Irf5 deficiency in macrophages promotes beneficial adipose tissue expansion and insulin sensitivity during obesity

Journal name:
Nature Medicine
Volume:
21,
Pages:
610–618
Year published:
DOI:
doi:10.1038/nm.3829
Received
Accepted
Published online

Abstract

Accumulation of visceral adipose tissue correlates with elevated inflammation and increased risk of metabolic diseases. However, little is known about the molecular mechanisms that control its pathological expansion. Transcription factor interferon regulatory factor 5 (IRF5) has been implicated in polarizing macrophages towards an inflammatory phenotype. Here we demonstrate that mice lacking Irf5, when placed on a high-fat diet, show no difference in the growth of their epididymal white adipose tissue (epiWAT) but they show expansion of their subcutaneous white adipose tissue, as compared to wild-type (WT) mice on the same diet. EpiWAT from Irf5-deficient mice is marked by accumulation of alternatively activated macrophages, higher collagen deposition that restricts adipocyte size, and enhanced insulin sensitivity compared to epiWAT from WT mice. In obese individuals, IRF5 expression is negatively associated with insulin sensitivity and collagen deposition in visceral adipose tissue. Genome-wide analysis of gene expression in adipose tissue macrophages highlights the transforming growth factor β1 (TGFB1) gene itself as a direct target of IRF5-mediated inhibition. This study uncovers a new function for IRF5 in controlling the relative mass of different adipose tissue depots and thus insulin sensitivity in obesity, and it suggests that inhibition of IRF5 may promote a healthy metabolic state during this condition.

At a glance

Figures

  1. Irf5 is specifically induced in visceral adipose tissue macrophages during obesity.
    Figure 1: Irf5 is specifically induced in visceral adipose tissue macrophages during obesity.

    (a) Irf5 gene profiling in ingWAT, epiWAT and retWAT of mice fed with a NCD or HFD for 12 weeks (n = 5 in each group). (b) Protein levels of Irf5 in ingWAT and epiWAT from lean and diet-induced obese mice (after 12 weeks of NCD or HFD, respectively). (c) Gene expression of Irf5 in the epiWAT of lean mice after 12 weeks on the NCD and diet-induced obese mice subjected to 4, 8 and 12 weeks of a HFD (n = 5). (d) Gene profiling of Irf5 in epiWAT fractions including adipocytes (Adipo) and the SVF composed of macrophages (immunoselected F4/80+ cells) and F4/80 cells (Neg) (n = 5 in each group). (e) Measurement of Irf5 mRNA levels in isolated adipose-tissue macrophages from the ingWAT and epiWAT of lean and diet-induced obese mice after 12 weeks of NCD or HFD (n = 5). (f) Flow cytometry analyses of stromal fractions isolated from ingWAT and epiWAT of mice fed a NCD or HFD for 12 weeks (n = 5). Macrophages were identified as Cd11b+F4/80+ cells. Differences between groups (NCD versus HFD) were determined by non-parametric unpaired Mann–Whitney comparative tests or analysis of variance (ANOVA), Bonferroni's post hoc test; *P < 0.05, **P < 0.01, ***P < 0.001 (from post hoc test). Data represent mean values ± s.e.m. (g) Correlative analysis between Irf5 mRNA levels and percentages of pro-inflammatory M1-like macrophages identified as Cd11b+F4/80+Cd11c+ cells in the same epiWAT of mice after 8–12 weeks of HFD (n = 16). Correlation was assessed by non-parametric Spearman's test (r = 0.98, P < 0.001). mRNA levels are relative to 18S rRNA for all experiments.

  2. Irf5 deficiency promotes intra-abdominal adipose tissue remodeling and type 2 immune responses while enhancing inguinal adiposity.
    Figure 2: Irf5 deficiency promotes intra-abdominal adipose tissue remodeling and type 2 immune responses while enhancing inguinal adiposity.

