Fat-Specific Sirt6 Ablation Sensitizes Mice to High-Fat Diet–Induced Obesity and Insulin Resistance by Inhibiting Lipolysis

Obesity is characterized by excessive triglyceride storage that enlarges adipose tissue. The growth of adipose tissue could result from adipocyte hyperplasia (cell number increase) and/or hypertrophy (cell size increase). Hypertrophy, a result of imbalanced triglyceride synthesis and breakdown, occurs before hyperplasia to meet the need for additional fat storage capacity during the progression of obesity (1,2).

Lipolysis is a process of triglyceride breakdown that liberates glycerol and free fatty acid for other organs to use as energy substrates. Impaired lipolysis results in excess storage of triglycerides that causes adipocyte hypertrophy and obesity (1,3,4). Lipolysis in adipose tissue is tightly regulated by complex regulatory mechanisms involving lipase themselves and hormonal and biochemical signals (1). Lipolysis is stimulated by catecholamines that activate cAMP-dependent protein kinase (PKA) and results in phosphorylation of perilipin 1 (Plin1) and hormone-sensitive lipase (HSL) (1). When phosphorylated, Plin1, which coats the surface of lipid droplets, alleviates the barrier function of this protein and promotes its active participation in the lipolytic process (5). The phosphorylation of HSL causes activation and subsequent translocation of this lipase from the cytosol to lipid droplets (5).

The full lipolytic reaction requires adipose triglyceride lipase (ATGL), which is now considered the rate-limiting lipolytic enzyme (6–9). ATGL catalyzes the initial step, converting triglycerides to diacylglycerols, whereas HSL and other lipase mainly hydrolyzes the diacylglycerols to monoacylglycerol and glycerol. Loss of ATGL reduces lipolysis in adipocyte and was accompanied by increased adipose tissue weight (10,11), whereas its overexpression promotes lipolysis (12). The expression of ATGL is controlled by FoxO1 and peroxisome proliferator–activated receptor γ (PPARγ) as a target gene (9,13). Insulin-inhibited lipolysis is believed to be mediated by suppressing FoxO1 activity and thereby reducing ATGL expression (9).

Sirt6, an NAD⁺-dependent protein deacetylase, plays a key role in metabolism. Sirt6-deficient mice show severe hypoglycemia, which causes their premature death within 4 weeks (14). Sirt6 expression is induced in liver by caloric restriction and suppressed in fatty liver disease (15). Hepatic-specific deletion of Sirt6 results in triglyceride accumulation in liver by increasing fatty acid uptake and reducing fatty acid oxidation (16). Whole-body overexpression of Sirt6 was associated with improved blood lipid profiles in mice fed a high-fat diet (HFD) (17). Recent reports indicated a possible function of Sirt6 in adipose tissue. Sirt6 expression was reduced in adipose tissue of db/db mice, a model of obesity (18). In contrast, Sirt6 expression was increased in adipose tissue of human subjects with weight loss (19). Although these studies suggested that Sirt6 might play a role in adipose tissue, the exact function and underlying mechanism remains unclear.

In this study, we investigated the function of Sirt6 by using fat-specific Sirt6 knockout (FKO) mice and Sirt6 KO adipocytes. Fat-specific Sirt6 KO promoted HFD-induced obesity because of impaired lipolytic activity. The impaired lipolysis in FKO mice is attributed to reduced expression of ATGL. Mechanistically, Sirt6 deficiency increased the acetylation of FoxO1, promoted FoxO1 nuclear export, and reduced its positive regulation on ATGL. In addition, FKO was associated with increased inflammation in adipose tissue and exacerbated HFD-induced insulin resistance and fatty liver disease. Finally, Sirt6 expression was suppressed in obese patients, which was associated with decreased expression of ATGL. Our results reveal an essential role of Sirt6 in obesity and related metabolic disorders.

All animal protocols were reviewed and approved by the Animal Care and Use Committee of Sichuan University. Sirt6 heterozygous mice, adipoq-Cre mice, and Sirt6^Loxp/Loxp mice were purchased from The Jackson Laboratory (Bar Harbor, ME). For the HFD-induced obesity model, 6-week-old mice received an HFD (D12492; Research Diets) for 8–12 weeks as indicated. The adipocyte surface areas were quantified as described (20).

Glucose tolerance test (GTT) and insulin tolerance test (ITT) were performed as we previously described (21).

For immunoprecipitation, cells were lysed into nondenaturing lysis buffer as described (22). Antibodies are listed in Supplementary Table 1.

Embryos were harvested at 12.5 to 13.5 days, and mouse embryonic fibroblasts (MEFs) were plated as monolayer cultures in DMEM with 10% FBS. The MEFs were differentiated into adipocytes as described (23).

White adipose tissue (WAT) explants (∼20 mg) were isolated from epididymal adipose tissue and washed with PBS three times as previously described (24). Glycerol released from explants into the culture medium was determined as an index of lipolysis (25).

To detect macrophage infiltration, stromal vascular fraction (SVF) was stained with fluorescence-labeled primary antibodies and analyzed by using CytoFlex (Beckman Coulter, Indianapolis, IN).

Adipose tissue samples were isolated from freshly excised human abdominal fat tissue from donors who underwent elective operations (e.g., abdominoplasty) at the Department of Surgery, West China Hospital, Sichuan University. The names of the patients were kept anonymous, and all patients used for experiments were without diabetes and aged 28–47 years.

Real-time PCR and chromatin immunoprecipitation (ChIP) assay were performed as described (26,27). Primer sequences are listed in Supplementary Tables 2 and 3.

Statistical significance was determined by Student unpaired two-tailed t test or one-way ANOVA multiple-comparison test. A P value <0.05 was considered statistically significant.

To determine the function of Sirt6 in the adipose tissue, we first evaluated whether the adipose expression of Sirt6 could be changed by nutritional challenges such as fasting and overnutrition. Upon 24-h fasting, the adipose expression of Sirt6 was significantly induced (Fig. 1A, left panel). In contrast, the adipose expression of Sirt6 was greatly suppressed in HFD-fed mice and ob/ob mice (Fig. 1A, middle and right panels). Interestingly, the expression of Sirt6 was also suppressed in adipose tissue of aged mice (Supplementary Fig. 1A). The expression of Sirt6 seemed to be positively regulated by Sirt1, because overexpression of Sirt1 or activation of Sirt1 by resveratrol increased the expression of Sirt6 in adipocytes (Supplementary Fig. 1B and C). The dynamics of Sirt6 expression prompted us to explore whether this deacetylase could contribute to energy metabolism in adipose tissue.

Adipose tissue mass was analyzed in whole-body Sirt6 KO mice. Sirt6 KO mice died shortly after weaning because of severe hypoglycemia (14); however, this partial postnatal lethality could be rescued by feeding the mice with 10% glucose-containing water (28). Therefore, we compared adipose tissue weight in Sirt6 KO mice and wild-type (WT) mice fed with water-containing glucose for 6 weeks. Surprisingly, Sirt6 KO mice showed a significant increase in the ratio of WAT weight to body weight (Fig. 1B). To determine the role of Sirt6 specifically in the adipose tissue, we generated FKO mice by crossbreeding the floxed Sirt6^Loxp/Loxp mice with the adiponectin-derived Cre-transgenic mice (29). The mRNA expression of Sirt6 was decreased by ∼80% in the mature adipocytes of epididymal adipose tissue from FKO mice, but not in the SVFs, which are a source of preadipocytes and macrophages (Fig. 1C). Consistently, the protein level of Sirt6 was also significantly decreased in mature adipocytes from both epididymal and subcutaneous adipose tissue of FKO mice (Fig. 1D). As expected, the expression of Sirt6 was not changed in liver, muscle, or other tissues of FKO mice (Supplementary Fig. 1D).

FKO mice were grossly normal and displayed comparable levels of blood glucose under ad libitum and fasting conditions (Supplementary Fig. 2A and B), which suggests that the hypoglycemia and lethal phenotype observed in Sirt6 KO mice (14) may not be because of the lack of Sirt6 in the adipose tissue. FKO mice on a chow diet showed modestly higher epididymal adipose weight despite having a similar body weight (Supplementary Fig. 2C and D). When challenged with an HFD, FKO mice gained more body weight than their Loxp counterparts (Fig. 1E). Necropsy revealed a significant increase in epididymal and subcutaneous adipose weight in FKO mice (Fig. 1F and G and Supplementary Fig. 3A). The ambulatory activity and food intake were similar between Loxp and FKO mice (Supplementary Fig. 3B and C). Together, these data indicate that fat-specific ablation of Sirt6 sensitizes mice to HFD-induced obesity.

