Thyroid-hormone-induced browning of white adipose tissue does not contribute to thermogenesis and glucose consumption

Article Thyroid-Hormone-Induced Browning of White Adipose Tissue Does Not Contribute to Thermogenesis and Glucose Consumption Graphical Abstract Highlights dThyroid hormone induces browning independent of sympathetic nervous system through TRb dDespite high UCP1, the beige fat lacks adrenergic input and is inactive dThyroid hormone’s metabolic and thermogenic effects are maintained in UCP1 knockout mice dThyroid hormone induces hyperthermia and elevates the body temperature setpoint Authors Kornelia Johann, Anna Lena Cremer, Alexander W. Fischer, ..., Jeffrey W. Dalley, Heiko Backes, Jens Mittag Correspondence [email protected] In Brief Thyroid hormone induces browning of white fat, but it is unclear whether this contributes to thermogenesis. Here, Johann et al. show that thyroid-hormone- induced beige fat is metabolically inactive due to lack of central stimulation and that the metabolic and thermogenic effects of the hormone are independent of UCP1. Johann et al., 2019, Cell Reports 27, 3385–3400 June 11, 2019 ª2019 The Author(s). https://doi.org/10.1016/j.celrep.2019.05.054 Cell Reports Article Thyroid-Hormone-Induced Browning of White Adipose Tissue Does Not Contribute to Thermogenesis and Glucose Consumption Kornelia Johann, 1 Anna Lena Cremer, 2 Alexander W. Fischer, 3 Markus Heine, 3 Eva Rial Pensado, 4,5 Julia Resch, 1 Sebastian Nock, 1 Samuel Virtue, 6 Lisbeth Harder, 1 Rebecca Oelkrug, 1 Mariana Astiz, 7 Georg Brabant, 1 Amy Warner, 6 Antonio Vidal-Puig, 6 Henrik Oster, 7 Anita Boelen, 8 Miguel Lo ´pez, 4,5 Joerg Heeren, 3 Jeffrey W. Dalley, 9,10 Heiko Backes, 2 and Jens Mittag 1,11, * 1 Internal Medicine I, Molecular Endocrinology, Center of Brain, Behavior and Metabolism, University of L€ ubeck, 23562 L€ ubeck, Germany 2 Multimodal Imaging of Brain Metabolism, Max Planck Institute of Metabolism Research, 50931 Cologne, Germany 3 Department of Biochemistry and Molecular Cell Biology, University Medical Center Hamburg-Eppendorf, 20246 Hamburg, Germany 4 NeurObesity Group, Department of Physiology, CIMUS, University of Santiago de Compostela-Instituto de Investigacio ´n Sanitaria, Santiago de Compostela 15782, Spain 5 CIBER Fisiopatologı ´a de la Obesidad y Nutricio ´n (CIBERobn), Santiago de Compostela 15782, Spain 6 University of Cambridge Metabolic Research Laboratories, Wellcome Trust-MRC Institute of Metabolic Science, Cambridge CB2 0QQ, UK 7 Institute of Neurobiology, Center of Brain, Behavior and Metabolism, University of L€ ubeck, 23562 L€ ubeck, Germany 8 Laboratory of Endocrinology, Amsterdam University Medical Centers, 1105 Amsterdam, the Netherlands 9 Department of Psychology, University of Cambridge, Cambridge CB2 3EB, UK 10 Department of Psychiatry, University of Cambridge, Cambridge CB2 2QQ, UK 11 Lead Contact *Correspondence: jens.mitt[email protected] https://doi.org/10.1016/j.celrep.2019.05.054 SUMMARY Regulation of body temperature critically depends on thyroid hormone (TH). Recent studies revealed that TH induces browning of white adipose tissue, possibly contributing to the observed hyperthermia in hyperthyroid patients and potentially providing metabolic benefits. Here, we show that browning by TH requires TH-receptor band occurs independently of the sympathetic nervous system. The beige fat, however, lacks sufficient adrenergic stimulation and is not metabolically activated despite high levels of uncoupling protein 1 (UCP1). Studies at different environmental temperatures reveal that TH instead causes hyperthermia by actions in skeletal muscle combined with a central body temperature set-point elevation. Consequently, the metabolic and thermogenic effects of systemic hyperthyroidism were maintained in UCP1 knockout mice, demonstrating that neither beige nor brown fat contributes to the TH-induced hyperthermia and elevated glucose consumption, and underlining that the mere presence of UCP1 is insufficient to draw conclusions on the therapeutic potential of browning agents. INTRODUCTION Obligatory thermogenesis is defined as heat resulting from maintenance of vital functions. In homeothermic species living in thermoneutral conditions, this type of thermogenesis is sufficient to maintain body temperature. Whenever the ambient temperature is reduced, heat-saving mechanisms, such as vasoconstriction, decreased movement, and piloerection, are activated, and adaptive (or facultative) thermogenesis is induced. Among these adaptive thermogenic mechanisms is the activation of non-shivering thermogenesis in brown adipose tissue (BAT). BAT thermogenesis relies on the expression of uncoupling protein 1 (UCP1) that dissociates mitochondrial substrate oxidation from ATP synthesis by introducing a proton leak, resulting in heat release (Cannon and Nedergaard, 2004). BAT is activated by the sympathetic nervous system (SNS) releasing norepinephrine (NE) that binds to b3-adrenergic receptors (ADRB3) on brown adipocytes, leading to an increase in intracellular cyclic AMP (cAMP) (Zhao et al., 1994). This rise in cAMP activates, for instance, the conversion of the thyroid hormone thyroxine (T4) to the more biologically active form 3,30,5-triiodothyronine (T3) by the enzyme deiodinase type 2 (DIO2) (Silva and Larsen, 1983). T3 in turn increases Ucp1 expression (Rabelo et al., 1996). BAT thermogenesis has been shown to have numerous metabolic benefits, like increased glucose and lipid uptake (Bartelt et al., 2011; Berbe ´eetal., 2015; Greco-Perotto et al., 1987; Shibata et al., 1989; Heine et al., 2018), anti-diabetic effects, and amelioration of obesity (Bartelt et al., 2018; Kajimura et al., 2015; Svensson et al., 2016). Additionally, it was found that sustained cold exposure or certain compounds can induce brown adipocyte-like (brite or beige) cells in subcutaneous white adipose tissue of mice, a process called browning (Bartelt and Heeren, 2014; Ishibashi and Seale, 2010; Petrovic et al., 2010; Young et al., 1984). It is currently assumed that this beige fat might contribute to thermogenesis and may have beneficial metabolic effects, although the majority of studies have failed to demonstrate whether increased UCP1 in beige fat also results in higher thermogenesis or metabolism on the systemic level (Bartelt and Heeren, Cell Reports 27, 3385–3400, June 11, 2019 ª2019 The Author(s). 