    (a) Weight of WT and KO mice during 12 weeks of NCD or HFD. (b) Left, fat mass in WT and KO mice. Right, weights of ingWAT, epiWAT, retWAT from WT and KO mice, and the sum of these weights (total WAT). (c) Representative images (left; n = 3 images per mouse) and quantification (top right) of collagen deposition in WAT from WT or KO mice. Scale bars, 100 μm. Collagen mRNA levels (Col1a1, Col3a1 and Col4a1) in WAT from WT or KO mice (bottom right). (d) Left, representative H&E stain images (n = 3 images per mouse) of WAT. Right, quantification of adipocyte size in WAT of WT and KO HFD-fed mice (right). Scale bars, 100 μm. (e) Correlative analyses between Irf5 mRNA levels and adipocyte diameters in the epiWAT of WT mice (n = 12). Correlations were assessed by non-parametric Spearman's test (r = 0.87, P = 0.004). (f) Representative images (n = 3 images per mouse) of immunohistochemical staining for Cd68 in epiWAT from WT and KO mice. Right, higher-magnification images of the boxed regions. Scale bars, 100 μm. (g) Left, representative flow cytometry plots showing frequencies of Cd11b+F4/80+ macrophages among stromal Cd45+ cells isolated from the epiWAT of WT (33.9% ± 7.7) and KO (56.2% ± 5.1) mice. Right, macrophages expressed per gram of epiWAT. (h) Quantification of Cd11b+F4/80+ Cd206+ (left) or Cd11c+ (right) cells isolated from epiWAT of WT or KO mice. (i) Concentrations of cytokines in epiWAT culture media of WT or KO mice. (j) Left, quantification of Cd45+Cd3+ lymphocytes among stromal Cd45+ cells isolated from the epiWAT of WT or KO mice. Right, concentrations of cytokines in epiWAT culture medium of WT or KO mice. Unless otherwise indicated, n = 5 for WT, 6 for KO and measurements taken after 12 weeks on the HFD. Data represent mean values ± s.e.m. Differences between WT and KO were determined by non-parametric unpaired Mann–Whitney comparative tests or ANOVA with Bonferroni's post hoc test; *P < 0.05, **P < 0.01, ***P < 0.001 (from post hoc test).

  3. Myeloid cell-specific Irf5 deletion promotes subcutaneous adiposity and intra-abdominal adipose tissue remodeling.
    Figure 3: Myeloid cell–specific Irf5 deletion promotes subcutaneous adiposity and intra-abdominal adipose tissue remodeling.

    (a) Weight gain of MacWT (n = 5) and MacKO (n = 6) mice during 12 weeks of the HFD. (b) Left, fat mass of MacWT and MacKO mice after 12 weeks of the HFD. Right, weights of ingWAT, epiWAT and retWAT from MacWT (n = 5) and Irf5 MacKO (n = 6), and the sum of these weights (total WAT) after 12 weeks of the HFD (right). (c) Representative images (left; n = 3 images per KO or WT mice) and quantification of collagen deposition (right) in epiWAT from MacWT (n = 5) or MacKO (n = 6) mice after 12 weeks of the HFD. Scale bars, 100 μm. (d) Quantification of stromal Cd45+F4/80+Cd11b+ cells expressed per gram of epiWAT (left) and Cd206+ cells expressed as percentage among Cd45+F4/80+Cd11b+ cells isolated from epiWAT of MacWT (n = 5) or MacKO (n = 6) mice on the HFD (right). (e) Quantification of stromal Cd45+Cd3+ cells (n = 5 for WT, 6 for KO). (f) Left, representative flow cytometry plots showing frequencies of Gata3+IL-13+ TH2 cells among Cd45+Cd3+ cells isolated from the epiWAT of HFD-fed WT (8.6% ± 1.5; n = 5) and KO (20.8% ± 2.5; n = 5) mice (left). Right, TH2 cells expressed per gram of epiWAT. (g) Concentrations of IL-5 and IL-13 in epiWAT culture medium of HFD-fed MacWT (n = 5) or MacKO mice (n = 6). (h) Left, representative flow cytometry plots showing frequencies of Cd11b+Siglec-F+ eosinophils among stromal Cd45+ cells isolated from the epiWAT of HFD-fed WT (0.8% ± 0.27; n = 5) or KO (2.05% ± 0.33; n = 6) mice. Right, eosinophils expressed per gram of epiWAT. Data represent mean ± s.e.m. Differences between MacWT and MacKO were determined by non-parametric unpaired Mann–Whitney comparative tests or ANOVA with Bonferroni's post hoc test; *P < 0.05, **P < 0.01, ***P < 0.001 (from post hoc test).

  4. Irf5 depletion preserves glucose homeostasis in obesity.
    Figure 4: Irf5 depletion preserves glucose homeostasis in obesity.

    (a,b) Fasting blood concentrations of glucose (a) and insulin (b) in HFD-fed WT (n = 5 in a and 7 in b), whole-body KO (n = 6), MacWT (n = 6 in a and 7 in b) and MacKO (n = 6) mouse models. (c) OGTT in WT (n = 5) and KO (n = 6) (left) and MacWT (n = 5) and MacKO (n = 6) (right) mice fed a HFD for 12 weeks. Results are representative of three independent experiments. (d) Insulin tolerance test in MacWT and MacKO adjusted on the percentage of fat mass gained after 12 (MacKO) or 16 (MacWT) weeks of HFD (n = 5). (e) HOMA-IR index normalized to the fat mass of WT and KO mice and MacWT and MacKO models (n = 5). (f) MacWT and MacKO HFD-fed mice were injected intraperitoneally with 1 U/kg insulin. Immunoblot analyses of epiWAT (left) and ingWAT (right) tissue samples using antibodies specific to AKT, phospho-AKT (pAKT-S473) and phospho-Gsk3 (pGsk3). Results are representative of at least four independent experiments. (g,h) ingWAT and epiWAT explants from HFD-fed MacWT or MacKO mice stimulated ex vivo with insulin (100 nM) for 10 min. (g) Quantification of intracellular radiolabeled [3H]-2-deoxyglucose from both depots (n = 3). (h) Insulin signaling (AKT, pAKT-S473, pGsk3) in epiWAT under conditions similar to those described in f. Results are representative of at least four independent experiments. (i) Adiponectin concentrations in epiWAT culture medium and serum of MacWT and MacKO (n = 5). Data represent mean values ± s.e.m. Differences between groups (MacWT versus MacKO) were determined by non-parametric unpaired Mann–Whitney comparative tests or ANOVA with Bonferroni's post hoc test; *P < 0.05, **P < 0.01, ***P < 0.001 (from post hoc test).

  5. IRF5 expression in macrophages is associated with adipose tissue collagen deposition in human obesity.
    Figure 5: IRF5 expression in macrophages is associated with adipose tissue collagen deposition in human obesity.

    (a) Representative images of immunohistochemical staining for IRF5, CD68 and CD11c in the viscWAT of obese subjects (n = 4). Scale bars, 100 μm (top) and 50 μm (bottom). (b) IRF5 mRNA in viscWAT immunoselected cell fractions and the double-negative fraction isolated from morbidly obese subjects (n = 10; population 1). Adipo, adipocytes; CD14+, macrophages; CD3+, lymphocytes; Neg, double-negative cells. (c,d) IRF5 mRNA (c) and representative western blot analyses (d) of IRF5 protein from immunoselected CD14+ macrophages isolated from the viscWAT of non-obese (n = 4) and morbidly obese (n = 21, population 1) subjects. Images are representative of two non-obese and five obese subjects. (e) IRF5 gene profiling in viscWAT from lean (LE; n = 12), overweight (OV; n = 12), obese (OB; n = 14) subjects and patients with metabolic syndrome (MS; n = 14) (population 2). (fi) Correlation analyses between IRF5 mRNA levels and fat mass (r = 0.52, P < 0.0001) (f), waist circumference (r = 0.60, P < 0.0001) (g), glucose disposal rate (r = −0.46, P = 0.0005) (h) and HOMA-IR (r = −0.77, P = 0.0012) (i). (j) Correlative analysis of IRF5 mRNA levels and percentages of collagen accumulation in viscWAT from obese subjects (r = −0.77, P < 0.0001, n = 15). Correlations were assessed by non-parametric Spearman's test. Data represent mean values ± s.e.m. Differences between groups (non-obese versus obese or LE versus OV versus OB versus MS) were determined by non-parametric unpaired Mann–Whitney comparative tests or ANOVA, Bonferroni's post hoc test; *P < 0.05, **P < 0.01, ***P < 0.001 (from post hoc test).

