Overexpression of Adiponectin Receptor 1 Inhibits Brown and Beige Adipose Tissue Activity in Mice

After cold induction, distinct white and brown adipose depots including subcutaneous inguinal white adipose tissues (iWAT), gonadal/epididymal white adipose tissues (gWAT), interscapular brown adipose tissues (BAT), and suprascapular white adipose tissues (supWAT) were collected. Transgenic AdipoR1 male mice had lower gWAT weights and greater iBAT weights compared with tissues in WT male mice. Transgenic AdipoR1 female mice had lower iWAT and gWAT weights compared with tissues in WT female mice (Figure 1B). These results showed different fat distribution under cold induction in WT or AdipoR1 mice in both sexes.

Regarding food intake and body weight changes, overall, during the whole period of cold-induction, AdipoR1 male or female mice had greater accumulated food intake compared with WT mice, especially during periods of D0 to D4 and D0 to D7 (g per mouse per d / body weights) (Figure 1C). However, examination of body weight changes under cold environments indicated that both AdipoR1 male and female mice showed significant enormous body weight losses compared to WT mice, especially during periods of D0 to D4 and D0 to D7. WT female mice even gained body weights under cold exposure in D0-D7 (Figure 1D). These results showed that although transgenic AdipoR1 mice had increased food intake per unit body weight, AdipoR1 mice still had negative body weight changes during cold stimulation, indicating signs of cold-intolerance symptoms.

To record live glucose uptake in vivo for BAT activity during cold stimulation in BAT and browned WAT, a PET/CT scanning with 2-deoxy-2-[¹⁸F]fluoro-D-glucose (¹⁸F-FDG) injection to the WT and transgenic AdipoR1 mice was applied after cold-stimulation. AdipoR1 mice showed significant reduction of cold-induced radio-labeled glucose uptake in vivo in classical iBAT, browned supWAT and browned iWAT compared to WT mice (Figure 2). Constructed 3D videos showed panoramic views of inhibited cold-induced glucose uptake in AdipoR1 mice (Supplementary Materials). Both male and female AdipoR1 mice showed strong diminution, especially pronounced in AdipoR1 female mice. These results demonstrated AdipoR1 mice clearly exhibit decreased cold-stimulated glucose uptake in BAT, browned supWAT, and browned iWAT.

To confirm the inhibited cold-induced thermogenesis by PET/CT scanning in AdipoR1 mice, the structure of BAT and browned iWAT (the primary WAT browning depot) were examined by hematoxylin and eosin (H&E) staining. Higher BAT activity was revealed in WT female mice versus WT male mice by PET/CT scanning (Figure 2). Therefore, we focus on comparing WT female mice with AdipoR1 female mice. AdipoR1 female mice both at room temperature (RT, Figure 3A) and under cold exposure (Figure 3B) showed increased adipocyte size with larger lipid droplets both in BAT and browned iWAT than WT female mice. These results indicated AdipoR1 mice acquire lower browning/beiging effects and maintain relative “whitening” phenotypes in classical BAT and browned iWAT than tissues in WT mice both at RT and under cold environments.

To confirm the diminished cold-induced thermogenesis by PET/CT scanning in AdipoR1 mice, the surface brown adipose depot temperature and surface body temperature was examined by infrared thermometer and camera. AdipoR1 female mice under cold stimulation showed reduced surface BAT temperature and surface body temperature both before (RT) and after cold exposure (especially, 6 h) than WT mice (Figure 4A). Thermographic imaging also showed temperature distribution of surface body temperature in WT female mice was warmer than in AdipoR1 female mice (Figure 4B). These results, in line with PET/CT scanning and H&E staining, indicated AdipoR1 mice have reduced BAT and browned WAT activity.

To evaluate expression of thermogenesis related genes in BAT and WAT beiging/browning for energy expenditure, multiple adipose tissues were assessed. White adipose tissues in certain depots possess the ability to be browned, termed beige adipocytes, according to the guide of an adipose tissue atlas [25]. Therefore, iBAT, supWAT, iWAT (two potential WAT browning depots), and gWAT (minimal potential for browning) were sampled.

