Insulin resistance and white adipose tissue inflammation are uncoupled in energetically challenged Fsp27-deficient mice - PubMed (original) (raw)

Shi-Young Park 2, Li Xu 3, Xiayu Xia 1, Jing Ye 4, Lu Su 1, Kyeong-Hoon Jeong 2, Jang Ho Hur 2, Hyunhee Oh 2, Yoshikazu Tamori 5, Cristina M Zingaretti 6, Saverio Cinti 6, Jesús Argente 7, Miao Yu 1, Lizhen Wu 1, Shenghong Ju 8, Feifei Guan 9, Hongyuan Yang 10, Cheol Soo Choi 11, David B Savage 12, Peng Li 1

Affiliations

Insulin resistance and white adipose tissue inflammation are uncoupled in energetically challenged Fsp27-deficient mice

Linkang Zhou et al. Nat Commun. 2015.

Abstract

Fsp27 is a lipid droplet-associated protein almost exclusively expressed in adipocytes where it facilitates unilocular lipid droplet formation. In mice, Fsp27 deficiency is associated with increased basal lipolysis, 'browning' of white fat and a healthy metabolic profile, whereas a patient with congenital CIDEC deficiency manifested an adverse lipodystrophic phenotype. Here we reconcile these data by showing that exposing Fsp27-null mice to a substantial energetic stress by crossing them with ob/ob mice or BATless mice, or feeding them a high-fat diet, results in hepatic steatosis and insulin resistance. We also observe a striking reduction in adipose inflammation and increase in adiponectin levels in all three models. This appears to reflect reduced activation of the inflammasome and less adipocyte death. These findings highlight the importance of Fsp27 in facilitating optimal energy storage in adipocytes and represent a rare example where adipose inflammation and hepatic insulin resistance are disassociated.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Reduced fat mass in the _ob/ob/Fsp27_−/− mice.

Four-month-old ob/ob and _ob/ob/Fsp27_−/− mice were maintained on a chow diet for the analyses shown in (a–i). (a) computed tomography (CT) scan analysis of the mice; fat is shown in yellow. (b) GWAT of ob/ob and _ob/ob/Fsp27_−/− mice. (c) Body composition of ob/ob (_n_=5) and _ob/ob/Fsp27_−/− mice (_n_=6). (d) Decreased TAG content in the GWAT of the ob/ob and _ob/ob/Fsp27_−/− mice (_n_=5). (e) Abdominal magnetic resonance imaging (MRI) of ob/ob and _ob/ob/Fsp27_−/− mice (upper panel). Fat is shown in white in these MRI images. GWAT morphology (middle and lower panels). H&E, haematoxylin and eosin staining; EM, electron microscope. Scale bar, 64 and 2 μm for H&E staining and EM, respectively. (f) LD number per adipocyte in ob/ob and _ob/ob/Fsp27_−/− mice. The number of LDs in ten adipocytes was measured. (g) The average LD diameter in the GWAT of ob/ob and _ob/ob/Fsp27_−/− mice. The diameter of LDs in ten adipocytes was measured. (h) The distribution of fat cell size in the GWAT of ob/ob and _ob/ob/Fsp27_−/− mice. The fat cell area from 400 adipocytes was measured. (i) A representative western blotting showing the expression pattern of Fsp27, Perilipin1, Perilipin2, Cidea, Cyto C, Cox4, ATGL, CGI58, HSL, CEBPβ and PPARγ in the GWAT of ob/ob and _ob/ob/Fsp27_−/− mice. Actin was used as a loading control. Quantitative data are presented as mean±s.e.m. Significance was established using a two-tailed Student’s _t_-test. Differences were considered significant at P<0.05. ***P<0.001.

Figure 2

Figure 2. Reduced WAT inflammatory response in _ob/ob/Fsp27_−/− mice.

Four-month-old chow-fed ob/ob and _ob/ob/Fsp27_−/− mice were used for the analyses in (a–i). (a) Gene expression profiling in the GWAT of ob/ob and _ob/ob/Fsp27_−/− mice. Green circles represent downregulated genes in the _ob/ob/Fsp27_−/− mice compared with ob/ob mice. The intensity of the green colour indicates the extent of downregulation. (b) Relative mRNA expression of F4/80, CD11c and other pro-inflammatory genes, or (c) anti-inflammatory genes in the GWAT of ob/ob and _ob/ob/Fsp27_−/− mice (_n_=4 per group). (d) F4/80 immunohistochemical analysis in the GWAT of ob/ob and _ob/ob/Fsp27_−/− mice. Scale bar, 64 μm. (e) Serum concentration of IL-6 (_n_=7 per group). (f) Serum concentration of TNFα (_n_=7 for ob/ob and _n_=9 for _ob/ob/Fsp27_−/−). (g) Serum concentration of adiponectin (_n_=8 per group). (h) TNFα immunohistochemical analysis in the GWAT of ob/ob and _ob/ob/Fsp27_−/− mice showing the CLSs (left). Right: statistic analysis of the CLSs per 1,000 adipocytes. Scale bar, 50 μm. (i) Relative mRNA levels of ASC, NLRP3, Caspase-1 and TXNIP in the GWAT of ob/ob and _ob/ob/Fsp27_−/− mice (n_=3 per group). Quantitative data are presented as mean±s.e.m. Significance was established using a two-tailed Student’s t_-test. Differences were considered significant at P<0.05. *P<0.05, **P<0.01, ***P<0.001.