3385 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). (legend on next page) 3386 Cell Reports 27, 3385–3400, June 11, 2019 2014; Harms and Seale, 2013; Kajimura et al., 2015; Keipert and Jastroch, 2014; Warner and Mittag, 2016). Thyroid hormones are essential in the regulation of metabolic and thermoregulatory pathways in mammals (Mullur et al., 2014; Silva, 2006; Yen, 2001). This becomes evident in hyperthyroid patients, which show impaired heat tolerance and increased body temperature (Silva, 2003). The effects of thyroid hormone on thermogenesis have been studied for decades, establishing the paradigm that the hormone increases obligatory thermogenesis in several tissues, including muscle, which leads to a higher body temperature (Silva, 2006). Thyroid hormones exert their actions by the nuclear thyroid hormone receptors (TRs) alpha1 (TRa1), beta1 (TRb1), or beta 2 (TRb2), altering gene expression (Sap et al., 1986; Tata, 1986; Thompson et al., 1987). TRs are expressed in almost every tissue and often have overlapping functions; however, there are tissues that predominantly rely on TRa1orTRb, respectively (Yen, 2001). With regard to thermogenesis, TRbhas been shown to play a role in the regulation of adaptive thermogenesis by regulating Ucp1 mRNA expression, and TRa1 seems to modulate adrenergic sensitivity (Ribeiro et al., 2001; Weiss et al., 1998). However, mice devoid of all TRs still display competent BAT recruitment and Ucp1 gene expression but depressed thermogenesis, suggesting a complex interplay between thyroid hormone and body temperature regulation (Golozoubova et al., 2004). Moreover, it was recently discovered that TRa1 also controls tail heat loss with secondary effects for BAT thermogenesis (Warner et al., 2013). Therefore, thyroid hormones are relevant for almost all aspects of thermoregulation, including thermal conductance as well as obligatory and facultative thermogenesis. In addition to the peripheral actions, recent studies revealed that thyroid hormone also induces facultative thermogenesis through central mechanisms, as central hyperthyroidism leads to direct activation of BAT in an AMP-activated protein kinase (AMPK)-dependent manner in rats (Alvarez-Crespo et al., 2016; Lo ´pez et al., 2010; Martı ´nez- Sa ´nchez et al., 2017b) and induces browning in mice (Martı ´- nez-Sa ´nchez et al., 2017a). Most interestingly, a recent study revealed that the TRbagonist GC-1 directly triggers browning (Lin et al., 2015), an effect that has been reported for thyroid hormone as well (Hoefig et al., 2016; Weiner et al., 2016). The data suggested that the beige fat recruitment could contribute to the body temperature increase observed in hyperthyroidism and might provide favorable metabolic effects. The findings therefore challenge the well-established model that thyroid hormone only modulates basal metabolic rate and obligatory thermogenesis (Silva, 2006) and indicate that beige fat might be a prominent player in the thermogenic and metabolic effects of the hormone. Therefore, we aimed to test the role of thermogenic adipose tissue in body temperature regulation and for the metabolic benefits observed in systemic hyperthyroidism. RESULTS Hyperthyroidism Improves Glucose Tolerance and Induces Browning of Inguinal White Adipose Tissue To establish the effects of hyperthyroidism, wild-type (WT) mice were treated for 14 days with either 0.5 mg/L T3 or 1 mg/L T4 in drinking water (Figure 1A). Treatment with T3 induced a substantial hyperthyroidism with an 8.5-fold increase in serum total T3 (tT3) and a 90% decrease in serum tT4 (Figures 1B and 1C). Hyperthyroidism was fully achieved after 24 h (not shown), persistently observed over the course of a day (Figures S1A and S1B), and leads to a complete suppression of pituitary thyroid-stimulating hormone (TSH) (Figure S1C). It was accompanied by an increase in energy expenditure, a lower respiratory quotient, and an increase in body weight and length as expected (Neilsen, 1953; Rakov et al., 2016; Green, 1975), as well as hyperphagia (Figures S1D–S1H). Our model for T4-induced hyperthyroidism was milder, with no significant elevation in tT3 and a 7.4-fold increase in serum tT4 (Figures 1B and 1C) but no significant changes in body weight, length, or food intake (Figures S1F–S1H). As hyperthyroidism is associated with improved glucose tolerance, we performed an intraperitoneal glucose tolerance test (ipGTT) and found reduced blood glucose levels in T3-treated, but not T4-treated, mice (Figures 1D and 1E; p < 0.01 for T3; p = 0.34 for T4; repeated measures [RM]-ANOVA). Core body temperature in thyroid-hormone-treated mice started to increase after 1 week of treatment (Figure 1F) and was significantly elevated in T3-treated, but not T4-treated, mice after 2 weeks (T3 p < 0.01; T4 p = 0.09; Figure 1G). Skin temperature above the interscapular brown adipose tissue (iBAT) as measured by infrared thermography did not significantly change Figure 1. Hyperthyroidism Improves Glucose Tolerance and Induces Browning of ingWAT (A) Scheme of treatment regimen for induction of hyperthyroidism in mice. (B) Serum total T4 levels after 14 days of treatment. (C) Serum total T3 levels after 14 days of treatment. (D) ipGTT after 12 days of treatment with THs. (E) Area under curve (AUC) of ipGTT (D). (F) Rectal temperature of control and T3- and T4-treated mice over 14 days of treatment. (G) Rectal temperature at the end of treatment. (H) iBAT temperature as measured by infrared thermography over 14 days of treatment. (I) iBAT temperature as measured by infrared thermography at the end of treatment normalized to rectal temperature. (J) Tail temperature of control and T3- and T4-treated mice as measured by infrared thermography over 14 days of treatment. (K) Hepatic PK activity in control and T3- and T4-treated mice. (L) Hepatic PEPCK activity in control and T3- and T4-treated mice. (M) Hepatic glycogen content of control and T3- and T4-treated mice. (N) Representative H&E stainings of iBAT and ingWAT of control and T3- and T4-treated mice. Scale bar 10 mm. (O and P) Gene expression analysis in (O) ingWAT and (P) iBAT in control and T3- and T4-treated mice. (Q) Direct comparison of ingWAT UCP1 protein expression of T3-treated mice to control iBAT. Data are presented as mean ±SEM; n = 6–8. *p < 0.05; **p < 0.01; ***p < 0.001. See also Figure S1. Cell Reports 27, 3385–3400, June 11, 2019 3387 over the course of the 14 days of treatment (Figures 1H and S1I) but was significantly decreased upon T3 at the end of the treatment, in absolute values but also when normalized to body temperature (Figures 1H and 1I). In parallel, skin temperature above the inguinal white adipose tissue (ingWAT) was unchanged (Figure S1J). Interestingly, tail temperature was significantly elevated over the course of the T3 treatment indicative of vasodilation to dissipate excess heat starting at day 4 of the treatment (Figure 1J). To further investigate glucose metabolism, we measured the activities of rate-limiting steps in glycolysis (pyruvate kinase [PK]) and gluconeogenesis (phosphoenolpyruvate-carboxykinase [PEPCK]) in liver. PK activity was significantly decreased in T3- and T4-treated mice (Figure 1K), and PEPCK activity was increased in T3-treated mice (Figure 1L). Hepatic glycogen content was decreased in T3-treated mice (Figure 1M), indicating a higher systemic need for glucose. Organ weights determined at the time of sacrifice showed the expected cardiac hypertrophy in T3-treated, but not T4-treated, mice (Figures S1K and S1L). In both T3- and T4-treated animals, iBAT depots were significantly larger and appeared paler, suggesting higher lipid deposition (Figure S1M), which was confirmed by H&E staining (Figure 1N). Interestingly, when we investigated ingWAT morphology (Figure 1N), an increase in smaller, multilocular fat cells was observed, indicating induction of browning by thyroid hormone, as reported previously (Alvarez-Crespo et al., 2016; Finan et al., 2016; Weiner et al., 2016). To test this on the molecular level, we measured gene expression of thermogenic marker genes (uncoupling protein 1 [Ucp1]; deiodinase 2 [Dio2]; peroxisome proliferator-activated receptor gamma coactivator 1 alpha [Ppargc1a]; mitochondrial transcription factor A [Tfam]; PR domain containing 16 [Prdm16]; cell-death-inducing DFFA-like effector a [Cidea]; b3-adrenergic receptor [Adrb3]; Figures 1O and 1P) and UCP1 protein expression (Figure 1Q). In ingWAT, we found an elevation of several thermogenic markers, including Ucp1,Dio2,Ppargc1a, and Cidea in T3- and T4-treated animals, and Adrb3 mRNA was reduced upon T3 treatment (Figure 1O). In contrast, iBAT gene expression analysis showed a decrease in most thermogenic and brown fat marker genes (Figure 1P). Interestingly, despite reduced mRNA levels, UCP1 protein was elevated in T3-treated iBAT with no change in the protein level of the mitochondrial respiratory chain complexes (Figures S1N–S1P). In T3-treated ingWAT, UCP1 protein reached levels comparable to about one-third of the amount found in untreated iBAT (Figure 1Q), suggesting a possibly high thermogenic potential. Browning was also induced by T3 and T4 in gonadal WAT (gWAT) on the mRNA level (Figure S1Q); however, no significant amount of UCP1 protein was detectable in this tissue (Figure S1R). Given that the T3 treatment entirely suppressed TSH, thus potentially blocking the oscillating rhythm of thyroid hormone secretion, we tested whether the lack of a circadian rhythm would by itself cause browning. Using clock-deficient Per1/Per2 double mutant (double knockout [DKO]) mice, we confirmed that this was not the case, as Ucp1 mRNA expression was not induced in ingWAT (Figure S1S). Taken together, these data show that T3 and T4 are capable of browning ingWAT on the molecular level; however, the systemic effects of hyperthyroidism are more pronounced in the T3-treated group. Ucp1 Expression in ingWAT Is Dependent on TRb As it was reported that browning of ingWAT can be induced by TRb-selective compounds (Lin et al., 2015), we investigated the metabolic phenotype of global TRbknockout (KO) mice (Figure 2A). These mice showed elevated serum tT3 and tT4 levels (Figures S2A and S2B) due to the impaired feedback of the hypothalamus-pituitary-thyroid axis (Forrest et al., 1996). Although ipGTT was not altered in TRbKO mice compared to their WT littermates (Figures 2B and 2C; p = 0.41 for TRb; RM-ANOVA), their basal metabolic rate (BMR) (defined here as metabolism at thermoneutrality) was significantly increased (Figures 2Dand 2E). This did, however, not translate to an increase in daily energy expenditure at room temperature (Figures S2C and S2D), suggesting a compensatory adaptation of non-shivering thermogenesis. Despite being hyperthyroid, TRbKO mice did not have elevated body or iBAT temperature (Figures 2F–2H), and the histology of iBAT and ingWAT was similar to controls (Figure 2I). When investigating gene expression in ingWAT, we found that Dio2 and Adrb3 expression were not altered in hyperthyroid TRbKO mice, but Ucp1 mRNA expression was even lower than in controls (Figure 2J). In iBAT, Ucp1 expression was not altered on mRNA (Figure 2K) or protein level (Figures S2E and S2F). These data demonstrate that browning of ingWAT by thyroid hormone on the molecular level depends on intact TRbsignaling. Hyperthyroidism Decreases Adipose Tissue Glucose and Lipid Uptake Activation of thermogenesis in adipose tissues is associated with increased glucose uptake (Cannon and Nedergaard, 2004). To test this directly, we performed [18]fluoro-2-deoxyglucose– positron emission tomography ( 18 FDG-PET)/computed tomography (CT) scans of T3- and T4-treated mice. Despite elevated UCP1 levels, ingWAT glucose uptake was not increased but decreased upon systemic T3 or T4 treatment (Figures 3Aand 3B). Likewise, glucose uptake in iBAT was reduced (Figures 3C and 3D), and no change was observed in gWAT or soleus muscle (Figures S2G and S2H). 18 FDG-PET/CT may, however, not accurately reflect brown or beige fat thermogenic activity (Hankir et al., 2017; Olsen et al., 2017), as thermogenesis is mainly fueled by fatty acids (Bartelt et al., 2011; Cannon and Nedergaard, 2004). We therefore also measured uptake of triglyceride-rich lipoproteins (TRLs) in hyperthyroid mice. In iBAT, TRL uptake significantly decreased to about one-third in systemically T3- and T4-treated mice, and no difference was observed in ingWAT (Figure 3E). Given the lack of metabolic activation in the presence of elevated UCP1 protein levels, we hypothesized that the thermogenic capacity of the tissue might not be fully exploited due to insufficient adrenergic stimulation. Therefore, we injected mice with 1 mg/kg NE to stimulate maximum non-shivering thermogenesis. Control and T3-treated animals showed an elevation in oxygen consumption after NE injection (Figure 3F); however, the response was more pronounced in T3-treated mice, indicating a higher thermogenic capacity. Our hypothesis of insufficient adrenergic stimulation was further supported on the molecular level, as we found no significant elevation in ingWAT and iBAT of intracellular cAMP (Figures 3G and 3H), phosphohormone sensitive lipase (HSL; Figures 3I and 3J), or free fatty acids (Figures 3K and 3L), which are required for activation of 3388 Cell Reports 27, 3385–3400, June 11, 2019 UCP1 (Shabalina et al., 2010). Interestingly, expression of genes involved in lipogenesis and lipolysis (peroxisome proliferatoractivated receptor gamma [Pparg]; fatty acid translocase [Cd36]; carbohydrate-responsive element-binding protein beta [Chrebpb]; lipoprotein lipase [Lpl]; leptin [Lep]; lipase A [Lipa]; stearoyl-coen- zyme A [CoA]desaturase 1 [Scd1]; fatty acid synthase [Fasn]) was mostly unaltered in ingWAT (Figure S2I) while being reduced in iBAT of hyperthyroid mice (Figure S2J). Collectively, these data suggest that, despite a molecular browning fingerprint including elevated UCP1 protein, the metabolic turnover is not elevated in thermogenic adipose tissues of hyperthyroid animals. Thyroid-Hormone-Induced Hyperthermia Is Independent of SNS Signaling in Adipose Tissues T3 centrally activates thermogenesis via the SNS (Alvarez- Crespo et al., 2016; Lo ´pez et al., 2010; Martı ´nez-Sa ´nchez et al., 2017b). Interestingly, we observed a downregulation of Adrb3 expression in iBAT and ingWAT in systemic hyperthyroidism, suggesting reduced adrenergic responsiveness of these tissues. We then measured NE turnover in hyperthyroid mice (Figures S3A and S3B). The data revealed that baseline NE levels (0 h) and NE turnover were not significantly different between controls and T3- or T4-treated mice at room temperature in ingWAT but lower in iBAT (Figures 4A, 4B, S3C, and S3D). This was accompanied by normal levels of tyrosine hydroxylase protein in both tissues (Figures S3E–S3G) and suggests that the elevation of body temperature in systemically hyperthyroid animals might not be caused by a central activation of the SNS. To elucidate this in greater detail, we housed mice at thermoneutrality (TN) (Figure 4C), where thermogenic adipose tissue is only minimally innervated due to the lack of facultative thermogenesis (Cannon and Nedergaard, 2004; Sjo ¨gren et al., 2007). Figure 2. Ucp1 Expression in ingWAT Is Dependent on TRbExpression (A) Adult TRbKO mice and their WT littermates were metabolically characterized. (B) ipGTT of WT and TRbKO mice. (C) Area under curve of ipGTT (B). (D) BMR of WT and TRbKO mice normalized to body weight. (E) BMR of WT and TRbKO mice plotted against body weight. (F) Rectal temperature of WT and TRbKO mice. (G) iBAT temperature as measured by infrared thermography. (H) Representative infrared pictures of iBAT of WT and TRbKO mice. (I) Representative H&E stainings of ingWAT and iBAT of WT and TRbKO mice. Scale bar 10 mm. (J and K) Gene expression analysis of (J) ingWAT and (K) iBAT of WT and TRbKO mice. Data are presented as mean ±SEM; n = 5. *p < 0.05; **p < 0.01; ***p < 0.001. See also Figure S2. Cell Reports 27, 3385–3400, June 11, 2019 3389 Figure 3. Hyperthyroidism Decreases Adipose Tissue Glucose and Lipid Uptake (A) Rate of glucose metabolism in ingWAT of control and hyperthyroid mice after 14 days of treatment as measured by 18 FDG-PET/CT. n = 7–14. (B) Representative 18 FDG-PET/CT pictures of ingWAT of control and hyperthyroid mice. White arrows indicate ingWAT depot analyzed. (C) Rate of glucose metabolism in iBAT of control and hyperthyroid mice after 14 days of treatment as measured by 18 FDG-PET/CT. n = 7–14. (D) Representative 18 FDG-PET/CT pictures of iBAT of control and hyperthyroid mice. White arrows indicate iBAT depot analyzed. (E) Uptake of radioactively labeled TRLs into ingWAT and iBAT of control and T3- and T4-treated mice after 14 days of treatment. n = 7–8. (legend continued on next page) 3390 Cell Reports 27, 3385–3400, June 11, 2019 Interestingly, TN-housed mice still displayed a significantly elevated body temperature already after 1 day of treatment with T3 (Figures 4D, 4E, S3H, and S3I). As expected, NE levels in ingWAT (Figures 4F and S3J) as well as iBAT (Figures 4G and S3K) were strongly reduced at TN as compared to room temperature. Only when normalized to tissue weight, but not protein content, T3 caused a significantly further decline of NE in iBAT at both temperatures (Figures 4G and S3K). In both tissues, no significant change was observed in intracellular cAMP (Figures 4H and 4I) and free fatty acids (Figures 4J and 4K). Interestingly, ingWAT of T3-treated mice at TN still showed histological signs of browning (Figure 4L), together with increased expression of Ucp1, albeit at much lower levels (Figure 4M; Table S1). As expected, we found ‘‘whitened iBAT’’ in control and T3-treated mice at TN characterized by bigger, unilocular cells (Figure 4L). The lower Ucp1 gene expression in iBAT by systemic T3 treatment was reversed at TN (Figure 4N; Table S1), which together with elevated UCP1 protein (Figure 4O) suggests a ‘‘re-browning’’ of the whitened iBAT by T3. Together, our data implicate that the increase in body temperature and the browning of ingWAT in systemic hyperthyroidism occur independently of SNS activation. Moreover, the beige fat seems to lack thermogenic activity due to insufficient sympathetic stimulation. Interestingly, this is opposite to what is observed in central hyperthyroidism, which is characterized by marked SNS activation (Lo ´pez et al., 2010; Martı ´nez-Sa ´nchez et al., 2017b). We therefore tested whether the AMPK pathway in the ventromedial nucleus of the hypothalamus (VMH) is triggered by the systemic T3 treatment. Our data revealed that there was no effect of the systemic T3 treatment at 23Cor30 C on AMPKaphosphorylation in the VMH (Figures 4P–4S), despite T3 reaching the brain as evidenced by the T3-induced hypothalamic genes Dio3,Hr, and Klf9 (Figure S3L), as well as significantly increased hypothalamic T3 content (Figure S3M). Thyroid-Hormone-Mediated Increase in Body Temperature Persists in the Cold To test whether alleviating the heat stress of the T3 treatment normalizes body temperature regulation, we repeated the treatment at 10C(Figures 5A, S3N, and S3O). Interestingly, the elevation of body temperature by T3 persisted in these animals (Figure 5B) with normal iBAT (Figures 5C and 5D) and tail temperature (Figure 5E), indicating the absence of hyperthermia and a centrally elevated body temperature setpoint. Interestingly, T3 still induced browning of ingWAT at the histological level, and iBAT contained more fat droplets, comparable to the situation at higher temperatures (Figure 5F). However, on the molecular level (Figures 5G and S3P), Ucp1 mRNA was not significantly higher in the T3-treated group, most likely because the ingWAT of control animals was browned by the lower temperature. Only Dio2 mRNA levels were still elevated by the T3 treatment (Figure 5G). In contrast, iBAT showed lower expression of thermogenic genes and Adrb3 in the T3 group (Figures 5H and S3Q), with normal UCP1 protein (Figure 5I). The NE stimulation test showed a stronger response in both groups and a smaller difference between T3-treated and untreated animals (Figure 5J), suggesting a lower additional effect of T3 on thermogenic capacity at 10C as compared to 23C. We observed comparable levels of NE and intracellular cAMP in ingWAT and iBAT of both groups (Figures 5K–5N, S3R, and S3S). Interestingly, free fatty acids were significantly elevated in ingWAT of T3-treated mice, and no difference was found in iBAT (Figures 5O and 5P). When we analyzed the VMH, we observed that now, in the absence of heat stress as evidenced by the normal tail temperature, AMPKaphosphorylation was reduced by the T3 treatment (Figures 5Q and 5R). Thyroid-Hormone-Mediated Increase in Body Temperature Is Independent of UCP1 As neither beige nor brown fat appeared to be active in hyperthyroid mice at room temperature, we hypothesized that the observed hyperthermia might not depend on UCP1-mediated thermogenesis. To test this directly, we treated UCP1 KO mice and their WT littermates at room temperature with T3 (Figure 6A), resulting in both genotypes in a decrease in serum tT4 and a strong increase in serum tT3 (Figures S4A and S4B) and reduced blood glucose levels in the ipGTT (Figures 6B and 6C; ipGTT: p genotype = 0.466, p T3 < 0.0006, and p interaction = 0.892). The relative amount of lean mass was increased in T3-treated WT and UCP1 KO mice, and the relative amount of fat mass was not significantly altered (Figures S4C and S4D; lean mass: p genotype = 0.721, p T3 < 0.0008, and p interaction = 0.065; fat mass: p genotype = 0.458, p T3 = 0.565, and p interaction = 0.069). Body temperature was significantly elevated in both groups after treatment with T3 (Figure 6D; body temperature: p genotype = 0.794, p T3 < 0.0001, and p interaction = 0.213). Interestingly, we found multilocular cells in ingWAT of UCP1 KO mice after T3 treatment as a histological sign of browning, and the multilocular cells in the iBAT of these mice appeared bigger (Figure 6E). Moreover, we found an increased expression of Cidea after T3 treatment in UCP1 KO mice, and most other markers of browning were not altered in ingWAT (Figure 6F; Table S2). iBAT gene expression of WT and UCP1 KO mice showed a decrease in Adrb3,Dio2, and Ppargc1ain systemic hyperthyroidism (Figure 6G; Table S2). It was reported that ingWAT features other thermogenic mechanisms, like calcium or creatine cycling (Ikeda et al., 2017; Kazak et al., 2015, 2017); however, we only found a minor increase in Serca2 (Atp2a2) gene expression in ingWAT of T3-treated WT (F) Change in oxygen consumption of control and T3-treated mice housed at 23C after NE injection. n = 6. (G) cAMP levels in ingWAT of mice housed and treated at 23C. n = 4. (H) cAMP levels in iBAT of mice housed and treated at 23C. n = 4. (I) Protein expression of HSL and phospho HSL (pHSL) in ingWAT of mice housed and treated at 23C. n = 4. (J) Protein expression of HSL and pHSL in iBAT of mice housed and treated at 23C. n = 4. (K) Free fatty acids in ingWAT of mice housed and treated at 23C. n = 4. (L) Free fatty acids in iBAT of mice housed and treated at 23C. n = 4. Data are presented as mean ±SEM. *p < 0.05; **p < 0.01; ***p < 0.001. See also Figure S2. Cell Reports 27, 3385–3400, June 11, 2019 3391 Figure 4. Thyroid-Hormone-Induced Hyperthermia Is Independent of SNS Signaling (A and B) NE turnover in (A) ingWAT and (B) iBAT of control and T3- and T4-treated mice after 14 days of treatment normalized against tissue weight. n = 4 per time point and group. (legend continued on next page) 3392 Cell Reports 27, 3385–3400, June 11, 2019 Little, A.G., and Seebacher, F. (2014). The evolution of endothermy is explained by thyroid hormone-mediated responses to cold in early vertebrates. J. Exp. Biol. 217, 1642–1648. Lo ´pez, M., Varela, L., Va ´zquez, M.J., Rodrı ´guez-Cuenca, S., Gonza ´lez, C.R., Velagapudi, V.R., Morgan, D.A., Schoenmakers, E., Agassandian, K., Lage, R., et al. (2010). Hypothalamic AMPK and fatty acid metabolism mediate thyroid regulation of energy balance. Nat. Med. 16, 1001–1008. Lo ´pez, M., Alvarez, C.V., Nogueiras, R., and Die ´guez, C. (2013). Energy balance regulation by thyroid hormones at central level. Trends Mol. Med. 19, 418–427. Martı ´nez-Sa ´nchez, N., Moreno-Navarrete, J.M., Contreras, C., Rial-Pensado, E., Fernø, J., Nogueiras, R., Die ´guez, C., Ferna ´ndez-Real, J.M., and Lo ´pez, M. (2017a). Thyroid hormones induce browning of white fat. J. Endocrinol. 232, 351–362. Martı ´nez-Sa ´nchez, N., Seoane-Collazo, P., Contreras, C., Varela, L., Villarroya, J., Rial-Pensado, E., Buque ´, X., Aurrekoetxea, I., Delgado, T.C., Va ´zquez-Mar- tı ´nez, R., et al. (2017b). Hypothalamic AMPK-ER stress-JNK1 axis mediates the central actions of thyroid hormones on energy balance. Cell Metab. 26, 212–229.e12. Mullur, R., Liu, Y.Y., and Brent, G.A. (2014). Thyroid hormone regulation of metabolism. Physiol. Rev. 94, 355–382. Nedergaard, J., and Cannon, B. (2013). UCP1 mRNA does not produce heat. Biochim. Biophys. Acta 1831, 943–949. Neilsen, E.L. (1953). Studies on hereditary dwarfism in mice. XIV. Effect of thyroxin and growth hormone on growth. Acta Pathol. Microbiol. Scand. 