|Home | About | Journals | Submit | Contact Us | Français|
Obesity develops due to altered energy homeostasis favoring fat storage. Here we describe a novel transcription co-regulator for adiposity and energy metabolism, TRIP-Br2 (also called SERTAD2). TRIP-Br2 null mice are resistant to obesity and obesity-related insulin resistance. Adipocytes of the knockout (KO) mice exhibited greater stimulated lipolysis secondary to enhanced expression of hormone sensitive lipase (HSL) and β3-adrenergic (Adrb3) receptors. The KOs also exhibit higher energy expenditure due to increased adipocyte thermogenesis and oxidative metabolism by up-regulating key enzymes in respective processes. Our data show for the first time that a cell cycle transcriptional co-regulator, TRIP-Br2, modulates fat storage through simultaneous regulation of lipolysis, thermogenesis and oxidative metabolism. These data together with the observation that TRIP-BR2 expression is selectively elevated in visceral fat in obese humans suggests that this transcriptional co-regulator is a novel therapeutic target for counteracting the development of obesity, insulin resistance and hyperlipidemia.
Obesity and its related complications are associated with high morbidity and mortality making it a pathologic condition of global significance1,2. Understanding the regulation of the factors that control storage, mobilization, and utilization of excess energy in adipocytes may lead to the development of potential therapies for obesity and related diseases.
In mammals, white adipose tissue (WAT) is the primary energy reservoir and a major source of metabolic fuel and regulatory adipokines3–5. On the other hand, brown adipose tissue (BAT) is specialized in energy expenditure and thermogenesis through high levels of fatty acid oxidation in numerous mitochondria, and the expression of uncoupling protein 1 (UCP-1)6.
Transcriptional control of cellular energy metabolic pathways in mammals is achieved by the coordinated action of numerous transcription factors and associated co-regulators7, which integrate signals from dietary, metabolic and endocrine pathways to control target gene expression8. The co-regulators, which can have either positive (co-activators such as PGC1α and β) or negative (co-repressors such as SRC-2 and RIP140) gene transcriptional effects on downstream targets9,10, play a crucial role in controlling fat accretion8 primarily through the modulation of PPARγ activity, the master regulator of adipocyte differentiation and lipid storage11.
We identified a novel co-regulator of the E2F1/DP1/RB gene transcription complex involved in cell cycle progression. TRIP-Br2 (also known as SERTAD2-SERTA domain containing 2)12, is a member of a novel family of mammalian transcriptional co-regulators comprised of five members13, four of which have been shown to modulate E2F-dependent transcriptional activities12,14,15. While the role of TRIP-Br2 in adipose tissue is virtually unknown, it has been proposed to recruit PHD zinc finger- and/or bromodomain-containing transcriptional co-regulators12,16, such as p300/CBP17 and KRIP-118, to E2F1/DP1 transcription complexes assembled on E2F-responsive promoters. Recent evidence suggests that besides its role in cell cycle regulation, the E2F1/DP1 transcriptional complex regulates adipocyte biology and energy metabolism19,20.
We report that the expression of TRIP-Br2 in adipose tissues is necessary for the development of diet- and age-induced obesity while its ablation enhances adipocyte lipolysis, thermogenesis and oxidative metabolism without ectopic TG storage or increase in serum nonesterified fatty acid levels. To our knowledge, this is the first report of a cell cycle transcriptional co-regulator that modulates fat storage by simultaneously regulating multiple processes including lipolysis, thermogenesis and oxidative metabolism leading to enhanced fatty acid turnover in adipocytes, in contrast to the other known described co-regulators including PGC1α, 1β, SRC-2 and RIP140 which act by regulating adipocyte lipogenesis and differentiation. The elevated expression of TRIP-Br2 in visceral fat in obese humans and its significant correlation with indices of glucose homeostasis suggests that this transcriptional co-regulator is a novel target to treat obesity and its related complications.
Consistent with a potential role in adiposity, gene and protein expression of TRIP-Br2 was significantly and selectively up-regulated in BAT, epididymal (visceral) and subcutaneous WAT in high fat diet (HFD) fed male C57BL/6J mice compared to low fat diet (LFD) fed controls (Fig. 1a; Supplementary Fig. 1a). Similarly, TRIP-Br2 expression was selectively up-regulated in epididymal and subcutaneous adipose tissues from 12 week old obese ob/ob mice (Fig. 1b, Supplementary Fig. 1b), and down-regulated in mice on a calorie restricted (CR) diet (Fig 1c). Interestingly, a significantly higher expression of TRIP-Br2 was evident in the visceral (omental) fat from obese individuals, especially those with predominant visceral fat accumulation (Fig 1d, Supplementary Fig. 2a). TRIP-Br2 expression in visceral fat was significantly correlated with visceral but not subcutaneous fat area (Fig. 1e, Table 1) in both genders (Supplementary Fig. 2b, Supplementary Table 1, Supplementary Table 2) and independently of the presence of impaired glucose tolerance or diabetes, or the treatment modalities of these conditions. The correlation of TRIP-Br2 expression in visceral fat, even after adjustment for visceral fat area, remained significant for fasting plasma insulin, glucose utilization during euglycemic-hyperinsulinemic clamp, and HbA1c, suggesting independent dual roles of TRIP-Br2 in the modulation of adiposity and glucose homeostasis (Table 1, Supplementary Table 2). In both genders, the obesity-adjusted odds of diabetes progressively increased from the first to the fourth quartile of TRIP-Br2 visceral fat expression, with OR of 4.6 (95% CI 1.1–18.6) for Q2, 5.1 (1.2–21.4) for Q3, and 9.5 (2.2–41.1) for Q4 (Fig. 1f).
To elucidate the physiological role of TRIP-Br2, we initiated studies in mice due to the limitations in performing mechanistic studies in human tissues. We investigated gene-trap targeted TRIP-Br2 knockout (KO) males on the original 129SvJ founder background21 and on the C57BL/6J background and used the latter for our studies. Successful ablation was confirmed by the virtual absence of TRIP-Br2 gene and protein expression, while the expression of TRIP-Br 1/3/4 remained unaltered (Supplementary Fig. 3). To potentially unmask the phenotype, we placed the mice on either a LFD (10% fat by kcal) or HFD (60% fat by kcal) for 12 weeks. The KO mice on the LFD exhibited a trend to reduce (~10%, p=0.05) their body weights (Fig. 1g, left). However, on the HFD, the KO mice were significantly leaner (Fig. 1g, right). At the end of the HFD, the KO mice showed little change in their body weights and were similar to that of the WTs on the LFD (p=0.92) (Fig. 1g). We also observed a protection against weight gain in the KOs even when maintained on the 129Sv background, which is known to resist diet-induced obesity22 (Fig. 1h). Interestingly, knocking out TRIP-Br2 also protected the mice from age-induced obesity (WT: 51.2 ± 2.7; Het: 47.6 ± 3.9; KO: 43.9 ± 1.9; WT vs KO, p<0.01). Food intake (Supplementary Fig. 4a), gene expression of appetite regulating peptides in the hypothalamus (Supplementary Fig. 4b) and the ability of the intestine to absorb fat23 were similar between groups (Supplementary Fig. 5).
The lower body weight of TRIP-Br2 KO mice was largely accounted for by a reduction in WAT weight (Fig. 1i, Supplementary Fig. 6) without an alteration in lean body mass and body length in HFD fed mice (Fig. 1i, 1j). On the LFD, the lack of a significant decrease in body weight in the KOs (Fig. 1g) was likely masked by an increase in their lean body mass despite a significant reduction in the percent of fat mass (Supplementary Fig. 7) indicating protection from diet-and age-induced obesity primarily due to lower fat mass.
TRIP-Br2 KO mice fed a HFD exhibited improved glucose tolerance (Fig. 2a, left), and insulin sensitivity (Fig. 2a, right), and lower fasting glucose and insulin levels (Fig. 2b). Further, the fed serum adiponectin levels, 24, were significantly higher in KO mice on the HFD (Fig. 2b). Consistently, adiponectin mRNA levels were higher in the HFD KO mice compared to WT (Fig. 2b). Increased adipocyte volume and number are positively correlated with leptin production25. The HFD-induced 4.5-fold increase in serum leptin levels in WT mice (Fig. 2b) was also reflected by higher adipose leptin mRNA levels, while serum leptin levels remained low in the HFD-fed KOs. TRIP-Br2 KO mice exhibited significantly lower fed and fasting serum cholesterol and a trend to decrease triglyceride (TG) levels on HFD (Fig. 2b, Supplementary Fig 9b). Despite the increased lipolysis in TRIP-Br2 KO mice (discussed below), serum non-esterified fatty acid (NEFA) levels remained in the normal range in KO mice on HFD diet (Supplementary Fig. 9b).
In contrast, the TRIP-Br2 KO mice fed a LFD showed a trend toward improved insulin sensitivity and a mild metabolic phenotype probably due to an insignificant reduction in their body weights (Fig. 1g, Supplementary Fig. 8, Supplementary Fig. 9a, 9b). To confirm the effect of TRIP-Br2 KO on insulin sensitivity, we stimulated LFD-fed mice in vivo or differentiated adipocytes over-expressing TRIP-Br2 in vitro with insulin and observed that expression of proteins in the insulin signaling pathway were not significantly different between the two groups in the liver and muscle (Supplementary Fig 10) as well as in differentiated adipocytes (Supplementary Fig. 11).
Chronic exposure of mice to HFD causes fatty liver (steatosis)26. Feeding a HFD increased liver mass only in the WTs (Fig. 1g) while in the KOs, the livers exhibited smaller lipid-containing vacuoles (Fig. 2c, left) and reduced TG content (Fig. 2c, right).
Obesity is associated with an increase in macrophage infiltration in visceral adipose tissue27,28. Adipose tissues from TRIP-Br2 KO mice on the HFD exhibited 57% less adipose tissue macrophage (ATM) infiltration (Fig. 2d, left and middle). Consistently, the mRNA expression of macrophage marker, F4/80 and monocyte chemoattractant protein-1 (MCP-1) was significantly lower in WAT of TRIP-Br2 KO mice (Fig. 2d, right). These data suggest TRIP-Br2 is one factor that mediates the detrimental effects of HFD-induced obesity such as glucose intolerance, insulin resistance, changes in adiponectin, leptin and serum lipid levels, hepatic steatosis and adipose tissue inflammation.
A decrease in adipose tissue mass could result either from a reduction in adipocyte size and/or a reduction in adipocyte number due to impaired differentiation29. The mRNA levels of adipogenic transcription factors30 as well as late markers of adipocyte differentiation were not significantly different in WAT between groups suggesting normal adipocyte differentiation in both fat depots (Supplementary Fig. 12a). In addition, no significant differences were observed in morphology or lipid accumulation, nor in mRNA expression of early and late adipogenesis markers in TRIP-Br2 over-expressing or TRIP-Br2 KD 3T3-L1 cells differentiated in vitro (Supplementary Fig. 12b, 12c). These results suggest that TRIP-Br2 is not important for adipocyte differentiation. A down-regulation of endogenous TRIP-Br2 expression during the differentiation of WT adipocyte (Supplementary Fig. 13) argues for a putative repressor role in mature adipocytes.
Instead, histological analyses revealed a significantly greater frequency of small adipocytes and a lower frequency of midsized and large adipocytes in epididymal WAT from the KO mice on HFD (Fig. 2e, 2f, Supplementary Fig. 14). Thus, TRIP-Br2 ablation leads to adipocyte hypotrophy due to lower TG accumulation (Fig. 2f, left) and smaller adipocyte size rather than impaired adipocyte differentiation.