  6. IRF5 controls transcription of the TGFB1 gene in human adipose-tissue macrophages.
    Figure 6: IRF5 controls transcription of the TGFB1 gene in human adipose-tissue macrophages.

    (a) Regulatory model of IRF5 activity based on pangenomic profiling of human adipose tissue macrophages isolated from the viscWAT CD14+ cells of obese patients (n = 21; population 1). (b) Correlative analysis between mRNA levels of IRF5 and TGFB1 in viscWAT CD14+ cells from obese subjects (r = −0.85, P < 0.0001, n = 21; population 1). (c) Correlative analyses of TGFB1 mRNA levels and percentages of collagen accumulation in viscWAT from obese subjects (r = 0.68, P = 0.006, n = 15; population 1). (d) mRNA quantification of the TGFB1 gene with adenoviral overexpression of IRF5 or GFP in immunoselected CD14+ isolated from the viscWAT of obese subjects (n = 5; population 1). (e) mRNA quantification of the TGFB1 gene with IRF5-specific (siIRF5) or control (siCTL) siRNA in human macrophage–derived monocytes (HMDMs) stimulated by adipose tissue–conditioned medium (ATCM) for 24 h (n = 5 for each group). (f) Tgfb1 gene profiling in epiWAT stromal cells isolated from KO (n = 6), MacKO (n = 6) and WT littermates (n = 5). (g,h) Recruitment of IRF5 and RNA polymerase II (POL2) to the TGFB1 gene promoter of HMDMs overexpressing IRF5 or GFP and stimulated with ATCM (g) and viscWAT stromal cells isolated from obese subjects (h) (n = 4; population 1). (i) mRNA quantification of collagen genes in epiWAT explants from obese MacWT and MacKO treated with a neutralizing antibody against TGF-β1 or IgG (control) for 24 h (n = 4). Data represent mean values ± s.e.m. Differences between groups were determined by non-parametric unpaired Mann–Whitney comparative tests or ANOVA with Bonferroni's post hoc test; *P < 0.05, **P < 0.01, ***P < 0.001 (from post hoc test). (j) Schematic illustration depicting the role of IRF5 in adipose tissue remodeling and glucose tolerance.