An increased mass of adipose tissue can result from an induction of adipocyte size, number because of abnormal differentiation, or both (25). To understand the mechanism of increased adiposity seen in FKO mice, we first measured the adipocyte size in adipose tissue of the HFD-fed Loxp and FKO mice. Hematoxylin and eosin (H&E) staining showed that the size of adipocyte was larger in both epididymal and subcutaneous adipose tissue from FKO than Loxp mice (Fig. 2A and Supplementary Fig. 3D). The increased adipocyte size in FKO adipose tissue was further supported by cell size quantification (Fig. 2B and Supplementary Fig. 3D). Despite the adipocyte hypertrophy, the expression of adipogenic transcription factors, C/EBPα/β/δ, and PPARγ, was not changed in the FKO mice (Fig. 2C, left panel). The expression of markers of adipocyte differentiation such as adipocyte fatty acid–binding protein (Fabp4) and lipoapoprotein lipase (Lpl) were not changed either (Fig. 2C, right panel). The mRNA expression of genes involved fatty acid uptake was not changed (Supplementary Fig. 4A, left panel). Interestingly, the β-oxidative gene expression in adipose tissue was slightly higher in FKO mice, probably an in vivo compensatory mechanism aimed at restraining excessive body weight gain caused by Sirt6 deletion in fat tissue (Supplementary Fig. 4A, right panel).

The effect of Sirt6 ablation on adiposity was also supported by in vitro results. In this experiment, MEFs were isolated from WT and Sirt6 KO embryos and then subjected to in vitro differentiation to adipocytes. During differentiation of WT MEFs, Sirt6 expression was gradually increased, which suggests a role of Sirt6 in lipid metabolism (Fig. 2D). The Sirt6 KO MEFs differentiated normally as supported by their comparable gene expression of adipogenic markers, fatty acid uptake, and β-oxidation, except acyl-CoA oxidase (Fig. 2E and F and Supplementary Fig. 4B). Oil Red O staining indicated Sirt6 KO adipocytes showed higher lipid accumulation than WT cells (Fig. 2G and H). Similarly, Sirt6 overexpression did not affect adipogenic markers, fatty acid uptake, and oxidative gene expression (Fig. 2I and J and Supplementary Fig. 4C). Thus, our in vivo and in vitro results suggested Sirt6 ablation might cause adipocyte hypertrophy because of greater triglyceride accumulation and larger adipocyte size rather than abnormal adipocyte differentiation.

To determine whether the markedly increased adiposity observed in FKO mice was a result of impaired lipolytic activity, we compared the lipolysis in explants of WAT from HFD-fed Loxp and FKO mice. The basal rate of glycerol release was similar in adipose explants of Loxp and FKO mice, but when treated with isoproterenol (ISO), which stimulates lipolysis by activating the PKA pathway, the lipolytic rate was significantly lower in epididymal adipose explants from FKO than control mice (Fig. 3A). The decreased lipolysis was further supported by reduced expression of ATGL and phosphorylation of HSL and Plin1 in FKO mice compared with Loxp mice fed with HFD in both epididymal and subcutaneous adipose tissue (Fig. 3B and Supplementary Fig. 5). The decreased expression of ATGL was also observed in both epididymal and subcutaneous adipose tissue of the FKO mice that were maintained on chow diet (Supplementary Figs. 6 and 7).

Because of the heterogeneity of cell populations within adipose explants, we also examined the effect of Sirt6 ablation on lipolysis directly in adipocytes differentiated from MEFs. In agreement with the results from WAT explants, both basal and stimulated lipolysis were significantly decreased in Sirt6 KO adipocytes (Fig. 3C). Again, the protein level of ATGL and phosphorylation of HSL and Plin1 were decreased in Sirt6 KO adipocytes (Fig. 3D and Supplementary Fig. 8A).

We then used the Sirt6 gain-of-function model to further demonstrate the effect of Sirt6 on lipolysis. Under the basal condition, Sirt6 had no effect on lipolysis, which is probably because of a low basal rate of lipolytic activity (Fig. 3E). However, when the adipocytes were stimulated with ISO, lipolysis was significantly higher in Sirt6-overexpressing adipocytes than the control adipocytes (Fig. 3E). Consistently, the protein level of ATGL and the phosphorylation of HSL and Plin1 were higher in Sirt6-overexpressing adipocytes (Fig. 3F and Supplementary Fig. 8B). To further confirm that Sirt6 can regulate ATGL expression in vivo, we injected the Sirt6-expressing adenovirus into the adipose tissue of the ob/ob mouse with overnutrition. As shown in Supplementary Fig. 9A and B, the expression of ATGL was greatly suppressed in ob/ob mice. In the same Ad-Sirt6–infected adipose tissue, the expression of ATGL was increased, and the size of adipocyte (Supplementary Fig. 9A and B) was subsequently decreased (Supplementary Fig. 9C and D). These results together suggested that Sirt6 is both necessary and sufficient to affect lipolysis, which was associated with the regulation of ATGL expression and phosphorylation of HSL and Plin1.

To understand the mechanism by which Sirt6 regulates the expression of ATGL, we evaluated the localization, expression, and transcriptional activity of FoxO1, a well-known positive transcriptional regulator of ATGL (9). Immunofluorescence staining revealed that Sirt6 KO changed the subcellular localization FoxO1, which is critical for FoxO1 activity (Fig. 4A). The results of immunofluorescence were further confirmed by Western blot analysis showing that Sirt6 ablation promoted FoxO1 nuclear export and its cytoplasm accumulation (Fig. 4B), suggesting a decrease in FoxO1 transcriptional activity. Previous reports showed that the subcellular localization of FoxO1 was regulated by its phosphorylation and acetylation (30,31). Sirt6 ablation increased the phosphorylation of FoxO1 (Fig. 4C). Conversely, overexpression of the WT Sirt6, but not the catalytic inactive mutant Sirt6 (H133y), decreased the phosphorylation of FoxO1 (Supplementary Fig. 10A and B). To understand whether Sirt6 can directly deacetylate FoxO1, coimmunoprecipitation was performed and showed that Sirt6 physically interacted with FoxO1 in HEK-293 cells (Fig. 4D). The interaction between the endogenous Sirt6 and FoxO1 in adipocytes was also confirmed by immunoprecipitation and Western blot analysis (Supplementary Fig. 10C). Furthermore, FoxO1 acetylation is increased in the Sirt6 KO cells, whereas its acetylation is decreased in Sirt6-overexpressing cells (Fig. 4E and F), suggesting Sirt6 directly deacetylates FoxO1. The deacetylation of FoxO1 appeared to be dependent on Sirt6 catalytic activity, because the mutant Sirt6 did not have these effects (Supplementary Fig. 10D and E).

To further confirm whether the activity of FoxO1 could be affected by its altered phosphorylation and acetylation, we used ChIP assay to measure FoxO1 transcriptional activity on the ATGL gene promoter. The recruitment of FoxO1 to ATGL promoter was significantly decreased in Sirt6 KO adipose tissue (Fig. 4G). In agreement with the ChIP results, ATGL natural promoter luciferase reporter assay showed that overexpression of the WT Sirt6, but not the mutant Sirt6, increased the activity of the ATGL reporter gene (Supplementary Fig. 10F, left panel). Mutation of the FoxO1 acetylation site lysine (FoxO1 6KQ, which mimics the acetylated state; or FoxO1 6KR, which prevents its acetylation) abolished the regulatory effect of Sirt6 (Supplementary Fig. 10F, right panel). Consistently, the ATGL mRNA expression was reduced in Sirt6 KO adipose tissue (Fig. 4H). Thus, ablation of Sirt6 promoted FoxO1 cytosolic localization and inhibited its occupancy on the ATGL gene promoter by increasing the phosphorylation and acetylation of FoxO1.

Obesity is often associated with macrophage infiltration in adipose tissue, which could cause chronic inflammation. Immunochemistry revealed increased staining for CD11b⁺ and F4/80 in adipose tissue of FKO mice fed an HFD, which indicated macrophage infiltration (Fig. 5A). Flow cytometry studies confirmed that CD11b⁺/F4/80 double-positive adipose tissue macrophages (ATMs) were increased ∼60% (Fig. 5B, top left panel). ATMs can be further classified into classically activated macrophages (M1), which produce proinflammatory cytokines, and alternatively activated macrophages (M2), which produce anti-inflammatory cytokines. The M1-like polarized CD11c⁺/CD11b⁺/F4/80 and triple-positive ATMs were markedly increased in adipose tissue from FKO mice (Fig. 5B, top right panel). The M2-like polarized CD206⁺/CD11b⁺/F4/80 triple-positive ATMs were not changed (Fig. 5B, bottom left panel). As a result, the ratio of M1/M2 was significantly higher in adipose tissue of FKO mice (Fig. 5B, bottom right panel). Consistent with these results, mRNA expression of macrophage marker and inflammatory genes such as CD11b⁺, Cxcl2, CD68, tumor necrosis factor-α (TNF-α), Mcp-1, and interleukin-6 (IL-6) was induced in FKO adipose tissue (Fig. 5C and D). Sirt6 KO also substantially increased the expression of inflammatory Mcp-1, IL-6, and TNF-α in the in vitro–differentiated adipocytes (Fig. 5E). Previous studies showed that c-Jun N-terminal kinase (JNK) was required for proinflammatory macrophage polarization in adipose tissue (32). The JNK was activated in adipose tissue of FKO mice (Fig. 5F) and in the Sirt6 KO adipocytes (Fig. 5G). ChIP assay also showed Sirt6 ablation increased the occupancy of c-Jun, downstream of JNK, on the gene promoters of IL-6 and Mcp-1, two proinflammatory genes that were dramatically increased in the Sirt6 knockout adipose tissue (Fig. 5H). Thus, Sirt6 deficiency may have increased inflammation in adipose tissue of mice.