We next examined whether the marked decrease in adipocyte cell size in the KO mice is due to alterations in lipolysis or lipogenesis. While no differences were detected either in the rate of lipogenesis or gene expression of lipogenesis markers31 in epididymal adipocytes between groups (Supplementary Fig. 15), β-adrenergic agonist (isoproterenol)-stimulated lipolysis was significantly higher in adipocytes from TRIP-Br2 KO mice fed a LFD or HFD (Fig. 3a, Supplementary Fig. 16). Further, while HFD feeding blunted the isoproterenol-stimulated lipolytic response of WT adipocytes, the adipocytes from TRIP-Br2 KOs fed either diet remained sensitive (Supplementary Fig. 16). This observation was confirmed by lipolysis assay in in vitro differentiated TRIP-Br2 KD or over-expressing 3T3-L1 cells (Fig. 3b). To further confirm the lipolysis changes observed in vitro or ex vivo, we measured the lipolysis rate in vivo either by IP injection of the CL316342 compound32, a β3-adrenergic agonist, or by the heavy water labeling technique33,34.The two approaches independently confirmed a significant increase in in vivo lipolysis rate in the TRIP-Br2 KO mice (Fig. 3c, 3d). Analyses of key lipolytic genes, in epididymal fat of TRIP-Br2 KO mice placed on either diet, revealed unchanged levels of adipose triglyceride lipase (ATGL)35 or its co-regulators including CGI-58 and G0S2 (Fig 3e, 3f, Supplementary Fig. 17, 18), but a significant increase in another key lipolytic enzyme, hormone sensitive lipase (HSL) at both mRNA and protein levels36 (Fig. 3e, 3f, Supplementary Fig. 19), but not one of the HSL regulators, Perilipin A. Interestingly, we observed an increased in Perilipin A gene expression (Supplementary Fig. 17) but not in the protein level (Supplementary Fig. 18) in the epididymal fat from the HFD fed KO mice. We also observed a significant increase in the mRNA and protein expression of the β3-adrenergic receptor (Adrb3) in all adipose depots from HFD fed KO mice (Fig. 3e, 3f).
Stimulation of Adrb3 signaling phosphorylates HSL and increases HSL-mediated lipolysis leading to reduced fat mass37. Consistently, we observed an increase in HSL phosphorylation in TRIP-Br2 KO mice (Fig. 3f). In the KO mice fed a LFD, a significant increase in Adrb3 and HSL mRNA levels was detected in the epididymal WAT (Supplementary Fig. 19). Consistent with the in vivo data, the gene and protein expression of HSL and Adr3b were downregulated in differentiated 3T3-L1 adipocytes over-expressing TRIP-Br2 and again ATGL was found to be unchanged (Fig. 3g).
Further, treatment of TRIP-Br2 KO adipocytes with a HSL specific inhibitor, 4-isopropyl-3-methyl-2-[1-(3-(S)-methyl-piperidin-1-yl)-methanoyl]-2H-isoxazol-5-one (BAY)38, completely prevented the increase in isoproterenol-stimulated lipolysis in TRIP-Br2 KO adipocytes (Fig. 3h), confirming the role of HSL as the major lipolytic enzyme regulated by TRIP-Br2 in the adipocyte.
To dissect the relative importance of β3-adrenergic receptor and HSL in the TRIP-Br2 KO adipocyte, we stimulated isolated primary adipocytes with isoproterenol, forskolin or IBMX to either activate the β3-adrenergic receptor or increase the cAMP level to indirectly enhance HSL enzyme activity. Interestingly, isoproterenol-stimulated lipolysis showed a tendency to be consistently higher compared to stimulation with forskolin or IBMX (Fig. 3i, upper panel) suggesting that the primary site for action is an increase in β3-adrenergic receptors in the TRIP-Br2 KO adipocyte. This observation was confirmed in the differentiated TRIP-Br2 KD 3T3-L1 cell line (Fig. 3i, lower panel).
These results suggest that increased lipolysis in adipocytes and reduced adiposity in the KO mice is mediated by specific regulation of β3-adrenergic receptors and HSL.
Reduction in adipose tissue mass without alterations in food intake (Supplementary Fig 4a) suggested a potential increase in energy expenditure in the KOs. Indeed, measurement of oxygen consumption (VO2) rates over 48 h (light and dark phases) revealed ~20% higher levels in the KOs (Fig. 4a) which could not be attributed to changes in activity (Supplementary Fig. 20a). The rate of carbon dioxide elimination (VCO2) was unchanged (Supplementary Fig. 20b). Consistent with the higher VO2, the KOs generated more heat (Supplementary Fig. 20c), a ~0.7°C higher basal core body temperature (Fig. 4b, left panel), and a higher body temperature for up to 3 h even in a cold environment (4°C) (Fig. 4b, right panel), suggesting enhanced adaptive thermogenesis. The generally lower respiratory quotient (RQ = VCO2/VO2) in the TRIP-Br2 KO mice indicated higher fat utilization (Supplementary Fig. 20d). These data indicating increased thermogenesis in the TRIP-Br2 KO mice, prompted us to examine BAT. Notably, although we did not detect significant alterations in the gross appearance or weight of interscapular BAT in the KO mice, we observed a significant increase in the number of smaller adipocytes associated with smaller lipid droplets (Fig. 1i, ,2e).2e). These results indicate that absence of TRIP-Br2 increases thermogenesis, in part, due to an increase in brown adipocyte number.
Consistent with the increase in thermogenesis and VO2 in the KOs, we observed an up-regulation of gene expression for uncoupling protein 1 (UCP1), PPARγ coactivator-1α (PGC-1α), peroxisomal enoyl coenzyme A hydratase 1 (ECH1), cytochrome C (Cyc1) and cytochrome c oxidase subunit VIIIb (Cox8b), which are genes involved in thermogenesis, oxidative metabolism and mitochondrial biogenesis (Fig. 4c, left panel, Supplementary Fig 21). Further, examination of BAT in mice exposed to a cold environment revealed a significant up-regulation in the expression of HSL and Adrb3 (Fig. 4c, right panel), consistent with observations under ambient conditions (Fig. 3e). The cold exposed BAT from the KO mice also showed a significant up-regulation of genes involved in thermogenesis and in both mitochondrial and peroxisomal fatty acid oxidation (Fig. 4c, right panel). Thus, ablation of TRIP-Br2 in brown adipocytes leads to up-regulation of Adrb3 expression and genes essential for adaptive thermogenesis and oxidative metabolism. Considering Adrb3 signaling regulates thermogenesis and fatty acid oxidation pathways39, the significant up-regulation of Adrb3 expression upon cold-exposure could potentially up-regulate the genes involved in thermogenesis and fatty acid oxidation observed in BAT in the KOs. These data were further supported by our observation that the BAT in TRIP-Br2 KOs exhibits a significantly higher oxygen consumption rate (OCR) (Fig. 4d) and increased ratio of OCR to extracellular acidification rate (ECAR) (Fig. 4e) suggesting that the BAT lacking TRIP-Br2 has enhanced oxidative metabolism compared to BAT in WT controls. To further determine its role, we denervated the BAT and interestingly observed that this manipulation only partially reverses the TRIP-Br2 obesity resistant phenotype (Fig. 4f) suggesting contributions from other tissues.