Accession codes

Primary accessions

Gene Expression Omnibus

References

  1. Tchernof, A. & Despres, J.P. Pathophysiology of human visceral obesity: an update. Physiol. Rev. 93, 359404 (2013).
  2. Klöting, N. et al. Insulin-sensitive obesity. Am. J. Physiol. Endocrinol. Metab. 299, E506E515 (2010).
  3. Abate, N., Garg, A., Peshock, R.M., Stray-Gundersen, J. & Grundy, S.M. Relationships of generalized and regional adiposity to insulin sensitivity in men. J. Clin. Invest. 96, 8898 (1995).
  4. Lawrence, T. & Natoli, G. Transcriptional regulation of macrophage polarization: enabling diversity with identity. Nat. Rev. Immunol. 11, 750761 (2011).
  5. Odegaard, J.I. & Chawla, A. Pleiotropic actions of insulin resistance and inflammation in metabolic homeostasis. Science 339, 172177 (2013).
  6. Liao, X. et al. Kruppel-like factor 4 regulates macrophage polarization. J. Clin. Invest. 121, 27362749 (2011).
  7. Odegaard, J.I. et al. Macrophage-specific PPAR-γ controls alternative activation and improves insulin resistance. Nature 447, 11161120 (2007).
  8. Graham, R.R. et al. A common haplotype of interferon regulatory factor 5 (IRF5) regulates splicing and expression and is associated with increased risk of systemic lupus erythematosus. Nat. Genet. 38, 550555 (2006).
  9. Krausgruber, T. et al. IRF5 promotes inflammatory macrophage polarization and TH1-TH17 responses. Nat. Immunol. 12, 231238 (2011).
  10. Bertola, A. et al. Identification of adipose tissue dendritic cells correlated with obesity-associated insulin-resistance and inducing Th17 responses in mice and patients. Diabetes 61, 22382247 (2012).
  11. Dalmas, E. et al. T cell-derived IL-22 amplifies IL-1β-driven inflammation in human adipose tissue: relevance to obesity and type 2 diabetes. Diabetes 63, 19661977 (2014).
  12. Fabbrini, E. et al. Association between specific adipose tissue CD4+ T-cell populations and insulin resistance in obese individuals. Gastroenterology 145, 366374 e1–3 (2013).
  13. Bourlier, V. et al. Remodeling phenotype of human subcutaneous adipose tissue macrophages. Circulation 117, 806815 (2008).
  14. Xu, X. et al. Obesity activates a program of lysosomal-dependent lipid metabolism in adipose tissue macrophages independently of classic activation. Cell Metab. 18, 816830 (2013).
  15. Kim, J.Y. et al. Obesity-associated improvements in metabolic profile through expansion of adipose tissue. J. Clin. Invest. 117, 26212637 (2007).
  16. Wander, P.L. et al. Change in visceral adiposity independently predicts a greater risk of developing type 2 diabetes over 10 years in Japanese Americans. Diabetes Care 36, 289293 (2013).
  17. Arner, P., Arner, E., Hammarstedt, A. & Smith, U. Genetic predisposition for type 2 diabetes, but not for overweight/obesity, is associated with a restricted adipogenesis. PLoS ONE 6, e18284 (2011).
  18. Hoffstedt, J. et al. Regional impact of adipose tissue morphology on the metabolic profile in morbid obesity. Diabetologia 53, 24962503 (2010).
  19. Sun, K., Tordjman, J., Clement, K. & Scherer, P.E. Fibrosis and adipose tissue dysfunction. Cell Metab. 18, 470477 (2013).
  20. Henegar, C. et al. Adipose tissue transcriptomic signature highlights the pathological relevance of extracellular matrix in human obesity. Genome Biol. 9, R14 (2008).
  21. Chun, T.H. et al. A pericellular collagenase directs the 3-dimensional development of white adipose tissue. Cell 125, 577591 (2006).
  22. Vila, I.K. et al. Immune cell toll-like receptor 4 mediates the development of obesity- and endotoxemia-associated adipose tissue fibrosis. Cell Rep. 7, 11161129 (2014).
  23. Khan, T. et al. Metabolic dysregulation and adipose tissue fibrosis: role of collagen VI. Mol. Cell. Biol. 29, 15751591 (2009).
  24. Lee, J.T. et al. Macrophage metalloelastase (MMP12) regulates adipose tissue expansion, insulin sensitivity, and expression of inducible nitric oxide synthase. Endocrinology 155, 34093420 (2014).
  25. Divoux, A. et al. Fibrosis in human adipose tissue: composition, distribution, and link with lipid metabolism and fat mass loss. Diabetes 59, 28172825 (2010).
  26. Arner, E. et al. Adipocyte turnover: relevance to human adipose tissue morphology. Diabetes 59, 105109 (2010).
  27. Tordjman, J. et al. Structural and inflammatory heterogeneity in subcutaneous adipose tissue: relation with liver histopathology in morbid obesity. J. Hepatol. 56, 11521158 (2012).
  28. Halberg, N. et al. Hypoxia-inducible factor 1α induces fibrosis and insulin resistance in white adipose tissue. Mol. Cell. Biol. 29, 44674483 (2009).
  29. Wernstedt Asterholm, I. et al. Adipocyte inflammation is essential for healthy adipose tissue expansion and remodeling. Cell Metab. 20, 103118 (2014).
  30. Wynn, T.A. & Barron, L. Macrophages: master regulators of inflammation and fibrosis. Semin. Liver Dis. 30, 245257 (2010).
  31. Song, E. et al. Influence of alternatively and classically activated macrophages on fibrogenic activities of human fibroblasts. Cell. Immunol. 204, 1928 (2000).
  32. Keophiphath, M. et al. Macrophage-secreted factors promote a profibrotic phenotype in human preadipocytes. Mol. Endocrinol. 23, 1124 (2009).
  33. Lazear, H.M. et al. IRF-3, IRF-5, and IRF-7 coordinately regulate the type I IFN response in myeloid dendritic cells downstream of MAVS signaling. PLoS Pathog. 9, e1003118 (2013).
  34. Takaoka, A. et al. Integral role of IRF-5 in the gene induction programme activated by toll-like receptors. Nature 434, 243249 (2005).
  35. Spencer, M. et al. Adipose tissue extracellular matrix and vascular abnormalities in obesity and insulin resistance. J. Clin. Endocrinol. Metab. 96, E1990E1998 (2011).
  36. Seki, E. et al. TLR4 enhances TGF-β signaling and hepatic fibrosis. Nat. Med. 13, 13241332 (2007).
  37. Wu, D. et al. Eosinophils sustain adipose alternatively activated macrophages associated with glucose homeostasis. Science 332, 243247 (2011).
  38. Petruschke, T., Rohrig, K. & Hauner, H. Transforming growth factor beta (TGF-beta) inhibits the differentiation of human adipocyte precursor cells in primary culture. Int. J. Obes. Relat. Metab. Disord. 18, 532536 (1994).
  39. Ignotz, R.A. & Massague, J. Type beta transforming growth factor controls the adipogenic differentiation of 3T3 fibroblasts. Proc. Natl. Acad. Sci. USA 82, 85308534 (1985).
  40. Klimčáková, E. et al. Worsening of obesity and metabolic status yields similar molecular adaptations in human subcutaneous and visceral adipose tissue: decreased metabolism and increased immune response. J. Clin. Endocrinol. Metab. 96, E73E82 (2011).
  41. Klimcáková, E. et al. Macrophage gene expression is related to obesity and the metabolic syndrome in human subcutaneous fat as well as in visceral fat. Diabetologia 54, 876887 (2011).
  42. Krausgruber, T. et al. IRF5 promotes inflammatory macrophage polarization and TH1-TH17 responses. Nat. Immunol. 12, 231238 (2011).
  43. Joly-Amado, A. et al. Hypothalamic AgRP-neurons control peripheral substrate utilization and nutrient partitioning. EMBO J. 31, 42764288 (2012).
  44. Berglund, E.D. et al. Glucose metabolism in vivo in four commonly used inbred mouse strains. Diabetes 57, 17901799 (2008).
  45. Hajduch, E., Darakhshan, F. & Hundal, H.S. Fructose uptake in rat adipocytes: GLUT5 expression and the effects of streptozotocin-induced diabetes. Diabetologia 41, 821828 (1998).
  46. Kammoun, H.L. et al. GRP78 expression inhibits insulin and ER stress-induced SREBP-1c activation and reduces hepatic steatosis in mice. J. Clin. Invest. 119, 12011215 (2009).
  47. Divoux, A. et al. Fibrosis in human adipose tissue: composition, distribution, and link with lipid metabolism and fat mass loss. Diabetes 59, 28172825 (2010).
  48. Rouault, C. et al. Roles of chemokine ligand-2 (CXCL2) and neutrophils in influencing endothelial cell function and inflammation of human adipose tissue. Endocrinology 154, 10691079 (2013).
  49. Prieur, X. et al. Differential lipid partitioning between adipocytes and tissue macrophages modulates macrophage lipotoxicity and M2/M1 polarization in obese mice. Diabetes 60, 797809 (2011).
  50. Toubal, A. et al. SMRT-GPS2 corepressor pathway dysregulation coincides with obesity-linked adipocyte inflammation. J. Clin. Invest. 123, 362379 (2013).
  51. Storey, J.D. in International Encyclopedia of Statistical Science 1st edn, (ed. Lovric, M.) False Discovery Rate (Springer, 2011).
  52. Prifti, E., Zucker, J.D., Clement, K. & Henegar, C. FunNet: an integrative tool for exploring transcriptional interactions. Bioinformatics 24, 26362638 (2008).
  53. Arch, J.R., Hislop, D., Wang, S.J. & Speakman, J.R. Some mathematical and technical issues in the measurement and interpretation of open-circuit indirect calorimetry in small animals. Int. J. Obes. (Lond) 30, 13221331 (2006).
  54. Even, P.C., Mokhtarian, A. & Pele, A. Practical aspects of indirect calorimetry in laboratory animals. Neurosci. Biobehav. Rev. 18, 435447 (1994).
  55. Weir, J.B. New methods for calculating metabolic rate with special reference to protein metabolism. 1949. Nutrition 6, 213221 (1990).

