucagon, leptin, adiponectin and FGF21 levels were assessed. The glucose content, PEPCK activity, TG concentration and FFA content were measured in liver tissue, and the expression of genes involved in gluconeogenesis (PEPCK and G6Pase), ketogenesis (PPARα, CPT-1a and HMGCS2) and the protein expression of nutrient-sensing signaling molecules (AMPK, mTOR and SIRT1) were determined by RT-qPCR and western blotting, respectively.

Results: Fasting significantly decreased the body weight, which was totally recovered to baseline after 3 days of refeeding. A 4-day fast triggered an energy metabolic substrate shift from glucose to ketones and caused serum hormone changes and changes in the protein expression levels of nutrient-sensing signaling molecules. Glycogenolysis served as the primary fuel source during the first 24 h of fasting, while gluconeogenesis supplied the most glucose thereafter. Serum FFA concentrations increased significantly with 48 h of fasting. Serum FFAs partly caused high serum β-hydroxybutyrate levels, which became an important energy source with the prolongation of the fasting duration. One day of refeeding quickly reversed the energy substrate switch. Nutrient-sensing signaling molecules (AMPK and SIRT1 but not mTOR signaling) were highly expressed at the beginning of fasting (in the first 4 h). Serum insulin and leptin decreased with fasting initiation, and serum glucagon increased, but adiponectin and FGF21 showed no significant changes. Herein, we depicted in detail the timing of the metabolic response and adaptation of the liver to a 4-day water-only fast and subsequent refeeding in rats, which provides helpful support for the design of safe prolonged and intermittent fasting regimens.

Keywords: Gluconeogenesis; Hormone; Ketogenesis; Lipolysis; Metabolic syndrome; Nutrient-sensing signaling molecules; Prolonged fasting.

Figure 1. Body weight changes in response to fasting and refeeding.

Body weight was recorded before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh), and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. Data were presented as means ± SD (n = 6). Two-way ANOVA was used with multiple comparisons using post hoc Tukey’s test. F, Fasting; R, Refeeding. Significance was determined as *, #, $, p < 0.05; **, ##, $, p < 0.01, *, VS, respectively.

Figure 2. Blood biochemical parameters changed during 4-day fasting and subsequent refeeding.

Blood glucose concentration (A), triglycerides (B), free fatty acids (C) and serum β-hydroxybutyrate (D) concentration in rats before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh), and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. Data were presented as means ± SD (n = 6). One-way ANOVA was used with multiple comparisons using post hoc Tukey’s test. F, Fasting; R, Refeeding. Significance was determined as *, #, $, p < 0.05; **, ##, $, p < 0.01, *, VS, respectively.

Figure 3. Hormonal parameters of rats under 4-day fasting and subsequent refeeding.

Serum insulin (A), glucagon (B), glucagon/insulin (C), leptin (D), adiponectin (E), and FGF21 (F) concentration in rats before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh), and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. Data were presented as means ± SD (n = 6). One-way ANOVA was used with multiple comparisons using post hoc Tukey’s test. F, Fasting; R, Refeeding. Significance was determined as *, #, $, p < 0.05; **, ##, $, p < 0.01, *, VS, respectively.

Figure 4. Hepatic glucoregulatory, lipid storage and metabolites responses to fasting and refeeding.

Hepatic glycogen (A), PEPCK activity (B), Triglyceride (C), and Free fatty acid (D) in rats before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh), and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. Data were presented as means ± SD (n = 6). One-way ANOVA was used with multiple comparisons using post hoc Tukey’s test. F, Fasting; R, Refeeding. Significance was determined as *, #, $, p < 0.05; **, ##, $, p < 0.01, *, VS, respectively.

Figure 5. mRNA and protein expression in hepatic related with gluconeogenesis responses to fasting and refeeding.

Hepatic PEPCK (A) and G6Pase (D) mRNA expression before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh), and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting; PEPCK protein (B) expression before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh); PEPCK protein (C) expression before fasting (F0h) and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. Data were presented as means ± SD (n = 6) One-way ANOVA was used with multiple comparisons using post hoc Tukey’s test. F: Fasting; R: Refeeding. Significance was determined as *, #, $, p < 0.05; **, ##, $, p < 0.01, *, VS, respectively.

Figure 6. mRNA and protein expression in liver related with ketogenesis responses to fasting and refeeding.

Liver PPAR α (A), CPT-1a (B), HMGCS2 (C) mRNA expression before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh), and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. HMGCS2 protein expression before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh) (D); and HMGCS2 protein expression before fasting (F0h) and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting (E). Data were presented as means ± SD (n = 6). One-way ANOVA was used with multiple comparisons using post hoc Tukey’s test. F, Fasting; R, Refeeding. Significance was determined as *, #, $, p < 0.05; **, ##, $, p < 0.01, *, VS, respectively.

Figure 7. Nutrient-sensing signal molecules responses to fasting and refeeding.

A is the collection of p-AMPK, AMPK, p-mTOR, mTOR and SIRT1 protein bands before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh). The expression ratio of AMPK (B), mTOR (D), SIRT1(F) to β-actin before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh); The ratio of p-AMPK thr172 to total AMPK protein (C), p-mTOR ser2448 to total mTOR protein (E) were analyzed before fasting (F0h) and fasting for 4, 8, 12, 24, 48, 72 or 96 h (Fxh). G is the collection of p-AMPK, AMPK, p-mTOR, mTOR and SIRT1 protein bands before fasting (F0h) and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. The expression ratio of AMPK (H), mTOR (J), SIRT1 (L) to β-actin before fasting (F0h) and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. The ratio of p-AMPK thr172 to total AMPK protein (I), p-mTOR ser2448 to total mTOR protein (K) were analyzed before fasting (F0h) and after 1-, 3-, or 6-day refeeding (Rxd) following 96 h fasting. Data were presented as means ± SD (n = 6). One-way ANOVA was used with multiple comparisons using post hoc Tukey’s test. F: Fasting; R: Refeeding. Significance was determined as *, #, $, p < 0.05; **, ##, $, p < 0.01, *, VS, respectively.