In epididymal WAT of TRIP-Br2 KO mice, we detected an elevated expression of genes (Fig. 4g, Supplementary Fig. 22), involved in promoting fatty acid oxidation. Consistently, the basal and isoproterenol-stimulated fatty acid oxidation was enhanced in the isolated primary TRIP-Br2 KO adipocytes (Fig. 4h). This observation was confirmed in vitro where the basal and isoproterenol-stimulated fatty acid oxidation rates were reduced in the differentiated 3T3-L1 adipocytes over-expressing TRIP-Br2, and conversely, enhanced in adipocytes with a TRIP-Br2 knockdown (Figure 4i). The up-regulation of UCP1 and oxidative metabolism genes are expected to result in the uncoupling of fatty acid oxidation and the enhancement of in situ fatty acid utilization, which provide a plausible explanation for the lack of increase in serum NEFA despite an elevated lipolytic activity in the WAT of TRIP-Br2 KO mice. This hypothesis is further supported by unchanged fatty acid oxidation in TRIP-Br2 KO skeletal muscle (Supplementary Fig. 23) a tissue known to be a major organ to utilize circulating free fatty acids. These findings suggest a model in which normal TRIP-Br2 expression is permissive for the development of obesity and its related complications through its effect on reducing energy expenditure and fat oxidation in adipocytes by down-regulating the expression of key proteins involved in adipocyte lipolysis, thermogenesis and oxidative metabolism (Fig. 5d).
We next used in silico sequence analysis to identify putative E2F consensus binding sites in the promoter sequences of the mHSL and mAdrb3 (Supplementary Fig. 24) that were conserved between the murine and human orthologs. Co-transfection of NIH-3T3 cells with expression constructs of E2F1, E2F4, DP1 or TRIP-Br2 with luciferase reporter constructs (pGL3) driven by the promoters of the mHSL or mAdrb3 genes revealed that over-expression of either E2F1/DP1 or E2F4/DP1 transcriptional complex stimulates both HSL and Adrb3 promoter-driven luciferase expression while the co-expression of TRIP-Br2 consistently repressed both promoter activities (Fig. 5a, Supplementary Fig. 25). These results were confirmed by partial deletion and by site-directed mutagenesis of putative E2F consensus binding sites. Our results show that the first two E2F sites in the HSL promoter (−445 and −240), and both of the E2F sites in the Adrb3 promoters are responsive to both E2F1/DP1 and E2F4/DP1 mediated transcriptional activation and TRIP-Br2 co-regulator mediated transcriptional inhibition (Fig. 5a, Supplementary Fig. 25).
To confirm the binding of TRIP-Br2 on the mHSL and mAdrb3 promoters in vivo, we performed chromatin immunoprecipitation (ChIP) in differentiated 3T3-L1 adipocytes stably expressing TRIP-Br2-FLAG or empty vector control. Fragments of mHSL and mAdrb3 promoters containing the E2F binding sites were enriched in the genomic DNA immunoprecipitated from mTRIP-Br2-FLAG over-expressing adipocytes using an anti-FLAG antibody and detected by qRT-PCR (Fig. 5b, left panel). Conversely, neither GAPDH (negative control) nor the respective promoter regions lacking the E2F sites could be detected in the immunoprecipitated genomic DNA samples (Supplementary Fig. 26). The luciferase promoter analyses demonstrating co-repressor activity of TRIP-Br2 on the transcriptional read-outs of the mHSL and mAdrb3 promoters were further confirmed by showing decreased RNA polymerase II binding at the mHSL and mAdrb3 promoters in mTRIP-Br2-FLAG over-expressing differentiated adipocytes (Fig. 5b, right panel).
To confirm the functionality of the E2F binding sites on the HSL promoter, we mutagenized the first two endogenous E2F binding sites on the HSL promoter in 3T3-L1 cells using a newly developed genome modification technique, TALEN (Transcription Activator-Like Effector Nucleases)40. Mutagenesis of the HSL promoter E2F binding sites abolished the enhanced isoproterenol-stimulated lipolysis observed in the differentiated TRIP-Br2 KD 3T3-L1 adipocytes (Fig. 5c), confirming the functionality and critical role of TRIP-Br2/E2F regulation on HSL in the adipocyte.
Taken together, these data suggest that TRIP-Br2 is recruited as a co-repressor to novel E2F1/DP1 or E2F4/DP1 complexes assembled on E2F consensus binding sites in the mHSL and mAdrb3 promoters.
Over the past several decades, adipocytes have been the target of intense investigation due to the dramatic emergence of obesity as a serious public health problem. Furthermore, adipocytes have been shown to be critical integrators of various physiological pathways regulating whole body glucose and energy homeostasis3. Identifying points of regulatory convergence that can serve as therapeutic targets to improve glucose and energy homeostasis is therefore of great necessity.