Figure 3

Figure 3. Hepatic steatosis in _ob/ob/Fsp27_−/− mice.

Four-month-old chow-fed ob/ob and _ob/ob/Fsp27_−/− mice were used (a–l). (a) Serum TAG concentrations (_n_=6 per group). (b) TAG content in the skeletal muscle (SM, gastrocnemius), heart and kidney (_n_=5 per group). (c) Photograph of the liver (top panel) and magnetic resonance imaging analysis of the liver section (lower panel). (d) Liver TAG content (_n_=8 for ob/ob and _n_=10 for ob/ob/Fsp27 −/− mice). (e) Liver CE content (_n_=5 per group). (f) Liver histology of ob/ob and _ob/ob/Fsp27_−/− mice. H&E, haematoxylin and eosin staining. EM, electron microscope image. Scale bar, 64 and 2 μm for HE and EM, respectively. (g) The average LD diameter in the liver of ob/ob and _ob/ob/Fsp27_−/− mice. The diameter of LDs in 50 cells was measured. (h) The distribution of LD size in the liver of ob/ob and ob/ob/Fsp27_−/− mice. (i–l) Relative mRNA expression levels in the livers of ob/ob and ob/ob/Fsp27_−/− mice (n_=4 per group). Quantitative data are presented as mean±s.e.m. Significance was established using a two-tailed Student’s t_-test. Differences were considered significant at P<0.05.*P<0.05, **P<0.01, ***P<0.001.

Figure 4

Figure 4. Insulin-resistant phenotype of _ob/ob/Fsp27_−/− mice.

(a) Fasting glucose and (b) fasting insulin concentrations of chow-fed 8.5-month-old ob/ob (_n_=6) and _ob/ob/Fsp27_−/−(_n_=7) mice. (c) Glucose tolerance tests (GTTs) and (d) ITTs in chow-fed 8.5-month-old ob/ob (_n_=7) and _ob/ob/Fsp27_−/− (_n_=8) mice. Peripheral and hepatic insulin sensitivity were assessed in 4-month-old male mice using hyperinsulinaemic–euglycaemic clamps (e–h, _ob/ob, n_=11; _ob/ob/Fsp27_−/−, n_=6). (e) Blood glucose concentrations during the clamp experiment. (f) Glucose infusion rates. (g) Peripheral glucose turnover. (h) Hepatic glucose output during basal and clamp conditions. (i) Relative mRNA level of G6pc and Pck1 in the liver (n_=4 per group). Insulin was injected in three pairs of anaesthetized 4-month-old male ob/ob and ob/ob/Fsp27_−/− mice. Representative images of basal and insulin-stimulated phospho-AKT (Ser473) levels in the muscle (j), BAT (k), gonadal fat (GWAT) (l) and liver (m). Quantitative data are presented as mean±s.e.m. Significance was established using a two-tailed Student’s t_-test. Differences were considered significant at P<0.05.*P<0.05, **P<0.01, ***P<0.001. Two-way repeated-measurement analyses of variance were used to evaluate the data in Fig. 4c,d,f (###P<0.001 in this figure indicates that the two groups respond differently following the intervention).

Figure 5

Figure 5. Hepatic steatosis and insulin resistance but reduced WAT inflammation in HFD-fed _Fsp27_−/− mice.

Three-month-old WT and _Fsp27_−/− mice were challenged with a HFD (D12331, 58% kcal of fat) for 3 months (a–g). (a) Morphology of WAT, BAT and liver in WT and _Fsp27_−/− mice. Scale bar, 64 μm for H&E (haematoxylin and eosin staining). Isolated hepatocytes were stained with bodipy 493/503. Scale bar, 10 μm. (b) TAG content in WAT, BAT and liver (_n_=5 per group). (c) Serum TAG concentrations (_n_=6 for WT and _n_=8 for _Fsp27_−/−). (d) Relative mRNA levels (_n_=4 per group). (e) Serum IL-6 concentrations (_n_=7 per group). (f) Serum adiponectin concentrations (_n_=8 per group). (g) Relative mRNA levels (_n_=4 per group). Three-month-old WT (_n_=6) and _Fsp27_−/− mice (_n_=6) were challenged with an HFD (D12492, 60% kcal of fat) for 6 weeks (h–n). Fasting blood glucose (h) and insulin (i) concentrations of WT and _Fsp27_−/− mice. Peripheral and hepatic insulin sensitivity were assessed using hyperinsulinaemic–euglycaemic clamps (j–l,n; WT, _n_=6; Fsp27_−/−, n_=6). (j) Blood glucose concentrations during the hyperinsulinaemic–euglycaemic clamp. (k) Glucose infusion rates. (l) Hepatic glucose output during the basal and hyperinsulinaemic clamp conditions. (m) Relative mRNA level of G6pc and Pck1 in the liver (n_=4 per group). (n) Peripheral glucose turnover. Quantitative data are presented as mean±s.e.m. Significance was established using a two-tailed Student’s t_-test. Differences were considered significant at P<0.05.*P<0.05, **P<0.01, ***P<0.001. Two-way repeated-measurement analyses of variance were used to evaluate the data in Fig. 5k (###P<0.001 in this figure indicates that the two groups respond differently following the intervention).

Figure 6

Figure 6. Reduced adipose inflammation but hepatic insulin resistance in BATless_/Fsp27_−/− mice.

Three-month-old BATless (n_=12) and BATless/Fsp27_−/− (n_=8) mice were challenged with HFD (D12492, 60% kcal of fat) for 6 weeks. (a) Body composition after the HFD. (b) Morphology of WAT (upper panel) and the liver (middle panel); Oil-red staining (lower panel) of the liver in BATless and BATless/Fsp27_−/−mice. Scale bar, 64 μm for H&E (haematoxylin and eosin staining). (c) Relative mRNA levels of macrophage markers F4/80 and MCP1 in WAT (BATless, n_=11; BATless/Fsp27_−/−, n_=7). (d) Plasma concentration of adiponectin (BATless, n_=11; BATless/Fsp27_−/−, n_=8). (e) Liver TAG content and (f) liver CE content (BATless, n_=9; BATless/Fsp27_−/−, n_=7). (g) Fasting glucose and (h) fasting insulin concentration in BATless and BATless/Fsp27_−/− mice. Peripheral and hepatic insulin sensitivity were assessed using hyperinsulinaemic–euglycaemic clamps (i–k). (i) Glucose infusion rates. (j) Hepatic glucose output during basal and hyperinsulinaemic clamp conditions. (k) Peripheral glucose turnover. Quantitative data are presented as mean±s.e.m. Significance was established using a two-tailed Student’s t_-test. Differences were considered significant at P<0.05.*P<0.05, **P<_0.01, ***P<0.001. Two-way repeated-measurement analyses of variance were used to evaluate the data in Fig. 6i (###P<0.001 in this figure indicates that the two groups respond differently following the intervention).

Figure 7

Figure 7. Liver-specific knockdown of Cidea alleviates hepatic stestosis in _ob/ob/Fsp27_−/− mice.

Four-month-old chow-fed ob/ob and _ob/ob/Fsp27_−/− mice were used to generate these data (a). Four-month-old chow-fed ob/ob or _ob/ob/Fsp27_−/− mice were injected with AD-shcontrol or AD-Cidea for 1 week before analysis (b–i). (a) Expression of the indicated proteins in the liver of ob/ob and _ob/ob/Fsp27_−/− mice. (b) Protein expression in the liver of _ob/ob/Fsp27_−/− mice injected with the indicated adenoviral vectors. (c) Liver H&E (haematoxylin and eosin) staining of _ob/ob/Fsp27_−/− mice injected with AD-shcontrol or AD-shCidea adenoviral vectors. Scale bar, 64 μm. (d) Liver TAG and (e) CE concentration in the livers of ob/ob and _ob/ob/Fsp27_−/− mice injected with the indicated adenovirusus (_n_=4). (f) Serum free fatty acid (_n_=7) and (g) TAG concentrations (n_=6). (h,i) Relative mRNA expression levels in the livers of adenovirus injected ob/ob and ob/ob/Fsp27_−/− mice (n_=4 per group). (j) Proposed model for the role of Fsp27 in regulating metabolism. Quantitative data are presented as mean±s.e.m. Significance was established using a two-tailed Student’s t_-test. Differences were considered significant at P<0.05.*P<0.05, **P<0.01, ***P<0.001.

Similar articles

Cited by

References

    1. Guilherme A., Virbasius J. V., Puri V. & Czech M. P. Adipocyte dysfunctions linking obesity to insulin resistance and type 2 diabetes. Nat. Rev. Mol. Cell Biol. 9, 367–377 (2008). - PMC - PubMed
    1. Puri V. et al.. Cidea is associated with lipid droplets and insulin sensitivity in humans. Proc. Natl Acad. Sci. USA 105, 7833–7838 (2008). - PMC - PubMed
    1. Gong J., Sun Z. & Li P. CIDE proteins and metabolic disorders. Curr. Opin. Lipidol. 20, 121–126 (2009). - PubMed
    1. Nordstrom E. A. et al.. A human-specific role of cell death-inducing DFFA (DNA fragmentation factor-alpha)-like effector A (CIDEA) in adipocyte lipolysis and obesity. Diabetes 54, 1726–1734 (2005). - PubMed
    1. Matsusue K. et al.. Hepatic steatosis in leptin-deficient mice is promoted by the PPARgamma target gene Fsp27. Cell Metab. 7, 302–311 (2008). - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources