Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2015 Dec 28;113(2):446–451. doi: 10.1073/pnas.1517142113

Gsα deficiency in adipose tissue improves glucose metabolism and insulin sensitivity without an effect on body weight

Yong-Qi Li a, Yogendra B Shrestha a, Min Chen a, Tatyana Chanturiya b, Oksana Gavrilova b, Lee S Weinstein a,1
PMCID: PMC4720348  PMID: 26712027

Significance

The G protein Gsα mediates the activation of lipolysis in white and brown adipose tissue (WAT and BAT) and thermogenesis in BAT by sympathetic nerves. In this study we show that Gsα deficiency in BAT and WAT leads to cold intolerance and inactive BAT in mice, as expected, but does not alter body weight or energy balance, associated with a reduction in both lipolysis and lipogenesis. Despite having similar adiposity, these mice had improved insulin sensitivity and glucose metabolism, possibly secondary to reduced release of free fatty acids and fatty acid binding protein 4 from adipose tissue. These findings show that Gsα in adipocytes is not required to maintain energy balance but still has an important effect on glucose metabolism.

Keywords: G proteins, adipocyte, insulin sensitivity, lipid metabolism, brown adipose tissue

Abstract

Gsα, the G protein that transduces receptor-stimulated cAMP generation, mediates sympathetic nervous system stimulation of brown adipose tissue (BAT) thermogenesis and browning of white adipose tissue (WAT), which are both potential targets for treating obesity, as well as lipolysis. We generated a mouse line with Gsα deficiency in mature BAT and WAT adipocytes (Ad-GsKO). Ad-GsKO mice had impaired BAT function, absent browning of WAT, and reduced lipolysis, and were therefore cold-intolerant. Despite the presence of these abnormalities, Ad-GsKO mice maintained normal energy balance on both standard and high-fat diets, associated with decreases in both lipolysis and lipid synthesis. In addition, Ad-GsKO mice maintained at thermoneutrality on a standard diet also had normal energy balance. Ad-GsKO mice had improved insulin sensitivity and glucose metabolism, possibly secondary to the effects of reduced lipolysis and lower circulating fatty acid binding protein 4 levels. Gsα signaling in adipose tissues may therefore affect whole-body glucose metabolism in the absence of an effect on body weight.

The sympathetic nervous system (SNS) regulates energy homeostasis and adiposity through several mechanisms, including activation of nonshivering thermogenesis in brown adipose tissue (BAT), browning (formation of BAT-like “beige” cells) of white adipose tissue (WAT), and stimulation of lipolysis. Although these processes have been shown to be potential targets in treating obesity and diabetes (1–3), ablation of sympathetic nerves (4–6) or their main effectors (norepinephrine and epinephrine) (7) does not result in obesity or insulin resistance. Although mice lacking β adrenergic receptors (β-less mice) do develop obesity (8), it is likely that this effect is not due only to loss of β-adrenergic signaling in adipose tissue.

The main mediator of SNS function in adipose tissues is Gsα (9, 10), a ubiquitously expressed G protein α-subunit that in adipose tissue couples adrenergic and other receptors, such as the adenosine A2A receptor (11), to the generation of intracellular cAMP. We have previously generated adipose-specific Gsα knockout mice (FGsKO) using fatty acid binding protein 4 (FABP4) (aP2)-cre and showed these mice to have significant early mortality and a severely lean phenotype (12). However, the usefulness of this model to examine the role of Gsα in mature adipocytes is limited due to both the lack of specificity of FABP4-cre expression in adipose tissue and the presence of a severe defect in adipogenesis due to expression of FABP4, and therefore loss of Gsα, during an early step in adipocyte differentiation.

Adiponectin is a mature adipocyte marker expressed late in adipocyte differentiation (13). The more recent availability of adiponectin-cre mouse lines (14, 15) has enabled us to generate adipose-specific Gsα knockout mice (Ad-GsKO) in which Gsα deletion is restricted to mature adipocytes. Despite having loss of BAT function or browning of WAT, Ad-GsKO mice failed to develop obesity on either standard chow or a high-fat diet (HFD). Moreover, Ad-GsKO mice had improved glucose tolerance and insulin sensitivity associated with a significant reduction of circulating FABP4. Our results show that thermogenesis in BAT and in beige adipocytes is not required for normal weight maintenance and that Gsα signaling in adipose tissue has an effect on whole-body glucose metabolism independent of adiposity.

Results

Ad-GsKO Mice Have Normal Energy Balance.

We generated mice with adipose tissue-specific Gsα deficiency (Ad-GsKO: adiponectin-cre+, Gsαfl/fl). Loss of Gsα expression in WAT and BAT adipocytes was confirmed by immunoblotting (Fig. 1A), whereas Gsα expression was unaffected in other tissues (brain, liver, or kidney; Fig. S1). The number of mutants alive at weaning was consistent with expected Mendelian ratios and no premature death was observed up until age 4 mo.

Fig. 1.

Fig. 1.

Open in a new tab

Energy balance is unaffected in Ad-GsKO mice. (A) Immunoblots of adipocytes isolated from interscapular BAT and Epi WAT of control and Ad-GsKO mice with Gsα and β-actin antibodies. (B and C) Body weight curves of male Ad-GsKO and control mice fed a standard diet (B) or HFD (C). HFD was started at 9 wk of age. (D and E) Total, fat, and lean mass in 4- to 5-mo-old male mice maintained on standard diet (D) or HFD (E). (F and G) Body length (F) and organ weights (G) (expressed as percent body weight) of 4- to 5-mo-old Ad-GsKO and control mice on standard diet. BAT, interscapular BAT; Retro, retroperitoneal WAT. (H and I) Mean adipocyte diameter (H) and adipocyte size (I) distribution in Epi WAT from 4-mo-old male Ad-GsKO and control mice on standard diet. (J) Food intake of 4-mo-old male Ad-GsKO mice and controls. (K and L) Resting and total energy expenditure rate (K) and ambulatory (Amb) and total activity levels (L) measured over 24 h at 24 °C and 30 °C in 4-mo-old male mice on standard diet (n = 6–8 per group). (M) Respiratory exchange ratios (vCO2/vO2) measured over three consecutive days from 1200 to 1800 hours in 4-mo-old male mice on standard diet. Data are mean ± SEM, n = 5–8 per group. *P < 0.05 vs. controls.

Fig. S1.

Fig. S1.

Open in a new tab

Gsα expression is unaffected in brain, liver, and kidney of Ad-GsKO mice. Immunoblots of cerebral cortex (brain), liver, and kidney extracts from control and Ad-GsKO mice with Gsα- and β-actin–specific antibodies. The Gsα doublet is due to long and short forms of Gsα resulting from alternative splicing.

Both male and female Ad-GsKO mice had normal body weight and composition on both standard diet and HFD (Fig. 1 B–E and Fig. S2 A and B), as well as normal body length (Fig. 1F). Interscapular BAT weight was significantly increased in Ad-GsKO mice on standard diet (Fig. 1G), consistent with this tissue’s being metabolically inactive and having enlarged adipocytes (Fig. 2 and discussed below). However, there were no differences in the weights of inguinal (Ing), retroperitoneal, or epididymal (Epi) WAT pads (Fig. 1G), and adipocytes in Epi WAT of Ad-GsKO mice on standard diet showed a normal mean diameter and size distribution (Fig. 1 H and I). Kidney, heart, and liver weights were also unaffected (Fig. 1G). Consistent with no change in adiposity, serum leptin levels were similar in control and Ad-GsKO mice (Table 1). Food intake (Fig. 1J), energy expenditure (Fig. 1K), and activity levels (Fig. 1L) were all unaffected in Ad-GsKO mice maintained on standard diet. Respiratory exchange ratios (RERs, vCO2/vO2) measured from 1200 to 1800 hours at a time when mice do not generally eat tended to be higher in Ad-GsKO mice (P = 0.08) (Fig. 1M), suggesting that Ad-GsKO mice probably metabolize a lower ratio of lipid to carbohydrate during this time period.

Fig. S2.

Fig. S2.

Open in a new tab

Female Ad-GsKO mice have normal body weight and composition and improved glucose tolerance. (A) Body weight curve and (B) total, fat, and lean mass in 3- to 4-mo-old female Ad-GsKO and control mice maintained on standard diet (n = 10–19 per group). (C) Results of glucose tolerance test in 3- to 4-mo-old female Ad-GsKO and control mice (n = 8 per group). Data are mean ± SEM.

Fig. 2.

Fig. 2.

Open in a new tab

Gsα is required for BAT thermogenesis and browning of WAT. (A) Histology of interscapular BAT (above) and Ing WAT (below) from a 5-mo-old male control and an Ad-GsKO mouse maintained on a standard diet (H & E). Areas featuring beiging in the Ing WAT of control mice only comprise less than 1/20 of the total random selected areas examined. (Scale bar, 100 µm.) (B) Ucp1 and Serpina3k mRNA expression in BAT from 3- to 4-mo-old male control and Ad-GsKO mice (n = 3 per group). (C) Rectal temperature in 3- to 4-mo-old male control and Ad-GsKO mice maintained on standard diet at room temperature (time 0) and hourly after being placed at 4 °C (n = 7 per group). (D) Ucp1 mRNA levels in BAT of male control and Ad-GsKO mice at baseline (24 °C) and after 3 h at 4 °C (Cold) (n = 3 per group). (E) O2 consumption in 4-mo-old female control and Ad-GsKO mice on a standard diet measured at 30 °C before and after administration of the β3-adrenergic receptor agonist CL316243 (0.01 mg/kg i.p.; n = 5–6 per group). (F) Respiratory exchange ratios in control and Ad-GsKO mice before and after CL316243 administration. Differences before and after treatment were significantly different between groups. Data are mean ± SEM *P < 0.05 or ***P < 0.001 vs. controls; #P < 0.05 vs. room temperature or baseline.

Table 1.

Serum chemistries in control and Ad-GsKO mice

Metabolite/factor Control Ad-GsKO
Random
 Glucose, mg/dL 90 ± 6 89 ± 4
 Insulin, ng/mL 0.58 ± 0.10 0.37 ± 0.05
 FFAs, mM 0.39 ± 0.05 0.23 ± 0.04*
 TGs, mg/dL 139 ± 6 104 ± 13*
 Cholesterol, mg/dL 91 ± 8 95 ± 11
 Leptin, ng/mL 11 ± 3 10 ± 2
 FABP4, ng/mL 223 ± 41 121 ± 19*
 FGF21, pg/mL 100 ± 11 92 ± 8
 RBP4, μg/mL 29 ± 1 27 ± 1
 AdipoQ, μg/mL 16.8 ± 1.8 11.0 ± 1.4*
 T4, μg/dL 3.46 ± 0.18 3.73 ± 0.24
 T3, ng/dL 90.0 ± 2.7 92.1 ± 4.2
Starved
 Glucose, mg/dL 48 ± 2 48 ± 2
 Insulin, ng/mL 0.49 ± 0.06 0.34 ± 0.01*
Open in a new tab

Data are mean ± SEM n = 6–8 per group. Blood was collected from 12- to 16-wk-old female mice. *P < 0.05 vs. controls. AdipoQ, adiponectin.

We also observed no differences in body weight, body composition, food intake, or energy expenditure when mice were maintained at thermoneutral temperature (30 °C) for 8 wk (Fig. S3). This is in contrast to Ucp1 knockout mice, which were shown to have increased weight gain and metabolic efficiency when maintained at thermoneutrality (2).

Fig. S3.

Fig. S3.

Open in a new tab

Energy balance is unaffected in Ad-GsKO maintained at thermoneutrality. (A, C, and E) Body weight (A), fat mass (C), and lean mass (E) in male control and Ad-GsKO mice before (at 12–16 wk) and 8 wk after being maintained at 30 °C. (B, D, and F) Gain in body weight (B), fat mass (D), and lean mass (F) of control and Ad-GsKO mice during 8 wk on standard diet. (G and H) Average daily food intake (G) and total energy expenditure (H) during 8 wk on standard diet. Data are mean ± SEM, n = 6 per group.

Gsα Is Required for BAT Thermogenesis and Browning of WAT.

Consistent with their increased weight (Fig. 1G), interscapular BAT pads from Ad-GsKO mice were grossly enlarged and pale in comparison with BAT from controls. Histology revealed that BAT from Ad-GsKO mice contained enlarged adipocytes with unilocular lipid droplets (Fig. 2A), which were morphologically similar to WAT adipocytes. Consistent with this, Ucp1 mRNA expression was decreased by ∼92% in BAT from Ad-GsKO, whereas expression of Serpina3k, a relatively WAT-specific marker (16), was increased >10-fold (Fig. 2B).

Next, we investigated the impact of Gsα deficiency on the process of browning of WAT, a feature normally most prominent in s.c. WAT. Histological examination of s.c. Ing WAT from control mice raised at 22 °C showed areas of browning (smaller adipocytes with multiple lipid droplets), whereas browning was absent in Ing WAT from Ad-GsKO mice (Fig. 2A). Consistent with this, Ucp1 mRNA could be detected in control Ing WAT by RT-PCR (Cт: 30.0 ± 1.8) but was undetectable in Ad-GsKO Ing WAT (Cт > 40).

BAT and beige adipose tissue are known to be required for mice to maintain normal body temperature (Tb) in a cold environment. In contrast to controls, Ad-GsKO mice were unable to maintain their Tb when placed at 4 °C (Fig. 2C), consistent with the presence of inactive BAT and absence of beige fat. Induction of BAT Ucp1 mRNA in the cold was almost completely absent (Fig. 2D), which would be expected because the absence of Gsα would make BAT resistant to SNS stimulation. The ability of a β3-adrenergic receptor agonist (CL31624), which is relatively specific for adipose tissue, to stimulate whole-animal energy expenditure (O2 consumption rate) was also markedly impaired in Ad-GsKO mice, although not completely lost (Fig. 2E), a result very similar to that observed in Ucp1 knockout mice (17). Compared with controls, mutants also showed a lower reduction in the RER (vCO2/vO2) in response to CL31624 (Fig. 2F), indicating a lower stimulation of fatty acid metabolism in response to the drug.

Interestingly, Ad-GsKO mice had a slightly higher baseline Tb at 22 °C that was statistically significant during the light phase (Table S1). There were no associated differences in activity levels during either the light or dark phase (Table S1), and quadratic fit of activity vs. Tb showed that Tb was higher in the dark phase independent of the activity level (Fig. S4A). When mice were placed in a temperature gradient during the light phase we observed no difference in temperature preference between Ad-GsKO mice and controls (Fig. S4 B and C), suggesting that the mutants prefer their slightly higher Tb and did not perceive this as hyperthermia (18). Although the mechanism responsible for this small shift in Tb is unclear, our results suggest that mice can chronically defend against hypothermia at a temperature moderately below thermoneutrality by BAT/beige fat-independent mechanisms. Our results are consistent with the small increase in Tb (0.1–0.3 °C) that has also been observed in Ucp1 knockout mice maintained at 20 °C (19). In summary, Ad-GsKO mice raised at 22 °C showed many metabolic and thermogenic features in common with Ucp1 knockout mice (normal energy balance on standard diet and HFD, similar loss of metabolic response to a β3 adrenergic agonist, small increase in baseline Tb, and inability to defend Tb acutely at 4 °C) (17, 19). These changes in temperature in Ad-GsKO mice were not associated with any changes in circulating levels of thyroid hormones [triiodothyronine (T3) and thyroxine (T4); Table 1].

Table S1.

Body temperature and physical activity in control and Ad-GsKO mice

Genotype Dark Tb, °C Dark activity Light Tb, °C Light activity
Control 37.12 ± 0.13 14.16 ± 2.30 36.01 ± 0.16 7.32 ± 1.49
Ad-GsKO 37.42 ± 0.25 15.95 ± 1.96 36.44 ± 0.16 7.89 ± 0.77
N 6 6 6 6
P 0.16 0.28 0.04 0.37
Open in a new tab

Data are mean ± SEM. Body temperature (Tb) and activity were measured in 3- to 4-mo-old male Ad-GsKO and control mice fed on a standard diet by Mini Mitter telemetry for 5 d with data collected each minute. The means of dark and light phase were calculated and analyzed. Activity is in arbitrary units.

Fig. S4.

Fig. S4.

Open in a new tab

Body temperature vs. activity levels and environmental preference testing. (A) Quadratic fit of Tb vs. activity levels for 3- to 4-mo-old male Ad-GsKO and control mice (n = 6 per group) analyzed separately during the light (Right) and dark (Left) phase. Mean Tb and activity in 10-min intervals were calculated (4,320 data points for dark and light phase of each genotype). Dotted lines show the 95% confidence intervals of the fitted curves. (B and C) Mean distance from the 36.5 °C end (B) and velocity (C) of Ad-GsKO and control mice (n = 5 per group) during a temperature preference test. Data are mean ± SEM.

Both Adipose Tissue Lipolysis and Lipogenesis Are Reduced in Ad-GsKO Mice.

Ad-GsKO mice had significant down-regulation of multiple genes in BAT required for lipolysis, including those encoding perilipin 1 (plin1), hormone-sensitive lipase (lipe), and adipose triglyceride lipase (ATGL, pnpla2) and its coactivator CGI-58 (abhd5) (20) (Fig. 3A). Similar reductions of plin1, lipe, and pnpla2 mRNA, but not abhd5, mRNA were also observed in Ad-GsKO Ing WAT (Fig. 3B). Consistent with markedly reduced expression of genes related to lipolysis, Ad-GsKO mice had reduced levels of serum free fatty acids (FFA; Fig. 3E and Table 1) and glycerol (Fig. 3F) at baseline, which probably reflect reduced lipolysis but could also be due to changes in FFA and triglyceride (TG) turnover. In contrast to controls, Ad-GsKO mice showed minimal induction of either plasma FFA or glycerol in response to CL31624 (Fig. 3 E and F). This latter observation is likely secondary to both loss of β-adrenergic responsiveness in adipose tissue due to Gsα deficiency and marked reduction in expression of lipolysis-related genes and is consistent with the impaired induction of energy expenditure and reduced lowering of RER in response to CL31624 that we also observed in these mice (Fig. 2 E and F), although direct measures of lipolytic protein and phosphorylation levels were not examined.

Fig. 3.

Fig. 3.

Open in a new tab

Reduced lipolysis and lipogenesis in adipose tissues of Ad-GsKO mice. (A and B) mRNA expression of lipolysis-related genes in BAT (A) and Ing WAT (B) from 3- to 4-mo-old male control and Ad-GsKO mice maintained on standard diet (n = 3–5 per group). Abhd5, abhydrolase domain containing 5 (CGI-58); Lipe, hormone-sensitive lipase; Plin1, perilipin 1; Pnpla2, adipose triglyceride lipase. (C and D) mRNA expression of glucose uptake and de novo lipogenesis related genes in interscapular BAT (C) and Ing WAT (D) from 3- to 4-mo-old male control and Ad-GsKO mice maintained on standard diet (n = 3–5 per group). Ascl1, acyl-CoA synthetase long-chain family member 1; Chrebp, carbohydrate responsive element binding protein; Fasn, fatty acid synthase; Glut4, glucose transporter 4; Me1 and Me2, malic enzymes 1 and 2; Pcx, pyruvate carboxylase; Scd1, stearoyl-CoA desaturase 1. (E and F) Plasma FFA (E) and glycerol (F) levels in 8-mo-old male control and Ad-GsKO mice after i.p. injection of saline or CL316243 (CL; n = 5–7 per group). (G) d-[14C(U)]-glucose incorporation into lipid in BAT and Epi WAT isolated from 4-mo-old male control and Ad-GsKO mice (n = 3–5 per group). Results are normalized to tissue weight and expressed as percent of control. Data are mean ± SEM *P < 0.05 vs. controls; #P < 0.05 vs. baseline.

We also observed markedly reduced expression of genes required for glucose uptake and fatty-acid synthesis, including genes encoding glucose transporter 4 (Glut4), pyruvate carboxylase (Pcx), malic enzyme 1 (Me1), acyl-CoA synthetase long-chain family member 1 (Ascl1), fatty acid synthase (Fasn), stearoyl-CoA desaturase 1 (Scd1), and carbohydrate response element binding protein (Chrebp) in BAT and Ing WAT of Ad-GsKO mice (Fig. 3 C and D). There was no change in expression of malic enzyme 2 (Me2) in BAT or Ing WAT of Ad-GsKO mice (Fig. 3 C and D). Consistent with these gene expression results, 2-deoxyglucose (DG) uptake in BAT after i.p. administration of insulin in vivo was significantly reduced in BAT (Fig. 4E) and tended to be lower in Ing and Epi WAT (Fig. 4 F and G) of Ad-GsKO mice. In addition, Ad-GsKO mice had reduced rates of incorporation of glucose into lipid in BAT and epididymal WAT in vitro (Fig. 3F).

Fig. 4.

Fig. 4.