Transcriptional co-regulators have been clearly identified as having dominant roles in metabolism and energy homeostasis8,41. Despite the identification of numerous co-activators and co-repressors (e.g. PGC1 and RIP140), most of the co-regulators have shown prominent roles in fat accretion and adipocyte differentiation, primarily through the modulation of the master regulator, PPARγ transcription factor. In this study, we identify a novel transcriptional co-regulator, TRIP-Br2, for the regulation of adiposity and energy metabolism and we also report a novel link between the cell cycle regulators (E2F family) and energy homeostasis. To our knowledge, this is the first report of a cell cycle transcriptional co-regulator capable of modulating fat storage by simultaneous regulation of multiple processes including lipolysis, thermogenesis and oxidative metabolism, leading to enhanced fatty acid turnover in adipocytes. TRIP-Br2 appears to be functionally different from the other previously described transcriptional co-regulators, which regulate adipocyte lipogenesis and differentiation. Our studies suggest that TRIP-Br2 mediates this metabolic phenotype, in part, through transcription regulation of HSL and Adrb3 expression.
We observed that β-adrenergic agonist-stimulated lipolysis was markedly increased in in vivo in the TRIP-Br2 KO mice as well as ex vivo in the isolated primary adipocytes from TRIP-Br2 KO mice. Consistently, expression of HSL and Adrb3 were both significantly enhanced in the adipose tissues from the KO mice suggesting that the enhanced stimulated lipolysis is due to an increase in the expression of key proteins in the β-adrenergic regulated lipolytic pathway. However, due to technical limitations in accurately monitoring the rate of lipolysis by release of FA or glycerol, we are unable to rule out the effects of TRIP-Br2 KO on FA re-esterification.
The enhanced lipolysis is consistent with the observation of smaller adipocytes and reduced lipid accumulation in WAT and BAT of the TRIP-Br2 KO mice on a HFD. This finding is consistent with other studies reporting a smaller fat pad and adipocyte size secondary to increased lipolytic activity in the adipose specific desnutrin/ATGL overexpressors34 and the adipose-specific phospholipase A2 (AdPLA) KO42 mice. Lipolysis in adipocytes is under tight hormonal regulation and dependent on nutritional status. For example, during fasting, increased levels of catecholamines bind to β-adrenergic receptors and promote the activation of hormone sensitive lipase (HSL), whereas, in the fed state, lipolysis is inhibited by insulin. Our data suggests that absence of TRIP-Br2 enhances the fasting stimulated β-adrenergic regulated lipolytic activities.
However, it is evident that enhanced lipolysis alone may not be adequate to account for all the observed phenotypes in the TRIP-Br2 KO mice, as demonstrated by a study in which HSL over-expression failed to induce a lean phenotype in mice fed a HFD43. However, unlike the simultaneous up-regulation of both HSL and Adrb3 due to ablation of TRIP-Br2, Adrb3 was not simultaneously over-expressed in the HSL over-expression study. We observed no significant increase in serum NEFA in the TRIP-Br2 KO mice. We hypothesize that the lack of an increase in serum NEFAs despite enhanced lipolysis in TRIP-Br2 KO mice reflects the use of alternative mechanisms that allows the utilization of excess fat and FA in the mutant mice. Indeed, we observed enhanced fatty acid oxidation in the isolated primary adipocytes from the TRIP-Br2 KO mice and in the differentiated 3T3-L1 adipocytes with reduced expression of TRIP-Br2. In addition, the finding of increased oxygen consumption and heat production in the KO mice also indicates an increase in energy expenditure in vivo. The higher core body temperature and superior cold tolerance, both indicate an increase in adaptive thermogenesis in the KOs. The energy expenditure phenotype is consistent with a substantial up-regulation in the expression of genes that regulate thermogenesis, mitochondrial and peroxisomal FA oxidation in both WAT and BAT. Further, ectopic expression of UCP1 in WAT has been reported to cause resistance to diet-induced obesity with increased fatty acid oxidation in adipocytes44. Since peroxisomal FA oxidation is associated with efficient uncoupling34,45, the elevated levels of peroxisomal FA oxidation observed in TRIP-r2 KO mice is consistent with a switch from ATP production to heat generation in WAT and BAT. In BAT, cold-induced PGC1α expression has been reported to regulate thermogenesis by promoting mitochondrial respiration and by uncoupling electron transfer from ATP synthesis through the induction of UCP146,47. In our study, we observed that numerous genes involved in thermogenesis and FA oxidation were up-regulated in cold-induced BAT in the KO mice. Together, our findings indicate that FA liberated by enhanced lipolysis could potentially be used as fuel by the increased FA oxidation and thermogenesis that together contribute to a leaner phenotype in the TRIP-Br2 KO mice.
Our data demonstrate that despite enhanced lipolysis in TRIP-Br2 KO mice, coordinate up-regulation of fatty acid oxidation and thermogenesis in both WAT and BAT lead to enhanced in situ fatty acid utilization as well as the absence of increased circulating NEFA levels and ectopic fat accumulation in metabolic organs such as liver – a finding that is clearly observed in WT mice fed with HFD. This hypothesis is further supported by a recent observation that an increase in lipolysis and release of free fatty acids by the adipocytes during weight loss or in fasting mice leads to an increase in inflammation in adipose tissues48. In the TRIP-Br2 KOs, despite a striking increase in lipolysis, we observed markedly less inflammation in the adipose tissues, consistent with our hypothesis that the free fatty acid generated from the enhanced lipolysis is potentially utilized in situ in the adipocyte rather than released into the circulation.
Cell cycle regulators have been increasingly implicated in the control of metabolism49. Our data provide the first evidence that E2F/DP transcription complexes together with TRIP-Br2 regulate the β-adrenergic regulated lipolytic and mitochondrial beta oxidative pathways. This is in contrast to the previous report that E2F1 and E2F4 modulation of PPARγ transcription regulates adipocyte differentiation19,20.
The striking resistance to high fat diet-induced obesity and lean phenotype exhibited by TRIP-Br2 KO mice and its selective elevation in visceral fat in obese humans makes it an attractive therapeutic target for the treatment of obesity and obesity-related metabolic diseases.