The increased adiposity and inflammation in FKO mice suggested that FKO mice might have compromised glucose metabolism and insulin sensitivity. FKO mice on a chow diet showed a compromised ITT, whereas GTT was similar between the Loxp and FKO mice (Supplementary Fig. 11). When fed with an HFD, the FKO mice showed worse performance in both the GTT and ITT (Fig. 6A and B). In understanding the mechanism by which Sirt6 ablation exacerbates insulin resistance, we assessed the insulin signaling in the liver, WAT, and skeletal muscle. Insulin-induced phosphorylation of AKT was specifically attenuated in the adipose tissue (Fig. 6C), but not in the liver or skeletal muscle of FKO mice (Supplementary Fig. 12), which suggests tissue-specific impairment of insulin sensitivity. Serum chemistry also supported the impaired insulin sensitivity in the HFD-fed FKO mice, including an increase in serum levels of leptin and decrease in serum adiponectin level (Fig. 6D).

Hepatic steatosis is associated with obesity and insulin resistance. Oil Red O staining showed increased lipid accumulation in the liver of FKO mice (Fig. 6E). Lipid analysis showed that hepatic triglyceride and cholesterol levels were significantly increased in FKO mice (Fig. 6F). The mRNA levels of sterol regulatory element–binding protein-1c (Srebp-1c), fatty acid transport protein 1 (Fatp1), Fatp5, and Cd36 were induced in the liver of FKO mice (Fig. 6G), but the expression of fatty acid β-oxidative genes was not changed (Supplementary Fig. 13A–C). Together, the increased hepatic steatosis in FKO mice fed an HFD could be partially explained by increased fatty acid uptake and lipogenesis. The serum lipid profiles were similar between Loxp and FKO mice, except that the serum FFA and triglyceride levels were lower in HFD-fed FKO mice (Supplementary Fig. 13D).

Finally, we want to know whether the expression of Sirt6 is changed in adipose tissue of obese patients. Compared with nonobese subjects whose average BMI is 24, the expression of Sirt6 was significantly suppressed in adipose tissue of obese patients whose BMI is >27 (Fig. 7A and B). The obese subjects also showed a parallel decrease in ATGL expression (Fig. 7A and B), consistent with the observations in the FKO mice. The expression of ATGL is well correlated with Sirt6 expression in these individuals (Fig. 7C). Together, these results indicated that Sirt6 might have a regulatory role of Sirt6 in ATGL expression and obesity in humans.

In this study, we found that Sirt6 ablation increased diet induced-obesity through adipocyte hypertrophy rather than abnormal adipocyte differentiation. Adipocyte hypertrophy might be attributed to the impaired lipolytic activity in FKO mice, which causes fat storage synthesis to exceed lipolysis, resulting in obesity. Indeed, a defect in lipolysis has been associated with obesity (25,33). Lipolytic activity was reduced in adipocytes from obese people as compared with nonobese control subjects on stimulation (25,33,34). Lipolytic activity is stimulated acutely during fasting by ISO-stimulated β-adrenergic signaling, leading to PKA phosphorylation of perilipin and HSL (35,36). We found that Sirt6 deficiency decreased the phosphorylation of Plin1 and HSL, with no change in Plin1 and HSL protein expression. Future studies are needed to determine the mechanism by which Sirt6 ablation decreases the phosphorylation level of Plin1 and HSL.

