Adipose tissue supports normalization of macrophage and liver lipid handling in obesity reversal

  1. Assaf Rudich1,2
  1. 1The Department of Clinical Biochemistry and Pharmacology, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel
  2. 2The National Institute of Biotechnology in the Negev (NIBN), Ben-Gurion University, Beer-Sheva, Israel
  3. 3The Shraga Segal Department of Microbiology, Immunology and Genetics, Ben-Gurion University, Beer-Sheva, Israel
  1. Correspondence should be addressed to A Rudich; Email: rudich{at}bgu.ac.il
  1. Figure 1

    Glucose homeostasis and insulin sensitivity upon dietary switch. Eight-week-old C57BL6 mice were fed normal-chow diet (NC, ~10% kcal from fat) for 10 weeks, high-fat feeding (HFF, ~60% kcal from fat) for 10 weeks or HFF for 8 weeks and then switched to NC for 2 weeks (‘reverse group’, HFF→NC). Total body weight (BW) dynamics (A), epididymal fat at 10 weeks (B), circulating fasting blood glucose (C) and the glucose-to-insulin ratio (homeostatic model of insulin resistance – HOMA-IR) index 7 days after dietary reversal (D) are shown. n = 16–21 per group. Dynamically, intra-peritoneal glucose tolerance test (i.p. GTT, E) and intra-peritoneal insulin tolerance test (i.p. ITT, F) were performed 7 days after the dietary switch by i.p. administration of 2 g/kg BW glucose after an overnight fast or 0.2 U/kg BW insulin after 4-h fast, respectively. Area under the curves (AUC) of the GTT and ITT were calculated as well. For (E) n = 8/group, (F) n = 5–8/group. *P < 0.05 compared to NC; #P < 0.05 compared to HFF; ***P < 0.001 compared to NC; ###P < 0.001 compared to HFF; n.s. – non-significant (P > 0.05) compared to NC.

  2. Figure 2

    Early hepatic response to obesity reversal. Representative images (A) and weights (B) of livers 14 days after dietary switch (n = 11–12/group). (C) Representative liver hematoxylin and eosin (H&E)-stained histological sections of the three groups. I, II and III are sections from 3 individual mice from the HFF→NC group, demonstrating the variability in this group. (D) Mean area identified as steatotic (n = 4 for NC, HFF and n = 8 for HFF→NC. (E) Quantitative real-time PCR of liver expression of lipid metabolism genes: acetyl CoA carboxylase1 (Acaca), HNG CoA reductase (Hmgcr) and peroxisome proliferator-activated receptor γ (PPARγ). Values are the ΔΔCT values adjusted to Rplp0 and Hprt and are presented as fold of mean NC values (n = 15–16/group, from 4 independent experiments). (F, G and H) Results of pyruvate, fructose and alanine tolerance tests (PTT, FTT and ATT, respectively): twelve days after dietary reversal, mice were injected intraperitoneally with 2 g/kg BW each of the respective metabolites after an overnight fast. Glucose excursions during the following 2 h are shown. Insets are graphs of the calculated area under the curve (AUC). n = 9–10/group for PTT, n = 5–6/group for FTT and ATT). (I and J) Representative blots and densitometry of insulin signaling molecules in the liver. Mice were fasted overnight, and killed 12 min after intraperitoneal injection of 0.2 U/kg insulin. n = 7–12/group, from 2 to 3 independent experiments. *P < 0.05 compared to NC; #P < 0.05 compared to HFF; ** or ##P < 0.01 compared to NC or HFF, respectively; *** or ###P < 0.001 compared to NC or HFF, respectively; n.s. – non-significant (P > 0.05) compared to NC.

  3. Figure 3

    Adipose tissue adaptation to 2 weeks obesity reversal. (A and B) Representative blots and densitometry (n = 7–12/group from 2 to 3 independent experiments) of insulin signaling proteins in epididymal adipose tissue of mice treated as detailed in Fig. 2I and J. (C and D) Representative adipose tissue histological sections stained with H&E, and estimated size distribution. Flow Cytometry analyses of adipose tissue stromal-vascular cell fraction: Leucocytes were identified as CD45+ (E), adipose tissue macrophages (ATMs) as CD11b+F4/80+ (F), and ATM lipid content was assessed using the neutral lipid fluorescent dye BODIPY (H). Results are mean ± s.e.m. of mice from 2 independent experiments. Quantitative real-time PCR analysis in epididymal adipose tissue: Emr-1 (F4/80) as a marker for macrophages (F), Tnfa was used as a classical pro-inflammatory cytokine and with Itgax (Cd11c) as M1 markers, and Cd163 and Mgl2 as markers of alternatively activated, M2 macrophages (G). Values are the ΔΔCT adjusted to Rplp0 and Hprt, and are represented as fold of mean value of the NC group (n = 14–17/group from 5 independent experiments). (I) Representative blots and densitometry of adipose tissue stress signaling proteins. Densitometry are mean ± s.e.m. of n = 11–16/group from 3 to 4 independent experiments, with the mean value of the NC group defined as a value of 1. n.s. or *P < 0.05 vs NC, #P < 0.05 vs HFF. A full colour version of this figure is available at http://dx.doi.org/10.1530/JOE-17-0007.

  4. Figure 4

    Changes in adipocytokine secretion from adipose tissue upon dietary switch. (A) FFA concentrations in CM of adipose tissue (100 mg tissue/mL medium). (B) Total FFA production by epididymal fat pads. (C) Adipocytokine array was used to screen for changes induced by dietary switch in the secretion of adipocytokines by cultured adipose tissue fragments. The levels of each adipocytokine were quantified by densitometry and expressed as fold of HFF for each adipokine. The detected factors were divided to those which qualitatively changed similarly (black bars) or differently (white bars) to their change induced in HFF adipose tissue by incubation with p38MAPK + JNK inhibitors (results of this analysis is presented in Supplementary Fig. 1). Results are mean ± s.d. of 2–3 independent experiments.

  5. Figure 5

    Functional reversal of increased liver cell and macrophage lipid storage by adipose tissue conditioned medium. (A) Insulin-stimulated Akt phosphorylation in Hepa-1c hepatoma cells exposed to conditioned media of adipose tissue from NC, HFF, and HFF→NC mice. Shown is a representative blot and densitometry analyses normalized to the signal in insulin stimulated cells pre-treated with conditioned media of NC mice. N = 4–20 individual wells from 2 independent experiments, *P < 0.05 vs NC. Hepa-1c hepatoma cells (B) and RAW264.7 macrophages (C) were treated with adipose tissue conditioned media (AT-CM) for 6 h. Lipid droplets (LDs) were stained with BODIPY 493/503 neutral lipid dye and imaged by Operetta high content imaging system. Graphs represent the relative number of LDs per cell, compared to NC treated cells. The results are expressed as mean ± s.e.m. of five (Hepa-1c) or four (RAW264.7) independent experiments, each preformed in triplicates. * or #P < 0.05 compared to NC or HFF, respectively.

| Table of Contents
OPEN ACCESS ARTICLE

This Article

  1. J Endocrinol 233 293-305
  1. Free via Open Access: OA
  2. Free via Creative Commons: CC
  3. OA Abstract
  4. Figures Only
  5. Supplementary Data
  6. All Versions of this Article:
    1. JOE-17-0007v1
    2. JOE-17-0007v2
    3. 233/3/293 most recent

Article Metrics