The creation of the original TRIP-Br2 knockout out bred mouse strain in the 129SvJ background and the generation of a TRIP-Br2 inbred mouse strain on a C57BL/6 background by over 10 generations of backcrossing, has been previously reported21. Mice were housed in pathogen-free facilities and maintained on a 12 hr light/dark cycle at the Foster Biomedical Research Laboratory of Brande is University in Waltham, MA. All protocols were approved by the Institutional Animal Care and Use Committee of the Joslin Diabetes Center and Brande is University and were in accordance with NIH guidelines.
Paired samples of visceral and subcutaneous adipose tissue were obtained from 178 Caucasian men (N=92) and women (N=86), who underwent abdominal surgery as described in detail elsewhere50. The age ranged from 16 to 82 years and body mass index from 20.8 to 54.1 kg/m2. In these subjects, abdominal visceral and subcutaneous fat area was calculated using abdominal MRI scans or computed tomography scans at the level of L4–L5. Percentage body fat was measured by dual-energy X-ray absorptiometry (DEXA). Individual assigned to all three cohorts (lean: BMI<25; visceral obesity: visceral/subcutaneous fat area > 0.4) fulfilled the following inclusion criteria: 1) Absence of any acute or chronic inflammatory disease as determined by a leukocyte count > 7000 Gpt/l, C-reactive protein (CRP) > 5.0 mg/dl or clinical signs of infection, 2) Undetectable antibodies against glutamic acid decarboxylase (GAD), 3) No medical history of hypertension, i.e. systolic blood pressure (SBP) was < 140mmHg and diastolic blood pressure (DBP) was < 85mmHg, 4) No clinical evidence of either cardiovascular or peripheral artery disease, 5) No thyroid dysfunction, 6) No alcohol or drug abuse, 7) No pregnancy. All subjects had a stable weight with fluctuations smaller than 2 % of the body weight for at least 3 months before surgery. Normal glucose tolerance (NGT) was defined as fasting plasma glucose <6.0mmol/l and a 120 min OGTT plasma glucose <7.8mmol/l. Impaired glucose tolerance (IGT) was defined as fasting plasma glucose <6.0mmol/l and a 120 min OGTT plasma glucose >7.8mmol/l and <11.1mmol/l. Type 2 diabetes was defined as fasting plasma glucose > 7.0mmol/l and/or a 120 min OGTT glucose >11.1mmol/l. The study was approved by the ethics committee of the University of Leipzig. All subjects gave written informed consent before taking part in the study.
Fasting plasma insulin was measured with an enzyme immunometric assay for the IMMULITE automated analyzer (Diagnostic Products Corporation). Plasma leptin levels were assessed by radioimmunoassay (Millipore). The OGTT was performed after an overnight fast with 75 g standardized glucose solution (Glucodex Solution 75g; Merieux) and insulin sensitivity was assessed with the euglycemic-hyperinsulinemic clamp method as described51.
Human TRIP-Br2 mRNA expression was measured by quantitative real-time RT-PCR in a fluorescent temperature cycler, and fluorescence was detected on an ABI PRISM7000 sequence detector (Applied Biosystems). Total RNA was isolated from paired subcutaneous and omentaladipose tissue samples using TRIzol (Life Technologies), and 1 μg RNA was reverse transcribed with standard reagents (Life Technologies). TRIP-Br2 mRNA expression was determined by a premixed assay on demand: Hs00207372_m1 SERTA domain containing 2 (Applied Biosystems). Samples were incubated in the ABI PRISM 7000 sequence detector for an initial denaturation at 95°C for 10 min, followed by 40 PCR cycles, each cycle consisting of 95°C for 15 s, 60°C for 1 min, and 72°Cfor 1 min. TRIP-Br2 mRNA expression was calculated relative to the mRNA expression of hypoxanthine phosphoribosyltransferase 1 (HPRT1), determined by a premixed assay on demand for human HPRT1 (Applied Biosystems). Amplification of specific transcripts was confirmed by melting curve profiles(cooling the sample to 68°C and heating slowly to 95°Cwith measurement of fluorescence) at the end of each PCR. The specificity of the PCR was further verified by subjecting the amplification products to agarose gel electrophoresis.
All analyses were conducted with the two genders considered together as well as separately. TRIP-Br2 expression in subcutaneous and visceral fat was compared between lean and obese subjects by means of t tests. The unadjusted correlation between TRIP-Br2 expression and metabolic parameters was evaluated by means of Pearson’s correlation coefficients. Adjusted correlation coefficients were derived from partial r2 obtained from ANCOVA models that included each of the metabolic parameters and the visceral fat area as predictors of TRIP-Br2 expression. The association between quartiles of TRIP-Br2 visceral fat expression and risk of type 2 diabetes was evaluated by logistic regression with and without adjustment for visceral and subcutaneous fat areas and expressed by means of Odds Ratios and their 95% CI.
For diet-induced obesity, all mice were fed a chow diet (21.6% fat, 23% protein and 55.4% carbohydrate by kcal; #5020, LabDiet) until 6 weeks of age. Subsequently, mice were assigned randomly to either a low-fat (10% fat, 20% protein, and 70% carbohydrate by kcal; D12450B, Research Diets) or a high-fat diet (60% fat, 20% protein, and 20% carbohydrate by kcal; D12492, Research Diets) until the end of the experimental protocol. Body weight was measured weekly until 18 weeks of age.
Adipocyte cross-sectional area from caveolin (BD Biosciences) stained adipose tissue images (150-200 adipocytes/mouse, 3 mice/genotype) was calculated using CellProfiler image analysis software (Broad Institute, http://www.cellprofiler.org/)52,53
Differentiated 3T3-L1 cells were treated with either vehicle or isoproterenol for 1 hr. Cells were then washed and incubated with assay media containing 2% fatty acid-free BSA, 0.30 mM L-Carnitine, and 3H-palmitic acid (3 uCi/well) in low glucose DMEM media for 1 hr. Fatty acid oxidation was determined by measuring 3H2O production as previously described and normalized with total cellular protein45.
Isolated white and brown adipocytes were used to determine fatty acid oxidation by measuring 14CO2 production from [U-14C] palmitic acid (0.2 uCi/ml) after incubation for 1 hr at 37°C with gentle shaking as previously described before normalized with total lipid content54.