32, 316–334. Olsen, J.M., Csikasz, R.I., Dehvari, N., Lu, L., Sandstro ¨m, A., O ¨berg, A.I., Nedergaard, J., Stone-Elander, S., and Bengtsson, T. (2017). b 3 -Adrenergically induced glucose uptake in brown adipose tissue is independent of UCP1 presence or activity: Mediation through the mTOR pathway. Mol. Metab. 6, 611–619. Oster, H., Damerow, S., Hut, R.A., and Eichele, G. (2006). Transcriptional profiling in the adrenal gland reveals circadian regulation of hormone biosynthesis genes and nucleosome assembly genes. J. Biol. Rhythms 21, 350–361. Petrovic, N., Walden, T.B., Shabalina, I.G., Timmons, J.A., Cannon, B., and Nedergaard, J. (2010). Chronic peroxisome proliferator-activated receptor gamma (PPARgamma) activation of epididymally derived white adipocyte cultures reveals a population of thermogenically competent, UCP1-containing adipocytes molecularly distinct from classic brown adipocytes. J. Biol. Chem. 285, 7153–7164. Puigserver, P., Herron, D., Gianotti, M., Palou, A., Cannon, B., and Nedergaard, J. (1992). Induction and degradation of the uncoupling protein thermogenin in brown adipocytes in vitro and in vivo. Evidence for a rapidly degradable pool. Biochem. J. 284, 393–398. Rabelo, R., Schifman, A., Rubio, A., Sheng, X., and Silva, J.E. (1995). Delineation of thyroid hormone-responsive sequences within a critical enhancer in the rat uncoupling protein gene. Endocrinology 136, 1003–1013. Rabelo, R., Reyes, C., Schifman, A., and Silva, J.E. (1996). Interactions among receptors, thyroid hormone response elements, and ligands in the regulation of the rat uncoupling protein gene expression by thyroid hormone. Endocrinology 137, 3478–3487. Rakov, H., Engels, K., Ho ¨nes, G.S., Strucksberg, K.H., Moeller, L.C., Ko ¨hrle, J., Zwanziger, D., and F€ uhrer, D. (2016). Sex-specific phenotypes of hyperthyroidism and hypothyroidism in mice. Biol. Sex Differ. 7,36. Ribeiro, M.O., Carvalho, S.D., Schultz, J.J., Chiellini, G., Scanlan, T.S., Bianco, A.C., and Brent, G.A. (2001). Thyroid hormone–sympathetic interaction and adaptive thermogenesis are thyroid hormone receptor isoform–specific. J. Clin. Invest. 108, 97–105. Green, E.L. (1975). Biology of the Laboratory Mouse (Dover Publications). Rubio, A., Raasmaja, A., and Silva, J.E. (1995). Thyroid hormone and norepinephrine signaling in brown adipose tissue. II: Differential effects of thyroid hormone on beta 3-adrenergic receptors in brown and white adipose tissue. Endocrinology 136, 3277–3284. Salvatore, D., Simonides, W.S., Dentice, M., Zavacki, A.M., and Larsen, P.R. (2014). Thyroid hormones and skeletal muscle–new insights and potential implications. Nat. Rev. Endocrinol. 10, 206–214. Santhanam, P., Ahima, R.S., Mammen, J.S., Giovanella, L., and Treglia, G. (2018). Brown adipose tissue (BAT) detection by 18F-FDG PET and thyroid hormone level(s)-a systematic review. Endocrine 62, 496–500. Sap, J., Mun ˜oz, A., Damm, K., Goldberg, Y., Ghysdael, J., Leutz, A., Beug, H., and Vennstro ¨m, B. (1986). The c-erb-A protein is a high-affinity receptor for thyroid hormone. Nature 324, 635–640. Shabalina, I.G., Ost, M., Petrovic, N., Vrbacky, M., Nedergaard, J., and Cannon, B. (2010). Uncoupling protein-1 is not leaky. Biochim. Biophys. Acta 1797, 773–784. Shabalina, I.G., Petrovic, N., de Jong, J.M., Kalinovich, A.V., Cannon, B., and Nedergaard, J. (2013). UCP1 in brite/beige adipose tissue mitochondria is functionally thermogenic. Cell Rep. 5, 1196–1203. Shibata, H., Pe ´russe, F., Vallerand, A., and Bukowiecki, L.J. (1989). Cold exposure reverses inhibitory effects of fasting on peripheral glucose uptake in rats. Am. J. Physiol. 257, R96–R101. Silva, J.E. (2003). The thermogenic effect of thyroid hormone and its clinical implications. Ann. Intern. Med. 139, 205–213. Silva, J.E. (2006). Thermogenic mechanisms and their hormonal regulation. Physiol. Rev. 86, 435–464. Silva, J.E., and Larsen, P.R. (1983). Adrenergic activation of triiodothyronine production in brown adipose tissue. Nature 305, 712–713. Sjo ¨gren, M., Alkemade, A., Mittag, J., Nordstro ¨m, K., Katz, A., Rozell, B., Westerblad, H., Arner, A., and Vennstro ¨m, B. (2007). Hypermetabolism in mice caused by the central action of an unliganded thyroid hormone receptor alpha1. EMBO J. 26, 4535–4545. Svensson, K.J., Long, J.Z., Jedrychowski, M.P., Cohen, P., Lo, J.C., Serag, S., Kir, S., Shinoda, K., Tartaglia, J.A., Rao, R.R., et al. (2016). A secreted Slit2 fragment regulates adipose tissue thermogenesis and metabolic function. Cell Metab. 23, 454–466. Tamucci, K.A., Namwanje, M., Fan, L., and Qiang, L. (2018). The dark side of browning. Protein Cell 9, 152–163. Tata, J.R. (1986). The search for the mechanism of hormone action. Perspect. Biol. Med. 29, S184–S204. Thompson, C.C., Weinberger, C., Lebo, R., and Evans, R.M. (1987). Identification of a novel thyroid hormone receptor expressed in the mammalian central nervous system. Science 237, 1610–1614. Vargas-Castillo, A., Fuentes-Romero, R., Rodriguez-Lopez, L.A., Torres, N., and Tovar, A.R. (2017). Understanding the biology of thermogenic fat: is browning a new approach to the treatment of obesity? Arch. Med. Res. 48, 401–413. Vujovic, M., Nordstro ¨m, K., Gauthier, K., Flamant, F., Visser, T.J., Vennstro ¨m, B., and Mittag, J. (2009). Interference of a mutant thyroid hormone receptor alpha1 with hepatic glucose metabolism. Endocrinology 150, 2940–2947. Warner, A., and Mittag, J. (2012). Thyroid hormone and the central control of homeostasis. J. Mol. Endocrinol. 49, R29–R35. Warner, A., and Mittag, J. (2016). Breaking BAT: can browning create a better white? J. Endocrinol. 228, R19–R29. Warner, A., Rahman, A., Solsjo ¨, P., Gottschling, K., Davis, B., Vennstro ¨m, B., Arner, A., and Mittag, J. (2013). Inappropriate heat dissipation ignites brown fat thermogenesis in mice with a mutant thyroid hormone receptor a1. Proc. Natl. Acad. Sci. USA 110, 16241–16246. Weiner, J., Kranz, M., Klo ¨ting, N., Kunath, A., Steinhoff, K., Rijntjes, E., Ko ¨hrle, J., Zeisig, V., Hankir, M., Gebhardt, C., et al. (2016). Thyroid hormone status defines brown adipose tissue activity and browning of white adipose tissues in mice. Sci. Rep. 6, 38124. Weiss, R.E., Murata, Y., Cua, K., Hayashi, Y., Seo, H., and Refetoff, S. (1998). Thyroid hormone action on liver, heart, and energy expenditure Cell Reports 27, 3385–3400, June 11, 2019 3399 in thyroid hormone receptor beta-deficient mice. Endocrinology 139, 4945–4952. Yen, P.M. (2001). Physiological and molecular basis of thyroid hormone action. Physiol. Rev. 81, 1097–1142. Young, P., Arch, J.R., and Ashwell, M. (1984). Brown adipose tissue in the parametrial fat pad of the mouse. FEBS Lett. 167, 10–14. Zhang, Z., Bisschop, P.H., Foppen, E., van Beeren, H.C., Kalsbeek, A., Boelen, A., and Fliers, E. (2016). A model for chronic, intrahypothalamic thyroid hormone administration in rats. J. Endocrinol. 229, 37–45. Zhao, J., Unelius, L., Bengtsson, T., Cannon, B., and Nedergaard, J. (1994). Coexisting beta-adrenoceptor subtypes: significance for thermogenic process in brown fat cells. Am. J. Physiol. 267, C969–C979. 3400 Cell Reports 27, 3385–3400, June 11, 2019 STAR+METHODS KEY RESOURCES TABLE CONTACT FOR REAGENT AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact Prof. Dr. Jens Mittag ([email protected]). REAGENT or RESOURCE SOURCE IDENTIFIER Antibodies anti-UCP1 Jastroch et al., 2012 N/A anti-Serca2 Cell Signaling Technology Cat#4388; RRID:AB_2227684 anti-HSL Cell Signaling Technology Cat#4107; RRID:AB_2296900 anti-pHSL(Ser660) Cell Signaling Technology Cat#4126S; RRID:AB_490997 anti-OxPhos cocktail Invitrogen Cat#45-8099; RRID:AB_2533835 anti-Tyrosine Hydroxylase Sigma Cat#C2928 anti-b-actin Sigma Cat#A1978; RRID:AB_476692 anti-HSP90 Cell Signaling Technology Cat#4877; RRID:AB_2233307 anti-pAMPKa(Thr172) Cell Signaling Technology Cat#2535; RRID:AB_331250 anti-b-actin(AC-74) Sigma Cat#A5316; RRID:AB_476743 anti-rabbit polyclonal HRP-conjugated antibody DAKO, Denmark Cat#P0448; RRID:AB_2617138 anti-mouse polyclonal HRP-conjugated antibody DAKO, Denmark Cat#P0447; RRID:AB_2617137 Chemicals, Peptides, and Recombinant Proteins L-Thyroxine (T4) Sigma Aldrich, Germany Cat#T2376 3,30,5-Triiodo-L-thyronine (T3) Sigma Aldrich, Germany Cat#T6397 Arterenol (Norepinephrine) Sanofi Cat#03870227 a-methyl-DL-tyrosine methyl ester hydrochloride (AMPT) Sigma Aldrich, Germany Cat#M3281 Critical Commercial Assays Pierce BCA Kit Thermo Scientific, Germany Cat#23227 TGX Stain Free FastCast Acrylamide Kit Bio-Rad Laboratories, Germany Cat#1610183, Cat#1610185 Advansta WesternBright Quantum Advansta, USA Cat#K-12041-D20 Serum total T4 DRG Diagnostics, Germany Cat#EIA-1781 Serum total T3 NovaTec Immundiagnostica GmbH, Germany Cat#DNOV053 Amersham cAMP Biotrak EIA GE Healthcare Life Sciences Cat#RPN225 Free Fatty Acid Quantification Kit abcam plc, UK Cat#ab65341 Experimental Models: Organisms/Strains C57BL/6NCr Charles River Laboratories, Germany RRID:MGI:2160593 B6.129S1-Thrbtm1Df/J (TRbko) Forrest et al., 1996 RRID:IMSR_JAX:003462 B6.129-UCP1tmKZ/J (UCP1 ko) Enerba ¨ck et al., 1997 N/A Per 1/2 dko Husse et al., 2012 N/A Oligonucleotides Primer sequences see Table S3 N/A Software and Algorithms Phenomaster software TSE Systems, Germany N/A Minispec Plus Software 6.0 Bruker Corp., Billerica, MA, USA N/A Microsoft Office Excel 2013 Microsoft N/A Prism 7 GraphPad N/A CircWave v.1.4 Oster et al., 2006 N/A Cell Reports 27, 3385–3400.e1–e3, June 11, 2019 e1 EXPERIMENTAL MODEL AND SUBJECT DETAILS Unless stated otherwise, animals were housed in groups at 23 ±1C at constant 12 hour light/dark cycle with ad libitum access to food and water. Experiments were conducted in male mice at the age of three to six months. Wild-type C57/BL6NCr (RRID:MGI:2160593) were purchased from Charles River Laboratories (Charles River, Germany), TRbko mice (RRID:IMSR_JAX:003462), UCP1 ko mice and Per1/Per2 dko mice were generated as described before (Enerba ¨ck et al., 1997; Forrest et al., 1996; Husse et al., 2012) and knockout mice were always compared to WT littermate controls. All animal procedures were approved by the MELUR Schleswig-Holstein, LANUV Nordrhein-Westfalen and the BGV Hamburg, Germany. METHOD DETAILS Study design Hyperthyroidism was induced by treatment with THs in drinking water as recommended by the American Thyroid Association guidelines with a dose low enough to avoid cachexia (Bianco et al., 2014), i.e., 1 mg/L L-Thyroxine (T2376, Sigma Aldrich, Germany) in 0.01% BSA (A7906, Sigma Aldrich, Germany) or 0.5 mg/L 3,30,5-Triiodo-L-thyronine (T6397, Sigma Aldrich, Germany) in 0.01% BSA for 14 days (Figure 1A). Control mice received 0.01% BSA. Basic metabolic profiling was performed by measuring body weight, food and water intake, measurement of core body temperature using a rectal probe (BAT-12, Physitemp, USA), infrared thermography (T335, FLIR, Sweden) (Warner et al., 2013), and glucose tolerance tests (ipGTT, 2 g/kg body weight). Body composition was measured using Minispec LF110 and Minispec Plus Software 6.0 (Bruker Corp., Billerica, MA, USA). Organ collection was always performed in the middle of the light phase, unless stated otherwise. PET-CT [ 18 F]-FDG PET imaging and kinetic modeling was performed as described before (Jais et al., 2016). For the analysis of the parametric images of the metabolic rate of glucose (MRglc) volumes of interests (VOIs) containing iBAT or ingWAT, respectively were defined for each individual animal and each measurement. TRL Uptake Uptake of triglyceride-rich lipoproteins and [ 3 H]2-Deoxyglucose ( 3 HDG) in hyperthyroid C57/BL6NCr mice was investigated as described before (Bartelt et al., 2011). Briefly, mice were fasted for 4 hr before receiving an intravenous injection of 100 mL radiolabelled recombinant TRLs (80 mg triglycerides/kg) labeled with 14 C-triolein (0.6 MBq/kg) and 3 HDG. Organs from anesthetized mice were harvested after systemic perfusion with PBS-heparin (10 U/ml) via the left heart ventricle. Tissues were homogenized using Solvable (Perkin Elmer) and radioactivity was determined using liquid scintillation counting. Indirect Calorimetry and nonshivering thermogenic capacity Measurements of daily energy expenditure and basal metabolic rate in WT and TRbko littermates was performed in single-housed animals using Phenomaster (TSE Systems, Germany). Oxygen consumption (VO 2 ), carbon dioxide production (VCO 2 ), respiratory exchange rate (RQ), and energy expenditure (EE) were calculated with Microsoft Office Excel and Phenomaster software (TSE Systems, Germany). Basal metabolic rate was measured for 1 hour in animals fasted for 6 hours at 30C. Energy expenditure, respiratory exchange rate and nonshivering thermogenic capacity in fasted control and T3-treated mice were performed using the CaloBox system (PhenoSys GmbH, Germany). For