First, to uncover the role of adiponectin receptors and browning (beiging) effects between different sexes in cold environments, adiponectin, adiponectin receptors (AdipoR1, AdipoR2 and T-cadhrin), and brown adipocyte markers in BAT and in browned WAT were determined in WT male and female mice. Surprisingly, Tcad (mainly in hearts) was also detectable with Ct values of ~25 by qPCR detection. However, expression of Adipor1 and Adipor2 was still dominant in BAT, iWAT, supWAT, and gWAT under cold exposure compared to adipose Tcad expression (Figure 5A). In cold environments, WT female mice had greater Adipor1 expression in gWAT, increased Adipor2 expression in iWAT and in gWAT, enhanced Tcad expression in four adipose depots (BAT, iWAT, supWAT, and gWAT) and higher Adipoq expression in BAT and in gWAT compared with tissues of WT male mice (Figure 5B). Furthermore, WT female mice showed increased brown adipocyte markers, thermogenic Ucp1 in BAT and in gWAT and Prdm16, Cidea, Cox7a, and Cox8b in gWAT (Figure 5C). These data indicated that WT female mice possess higher expression of adiponectin receptors and increased brown adipocyte markers than WT male mice in cold environments.

Given that WT female mice possess enhanced cold-induced glucose uptake in vivo by PET/CT scanning, and increased expression of adiponectin receptors and brown adipocyte markers than WT male mice in cold environments, we focused on comparing WT female mice with AdipoR1 female mice under cold exposure. Compared with WT female mice, AdipoR1 female mice showed increased expression of Adipor1 in BAT, iWAT, and supWAT which confirmed that these transgenic mice were over-expressing adiponectin receptor 1 (Figure 6). Expression of Adipor2 was no different in WT and AdipoR1 female mice. However, AdipoR1 female mice in cold environments showed reduced expression of Tcad in iWAT and gWAT and decreased Adipoq in gWAT (Figure 6). These results indicate these AdipoR1 female mice overexpressed adiponectin receptor 1 and somewhat had compensatory reduced expression of T-cadherin in certain adipose tissue depots.

Compared with male mice, BAT activity by PET/CT scanning, expression of adiponectin receptors and expression of brown-adipocyte selective makers in BAT or in browned WAT were more pronounced in female mice. Therefore, we focused on classical BAT and subcutaneous iWAT in female mice for comparison of brown adipocyte markers in WT versus AdipoR1 mice. Brown adipocytes express highly distinct markers including thermogenic Ucp1, Prdm16, and Cidea and certain specific mitochondrial components Cox7a and Cox8b compared with white adipocytes. At RT (mild cold-induction), WT female mice already had greater expression of brown adipocyte markers, such as Ucp1 in BAT, iWAT, and gWAT, than AdipoR1 female mice with other brown-adipocyte marker genes in similar patterns (Figure 7A). After prolonged exposure to cold to activate BAT and browned WAT (beiging), expression of thermogenic Ucp1 in BAT, iWAT, and gWAT from AdipoR1 female mice was much lower compared to WT female mice. BAT-enriched mitochondrial marker Cox7a also had decreased expression in BAT, iWAT, and gWAT of AdipoR1 mice than in tissues of WT mice (Figure 7B). Other brown adipocyte markers Prdm16 and Cidea and BAT-mitochondrial marker Cox8b in AdipoR1 female mice showed a resembling curtailed fashion. Therefore, these results showed that, both at RT and in response to prolonged cold-exposure, AdipoR1 female mice had reduced expression of thermogenic Ucp1, general brown adipocyte markers and BAT-enriched mitochondria markers in BAT and in browned WAT.

To confirm inhibited cold-induced glucose uptake in vivo and reduced brown adipocyte markers in BAT and in browned iWAT of AdipoR1 female mice, gene regulation related to glucose uptake, to lipolysis, and to fatty acid transport, oxidation, and synthesis was determined. AdipoR1 female mice compared to WT female mice, upon cold exposure, had reduced expression of Glut4 in BAT (related to glucose uptake), Hsl in BAT, Atgl in BAT and in iWAT (both related to lipolysis), Cpt1a in BAT, Ppara in BAT and in iWAT (both related to fatty acid oxidation), and Srebpf1 in BAT and in iWAT (related to fatty acid synthesis) (Figure 8). However, expression of Fabp4, Cd36, Lpl, Acox1, Acc1, Acc2, or Fasn was not different between WT and AdipoR1 female mice (Figure 8). These results indicate AdipoR1 mice had decreased gene expression related to glucose uptake, lipolysis, and fatty acid oxidation that is congruent with in vivo results of lowered BAT activity of cold-induced glucose uptake and diminished brown adipocyte markers in AdipoR1 mice.

Inhibited cold-induced glucose uptake in vivo, reduced brown adipocyte markers and enlarged adipocyte size revealed by H&E staining, in BAT and in browned WAT of AdipoR1 mice, all implicate diminished BAT function in BAT and browned WAT. To confirm this phenotype, additional gene expression [26,27] related to adipose browning (highly expressed in BAT vs WAT) and adipose whitening (highly expressed in WAT vs BAT) was selected and determined. AdipoR1 mice showed reduced expression of browning genes, Cpt1b, Dio2, Elovl3, and Pgc1a in BAT and in iWAT and increased whitening genes Hmgn3 in iWAT than WT mice (Figure 9). These results indicate AdipoR1 mice maintain a relative adipose whitening phenotype in BAT and browned iWAT even after prolonged cold exposure.

Subcutaneous inguinal WAT is the largest white adipose depot in mice for isolating pre-adipocytes and also meets with retaining great potentials for differentiating into either traditional white or beige (brown-like) adipocytes, especially under PPARγ-agonist stimulation during differentiation process [9]. To further confirm the in vivo results of abated cold-induced glucose uptake and inhibited expression of brown adipocyte markers in BAT and browned WAT, we isolated pre-adipocytes from iWAT and differentiated them into beige adipocytes with or without synergistic browning agents of PPARγ-agonists, rosiglitazone. Rosiglitazone clearly augments beige (browning of white) adipocytes with increased gene expression of brown adipocyte markers and brown-adipocyte mitochondrial markers (Figure 10). However, mature beige adipocytes differentiated with rosiglitazone from AdipoR1 mice showed reduced expression of brown adipocyte markers, thermogenic Ucp1, and Cidea compared with rosiglitazone-supplemented beige adipocytes from WT mice (Figure 10A). Brown-adipocyte enriched mitochondrial makers, Cox7a and Cox8b, showed similar effects of diminution in beige adipocytes from AdipoR1 compared to WT mice (Figure 10B). Therefore, these results showed both in vivo and in vitro results of reduced adipose expression of brown adipocyte markers in AdipoR1 mice.

PET/CT scintigraphy was first widely applied in clinical cancer research because cancer cells favor large amounts of glucose consumption. Therefore, radio-labeled glucose, ¹⁸F-FDG helps physicians spot glucose-consuming cancer cells in a tumor area [28,29,30,31]. Later, PET/CT scanning was adopted by endocrinologists searching for potential BAT in adult humans [3,4,5].

Nowadays, ¹⁸F-FDG PET/CT scanning is also used in preclinical mouse models for evaluating BAT activity or for brain research such as Alzheimer’s disease. For example, a recent research for adipose depot classification in the mouse adipose-tissue atlas used ¹⁸F-FDG or [^123/125I]-β-methyl-p-iodophenyl-pentadecanoic acid coupled with PET/CT or SPECT/CT to image live cold-induced nutrient uptake in adipose tissues in vivo [25]. Depots of anatomically distinct brown, white, and beige (WAT browning) adipose tissues in mice are mapped using this method.

Based on the above mentioned mouse brown and white adipose-tissue atlas, in the current study, we choose (1) the interscapular region of brown adipose tissues (representing classical BAT), (2) the suprascapular and (3) the subcutaneous inguinal white adipose depots (both representing potential WAT browning sites), and (4) the gonadal region of white adipose tissues (representing dull WAT browning site) to evaluate cold-induced glucose uptake in BAT and in browning of WAT. By PET/CT scanning, our results showed inhibited cold-induced glucose uptake in BAT and browned WAT of AdipoR1 mice in both sexes.