Transcription Activator-Like Effector Nuclease (TALEN) plasmid DNA was designed, synthesized and tested by Cellectis Bioresearch. Exchange/integration matrix harboring the mutation was synthesized by GenScript. 3T3-L1 preadipocytes at 50-60% confluency were transfected with 1 μg of each TALEN plasmid and exchange matrix using Lipofectamine2000 (Invitrogen). Twenty-four h after transfection, cells were trypsinized and selected using puromycin. Colonies were then picked and expanded before screening with PCR and confirmed with sequencing.
Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were determined in mouse brown adipose tissue using a modified protocol55. Briefly, freshly isolated BAT from WT and TRIP-Br2 KO mice (n = 5 per genotype) were rinsed with unbuffered KHB medium containing 111 mM NaCl, 4.7 mM KCl, 2 mM MgSO4, 1.2 mM Na2HPO4, 0.5 mM carnitine and 2.5 mM glucose. Adipose tissue was cut into small pieces and rinsed with KHB medium, and 10 mg of tissue was placed in each well of a XF24-well Islet Flux plate (Seahorse Bioscience). Then, 450 μl of KHB medium was added to each well and samples were analyzed in an XF24 Extracellular Flux Analyzer (Seahorse Bioscience) at 37°C56. The XF24 Analyzer mixed the media in each well three times for 2 min before measurements to allow oxygen partial pressure to equilibrate. Basal OCR and ECAR were measured in all wells three times. Five tissue replicates from 5 mice per genotype were analyzed in independent experiments and results were normalized to tissue weight.
Isolated adipocytes were incubated with or without increasing doses of isoproterenol (10−6-10−9 M), isoproterenol (5μM), forskolin (20μM) or IBMX (0.2μM) for 90 min. Extracellular glycerol release was extracted and measured using Free Glycerol Reagent (Sigma) as an indicator of lipolysis57. For the assessment of lipogenesis, isolated adipocytes were incubated with [14C] deoxyglucose in the presence or absence of increasing concentrations of insulin (0.1-100 nM) for 90 min. Incorporated radiolabeled glucose was extracted and measured using liquid scintillation counting58.
Mouse 3T3-L1 preadipocytes were grown in DMEM supplemented with 10% fetal bovine serum (FBS). Two days after confluence, adipocyte differentiation was initiated with the addition of 1.7 μM insulin, 0.5 mM isobutylmethylxanthine (IBMX) and 1 μM dexamethasone in DMEM media supplemented with 10% FBS for 2 days, followed by 2 days in medium supplemented with insulin, and finally cultured for 4 days in normal growth medium.
Lipids from tissues were extracted with Folch solution consisting of a mixture of 2:1 (vol/vol) chloroform/methanol as previously described59. Lipids were solubilized in 1% Triton X-100 before evaporation under nitrogen gas. Triglyceride content was determined using Triglyceride Determination Kit (Sigma).
Complete ORF of TRIP-Br2 was amplified from 3T3-L1 preadipocytes cDNA with PCR using oligos carrying flag epitope-tag at the C-terminal from IDT and cloned into pBabe-puro retroviral or pCDNA3.1 vector60. Mouse hormone sensitive lipase (HSL) or β3-adrenergic receptor (Adrb3) promoter and respective deletion constructs were amplified from 3T3-L1 preadipocyte genomic DNA with PCR using oligos from IDT and cloned into pGL3 or pGL3-min-promoter luciferase vector respectively (Promega). All plasmids were sequenced verified. Expression plasmids for E2F1, E2F4 and DP1 were gifts from L. Fajas (Metabolism and Cancer Laboratory, INSERM, France)19.
3T3-L1 preadipocyte cell lines stably over-expressing TRIP-Br2-Flag were established by infecting 3T3-L1 cells using retrovirus containing pBabe-puro-TRIP-Br2-flag construct, followed by puromycin selection. The stable cell lines used in the experiments were mixed clones from at least three independent viral productions, infections, and selections.
Rabbit polyclonal anti-TRIP-Br2 was raised against the peptide with the amino acid sequence TRIP-Br2-25-39: DGPSKVSYTLQRQT and affinity purified with the antigen (YenZym Antibody). Chicken polyclonal anti-HSL, anti-Adrb3, rabbit anti-perilipin A and mouse anti-α-tubulin were obtained from Abcam. pHSL, pAkt, Akt, pERK, ERK, ATGL antibodies were from Cell Signaling. PPARγ, CGI-58 and G0S2 antibodies were from Santa Cruz.
Food intake, physical activity, oxygen consumption (VO2), carbon dioxide (VCO2) and heat production was measured using the Comprehensive Laboratory Animal Monitoring System (CLAMS; Columbus Instruments). The respiratory exchange ratio (VCO2/VO2) was calculated from the gas exchange data and all data were normalized to lean body mass. Body composition (fat and lean mass) was assessed by the Dual-Energy X-Ray Absorptiometry (DEXA).
Body temperatures were assessed in 18-week-old male mice using a RET-3 rectal probe for mice (Physitemp). Mice were exposed to an ambient temperature of 4°C in a cold room until their core body temperature dropped to 25°C. Body temperatures were measured at 30 min intervals using a digital thermometer.
Plasma insulin was measured with an ELISA kit (Millipore). NEFA, TG and cholesterol concentration in serum were measured with NEFA-C and Triglyceride E tests (Wako), respectively. Serum adiponectin and leptin levels were measured with ELISA kits from R&D Systems (Joslin DERC Assay Core).
Blood glucose was monitored with an automated glucose monitor (Glucometer Elite, Bayer). Glucose tolerance tests and insulin tolerance tests were performed 16 hr after fasting as described previously61. Mice were anesthetized, and tissues were rapidly dissected, weighed and processed for immunohistochemistry as described previously61.
