Ketogenic diet administration to mice after a high-fat-diet regimen promotes weight loss, glycemic normalization and induces adaptations of ketogenic pathways in liver and kidney

Abstract

Objective: The ketogenic diet (KD), characterized by very limited dietary carbohydrate intake and used as nutritional treatment for GLUT1-deficiency syndromes and pharmacologically refractory epilepsy, may promote weight loss and improve metabolic fitness, potentially alleviating the symptoms of osteoarthritis. Here, we have studied the effects of administration of a ketogenic diet in mice previously rendered obese by feeding a high fat diet (HFD) and submitted to surgical destabilization of the medial meniscus to mimic osteoarthritis.

Methods: 6-weeks old mice were fed an HFD for 10 weeks and then switched to a chow diet (CD), KD or maintained on a HFD for 8 weeks. Glycemia, β-hydroxybutyrate (BHB), body weight and fat mass were compared among groups. In liver and kidney, protein expression and histone post-translational modifications were assessed by Western blot, and gene expression by quantitative Real-Time PCR.

Results: After a 10 weeks HDF feeding, administration for 8 weeks of a KD or CD induced a comparable weight loss and decrease in fat mass, with better glycemic normalization in the KD group. Histone β-hydroxybutyrylation, but not histone acetylation, was increased in the liver and kidney of mice fed the KD and the rate-limiting ketogenic enzyme HMGCS2 was upregulated - at the gene and protein level - in liver and, to an even greater extent, in kidney. KD-induced HMGCS2 overexpression may be dependent on FGF21, whose gene expression was increased by KD in liver.

Conclusions: Over a period of 8 weeks, KD is more effective than a chow diet to induce metabolic normalization. Besides acting as a fuel molecule, BHB may exert its metabolic effects through modulation of the epigenome - via histone β-hydroxybutyrylation - and extensive transcriptional modulation in liver and kidney.

Keywords: HMGCS2; Histone PTMs; Ketogenesis; Ketogenic diet; β-hydroxybutyrate.

Figure 1

Experimental protocol diagram and metabolic parameters. (A) Schematic representation of the study protocol. Groups are defined by empty circles (HFD throughout the study), grey circles (CD, switch to a chow diet) and black circles (KD, switch to a ketogenic diet). (B) Mice body weight throughout the study protocol. Body weight area under the curve was determined from week 18 to week 24 and compared among the three groups by one-way Anova. (C) Body fat mass at weeks 15,20 and 23 as determined by DEXA analysis. Fat content differences among groups were significant at weeks 20 and 23. (D) Kidney weight at the end of the study. (E) Glycemia and BHB concentration measured in tail blood at week 23. (F) IGTT as performed at week 23 on the three experimental groups. Relevant pairwise statistically significant differences by one-way ANOVA and Tukey's post-hoc test are shown.

Figure 2

Effects of dietary switch on histone β-hydroxybutyrlation and histone acetylation. Acid-extracted histones from kidney (A) and liver (B) were immunoblotted with antibodies to β-hydroxybutyrylated histone H3 lysine 4 (H3K4–BHB, left graphs) and antibodies to acetylates histone H3 lysines 9/14 (H3-Ac, right graphs). Loading was assessed by coomassie staining. Immunoblotting signal quantification is relative to the coomassie blue signal. Immunoblotting membranes are shown below each graph and in Supplementary Fig. 1 (n = 8 for each experimental condition). All pairwise statistically significant differences by one-way ANOVA and Tukey's post-hoc test are shown.

Figure 3

Effects of HFD, KD and CD dietary switch on transcription of ketogenic (Hmgcs2) and ketolytic (Oxct1) rate limiting genes as well as ketolytic/ketogenic shared genes Bdh1 and Acat1 in kidney (A) and liver (B). Detection of cDNAs was performed by RT qPCR and gene expression was quantified by the ΔΔCt method using β-actin as housekeeping gene. All pairwise statistically significant differences by one-way ANOVA and Tukey's post-hoc test are shown.

Figure 4

Effects of HFD, KD and CD dietary switch on protein expression of ketogenic HMGCS2, ketolytic SCOT1 (coded by Oxct1) and ketolytic/ketogenic BDH1 enzymes in kidney (A) and liver (B). Immunoblotting membranes for (A) and (B) panels are shown in the figure and in Supplementary Fig. 2 (n = 8 for each experimental condition). (C) Direct comparison of SCOT1 expression between liver and kidney. Protein expression was normalized to β-actin or tubulin expression as indicated. In quantification graphs of panel (A) and (B) all pairwise statistically significant differences by one-way ANOVA and Tukey's post-hoc test are shown.

Figure 5

Effects of HFD, KD and CD dietary switch on genes regulating ketogenesis (Fgf21), fatty acid import within the mitochondrion (Cpt1a) and ketone bodies export (Slc16a1) in kidney (A) and liver (B). (C) Quantification of triglycerides and total cholesterol in liver and kidney. All pairwise statistically significant differences by one-way ANOVA and Tukey's post-hoc test are shown.

Figure 6

Effects of HFD, KD and CD dietary switch on the expression of lipogenic proteins and genes in the liver. (A) Immunoblotting membranes stained with antibodies directed to ACC, FASN, DGAT2 and tubulin and relative quantifications (normalized to tubulin). Immunoblotting membranes are shown in the figure and in Supplementary Fig. 4 (n = 8 for each experimental condition). (B) Liver weight at the end of the study. (C, D) Gene expression levels of lipogenic (C) and lipolytic (D) genes as performed by RT qPCR. All pairwise statistically significant differences by one-way ANOVA and Tukey's post-hoc test are shown.

Figure 7

Effects of HFD, KD and CD dietary switch on expression of inflammatory genes. Anti-inflammatory Il10 and suppressor of cytokine signaling 3 (Socs3) and pro-inflammatory Il1b and Tnfa genes were quantified in kidney (A) and liver (B). Anti-fibrotic gene Fndc5 was quantified in the liver (B). All pairwise statistically significant differences by one-way ANOVA and Tukey's post-hoc test are shown.

Figure 8

Mild modulation of insulin signalling in the dietary switch to control but not ketogenic diet. The effects of HFD, KD and CD dietary switch on the protein expression of the insulin receptor (IR, panel A); and phosphorylated PKB (on Serine 473, PKB-pSer473, panel B) were evaluated in kidney and liver. Immunoblotting membranes for IR, tubulin, PKB-pSer473 and total PKB are shown in the figure and in Supplementary Fig. 3 (n = 8 for each experimental condition). All pairwise statistically significant differences, determined by one-way ANOVA and Tukey's post-hoc test, are shown.