Open in a new tab

Improved glucose metabolism and insulin sensitivity in Ad-GsKO mice. (A and B) Results of glucose tolerance tests on 4- to 5-mo-old male control and Ad-GsKO mice maintained on standard diet (A) or after 8 wk of HFD (B). (C and D) Results of insulin tolerance tests on 4- to 5-mo-old control and Ad-GsKO mice maintained on standard diet (C) or 8 wk of HFD (D) (n = 5–8 per group). (E–G) In vivo 2-DG uptake in interscapular BAT (E), Epi WAT (F), Ing WAT (G), gastrocnemius muscle (H), and cardiac muscle (I) of 2-mo-old male control and Ad-GsKO mice after administration of insulin (Humulin, 0.75 mU/kg i.p.; n = 3–5 per group). (J) Blood glucose levels in 4-mo-old male control and Ad-GsKO mice before and after administration of pyruvate (2 g/kg i.p.; n = 5–8 per group). (K and L) Liver (K) and soleus muscle (L) TG content in 4-mo-old male control and Ad-GsKO mice (n = 6 per group). (M) mRNA expression of interleukin 6 (IL-6) and tumor necrosis factor-α (TNF-α) in Epi WAT from 3-mo-old male control and Ad-GsKO mice (n = 6 per group). Area under curves for glucose, insulin, and pyruvate tolerance tests were all significantly different between control and Ad-GsKO mice. Data are mean ± SEM *P < 0.05 vs. controls.

Ad-GsKO also had reduced expression of genes involved in fatty acid uptake [e.g., lipoprotein lipase (Lpl), fatty acid translocase (Cd36)] and esterification [e.g., phosphoenolpyruvate carboxykinase (Pepck), diacylglycerol acyltransferase 1 (Dgat1), and mitochondrial glycerol-3-phosphate acyltransferase (Gpam)] in both BAT (Fig. S5A) and Ing WAT (Fig. S5B). Pepck, a gene involved in glycerol-3-phosphate production, was also significantly down-regulated in Epi WAT in both fed and fasted Ad-GsKO mice (Fig. S5C). Consistent with these changes in gene expression, we showed that glucose incorporation into TGs and fatty acids (both free and associated with TG) were markedly reduced in explanted BAT and Ing and Epi WAT (Fig. S5D). The effect of insulin on glucose incorporation was not examined. Incorporation of extracellular oleic acid to TG in explanted BAT and Epi WAT of Ad-GsKO mice were also significantly reduced (Fig. S5E), which could be due to either reduced lipid uptake, esterification, or both. Overall our results suggest that glucose and lipid uptake, fatty acid synthesis, and esterification may all be reduced in adipose tissue of Ad-GsKO mice. At least for WAT, normal adipocyte cell size and lipid content may be the result of similar reductions in lipolysis and lipid synthesis, which counterbalance their opposing effects on lipid storage.

Fig. S5.

Fig. S5.

Open in a new tab

Reduced lipogenesis in adipose tissue of Ad-GsKO mice. (A and B) mRNA expression of fatty acids uptake and esterification-related genes in BAT (A) and Ing WAT (B) of 3- to 4-mo-old male control and Ad-GsKO mice on standard diet (n = 3 per group). CD36, fatty acid translocase; Dgat1 and Dgat2, diacylglycerol acyltransferases 1 and 2; Fabp4, fatty acid binding protein 4; Gpam, mitochondrial glycerol-3-phosphate acyltransferase; Lpl, lipoprotein lipase; Pepck, phosphoenolpyruvate carboxykinase. (C) Pepck mRNA expression levels in Epi WAT from 3- to 4-mo-old fed or 17-h-starved male control and Ad-GsKO mice (n = 4 per group). (D) d-[14C(U)]-glucose incorporation into TG and total fatty acids (FA; both free and triglyceride-related) in BAT, Ing WAT, and Epi WAT collected from 4-mo-old male control and Ad-GsKO mice (n = 5 per group). (E) [3H]9,10-oleic acid incorporation into TG in BAT and Epi WAT collected from 3- to 4-mo-old male control and Ad-GsKO mice (n = 5 per group). Data are mean ± SEM *P < 0.05 vs. controls, #P < 0.05 vs. fed.

Despite a reduction in adipose tissue lipid metabolism, Ad-GsKO mice were able to maintain a normal rate of whole-body fatty acid oxidation in vivo (Fig. S6A). To what extent increased metabolism in skeletal muscle prevented the development of obesity in Ad-GsKO mice is not fully clear. It is interesting to note that in these animals gastrocnemius muscle had increased insulin-stimulated glucose uptake (Fig. 4 F and G) and soleus muscle had increased expression of the peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α) and cytochrome c genes, although no increase in expression of the mitochondrial genes Tfam or Nrf1 was observed (Fig. S6B). We also did not observe an increase in expression of the gene encoding sarcolipin, a protein implicated in nonshivering thermogenesis in skeletal muscle (21), in soleus muscle of Ad-GsKO mice. Rather, its expression was decreased (Fig. S6B).

Fig. S6.

Fig. S6.

Open in a new tab

Whole-body fatty acid oxidation and muscle gene expression. (A) Whole-body fatty acid (oleic acid) oxidation measured in starved 2-mo-old male Ad-GsKO and control mice (n = 7 per group). Mice were of similar body weight during the study (Ad-GsKO 30.0 ± 0.9 g vs. control 29.4 ± 0.6 g). (B) mRNA expression of genes related to thermogenesis in soleus muscle of Ad-GsKO and control mice (n = 3 per group). Expression levels were normalized to β-actin. CycC, cytochrome C; Nrf1, nuclear respiratory factor 1; Tfam, transcription factor A mitochondrial; Sln, sarcolipin. Data are mean ± SEM *P < 0.05 vs. controls.

Gsα Deficiency in Fat Tissues Improved Glucose Metabolism and Insulin Sensitivity.

Ad-GsKO mice had normal fed and starved serum glucose levels, whereas serum insulin levels were significantly lower in the starved state and tended to be lower in the fed state (Table 1). Glucose and insulin tolerance tests showed Ad-GsKO mice to have significantly improved glucose tolerance and insulin sensitivity on both standard diet and HFD (Fig. 4 A–D and Fig. S2C). Consistent with this finding, insulin-stimulated 2-DG uptake was significantly increased in vivo in both skeletal (gastrocnemius) and cardiac muscle in Ad-GsKO mice (Fig. 4 H and I). In addition, Ad-GsKO mice showed a smaller rise in serum glucose after administration of pyruvate, consistent with reduced hepatic gluconeogenesis, although this difference could reflect increased pyruvate oxidation in liver to compensate for reduced energy substrate availability (Fig. 4J). Glycogen levels (measured at 1300 hours) were reduced by 15% in livers of Ad-GsKO mice whereas they were unchanged in soleus muscle (Fig. S7 A and B). AMP kinase phosphorylation, a downstream marker of insulin signaling, was unaffected in liver but increased by 41% in soleus muscle of Ad-GsKO mice (Fig. S7 C–F).

Fig. S7.

Fig. S7.

Open in a new tab

Glycogen levels and AMPK activation in liver and muscle. (A and B) Glycogen levels in liver (A) and soleus muscle (B) in 3- to 4-mo-old male control and Ad-GsKO mice maintained on standard diet (n = 4–6 per group, measured at 1300 hours). (C and D) Immunoblots of soleus muscle (C) and liver (D) from 3- to 4-mo-old male control and Ad-GsKO mice with antibodies against phosphorylated AMPK (p-AMPK), total AMPK, and β-actin. (E and F) Quantification of p-AMPK/AMPK ratios (normalized to control) in liver (E) and soleus muscle (F) (n = 4/group). Data are mean ± SEM *P < 0.05 vs. controls.

Improved insulin sensitivity in Ad-GsKO mice may be partially explained by improved lipid status, because Ad-GsKO mice had lower serum FFA and TG levels, as well as liver and muscle TG levels (Fig. 4 K and L). Expression of genes involved in lipid synthesis were either unchanged or reduced (e.g., Pcx, Chrebp, Cd36, and Gpam) in livers from Ad-GsKO mice (Fig. S8A) and liver TG secretion rates were significantly reduced in Ad-GsKO mice (Fig. S8B).

Fig. S8.

Fig. S8.

Open in a new tab

Ad-GsKO have reduced TG secretion. (A) mRNA expression of lipogenesis-related genes in livers of 3- to 4-mo-old male control and Ad-GsKO mice on standard diet (n = 4 per group). Ascl1, acyl-CoA synthetase long-chain family member 1; CD36, fatty acid translocase; Chrebp, carbohydrate responsive element binding protein; Dgat1 and Dgat2, diacylglycerol acyltransferases 1 and 2, respectively; Fasn, fatty acid synthase; Glut2, glucose transporter 2; Gpam, mitochondrial glycerol-3-phosphate acyltransferase; Me1 and Me2, malic enzymes 1 and 2; Pcx, pyruvate carboxylase; Pepck, phosphoenolpyruvate carboxykinase; Scd1, stearoyl-CoA desaturase 1. (B) Hepatic TG secretion rate in 8-mo-old male control and Ad-GsKO mice (n = 3 – 4 per group) on standard diet after receiving lipase inhibitor WR1339 (taloxopol, 100 μL of a 1:10 dilution in PBS i.v.). Data are mean ± SEM *P < 0.05 vs. controls.

We also examined several circulating factors that have been shown to affect glucose metabolism and insulin sensitivity. Serum retinol binding protein 4 (RBP4) and fibroblast growth factor 21 (FGF21) levels were unaffected (Table 1). Serum adiponectin levels were reduced in Ad-GsKO mice, consistent with our prior results showing that adiponectin expression in adipose tissue is induced by stimulation with a β3 agonist (22). However, reduced serum adiponectin levels are associated with impaired, rather than improved, peripheral glucose metabolism. Serum levels of FABP4, an adipokine that is markedly elevated in obese/diabetic patients and that stimulates hepatic gluconeogenesis (23–25), were markedly reduced by almost 50% in Ad-GsKO mice (Table 1), making this a possible mechanism contributing to the improved glucose metabolism present in Ad-GsKO mice. There were no differences in expression of the inflammatory genes interleukin-6 and tumor necrosis factor-α in Epi fat from Ad-GsKO and control mice (Fig. 4M), making it unlikely that differences in adipose tissue inflammation are contributing to the observed differences in glucose metabolism.

Discussion

In this paper we generated mice with loss of Gsα signaling (and therefore resistance to activation by SNS and other hormones) in mature BAT and WAT adipocytes and examined the effects on temperature maintenance, adipocyte metabolism, energy balance, and glucose homeostasis. Our results provide important insights into on how these physiological parameters are regulated by SNS activation of adipose tissue. As expected, Ad-GsKO mice showed marked reductions in BAT activation, cold tolerance, cold-induced Ucp1 up-regulation, and browning of WAT. This result is expected, given the known role of SNS activation of BAT and WAT in these processes, and is consistent with the cold intolerance observed in both our prior adipose-specific Gsα knockout model (10) and Ucp1 knockout mice (17). Similar to Ucp1 knockout mice (17), we observed a small increase in Tb at room temperature in Ad-GsKO mice. Presumably this reflects increased thermogenesis in other tissue(s), although the mechanisms responsible for these observations are unclear. Although it has been shown that loss of BAT function can be compensated for by increased browning of WAT (26), this is not an explanation in our model because WAT browning was absent. The observed absence of a detectable change in energy expenditure despite a small increase in Tb may reflect compensation by the small, though not significant, reduction in activity levels or may reflect the limitations of indirect calorimetry to detect small differences in energy expenditure.

Despite loss of BAT and WAT activation by SNS, Ad-GsKO mice had normal body weight and adiposity on both standard diet and HFD. One potential explanation for this observation is that loss of Gsα in adipocytes led not only to reduced lipolysis, a process known to be activated by Gsα and cAMP, but also to a similar reduction in expression of genes critical for glucose and lipid uptake, fatty acid synthesis, and esterification in both BAT and WAT, associated with reduced incorporation of glucose and fatty acids into TG. This finding is consistent with a recent study showing that β3-adrenergic stimulation increased both lipolysis and de novo lipogenesis in BAT and WAT and that induction of de novo lipogenesis may be coupled to lipolysis (27). Another study showed that although norepinephrine acutely inhibited lipogenesis in cultured BAT cells, longer treatments led to stimulation of lipogenesis (28). It is interesting to note that mutations of lipolytic genes encoding ATGL and hormone-sensitive lipase are associated with only small changes in adiposity and reduced lipogenesis in adipose tissue, suggesting that lipid anabolic and catabolic pathways may be coupled (29, 30). It is also possible that the reduced lipogenesis observed in Ad-GsKO mice may reflect fundamental changes in the adipogenic program, because a profound effect of Gsα on adipogenesis was observed in our prior FGsKO mouse model in which Gsα deletion occurred at an earlier stage of adipogenesis (12). Further studies will be required to understand the mechanisms by which Gsα seems to affect both lipolysis and lipogenesis in the same direction.

The role of SNS-induced BAT activation and browning of WAT in diet-induced thermogenesis has great appeal and stimulation of these responses has been proposed as a potential treatment for obesity (1, 2, 8, 9). However, whether BAT function is necessary for the maintenance of normal body weight is still not definitively answered. Ablation of sympathetic nerves (4–6) or their neurotransmitters (norepinephrine and epinephrine) (7) does not result in obesity, even when BAT is functionally inactive. Similarly, Ucp1 knockout mice fed on either a standard chow or HFD do not develop obesity relative to littermate controls at room temperature (17). Although Ucp1 knockout mice showed increased gain in fat mass and diet-induced thermogenesis when maintained at thermoneutrality and this was ascribed to loss of diet-induced thermogenesis (2), we observed no changes in fat mass or energy expenditure in Ad-GsKO maintained at thermoneutrality. The absence of greater adiposity of Ad-GsKO mice on either standard diet or HFD provides further direct evidence that activation of BAT or browning of WAT is not required for maintenance of normal energy balance.

Another striking finding in Ad-GsKO mice is significant improvement in insulin sensitivity and peripheral glucose metabolism despite the fact that BAT is inactive and the absence of improvement in whole-body adiposity. This effect of glucose metabolism is probably independent of changes in Ucp1 expression, because Ucp1 knockout mice on standard diet were shown to have no differences in glucose tolerance or insulin levels (31). We observed no differences in circulating FGF21 or RBP4 levels to help explain this effect, and circulating levels of adiponectin, a factor associated with improved glucose metabolism, were actually decreased in Ad-GsKO mice. In addition, we found no evidence for differences in WAT inflammation that might contribute to alterations in glucose metabolism. One potential mechanism contributing to improved insulin sensitivity and glucose metabolism in Ad-GsKO mice is reduced lipolysis in adipose tissue resulting in lower circulating FFA levels. It is well established that FFAs, the major product of lipolysis, cause deleterious effects on insulin sensitive organs (32, 33) and that genetic or pharmacologic inhibition of adipose tissue lipolysis in mice improves glucose metabolism and insulin sensitivity (29, 34). Moreover, high lipolysis rate is associated with low insulin sensitivity in humans (34). The lower TG content in liver present in Ad-GsKO may also contribute to improved glucose metabolism in these mice.

FABP4 is a recently identified adipokine released from adipocyte that leads to impaired glucose metabolism via increased hepatic glucose production and perhaps other mechanisms as well, although alternative mechanisms have not been directly examined (23, 25). FABP4 release from adipocytes is stimulated by β-adrenergic receptor agonists that activate Gsα, and circulating FABP4 levels have been shown to be markedly elevated in obese/diabetic patients (23, 24). Conversely, reduced plasma FABP4 levels are associated with improved glucose metabolism and insulin sensitivity (23, 35). Therefore, inhibiting Gsα signaling pathway in adipocytes may improve insulin sensitivity and glucose metabolism via reduced lipolysis, release of FABP4, and perhaps other additional mechanisms.

In conclusion, although the level of functional BAT is inversely correlated with adiposity in humans (36, 37) and positively correlated with improved insulin sensitivity and glucose metabolism in mice and humans (38–40), our results suggest that reduced BAT function alone may be insufficient for the development of obesity or insulin resistance. Moreover, specific inhibition of adipose tissue Gsα signaling seems to dramatically affect whole-body glucose metabolism in the absence of an effect on body weight.

Materials and Methods

All studies were approved by the National Institute of Diabetes and Digestive and Kidney Diseases Animal Care and Use Committee. Gsα exon 1-floxed mice (E1fl/fl) (41) were mated with adiponectin-cre mice (15) to generate Ad-GsKO mice (adiponectin-cre+, E1fl/fl). Animals were maintained on a 12-h light/dark cycle (6:00 AM/6:00 PM) and standard pellet diet (NIH-07, 5% fat by weight) unless indicated. HFD (Bio-Serv) consisted of 59.4 kcal% fat, 16.2 kcal% protein, and 24.5 kcal% carbohydrate. All data are expressed as mean ± SEM and analyzed by Student’s t test with differences considered significant at P < 0.05. Detailed methods are described in SI Materials and Methods.

SI Materials and Methods

Food Intake, Body Composition, and Metabolic Studies.

Body composition was measured with the Echo3-in1 NMR analyzer (Echo Medical Systems). To measure food intake mice were single-housed and allowed to acclimate for 1 wk before measurement of food intake during the subsequent 2 wk. O2 consumption and CO2 production was measured at 24 °C and 30 °C, each over a 24-h period in a CLAMS system (Columbus Instruments; 2.5-L chambers with plastic floors, using 0.6 L/min flow rate, one mouse per chamber) after 48-h adaptation period. Motor activity (total and ambulatory) was simultaneously measured by infrared beam interruption. Resting O2 consumption was calculated as the mean of the points with fewer than six ambulating beam breaks per minute. CL316243 (Sigma-Aldrich; 0.01 mg/kg i.p.) or saline vehicle were administered and O2 consumption was measured at 30 °C from 1 to 4 h after injection.

Temperature Studies.

For temperature studies at ambient temperature mice were anesthetized and a G2 E-mitter was inserted intraperitoneally. After 1 wk Tb and activity levels were continuously recorded via telemetry (Mini Mitter; Philips Respironics) with data collected each minute. For cold tolerance test, core body temperature was measured using a rectal probe (Thermalet TH-5) at ambient room temperature and during exposure to 4 °C. Food and water were provided for ad libitum consumption.

Glucose, Insulin, and Pyruvate Tolerance Tests.

For glucose tolerance tests, overnight-starved mice received glucose (2 g/kg i.p.). For insulin tolerance tests, mice were starved for 5 h before administration of insulin (Humulin, 0.75 mIU/g, i.p.). For pyruvate tolerance test, overnight-starved mice received pyruvate (2 g/kg i.p.). For all three tests glucose was measured from tail vein at indicated time points with a glucometer (Contour; Bayer).

Uptake of 2-DG.

After being food-deprived for 4.5 h, mice were injected with insulin (Humulin, 0.75 mU/kg i.p.) and 2-deoxy-d-[1-14C] glucose (10 μCi; PerkinElmer). Mice were killed 40 min after injection. Tissues were harvested and tissue 2-deoxy-d-[1-14C] glucose-6-phosphate content was measured using poly-prep prefilled chromatography columns (731-6211; Bio-Rad) as described (42).

In Vivo Lipolysis.

Randomly fed mice were injected i.p. saline or CL316243 (Sigma, 0.1 mg/kg) in a cross-over manner and blood was collected 20 min later. Plasma glycerol and FFAs were measured with reagents from Sigma and Thermo Scientific, respectively.

Serum and Tissue Chemistries.

Glucose was measured with a glucometer (Bayer). TGs, FFAs, and total cholesterol were measured with reagents from Roche, Thermo Scientific, and Fisher, respectively. Adiponectin, insulin, and leptin were measured by RIA (Millipore). T3 and T4 were also determined by RIA (MP Biomedicals). RBP4, FGF21, and FABP4 were measured by ELISA (AdipoGen, R&D Systems and BioVendor, respectively). Tissue TGs were measured by a kit (Cayman Chemical). Glycogen was measured in tissues from mice killed at 1500 hours using a kit (Sigma).

Gene Expression.

Total RNA was extracted using RNeasy kit (Qiagen) and DNase I-treated (Invitrogen) at room temperature for 15 min. Reverse transcription was performed using MultiScribe RT (Applied Biosystems). Gene expression levels were measured by real-time quantitative RT-PCR (StepOnePlus; Life Technologies). Standard curves were simultaneously generated with serial dilutions of cDNA ranging from 1 to 100 ng, and results were normalized to β-actin mRNA levels in each sample, which were determined simultaneously by the same method. Specificity of each RT-PCR product was checked by its dissociation curve and the presence of a single band of expected size on acrylamide gel electrophoresis. Primer sequences are available upon request.

Histology.

Tissues were fixed in 4% (wt/vol) paraformaldehyde and paraffin-embedded, and sections were H&E-stained. Images were captured using BZ-9000 all-in-one microscope (Keyence). Epi adipocyte size was determined by analyzing at least 300 random selected adipocytes with a BZ-II analyzer (Keyence).

Whole-Body in Vivo Fatty Acid Oxidation.

PerkinElmer [1-14C] oleic acid (1 μCi in 200 μL saline i.p.) was administered to overnight-starved mice and expired 14CO2 was measured over time as previously described (43)

Temperature Preference Study.