In anesthetized mice, the posterior aspect around the neck and intra-scapular region were first shaved and cleaned before a T-shaped incision (1/2 inch each direction) in the mid-scapular region was made. Under a dissecting microscope, the intra-scapular fat pad was carefully separated from the surrounding skin and a drop of 1% filtered sterilized toludine blue was applied to the fat pad to allow visualization of the nerves. After careful separation of the nerves from the surrounding tissues and vasculature around the fat pad, small nerves were cut in 3 different locations and around 3mm segments of the larger nerves were removed to prevent possible re-connection. Sham-operated animals were treated similarly using the same procedure except cutting the nerves. After recovery, operated animals were placed on HFD and monitored for their body weight and temperature until the end of the experiment.
3T3-L1 preadipocytes were cotransfected by Lipofectamine 2000 reagent (Invitrogen) with HSL and Adrb3 reporter constructs, various combinations of pCDNA3.1-TRIP-Br2-Flag and mammalian expression constructs expressing E2F1, E2F4, and DP1, along with a promoterless Renilla luciferase construct. Lysates were collected 36 h after transfection, and firefly and Renilla luciferase activities were measured with a Dual-Luciferase Reporter System (Promega).
To prepare the substrate for the TG hydrolase activity, triolein (Sigma) and [9,10-3H] triolein (Perkin Elmer) as radioactive tracer was emulsified with phosphatidylcholine/phosphatidylinositol (Sigma) using a sonicator. The final substrate concentration was 167 nmol of triolein/assay (8000 cpm of [9,10-3H] triolein/nmol). 100 μl with or without HSL specific inhibitor were incubated with 100μl of the substrate (adipose tissue extract) in a water bath at 37°C for 60 min. The reaction was terminated by adding 3.25 ml of methanol/chloroform/heptane (10:9:7) and 1 ml of 0.1 M potassium carbonate, 0.1 M boric acid, pH 10.5 before vortex twice for 30 s. After centrifugation at 3200 rpm for 15 min at room temperature, the radioactivity in 1 ml of the upper phase was determined by liquid scintillation counting.
Adipocytes derived from the Cont-3T3-L1 and TRIP-Br2-Flag-3T3-L1 cell lines were fixed at day six of their differentiation by addition of 37% formaldehyde to a final concentration of1% formaldehyde and incubated at room temperature for 10 min. Cross-linking was stopped by addition of glycine to a final concentration of 0.125 M. Cells were then scraped and samples were prepared using the EZ-Magna ChIP™ G Chromatin Immunoprecipitation Kit (Millipore) according to the manufacturer’s protocol. The chromatin fractions were incubated in each case with 10 μg of one of the following antibodies: anti-Flag M2 (F-1804; Sigma), anti-RNA polymerase II and Normal IgG mouse (both provided by Millipore Kit), Millipore) at 4 °C overnight with Magnetic Protein G Beads. After extensive washing and final elution, the product was treated for 4h at 65°C to reverse cross-linking. Input DNA and immunoprecipitated DNA were purified using kit column and analyzed by quantitative PCR using Maxima™ Sybr Green qPCR Master Mix (Fermentas Life Sciences) with different set of primers (both proximal and distal promoter regions) listed here. Mouse HSL proximal promoter (F 5′-gcggaatggaaacagcgtagtgaa-3′, R 5′-tgttacctgccattgcttcggaga-3′), mouse HSL distal promoter (F 5′-aacacagttcaagggctggagaga -3′, R 5′-caccatgtggttgctgggaattga -3′), mouse GAPDH promoter 62 (F 5′-tactcgcggctttacggg-3′, R 5′-tggaacagggaggagcag-3′), mouse β3 adrenergic receptor proximal (F 5′-cttggatggtttgggttgttcggt -3′, R 5′-agggcttctgctgcaaagaagaga -3′) and mouse β3 adrenergic receptor distal (F 5′-ttgtcccaaccaggacagaacaga -3′, R 5′-aagaaagccaggcacgttcacaag -3′). Data were normalized by input values (for each set of promoter oligonucleotides) and binding was expressed relative to the non-specific binding of IgG immunoprecipitated DNA content.
All data are presented as mean ± SEM and analyzed by unpaired two-tailed Student’s t-test or analysis of variance (ANOVA) as appropriate. A p-value less than 0.05 is considered significant.
The authors thank C. R. Kahn MD for providing reagents and discussions, E. Rosen MD PhD for discussions, E. Morgan and K. Parlee for excellent assistance in the preparation of this manuscript, H. Li for assistance with hormone assays, O. P. McGuinness PhD for mouse metabolic phenotyping and M. Mori PhD for providing samples, R. Zechner PhD for providing the protocol for TG hydrolase activitites. This work was supported by RO1 DK 67536 and the Graetz Bridge Funds (R.N.K.), R01 HL073168 (A.D.), K99 DK090210 (C.W.L.), DK51586 and DK58825 (S.R.F.) and the Joslin DERC Specialized Assay and Advanced Microscopy Cores (NIH P30 DK36836). The human studies were supported by a grant of the Deutsche Forschungsgemeinschaft Clinical Research group “Atherobesity” (KFO152; BL 833/1-1). C.W.L. was supported by a NIH Interdisciplinary training grant (1RL9EB008539-01) (SysCODE), K99 DK090210 and R00 DK090210. D.K. is the recipient of a Research Fellowship (Manpei Suzuki Diabetes Foundation, Japan), and a JDRF Post-doctoral Fellowship. S.I.H. and J.K.C were supported by the R. Glenn Davis (DCI) Endowed Professorship in Clinical and Translational Medicine and by the University of Florida, Division of Nephrology Gatorade Fund. S.I.H. was supported as a Scholar of the Clinical Translational Science Institute at the University of Florida.
Author ContributionsC.W.L., J.K.C., S.I.H. and R.N.K. conceived the project. C.W.L. and R.N.K. designed the experiments. A.D. analyzed human data. C.W.L., J.B., J.K.C., C.V., D.K., J.H., C.M., M.K.H., K.T. and H.J.K. performed experiments and analyzed data. M.B. contributed human samples and supervised human expression analysis. S.R.F. supervised experiments. D.L., S.I.H., Y.H.T. and L.G. contributed reagents. C.W.L. and R.N.K. wrote the paper. All authors discussed the results and commented on the manuscript.