Notably, this study also unveils that WT female mice possess greater amounts of cold-induced glucose uptake in vivo in BAT than WT male mice. Cold-acclimated WT female mice also had higher expression of thermogenic UCP1 in BAT and gWAT than WT male mice. These results are consistent with another study indicating that the β3-adrenergic receptor agonist CL316,243 induces enhanced brown adipocyte markers and mitochondrial respiratory chain proteins in gWAT of female mice than in gWAT of male mice, but without effects in iWAT [32]. Furthermore, estradiol, a major female sex hormone known to increase BAT activity, known to activate hypothalamic AMP-activated protein kinase (AMPK)-related [33] and augment bone morphogenetic protein 8b (BMP8b)-related thermogenic signaling [34], may account for the sex differences in BAT activity and consequently estradiol could elevate BAT activity in female mice.

Compared with WT male mice, cold-acclimated WT female mice also had elevated expression of adiponectin in BAT and gWAT, of AdipoR1 in gWAT, of AdipoR2 in iWAT and gWAT, and of T-cadherin in BAT, iWAT, supWAT, and gWAT. Female mice also had enhanced browning processes of cold-induced glucose uptake in BAT and thermogenic UCP1 in BAT and gWAT. Thus, our current study focused on female mice. Moreover, female mice with greater expression of adiponectin receptors and thermogenic UCP1 in adipose may be suitable for evaluating maximal effects of adiponectin receptor signaling on the cold-induced browning process in classical brown or browned white adipose tissues.

Without reducing food intake for cutting excessive supply of calories and impacting de novo lipogenesis, an alternative way to combat obesity-induced diabetes is increasing energy expenditure, potentially through activating brown adipose or browning/beiging of white adipose depots [35]. The thermogenic ability of BAT not only oxidizes fats but also consumes significant amounts of glucose as a metabolic “glucose sink”, especially activated by cold stimulation. For example, elevating BAT activity by cold exposure consumes significant amounts of glucose, which can be assessed by radio-labeled glucose coupled with PET/CT scanning, and also dramatically promotes clearance of plasma triglycerides and uptake into BAT through membrane receptor CD36 [36]. Therefore, activating BAT or WAT browning could be a viable way to defend against diet-induced obesity or type 2 diabetes.

In the clinically human translational application, β₃-adrenergic receptor agonists, cold or other agents robustly activate human BAT and increase thermogenesis and energy expenditure both in healthy or obese human subjects. This cold stimulation in human studies can be achieved by staying in a cold-room or by putting on a running cooling water-filled suit with set temperature [37,38,39,40,41,42]. Therefore, application of PET/CT scanning in these human clinical studies or in above mentioned preclinical mouse models clearly showed that human and mouse BAT can be activated by cold-stimulation, thereby resulting in elevated energy expenditure and contributing to body fat reduction.

This current study also established a cold-induction procedure for mice to activate BAT or browned WAT assessed by PET/CT scanning. As opposed to a study of mice staying at 4 °C for C57BL/6 mice [43], we found it is much safer to perform cold-induction around 10 °C in a cold-room for our WT or transgenic AdipoR1 mice with FVB/NJNarl background. Therefore, strains of mouse species and transgene effects may account for different temperatures and periods of varying lengths for cold-induced thermogenesis.

Adiponectin is a beneficial adipokine to resist diabetes and causes a “healthy obese” state by increasing white adipose expansion and by limiting lipotoxicity in other tissues during high-fat feeding. For example, a classic mouse model over-expressing adiponectin in the prone-obese ob/ob background generates a super-obese phenotype with twice the size of the controls of already obese ob/ob mice. Nevertheless, these super-obese mice over-expressing adiponectin are protected from glucose intolerance and insulin resistance [44,45]. Therefore, these studies indicated that adiponectin reduces energy expenditure and increases energy preservation in white adipose tissues. Other studies also support that adiponectin knockout mice showed a lean phenotype or fat reduction [46,47,48].

However, exogenous administration of adiponectin to mice decreases plasma glucose, free fatty acids and triglycerides, increases muscle fatty acid oxidation and induces weight loss [49]. One of the functions of adiponectin is carried out through the activation of AMPK [46,50]. Therefore, short-term induction versus long-term supply of adiponectin and exogenous administration versus gene overexpression/deletion may cause discrepancy and result in inconsistent results of fat reduction or expansion.

In the current study, cold-acclimated AdipoR1 mice exhibited fat reduction of gWAT weights and showed whitening phenotype of iBAT assessed by H&E staining and by expression levels. AdipoR1 mice coupled with these fat alterations cannot resist acute cold exposure and flaunted lowered surface body and surface BAT depot temperature, accompanied with significant bodyweight losses. Therefore, these cold-acclimated AdipoR1 mice indicate implications of cold intolerance.

Non-shivering thermogenesis primarily occurs in brown adipose tissues. BAT is enriched with abundant mitochondria identified with many brown-specific mitochondrial makers, such as Cox7a and Cox8b [51]. Thermogenic UCP1 in mitochondria is responsible for uncoupled respiration from ATP synthesis and generation of heat to maintain core body temperature, especially during cold-induced hypothermia and in diet-induced thermogenesis [52].

In this study, we found that these AdipoR1 female mice, after prolonged cold exposure, showed not only inhibited cold-induced BAT activity of glucose uptake assessed by PET/CT but also concomitant strong reduction of brown adipocyte markers, including thermogenic gene Ucp1, brown adipocyte markers Prdm16, Cidea, Dio2, Cpt1b, Elovl3, and Pgc1a, and BAT-specific mitochondrial markers Cox7a and Cox8b in classical BAT or browned iWAT. Furthermore, AdipoR1 female mice at RT already showed this diminished pattern which is even more pronounced in cold environments. In vitro gene expression analysis of brown adipocyte markers in differentiated beige adipocytes from iWAT of AdipoR1 mice also support in vivo results of PET/CT scanning and expression of brown adipocyte markers.

Cold induction is known to decrease serum adiponectin concentrations through sympathetic activation in mice [53], especially evidenced in atherogenic conditions [54]. However, studies related to adiponectin receptors on cold-induced thermogenesis are limited. Our results of adiponectin receptor 1 on inhibition of thermogenesis are consistent with studies revealing that administering adiponectin also suppresses cold-induced thermogenesis and thermogenic UCP1 expression in mouse BAT [55]. Although the same study also showed that ablating adiponectin in mice elevates while abating AdipoR1 or AdipoR2 in mice decreases cold-induce thermogenesis in vivo [55]. These results are confounding and inconclusive. One possible speculation is that reduced thermogenesis by adiponectin may be simultaneously through multiple adiponectin receptors or adiponectin signaling. Individually ablating each adiponectin receptor (AdipoR1, AdipoR2, or T-cadherin) separately may actually enhance adiponectin signaling by causing compensatory increases in other adiponectin receptors. Therefore, both over-expressing (in our mouse model) or ablating individual adiponectin receptor could all augment adiponectin signaling, thereby diminishing cold-induced thermogenesis. However, detailed mechanisms await further investigation.

Cold-induced thermogenesis requires fatty acids as energy fuels, largely attributed to WAT lipolysis. In response to cold exposure under hypothermia, the sympathetic nervous system sends signals to activate lipolysis and thermogenesis. Circulating liberated fatty acids from WAT lipolysis are taken up by BAT for mitochondrial oxidation and thermogenesis [56]. Mechanistically, adiponectin suppresses basal and catecholamine-induced lipolysis in white adipocytes mediated by inhibited protein-kinase-A signaling [57]. Therefore, adiponectin decreases lipolysis in WAT which could result in decreased fuel sources for other tissues including BAT.

Recently, a new study also showed adiponectin or receptor agonist AdipoRon suppresses cold-induced thermogenic gene expression and energy expenditure in vivo through limiting immune-cell ILC2s-activated thermogenesis in BAT or beige (browned WAT) adipose tissues [58]. Therefore, this current study showed that reduction of cold-induced thermogenesis in mice over-expressing AdipoR1 is congruent with other studies of inhibited thermogenesis in mice administered adiponectin and of increased thermogenesis in mice with ablated adiponectin. Therefore, the combination of above-mentioned results of inhibited BAT thermogenesis by adiponectin or receptor overexpression might be partly through fatty acid retention in WAT with diminished lipolysis in WAT, resulting in decreased supply of fatty acids as thermogenic substrates. Moreover, we also found that over-expressing AdipoR1 suppresses WAT browning both in vivo and in vitro. Hence, in our mouse model, AdipoR1 inhibits cold-induced thermogenesis in both classical BAT and browned WAT.

To conclude, the live ¹⁸F-FDG PET/CT scanning for measurements of cold-induced BAT activity in mice showed that mice over-expressing adiponectin receptor 1 exhibit reduced cold-induced glucose uptake in BAT and browned supWAT, accompanied with decreased surface body and surface BAT depot temperature and increased adipocyte size of whitening morphology in BAT and browned iWAT. Further in vitro and in vivo analyses indicated expression of thermogenic gene Ucp1, brown fat-specific markers and brown adipocyte-enriched mitochondrial markers are largely attenuated in BAT or browned WAT of AdipoR1 mice. However, white adipocyte markers in BAT or browned WAT are maintained or elevated in AdipoR1 mice. These combined results with live PET/CT scanning reveal the role of adiponectin receptor 1 in regulating energy homeostasis, especially in cold-induced thermogenesis. Unveil detailed mechanisms underlying adiponectin signaling on brown adipose tissues or browning/beiging of white adipose tissues may provide insights for therapeutic application in obesity-associated metabolic diseases.

Transgenic FVB/NJNarl mice over-expressing porcine adiponectin receptor 1 were generated through the pronuclear microinjection of fertilized eggs, and bred as previously reported [23,24]. Briefly, founder mice were crossed with WT mice to generate F1 heterozygous (AdipoR1^+/−) offspring. The F1 heterozygous mice were backcrossed with WT mice to generate the offsprings as F2 mice. F2 heterozygous mice were double-crossed to generate progeny of homozygous (AdipoR1^+/+) littermates. These F3 littermates were checked as homozygous mice via quantitative real-time PCR analyses using DsRed2 primers (Supplementary Table S1) and GFP/RFP-fluorescent checking systems. All experiments were performed on homozygous offsprings starting from the F4 or later generations housing at the mouse facility in our department.

Adult age-paired male and female mice (>12 weeks old) of WT or with the transgene, from the F4 or later generations, were randomly assigned for each experimental group (n ≥ 3 for each sex per group), maintained at room temperature or at a cold-room with a light-dark cycle of 12: 12 h, and provided with feed (LabDiet 5001; LabDiet, St Louis, MO, USA) and water ad libitum. The animal experiments were approved by the Institutional Animal Care and Use Committee at National Taiwan University, Taipei and Institute of Nuclear Energy Research, Taoyuan, Taiwan (NTU-103-EL-42, NTU-106-EL-00088, NTU-108-EL-00076).

First, a pre-test of cold-induction protocol for our mouse strain was implemented to evaluate cold-stimulated thermogenesis in mice. Because different genetic inbred or outbred backgrounds of mice may respond very differently regarding cold tolerance at discrete low temperature and exposure times, a pre-trial with small numbers of WT and transgenic AdipoR1 mice (n = 3) in the FVB/NJNarl genetic background were housed in a cold-room at 4, 7, or 10 °C. Below 10 °C, WT or transgenic mice cannot tolerate the cold for more than 7 d. This was especially prevalent at 4 °C (data not shown). Therefore, in the following experiments, cold-induced thermogenesis in WT or AdipoR1 of FVB/NJNarl mice will be evaluated at about 10 °C.

Adult WT or AdipoR1 transgenic mice were divided and maintained on a regular chow diet. For cold exposure experiments after the pre-test, mice were transferred into a cold-room (8–10 °C) for two weeks and provided with feed and water ad libitum, with the same light-dark cycle of 12 h–12 h. Mice were closely monitored in cold environments and checked for their health and behavior at least once per day to prevent widespread sudden death from cold intolerance. Early termination will be initiated if extensive sudden death is observed.

Food intake and body weights of each group were measured on d 0, 4, 7 and 11. For tissue sampling, mice were sacrificed using carbon dioxide. Peripheral adipose tissues, including subcutaneous inguinal, gonadal, and suprascapular white adipose tissues, and interscapular brown adipose tissues, were harvested, weights recorded, snap-frozen in liquid nitrogen, and stored at −70 °C for future analysis.

After two weeks of cold acclimation with at least 6 h of terminal fasting, mice were escorted to the Institute of Nuclear Energy Research, Taoyuan, Taiwan for the ¹⁸F-FDG PET/CT Scanning (Mediso NanoPET/CT, Mediso Medical Imaging Systems, Budapest, Hungary). Before radiotracer injection, mice were anesthetized with 2% isoflurane injected through a tail vein. Then, 150 μCi of locally synthesized ¹⁸F-FDG was injected into the tail vein of the mice and mice were held in cages for 1 h waiting for biodistribution before scanning. Dynamic emission scans for each WT or AdipoR1 mouse were acquired for 20 min. The mice were constantly anesthetized with 0.1% isoflurane in air during the imaging. Results of ¹⁸F-FDG PET/CT images were coregistrated (constructed) using a PMOD software (PMOD Technologies LLC, Zurich, Switzerland) to estimate the radioactivity concentration. Volumes of interest and the activity were expressed as %ID/g.