Temperature preference was measured by placing mice in a stainless steel pan (64 × 15 × 20 cm, length × width × height) spanning two hot/cold plates set at 45 and 20 °C. Ninety-minute sessions were conducted on two successive days, with the last 60 min of the second session used for analysis. Position was tracked with an overhead camera and video tracking software (Ethovision 9.0; Noldus) (18).

Thermoneutrality Study.

After obtaining body weights and compositions, 12- to 16-wk-old male mice were housed at 30 °C for 8 wk in a constant temperature chamber (HPP 750 Life; Memmert) with free access to water and standard diet. Food intake was measured weekly. Body weight and composition was determined at the end of 8 wk and total daily energy expenditure was calculated as previously described (44).

Liver TG Secretion Rate.

TG secretion rate was studied by measuring the increase in circulating TGs after lipase inhibition by WR1339. Mice were starved for 5 h. WR1339 (T-8761; Sigma, 100 μL of a 1:10 dilution in PBS) was injected via warmed tail vein and plasma TGs were measured at 0, 1, and 2 h.

Incorporation of [3H]9,10-Oleic Acid into TG.

Pieces of BAT or WAT (20–30 mg) were incubated in DMEM containing 1g/L d-glucose (Gibco), 1% BSA (fatty acid-free; Sigma-Aldrich) and 10 μCi/mL [3H]9,10-oleic acid (PerkinElmer) for 4 h at 37 °C. Total lipids were extracted from the fat pads as previously described (45). An aliquot of the lipid extract was analyzed by TLC and the lipids were visualized with iodine vapor. Spots corresponding to TG were cut out and quantitated by liquid scintillation.

Incorporation of d-[14C (U)] Glucose into Lipids.

To measure d-glucose incorporation into the lipid of fat pads, pieces of BAT or WAT (30–40 mg) were incubated in DMEM containing 1g/L d-glucose (Gibco), 1% BSA (fatty acids-free; Sigma-Aldrich), and 0.5 µCi/mL d-[14C (U)] glucose (PerkinElmer) for 4 h at 37 °C. Total lipids were extracted from fat pads as previously described (45). Aliquots of the total lipid extract were quantitated by liquid scintillation. TLC analysis was performed and TG was quantitated by liquid scintillation. To measure the incorporation of d-glucose into fatty acids (both free and TG-associated), the lipid extracts were saponified with a TG lipase (Candida rugosa, 6 U/mL; Sigma-Aldrich) for 4 h at 37 °C. Fatty acids were separated by TLC and quantitated.

Immunoblot Analysis.

Gsα expression was assessed by immunoblotting in isolated adipocyte extracts and other tissues as described (46). Tissue protein extracts were separated on 4% Bis-Tris gels and blots were incubated with AMP kinase (AMPK) (1:1,000; Cell Signaling), p-AMPK (1:1,000; Cell Signaling), and β-actin (1:3,000; Sigma-Aldrich) antibodies. Protein was visualized with horseradish peroxidase-conjugated anti-mouse or anti-rabbit second antibodies (GE Healthcare).

Acknowledgments

We thank M. Reitman and A. Cypess for suggestions and technical assistance. This work was supported by the Intramural Research Program of the National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health.

Footnotes

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517142113/-/DCSupplemental.