measurement of daily energy expenditure at 23C and resting metabolic rate at 30C, mice were acclimated to the chambers individually for at least 60 minutes, before basal O 2 consumption and CO 2 production was analyzed for up to 60 minutes in 20 s intervals. For analysis of nonshivering thermogenesis capacity in mice housed and treated at 10Cor23 C, respectively, mice were acclimated to chambers for at least 60 minutes. Then basal energy expenditure at 23C was measured for 30 minutes, before injecting mice subcutaneously with 1 mg/kg NE (Arterenol, Sanofi) and measuring O 2 consumption and CO 2 production in intervals of 20 s for at least 60 minutes after injection. The change in oxygen consumption (DVO 2 ) was calculated by subtracting basal O 2 consumption before NE injection from O 2 consumption after NE injection. NE turnover In order to determine NE turnover in hyperthyroid WT mice, n = 16 mice per group were treated for two weeks with T3 or T4, respectively. On the last day of treatment mice were injected with the tyrosine hydroxylase inhibitor a-methyl-DL-tyrosine methyl ester hydrochloride (i.p. 120 mg/kg) (AMPT, M3281, Sigma Aldrich, Germany) and sacrificed after 0, 2, 4 and 6 hours (n = 4 per group and time point), respectively. Catecholamines were extracted in 2% perchloric acid (PCA, 48%–50%, 44464, Alfa Aesar, MA, USA) and quantified as described before (Dalley et al., 2002). Protein content of the samples was quantified using DC Protein Assay Kit according to manufacturer’s protocol (Bio-Rad Laboratories, Inc.). Gene expression analysis For gene expression analysis, RNA was isolated using QIAGEN RNeasy Kits (QIAGEN, Germany), transcribed into cDNA (Molecular Biology RevertAid Strand cDNA Kit, Thermo Fisher Scientific, Germany) following manufacturer’s instructions. qPCR analysis was e2 Cell Reports 27, 3385–3400.e1–e3, June 11, 2019 performed using SYBR Green PCR Master Mix (Roche, Germany) and QuantStudio Applied Biosystems (Thermo Fisher Scientific, Germany). Efficiency of the PCR was calculated using standard curves and levels of gene expression were normalized to a housekeeping gene (ribosomal protein lateral stalk subunit P0,Rplp0 and peptidylprolyl isomerase a,Ppia for adipose tissues; Ppia and hypoxanthine phosphoribosyltransferase,Hprt for M. Soleus and M.Gastrocnemius, and Hprt for pituitary and hypothalamus) using the DDC T method. Primer sequences are listed in Table S3. Western Blot Protein isolation from snap-frozen tissue was performed by homogenizing tissues in RIPA buffer (150 mM NaCl, 50 mM Tris-HCl pH 7.5, 0.1% wt/vol SDS, 0.5% wt/vol sodium deoxycholate, 1% vol/vol Nonidet P40, 1 mM EDTA, 1 mM EGTA, 2.5 mM sodium pyrophosphate, 1 mM NaVO 4 and 10 mM NaF), with freshly added Protease inhibitors (5892970001, Roche Diagnostics GmbH, Germany). Protein concentration was determined using Pierce BCA Kit (23227, Thermo Scientific, Germany). Electrophoresis was performed using SDS Gels (TGX Stain Free FastCast Acrylamide Kit (1610183, 1610185, Bio-Rad Laboratories, Germany). Proteins were then transferred onto PVDF membrane (IPVH00010, Merck Millipore, Ltd, Cork, Ireland), blocked for 1 hour in 5% milk in TBS-T and incubated with primary antibodies (anti-UCP1; Jastroch et al., 2012); anti-Serca2, 4388, Cell Signaling Technology, Inc.; anti-HSL, 4107, Cell Signaling Technology, Inc.; anti-pHSL(Ser660), 4126S, Cell Signaling Technology, Inc.; anti-OxPhos cocktail, 45-8099, Invitrogen; anti-Tyrosine Hydroxylase, C2928, Sigma; anti-b-actin, A1978, Sigma; anti-HSP90, 4877, Cell Signaling Technology, Inc.) for 16 hours at 4C. Afterward membranes were washed 4x10 minutes, incubated with secondary antibodies (anti-rabbit polyclonal HRP-conjugated antibody, P0448, DAKO, Denmark; anti-mouse polyclonal HRP-conjugated antibody, P0447, DAKO, Denmark) for 1 hour at room temperature, and washed again for 4x10 minutes. Chemiluminescence was recorded using Advansta WesternBright Quantum (K-12041-D20, Advansta, USA) and ChemiDoc Touch Imaging System (Bio-Rad Laboratories, Germany). Quantification of band intensities was performed using ImageLab TM Software (Bio-Rad Laboratories, Germany). Protein expression analysis in the VMH (anti-pAMPKa(Thr172), 2535, Cell Signaling Technology, Inc.; anti-b-actin(AC-74), A5316, Sigma) was performed as described before (Alvarez-Crespo et al., 2016). Protein expression was normalized to total protein content transferred onto the membrane, or housekeeping proteins, respectively. Enzyme activity and glycogen measurement Glycogen content in liver and M. Gastrocnemius, as well as activities of Pyruvate Kinase and hepatic Phosphoenolpyruvate Carboxykinase of snap-frozen tissue were determined as described before (Vujovic et al., 2009). ELISA Serum levels of tT4 (EIA-1781, DRG Diagnostics, Germany) and tT3 (DNOV053, NovaTec Immundiagnostica GmbH, Germany) were determined according to manufacturer’s instructions. Hypothalamic T3 content was measured in pooled hypothalamic punches of the lateral and basal hypothalamus as described previously (Zhang et al., 2016). cAMP levels in ingWAT and iBAT were determined according to manufacturer’s instructions (RPN225, GE Healthcare, UK). Free fatty acids in ingWAT and iBAT samples were analyzed according to manufacturer’s instructions (ab65341, abcam plc; UK). Histology Tissues were fixed in 4% paraformaldehyde for 48 hours, dehydrated by washing in increasing concentrations of ethanol and xylol before embedding into paraffin. Tissues were cut in 5 mm slides and stained with Hemalun and Eosin according to the manufacturer’s protocol (X883.2, T865.3, Carl Roth GmbH&Co KG, Germany). QUANTIFICATION AND STATISTICAL ANALYSIS For analysis of all data Microsoft Office Excel and GraphPad Prism 7 software was used. Analysis of circadian rhythms was performed using CircWave v.1.4 software (Oster et al., 2006). The variances between groups investigated were similar and appropriate tests were performed to analyze differences between the respective groups. For experiments with repeated-measurements a RM ANOVA was performed with individual post-tests, the experiments of the UCP1 ko mice and T3 treatment were analyzed using a 2-WAY ANOVA with post-tests. All values are represented as mean ±SEM. Number of animals (n) per experiment are depicted in each figure. Cell Reports 27, 3385–3400.e1–e3, June 11, 2019 e3