Download references

Author information

  1. These authors contributed equally to this work.

    • Elise Dalmas,
    • Amine Toubal &
    • Fawaz Alzaid

Affiliations

  1. Sorbonne Universités, Université Pierre et Marie-Curie; INSERM UMR_S 1166-ICAN, Nutriomics, Paris, France.

    • Elise Dalmas,
    • Amine Toubal,
    • Maria Pini,
    • Patricia Ancel,
    • Yin Ling,
    • Omran Allatif,
    • Sébastien André,
    • Christine Poitou,
    • Judith Aron-Wisnewsky &
    • Karine Clément
  2. Institute of Cardiometabolism and Nutrition, Paris, France.

    • Elise Dalmas,
    • Amine Toubal,
    • Fawaz Alzaid,
    • Kristell Lebozec,
    • Maria Pini,
    • Isabelle Hainault,
    • Patricia Ancel,
    • Amélie Lacombe,
    • Yin Ling,
    • Omran Allatif,
    • Sébastien André,
    • Christine Poitou,
    • Fabienne Foufelle,
    • Judith Aron-Wisnewsky,
    • Karine Clément &
    • Nicolas Venteclef
  3. Sorbonne Universités, Université Pierre et Marie-Curie, INSERM, UMR_S 1138 Cordeliers Research Center, Paris, France.

    • Amine Toubal,
    • Fawaz Alzaid,
    • Kristell Lebozec,
    • Isabelle Hainault,
    • Fabienne Foufelle &
    • Nicolas Venteclef
  4. Kennedy Institute Trust of Rheumatology, University of Oxford, Oxford, UK.

    • Katrina Blazek,
    • Hayley L Eames &
    • Irina A Udalova
  5. INSERM, University of Toulouse, Paul Sabatier University, UMR 1048, Toulouse, France.

    • Emilie Montastier
  6. Department of Clinical Biochemistry, Toulouse University Hospitals, Toulouse, France.

    • Emilie Montastier &
    • Nathalie Viguerie
  7. Department of Nutrition, Toulouse University Hospitals, Toulouse, France.

    • Emilie Montastier &
    • Nathalie Viguerie
  8. Université Paris Diderot, Sorbonne Paris Cité, Unité de Biologie Fonctionnelle et Adaptative, CNRS UMR 8251, Paris, France.

    • Raphaël G P Denis,
    • Céline Cruciani-Guglielmacci &
    • Serge Luquet
  9. Heart and Metabolism Division, Pitié-Salpêtrière Hospital, Assistance Publique-Hôpitaux de Paris, Paris, France.

    • Christine Poitou,
    • Judith Aron-Wisnewsky,
    • Dominique Langin &
    • Karine Clément
  10. Department of Sports Medicine, Third Faculty of Medicine, Charles University in Prague, Prague, Czech Republic.

    • Vladimir Stich &
    • Dominique Langin
  11. Franco-Czech Laboratory for Clinical Research on Obesity, Third Faculty of Medicine, Charles University in Prague, Prague, Czech Republic.

    • Vladimir Stich
  12. Visceral Surgery Division, Pitié-Salpêtrière Hospital, Assistance Publique-Hôpitaux de Paris, Paris, France.

    • Alexandra Torcivia

Contributions

E.D., I.A.U. and N. Venteclef conceived the study and wrote the manuscript. E.D. performed part of the in vivo studies (human and mouse) and analyzed data. A. Toubal and F.A. performed the in vivo studies and data analyses, and assisted in the preparation of the manuscript. K.B., H.L.E., M.P., I.H., Y.L. and P.A. assisted with the mouse studies. S.A. and K.L. assisted with flow cytometry analyses. A.L., R.G.P.D. and C.C.-G. performed the metabolic analyses in vivo. E.M., N. Viguerie, C.P., V.S., A. Torcivia, J.A.-W., D.L. and K.C. contributed to the human data collection, data analyses and interpretation. O.A. and K.C. performed statistical analyses in population 1. F.F., S.L., J.A.-W., D.L. and K.C. interpreted and assisted in the writing of the manuscript. I.A.U. and N. Venteclef designed, analyzed and interpreted the studies.

Competing financial interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to:

Author details

Supplementary information

PDF files

  1. Supplementary Text and Figures (1,620 KB)

    Supplementary Figures 1–7 and Supplementary Table 1

Additional data