References

  • 1.Cohen P, et al. Ablation of PRDM16 and beige adipose causes metabolic dysfunction and a subcutaneous to visceral fat switch. Cell. 2014;156(1-2):304–316. doi: 10.1016/j.cell.2013.12.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Feldmann HM, Golozoubova V, Cannon B, Nedergaard J. UCP1 ablation induces obesity and abolishes diet-induced thermogenesis in mice exempt from thermal stress by living at thermoneutrality. Cell Metab. 2009;9(2):203–209. doi: 10.1016/j.cmet.2008.12.014. [DOI] [PubMed] [Google Scholar]
  • 3.Robidoux J, Martin TL, Collins S. β-adrenergic receptors and regulation of energy expenditure: A family affair. Annu Rev Pharmacol Toxicol. 2004;44:297–323. doi: 10.1146/annurev.pharmtox.44.101802.121659. [DOI] [PubMed] [Google Scholar]
  • 4.Desautels M, Dulos RA. Effects of neonatal sympathectomy on brown fat development and susceptibility to high fat diet induced obesity in mice. Can J Physiol Pharmacol. 1991;69(12):1868–1874. doi: 10.1139/y91-276. [DOI] [PubMed] [Google Scholar]
  • 5.Levin BE, Triscari J, Marquet E, Sullivan AC. Dietary obesity and neonatal sympathectomy. I. Effects on body composition and brown adipose. Am J Physiol. 1984;247(6 Pt 2):R979–R987. doi: 10.1152/ajpregu.1984.247.6.R979. [DOI] [PubMed] [Google Scholar]
  • 6.Tordoff MG. Influence of sympathectomy on body weight of rats given chow or supermarket diets. Physiol Behav. 1985;35(3):455–463. doi: 10.1016/0031-9384(85)90323-3. [DOI] [PubMed] [Google Scholar]
  • 7.Thomas SA, Palmiter RD. Thermoregulatory and metabolic phenotypes of mice lacking noradrenaline and adrenaline. Nature. 1997;387(6628):94–97. doi: 10.1038/387094a0. [DOI] [PubMed] [Google Scholar]
  • 8.Bachman ES, et al. βAR signaling required for diet-induced thermogenesis and obesity resistance. Science. 2002;297(5582):843–845. doi: 10.1126/science.1073160. [DOI] [PubMed] [Google Scholar]
  • 9.Cannon B, Nedergaard J. Brown adipose tissue: Function and physiological significance. Physiol Rev. 2004;84(1):277–359. doi: 10.1152/physrev.00015.2003. [DOI] [PubMed] [Google Scholar]
  • 10.Weinstein LS, Xie T, Qasem A, Wang J, Chen M. The role of GNAS and other imprinted genes in the development of obesity. Int J Obes. 2010;34(1):6–17. doi: 10.1038/ijo.2009.222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Gnad T, et al. Adenosine activates brown adipose tissue and recruits beige adipocytes via A2A receptors. Nature. 2014;516(7531):395–399. doi: 10.1038/nature13816. [DOI] [PubMed] [Google Scholar]
  • 12.Chen M, et al. Gsα deficiency in adipose tissue leads to a lean phenotype with divergent effects on cold tolerance and diet-induced thermogenesis. Cell Metab. 2010;11(4):320–330. doi: 10.1016/j.cmet.2010.02.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Cristancho AG, Lazar MA. Forming functional fat: A growing understanding of adipocyte differentiation. Nat Rev Mol Cell Biol. 2011;12(11):722–734. doi: 10.1038/nrm3198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Eguchi J, et al. Transcriptional control of adipose lipid handling by IRF4. Cell Metab. 2011;13(3):249–259. doi: 10.1016/j.cmet.2011.02.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Wang ZV, Deng Y, Wang QA, Sun K, Scherer PE. Identification and characterization of a promoter cassette conferring adipocyte-specific gene expression. Endocrinology. 2010;151(6):2933–2939. doi: 10.1210/en.2010-0136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Kajimura S, et al. Regulation of the brown and white fat gene programs through a PRDM16/CtBP transcriptional complex. Genes Dev. 2008;22(10):1397–1409. doi: 10.1101/gad.1666108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Enerbäck S, et al. Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese. Nature. 1997;387(6628):90–94. doi: 10.1038/387090a0. [DOI] [PubMed] [Google Scholar]
  • 18.Lateef DM, Abreu-Vieira G, Xiao C, Reitman ML. Regulation of body temperature and brown adipose tissue thermogenesis by bombesin receptor subtype-3. Am J Physiol Endocrinol Metab. 2014;306(6):E681–E687. doi: 10.1152/ajpendo.00615.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Liu X, et al. Paradoxical resistance to diet-induced obesity in UCP1-deficient mice. J Clin Invest. 2003;111(3):399–407. doi: 10.1172/JCI15737. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Zechner R, et al. FAT SIGNALS--lipases and lipolysis in lipid metabolism and signaling. Cell Metab. 2012;15(3):279–291. doi: 10.1016/j.cmet.2011.12.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Bal NC, et al. Sarcolipin is a newly identified regulator of muscle-based thermogenesis in mammals. Nat Med. 2012;18(10):1575–1579. doi: 10.1038/nm.2897. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Chen M, et al. Increased insulin sensitivity in paternal Gnas knockout mice is associated with increased lipid clearance. Endocrinology. 2004;145(9):4094–4102. doi: 10.1210/en.2004-0038. [DOI] [PubMed] [Google Scholar]
  • 23.Cao H, et al. Adipocyte lipid chaperone AP2 is a secreted adipokine regulating hepatic glucose production. Cell Metab. 2013;17(5):768–778. doi: 10.1016/j.cmet.2013.04.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Tso AW, et al. Serum adipocyte fatty acid binding protein as a new biomarker predicting the development of type 2 diabetes: A 10-year prospective study in a Chinese cohort. Diabetes Care. 2007;30(10):2667–2672. doi: 10.2337/dc07-0413. [DOI] [PubMed] [Google Scholar]
  • 25.Kralisch S, et al. Adipocyte fatty acid-binding protein is released from adipocytes by a non-conventional mechanism. Int J Obes. 2014;38(9):1251–1254. doi: 10.1038/ijo.2013.232. [DOI] [PubMed] [Google Scholar]
  • 26.Schulz TJ, et al. Brown-fat paucity due to impaired BMP signalling induces compensatory browning of white fat. Nature. 2013;495(7441):379–383. doi: 10.1038/nature11943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Mottillo EP, et al. Coupling of lipolysis and de novo lipogenesis in brown, beige, and white adipose tissues during chronic β3-adrenergic receptor activation. J Lipid Res. 2014;55(11):2276–2286. doi: 10.1194/jlr.M050005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Bianco AC, Carvalho SD, Carvalho CR, Rabelo R, Moriscot AS. Thyroxine 5′-deiodination mediates norepinephrine-induced lipogenesis in dispersed brown adipocytes. Endocrinology. 1998;139(2):571–578. doi: 10.1210/endo.139.2.5737. [DOI] [PubMed] [Google Scholar]
  • 29.Schreiber R, et al. Hypophagia and metabolic adaptations in mice with defective ATGL-mediated lipolysis cause resistance to HFD-induced obesity. Proc Natl Acad Sci USA. 2015;112(45):13850–13855. doi: 10.1073/pnas.1516004112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Zimmermann R, et al. Decreased fatty acid esterification compensates for the reduced lipolytic activity in hormone-sensitive lipase-deficient white adipose tissue. J Lipid Res. 2003;44(11):2089–2099. doi: 10.1194/jlr.M300190-JLR200. [DOI] [PubMed] [Google Scholar]
  • 31.Kontani Y, et al. UCP1 deficiency increases susceptibility to diet-induced obesity with age. Aging Cell. 2005;4(3):147–155. doi: 10.1111/j.1474-9726.2005.00157.x. [DOI] [PubMed] [Google Scholar]
  • 32.Boden G. Obesity, insulin resistance and free fatty acids. Curr Opin Endocrinol Diabetes Obes. 2011;18(2):139–143. doi: 10.1097/MED.0b013e3283444b09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Samuel VT, Petersen KF, Shulman GI. Lipid-induced insulin resistance: unravelling the mechanism. Lancet. 2010;375(9733):2267–2277. doi: 10.1016/S0140-6736(10)60408-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Girousse A, et al. Partial inhibition of adipose tissue lipolysis improves glucose metabolism and insulin sensitivity without alteration of fat mass. PLoS Biol. 2013;11(2):e1001485. doi: 10.1371/journal.pbio.1001485. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Furuhashi M, et al. Treatment of diabetes and atherosclerosis by inhibiting fatty-acid-binding protein aP2. Nature. 2007;447(7147):959–965. doi: 10.1038/nature05844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Cypess AM, et al. Identification and importance of brown adipose tissue in adult humans. N Engl J Med. 2009;360(15):1509–1517. doi: 10.1056/NEJMoa0810780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.van Marken Lichtenbelt WD, et al. Cold-activated brown adipose tissue in healthy men. N Engl J Med. 2009;360(15):1500–1508. doi: 10.1056/NEJMoa0808718. [DOI] [PubMed] [Google Scholar]
  • 38.Lee P, et al. Temperature-acclimated brown adipose tissue modulates insulin sensitivity in humans. Diabetes. 2014;63(11):3686–3698. doi: 10.2337/db14-0513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Stanford KI, et al. Brown adipose tissue regulates glucose homeostasis and insulin sensitivity. J Clin Invest. 2013;123(1):215–223. doi: 10.1172/JCI62308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Chondronikola M, et al. Brown adipose tissue improves whole-body glucose homeostasis and insulin sensitivity in humans. Diabetes. 2014;63(12):4089–4099. doi: 10.2337/db14-0746. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Chen M, et al. Increased glucose tolerance and reduced adiposity in the absence of fasting hypoglycemia in mice with liver-specific Gsα deficiency. J Clin Invest. 2005;115(11):3217–3227. doi: 10.1172/JCI24196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Kim JK, Gavrilova O, Chen Y, Reitman ML, Shulman GI. Mechanism of insulin resistance in A-ZIP/F-1 fatless mice. J Biol Chem. 2000;275(12):8456–8460. doi: 10.1074/jbc.275.12.8456. [DOI] [PubMed] [Google Scholar]
  • 43.Gautam D, et al. Beneficial metabolic effects of M3 muscarinic acetylcholine receptor deficiency. Cell Metab. 2006;4(5):363–375. doi: 10.1016/j.cmet.2006.09.008. [DOI] [PubMed] [Google Scholar]
  • 44.Ravussin Y, Gutman R, LeDuc CA, Leibel RL. Estimating energy expenditure in mice using an energy balance technique. Int J Obes. 2013;37(3):399–403. doi: 10.1038/ijo.2012.105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem. 1957;226(1):497–509. [PubMed] [Google Scholar]
  • 46.Chen H, et al. Protein-tyrosine phosphatases PTP1B and syp are modulators of insulin-stimulated translocation of GLUT4 in transfected rat adipose cells. J Biol Chem. 1997;272(12):8026–8031. doi: 10.1074/jbc.272.12.8026. [DOI] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES