Inhibition of citrate cotransporter Slc13a5/mINDY by RNAi improves hepatic insulin sensitivity and prevents diet-induced non-alcoholic fatty liver disease in mice

Abstract

Objective: Non-alcoholic fatty liver disease is a world-wide health concern and risk factor for cardio-metabolic diseases. Citrate uptake modifies intracellular hepatic energy metabolism and is controlled by the conserved sodium-dicarboxylate cotransporter solute carrier family 13 member 5 (SLC13A5, mammalian homolog of INDY: mINDY). In Drosophila melanogaster and Caenorhabditis elegans INDY reduction decreased whole-body lipid accumulation. Genetic deletion of Slc13a5 in mice protected from diet-induced adiposity and insulin resistance. We hypothesized that inducible hepatic mINDY inhibition should prevent the development of fatty liver and hepatic insulin resistance.

Methods: Adult C57BL/6J mice were fed a Western diet (60% kcal from fat, 21% kcal from carbohydrate) ad libitum. Knockdown of mINDY was induced by weekly injection of a chemically modified, liver-selective siRNA for 8 weeks. Mice were metabolically characterized and the effect of mINDY suppression on glucose tolerance as well as insulin sensitivity was assessed with an ipGTT and a hyperinsulinemic-euglycemic clamp. Hepatic lipid accumulation was determined by biochemical measurements and histochemistry.

Results: Within the 8 week intervention, hepatic mINDY expression was suppressed by a liver-selective siRNA by over 60%. mINDY knockdown improved hepatic insulin sensitivity (i.e. insulin-induced suppression of endogenous glucose production) of C57BL/6J mice in the hyperinsulinemic-euglycemic clamp. Moreover, the siRNA-mediated mINDY inhibition prevented neutral lipid storage and triglyceride accumulation in the liver, while we found no effect on body weight.

Conclusions: We show that inducible mINDY inhibition improved hepatic insulin sensitivity and prevented diet-induced non-alcoholic fatty liver disease in adult C57BL6/J mice. These effects did not depend on changes of body weight or body composition.

Keywords: 2-DG, 2-Deoxy-d-glucose; Citrate transport; EE, energy expenditure; EGP, endogenous glucose production; FA, fatty acids; FLD, fatty liver disease; GIR, glucose infusion rate; HE clamp, hyperinsulinemic-euglycemic clamp; HFD, high-fat diet; IEX, anion-exchange high-performance liquid chromatography; INDY, ‘I'm not dead Yet’; INDY/Slc13a5; Insulin resistance; KO, knockout; Lipid accumulation; ORO, oil red O; RER, respiratory exchange ratio; SCR, non-silencing scrambled control siRNA; SKM, skeletal muscle; Steatosis; T2D, type-2 diabetes; TCA, tricarboxylic acid; WAT, white adipose tissue; WD, western diet; e, epididymal; mINDY, Slc13a5/SLC13A5; p, perirenal; s, subcutaneous; siINDY, mINDY-specific siRNA; siRNA; solute carrier family 13, member 5.

Graphical abstract

Figure 1

In vitro (A–C) and in vivo (D and E) validation of mINDY-specific siRNA (siINDY). (A) Dose–response curves upon transfection of HEK293 cells stably overexpressing mINDY carrying a FLAG-tag (HEK293-mINDY) with siINDY and its unmodified form (siINDY*) (n = 2). (B) Transfection of HEK293-mINDY with siINDY reduced mINDY protein in a concentration-dependent manner, determined by anti-FLAG In-Cell Western Assay (n = 3). The black arrow indicates background fluorescence signal of HEK293 wild-type cells analyzed in parallel. (C) Transfection of HEK293-mINDY cells with siINDY reduced cellular citrate uptake (n = 3). (D) Pilot mouse study to evaluate the siRNA efficacy. mINDY mRNA expression in liver 2 and 6 days after injection normalized to GAPDH and control siRNA at day 2 (n = 3). (E) mINDY mRNA expression in liver, kidney, and muscle tissue after 8 weeks of siRNA treatment normalized to GAPDH and control siRNA in liver (n = 4). Data are represented as mean ± SEM. Two-way ANOVA with Bonferroni's multiple comparisons test. Scrambled control siRNA (SCR): open bars; off-target-specific mINDY-unrelated siRNA (OT): fasciated bars; mINDY-specific unmodified siRNA (siINDY*): grey triangle; mINDY-specific modified siRNA (siINDY): black bars.

Figure 2

Liver-selective mINDY-specific siRNA approach. (A) Schematic of the mouse experiment. (B) mRNA expression of indicated genes in liver of clamped mice normalized to GAPDH (n = 7, two-tailed Student's t-test with Welch's correction). (C–E) Body composition measurement for fat mass (C), free fluids (D), and lean mass (E) at start (0 wk), after 5 weeks (5 wk) and before catheterization (7 wk) (n = 16). (F) Weight gain during intervention (n = 16). Data represent mean ± SEM. Black arrows: Recording of body weight/siRNA injection, Grey arrows: Procedures: BC: Body composition, MCA: Metabolic cage analysis, OP: Catheterization.

Figure 3

Metabolic characterization of siRNA-treated mice during metabolic cage analysis (A–D, n = 10) and ipGTT (E–F). (A) Food intake (left) and liquid consumption (with 6% sucrose, right) during light- and dark-phase (repeated measurement ANOVA with Bonferroni's multiple comparisons test). (B) Locomotor activity. (C) Energy expenditure. (D) Respiratory exchange ratio (VCO₂/VO₂). (E) Plasma glucose during ipGTT (n = 16). (H) According insulin levels during ipGTT (n = 15). Data represent mean ± SEM. Grey background indicates dark-phase.

Figure 4

Improvement of insulin sensitivity with mINDY-specific siRNA during HE clamp. (A) Blood glucose (top) at basal and insulin-stimulated phase and according glucose infusion rate (GIR, bottom) to maintain plasma glucose at 120 mg/dl. (B) Insulin-mediated reduction of plasma NEFA. (C) Endogenous glucose production (EGP). (D). Suppression of EGP. (E) Whole-body glucose uptake during steady state. (F) 2-DG uptake of SKM (Gastrocnemius) and epididymal WAT. Data represent mean ± SEM. n = 7. Two-tailed Student's t-test with Welch's correction.

Figure 5

mINDY-specific siRNA decreased hepatic lipid content. (A) Weight of indicated organs and epididymal (e), perirenal (p), and subcutaneous (s) WAT after 16 h fasting. (B) Representative H&E staining of liver. (C) Representative ORO staining. (D) Quantification of ORO stainings by densitometric analysis. (E) Triglyceride content in liver lysates. Data represent mean ± SEM. n = 4. One-tailed Student's t-test with Welch's correction. Black scales indicate 50 μm.