To evaluate the structure of interscapular BAT and subcutaneous iWAT, adipose depots were harvested, fixed in 10% formalin and then embedded in paraffin. The paraffin-embedded specimens were sliced at 4 μm using a microtome. The sections were stained with H&E to observe the pathological morphology of mouse adipose tissues.

WT and transgenic AdipoR1 mice, before or after housing in the cold, were assessed for regional surface temperature on skins above brown adipose tissue areas and on the abdomen by an infrared thermometer (HFS-1000, Thermofinder Plus, HuBDIC, Gyeonggi-do, Korea). Readings were recorded with constant values at least 3 times.

Thermography was performed using an infrared camera (IRM-P384G, Ching Hsing Computer-Tech Ltd., Taipei, Taiwan) on WT and transgenic AdipoR1 mice acclimated at a cold room. No anesthesia was applied to avoid disturbing mouse body temperature with human-handling.

Total RNA was extracted from mouse tissues and cells by the TRIsure reagents (Bioline, Meridian Bioscience, London, UK), with contaminating genomic DNA later removed by the TURBO-DNase free kit (AM1907; Thermo Scientific, Waltham, MA, USA), followed by reverse transcription into cDNA using the High Capacity cDNA Reverse Transcription kit (Thermo Scientific). Quantitative real-time polymerase chain reaction (qPCR) was performed using a SensiFAST SYBR Hi-ROX Kit (Bioline) on an ABI StepOnePlus Real-Time PCR System (Applied Biosystems, Thermo Scientific) or a LightCycler 480 Instrument II (Roche Diagnostics, Indianapolis, IN, USA). qPCR was performed as follows; activate enzyme at 95 °C for 2 min, and then 40 cycles of 95 °C for 5 s and 60 °C for 30 s with a final extension of 60 °C for 1 min. Primers used for amplification of qPCR reaction are listed in. The relative value for each target gene was normalized to β-actin expression in the same sample. The threshold cycle (Ct) values were obtained, and relative gene expression was calculated using the comparative Ct method. Amplification of specific transcripts was further confirmed by melting-curve analysis and agarose-gel electrophoresis to check resulting product specificity. Supplementary Table S1

Mouse pre-adipocytes within the stromal/vascular fraction (SVF) from subcutaneous iWAT were isolated from adult WT or transgenic AdipoR1 mice. Detailed procedures for pre-adipocyte isolation, cell culture and differentiation into mature adipocytes were described previously with slight modification [59,60,61]. Briefly, the mouse subcutaneous inguinal region of WAT was surgically dissected, minced and digested with collagenase (C6885, Sigma-Aldrich, St. Louis, MO, USA) and then filtered through chiffon. Pre-adipocytes within SVF were derived through a series of centrifugation and washing steps with red-blood-cell-lysis before the final washing step. Then, pre-adipocytes were suspended and seeded on tissue-culture treated plates containing medium of Dulbecco’s Modified Eagle Medium/F12 (Gibco DMEM/F12, Thermo Scientific), 10% fetal bovine serum (Thermo Scientific) and 1% antibiotics (penicillin, streptomycin and amphotericin B; Biological Industries, Kibbutz Beit-Haemek, Israel). Cells were incubated at 37 °C in air containing 5% CO₂ until confluency. Confluent pre-adipocytes were subsequently switched to induction medium with or without the PPARγ-agonist rosiglitazone for adipocyte differentiation. After around 7 to 9 d, mature adipocytes with at least 70% differentiation were collected for further analysis.

Data were expressed as means ± standard error of the mean (SEM). Results of two groups were compared and determined for statistical significance by the Student’s t-test (Graphpad Prism 6; Graphpad Software, San Diego, CA, USA). The p-values ≤ 0.05 were considered statistically significant.

The data that support the findings of this study are available from the corresponding author upon reasonable request.