ATGL is the key lipase that hydrolyzes triglycerides into diglycerides, which are further catalyzed by HSL into monoglycerides and glycerols (6). We found ATGL expression is significantly suppressed in adipose tissue of FKO mice. Our results are consistent with a previous report that ATGL ablation was sufficient to compromise lipolysis (10). The downregulation of ATGL may result from the increased phosphorylation and acetylation of FoxO1. ATGL is a direct target of FoxO1 (9). Insulin, an antilipolytic hormone, inhibits the expression of ATGL in adipocytes by increasing FoxO1 phosphorylation, thereby promoting its nuclear exclusion and decreasing its transcriptional activity (9,31). Recent studies also suggested that FoxO1 activity is inactivated with its acetylation during nutrient excess (37). Sirt6 physically interacted with FoxO1, and Sirt6 KO increased the acetylation of FoxO1. Functional analysis revealed that Sirt6 KO decreased FoxO1 transcriptional activity in adipocytes. Together, our results provide a plausible mechanism for Sirt6-modulated ATGL expression by regulating FoxO1 activity. The expression of ATGL is also subjected to the regulation by PPARγ (9,13,38). However, we found the expression of PPARγ and its target genes was largely unchanged.

Similar to rodents, we found the expression of Sirt6 is also suppressed by obesity in humans. The suppressed expression of Sirt6 may explain the lower level of ATGL in obesity, which has been observed in other studies (39). The expression of Sirt6 is well correlated with ATGL expression in obese subjects. It is tempting to speculate that Sirt6 might be a therapeutic target for obesity.

Chronic inflammation is an important factor in obesity and contributes to insulin resistance and type 2 diabetes (40). Sirt6 overexpression attenuates nuclear factor-κB activity (41). Conversely, Sirt6 deficiency triggers inflammation by increasing hypoxia-induced factor-1α activity (42). We found that Sirt6 KO promoted inflammatory gene expression in vivo and in vitro. Mcp-1 was greatly induced in adipose tissue of FKO mice and Sirt6 KO adipocytes. Mcp-1 can directly trigger the recruitment of macrophages to adipose tissue, and these macrophages may in turn secrete various cytokines and chemokines that further induce a local inflammatory response (43). Indeed, M1 macrophage infiltration was increased in adipose tissue of FKO mice. In addition, the ratio of M1/M2 was elevated. Our results suggested that the increased inflammatory gene expression might be a result of increased occupancy of c-JNK onto the promoter of inflammatory genes. JNK, a signal that is associated with the inflammation response, is also activated in Sirt6 KO adipose tissue and adipocytes.

The insulin resistance seen in FKO mice likely resulted from combined effects of obesity and inflammation. The reduced lipolysis of FKO adipocytes leads to marked hypertrophy of these cells. These hypertrophic adipocytes may have a compromised response to insulin. The increased adipose inflammation could promote insulin resistance in FKO mice. Sirt6 KO in adipose tissue also sensitized mice to HFD-induced hepatic steatosis, which was associated with increased expression of genes involved in fatty acid uptake and lipogenesis. The fatty acid transporters such as Fatp1, Fatp5, and Cd36 were higher in the liver of the FKO mice. The expression of the lipogenic gene Srebp-1c in the liver was also increased. Therefore, we cannot exclude the possibility that the fatty liver in FKO mice might be a secondary effect of obesity because of the inability for the animals to store excess energy within the adipose tissues, so the excess lipids are being ectopically deposited in the liver.

In summary, we show that adipose Sirt6 level is decreased in obesity. Fat-specific Sirt6 KO promoted HFD-induced obesity by inhibiting the lipolytic activity. Mechanistically, the lowered lipolysis in FKO mice was caused by reduced lipolytic activity because of impaired ATGL expression. Sirt6 KO increased phosphorylation and acetylation of FoxO1, thereby decreasing its transcriptional activity on ATGL. Our results suggest Sirt6 as an attractive therapeutic target for treating obesity and obesity-related metabolic disorders.

Acknowledgments. The authors thank Dr. Konstantin V. Kandror (Boston University School of Medicine, Boston, MA) for the ATGL promoter reporter.

Funding. This work was supported by the National Natural Science Foundation of China (81471068, 81270926, 81370521, and 81320157), Distinguished Young Scientists of Sichuan Province (2014JQ0034), and the Young Scientist Fellowship of Sichuan University (2013SCU04A17).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. J.K. designed and performed experiments and wrote the manuscript. Y.Z., Q.L., J.S., S.P., S.C., L.C., H.L., T.W., R.L., Y.L., and M.Z. helped with experiments. Z.Z., W.J., G.X., A.Q., and W.X. contributed to the discussion and review of the manuscript. J.H. obtained funding, designed experiments, and wrote the manuscript. J.K. and J.H. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.