Obesity is a significant health challenge, with the latest

Transcription

1 Obesity Brown Adipose Tissue Mechanisms and Potential Therapeutic Targets Charmaine S. Tam, PhD; Virgile Lecoultre, PhD; Eric Ravussin, PhD Obesity is a significant health challenge, with the latest projections estimating that there will be 2.16 billion overweight and 1.12 billion obese individuals globally by In addition to social stigmatization and impaired quality of life, obese people are faced with significantly increased risk of cardiovascular disease, type 2 diabetes mellitus, and a number of cancers. Weight gain results from a sustained imbalance between energy intake (calories consumed) and energy expenditure (calories burned), resulting in positive energy balance. To tip energy balance toward weight loss, not only should food intake be decreased, but increasing physical activity may help. It is important to realize, however, that the secular rises in obesity over the past 3 to 4 decades can be explained by both physiological and environmental drivers and a number of putative factors (Figure 1). 1,2 At one level, the obesity epidemic is a classic gene-environment interaction in which the human genotype is susceptible to environmental influences that affect energy intake and expenditure, with the obesogenic environment being dominant. These environmental factors include the 2 obvious explanations, ie, reduced physical activity and increased energy intake from high-calorie food and drinks, with the latter likely being the dominant factor. 3 However, other less studied factors have been implicated, including longer time spent awake, increased mean age of mothers at first birth, decreased prevalence of smoking, presence of environmental pollutants, ingestion of many novel medications, and reduction in the variability of seasonal ambient temperature owing to the presence of almost ubiquitous air conditioning. 4 Fundamentally, the obesity epidemic is explained by a dysregulation of energy balance in our obesogenic environment. Understanding the cause of obesity requires the study of how genetic and environmental factors interact to produce long-term positive energy balance. As a consequence of the obesogenic environment, the treatment of obesity requires intensive lifestyle modification. However, lifestyle modification in obese individuals tends to provide only transient success, and pharmacological treatment of obesity has often been disappointing and is very contentious because of the many safety concerns. Other approaches such as bariatric surgery are costly and not without risk. Thus, there is a concerted effort to find novel strategies to reduce excess body weight. This includes the potential of increasing energy expenditure via the stimulation of brown adipose tissue (BAT), now known to be present in adult humans. We would expect that such physiological or pharmacological stimulation of thermogenesis (heat production) would lead to dissipation of at least some excess ingested calories. This review provides a historical overview on the research on BAT, highlights recent developments in the field, and concludes with a discussion of the relevant issues that need to be addressed before BAT is considered a therapeutic option for human obesity. Characteristics of BAT The majority of adipose tissue in the human body is white adipose tissue (WAT), which acts as a passive depot for energy storage and is an active endocrine organ, releasing free fatty acids and adipokines (Figure 2). However, small amounts of BAT may be found in the neck; in the supraclavicular and axillary regions; in the paravertebral, perirenal/ adrenal, and paraventral regions; and around the major vessels (the aorta and its main branches: carotids, subclavian, intercostals, and renal arteries; Figure 3). BAT can also be found within WAT and skeletal muscle tissues. 5 Notably, the histological studies described in humans thus far suggest that brown and white adipocytes are mixed together in all depots. 6 8 Brown fat cells are characterized by multilocular lipid droplets and an increased number of mitochondria, which express uncoupling protein 1 (UCP1). UCP1 is located in the inner membrane of the mitochondria and uncouples the rates of substrate oxidation and ATP production by favoring a loss of protons and thus energy release. Genes such as peroxisome proliferator activated receptor g coactivator 1, cytochrome c, deiodothyronine type 2, 3 -adrenergic receptor, and PR domain containing 16 (PRDM16) are highly expressed in brown adipose cells and highlight the common features and physiology of BAT, namely high oxidative capacity, activation by thyroid hormones and catecholamines, and cell differentiation from skeletal muscle precursor cells. 9 A Historical Perspective on BAT Traditionally, BAT was considered a thermogenic organ for maintaining core temperature during cold exposure in small mammals and newborns. In humans, BAT was believed to disappear rapidly with age. The notion that BAT thermogen- From Human Physiology, Pennington Biomedical Research Center, Baton Rouge, LA. Correspondence to Eric Ravussin, PhD, Pennington Biomedical Research Center, 6400 Perkins Rd, Baton Rouge, LA eric.ravussin@pbrc.edu (Circulation. 2012;125: ) 2012 American Heart Association, Inc. Circulation is available at DOI: /CIRCULATIONAHA

2 Tam et al Brown Adipose Tissue 2783 Tradi onal Environment 1900s Leptogenic Environment Social Environment Built Environment 2000s Obesogenic Environment The big 2: Increased energy intake and reduced physical ac vity Other factors: Radia on/chemotherapy, Longer me spent awake, Increased age of mothers at first birth, Reduced variability in seasonal ambient temperature (Keith et al, IJO, 2006) Obesity Suscep bility Obesity Suscep bility Environment Gene cs Obesity Resistant Gene cs Obesity Prone Leptogenic environment esis, beyond its role in the regulation of body temperature in cold environments, could dissipate excess calories during positive energy balance was first suggested 30 years ago 10 and remains controversial in both rodents and humans. 11 In the 1970s and 1980s, Rothwell and Stock 12 performed a series of seminal experiments examining rodents on human cafeteria diets, ie, palatable high-fat, high-sucrose diets. They showed that despite consuming 80% more energy, the cafeteria-fed rats only had a 27% higher weight gain than chow-fed rats. Importantly, similar to animals exposed long term to cold, cafeteria-fed rats demonstrated increased sympathetic nervous activity and metabolic rates in parallel with an increase in the activity of BAT. This study provided the first evidence suggesting a potential role of BAT in the maintenance of body weight and body composition. Cafeteria-fed animals had increased BAT mass, higher skin and abdominal temperatures, and 4-fold-higher norepinephrine-stimulated free fatty acid release. Moreover, in that same publication, Rothwell and Stock 12 appeared to confirm the presence of functional BAT in an adult man by demonstrating heat production at the skin surface using infrared thermography after the administration of ephedrine. Even though these results were highly criticized (effect of ephedrine on local blood flow), they raised the question about the clinical significance of BAT in humans and its role in the regulation of energy balance. Rodent studies showed that BAT can produce up to 300 W/kg, ie, a metabolism 2 orders of Obesogenic environment BMI (kg/m 2) Figure 1. The effects of genes and environment on adiposity. Effects of genes and environment on adiposity. We show the potential effect of genes and environment on adiposity assessed here by body mass index (BMI). Some of the concepts described in this figure were proposed by Bouchard et al 1 and Keith et al. 4 Our environment has evolved over the past century from a traditional environment to a new Westernized environment. The left side presents the long-lasting traditional environment in which food is scarce and energy expenditure is high mostly as a result of occupational physical activity. Such an environment leads to leptogenic behaviors in which the variability of BMI will depend on the individual s genetic propensity to weight gain. On the right side, the more recent modern social and built environment leads to obesogenic behaviors characterized by plenty of cheap, high-calorie food and little need for physical activity. The variability in BMI will also depend on the genetic propensity of the individual to weight gain, but the obesogenic environment exerts a much stronger influence. The distribution of BMI will have a higher mean and higher standard error in the obesogenic environment than in the leptogenic environment. Such a paradigm can be applied to populations with similar genetic backgrounds living in drastically different environments like the Pima Indians in Arizona and in Mexico. 2 magnitude higher than that of other tissues. 13,14 On this basis, Rothwell and Stock calculated that if maximally stimulated, 40 to 50 g of BAT would be sufficient to increase daily energy expenditure by 20% in humans. 14 However, a recent review by van Marken Lichtenbelt and Schrauwen 15 challenges the 20% magnitude increase, stating that maximal BAT stimulation is necessary and that mice data are not easily extrapolated to humans. Therefore, 50 g of BAT in humans is more likely to increase energy expenditure only up to a maximum of 5%. In the late 1970s, a large body of studies in rodents led to the belief that an impaired maintenance increment of energy expenditure (ie, a decreased uncoupling and therefore heat production) was pivotal to increasing metabolic efficiency leading to obesity. For example, the inability of young mice to maintain body temperature when exposed to cold predicted an obese phenotype later in life. 16 Indirect evidence that BAT may be involved in the regulation of energy balance and body weight in humans came from studies of patients with pheochromocytoma Lean et al 18,19 suggested that the high circulating noradrenaline concentrations typically seen in subjects with pheochromocytoma (re)activate BAT and contribute to the dramatic weight loss (10% 20% body weight in already lean individuals) associated with the disease. More specifically, they reported a 4- to 10-fold increase in UCP1 content in the adipose tissue of these patients and a basal mitochondrial oxygen uptake close to the fully uncoupled

3 2784 Circulation June 5, 2012 Figure 2. Histology of brown and white fat cells. Light microscopy image of mouse anterior subcutaneous fat depot in an area where white and brown adipose tissues are close each other. Uncoupling protein 1 (UCP1) immunoreactive fat cells corresponding to brown adipose tissue (left) and UCP1-negative white fat cells corresponding to white adipose tissue (right) are visible. Immunohistochemistry was performed with the ABC method with sheep anti-rat UCP1 primary antibody (1:500), which was kindly provided by Daniel Ricquier (Paris). The scale bar represents 40 m. This figure was prepared by Dr Saverio Cinti (University of Ancona, Ancona, Italy). state (measured by adding the uncoupling agent carbonyl cyanide paratrifluoromethoxyphenylhydrazone [ie, FCCP]). A recent study supports the role of BAT-mediated energy expenditure in pheochromocytoma patients because, after resection of the tumor, fluorodeoxyglucose (FDG) uptake in BAT was no longer apparent in 25 patients. 21 Wang et al 20 reported that 6 of 14 subjects with pheochromocytoma had BAT detected by 18 F-FDG position emission tomography (PET) with computed tomography (CT) and that circulating metanephrine, a surrogate measure of plasma catecholamines, was associated with the presence of BAT. Despite the lack of association between catecholamine levels and the presence of BAT in 1 study, 17 together these observations provide evidence that excessive catecholamines secreted by adrenal tumors stimulate thermogenesis in BAT, thereby increasing total energy expenditure and leading to the lean phenotype characteristic of patients with pheochromocytoma. In a broader context, examining patients with pheochromocytoma provides a unique integrated physiological model for overstimulation of BAT (mass and activity) and fat mass loss. However, pheochromocytoma-associated weight loss can be attributed not only to BAT activation but also to the impact of catecholamines on other tissues such as skeletal muscle, the liver, and even the central nervous system. In the 1990s, pharmaceutical companies developed compounds targeting adipose tissue selective 3 -adrenoceptors as a putative pharmacological strategy to increase energy expenditure by BAT activation and thus reduce body weight. In humans, 3 -adrenoceptors are located predominantly on BAT with few or no receptors on WAT. 3 -Agonists showed some positive metabolic effects such as improved insulin action and increased fat oxidation but in general failed to increase 24-hour energy expenditure and likely BAT activity. 22 More recently, a novel 3 -adrenoceptor (TAK-677) was shown to marginally increase 24-hour energy expenditure compared with placebo but also caused an increase in heart rate of 9 bpm. 23 Less specific 3 -agonists also had undesired side effects such as tremors or tachycardia, mediated by their affinity for 1 - and 2 -adrenoceptors The (Re)Discovery of BAT in Humans In the early 2000s, studies in nuclear medicine using 18 F-FDG PET/CT scanning revealed the presence of BAT in adult humans. This pattern of FDG uptake, initially called USA (uptake in supraclavicular area) fat, was initially considered an artifact by radiologists who were interested in identifying active tumors. Importantly, the radiologists noticed that the Figure 3. Anatomic location of discrete brown adipose tissue measured by positron emission tomography. Computed tomographic (CT; top) and positron emission tomographic (PET; with 18 F- fluorodeoxyglucose; middle) images from the neck and thoracic region in 1 lean individual. Bottom, The superimposition of the CT and PET scans. In the 3 panels, shown from left to right are a transverse slice at the level of the clavicles, a sagittal slice at the level of the spine, and a coronal slice of the thorax. Activated brown adipose tissue areas (red and green) are in the cervical-supraclavicular (most common), perirenal/adrenal, and paravertebral regions. This figure was prepared by Dr Wouter van Marken Lichtenbelt (Maastricht University, Maastricht, the Netherlands).

4 Tam et al Brown Adipose Tissue 2785 presence of BAT was related to ambient outdoor temperature 27 and that, unlike in tumors, 18 F-FDG uptake could be blocked pharmacologically by -blockers in rats and humans Then, in 2009, 5 independent groups used 18 F- FDG PET/CT to identify and characterize the presence and relevance of BAT in adult humans. 6,8,31 33 All 5 groups showed major depots of metabolically active BAT in the cervical-supraclavicular regions, and some confirmed the presence of true BAT by the expression of UCP1 and histological characteristics. Cypess et al 31 analyzed 3640 consecutive 18 F-FDG PET/CT scans from 1972 patients and reported positive BAT scans in 7.5% of women and 3.1% of men. Virtanen et al 8 reported that exposure to cold (2 hours in a 17 C 19 C room and intermittent dipping of 1 foot in ice water) increased glucose uptake 15-fold in paracervical and supraclavicular adipose tissue. Adipose tissue biopsies were obtained from 3 out of 5 participants and UCP1 mrna and protein levels immunohistochemistry confirmed that the tissue identified in the PET/CT scans was true BAT. Similarly, van Marken Lichtenbelt et al 32 used 18 F- FDG PET/CT to study 24 men under thermoneutral conditions (22 C) and during mild cold exposure (15 C for 2 hours). BAT activity was observed in 23 of 24 subjects during cold exposure but in no participants during thermoneutral conditions. The same investigators reported a positive relationship between resting metabolic rate and BAT activity at thermoneutrality or during cold exposure, whereas an inverse association between BAT amount/activity and body mass index was found. Finally, Saito et al 33 observed that cold-induced BAT (measured with 18 F-FDG PET/CT) was more prevalent in young (53%) than in elderly (8%) subjects and was less active with increasing obesity. Using a more direct approach, Zingaretti et al 6 examined adipose tissue taken from the neck of patients 18 to 82 years of age undergoing surgery for thyroid diseases. In 20 of 35 patients (57%), they observed distinct islands of UCP1- positive staining that was richly sympathetically innervated. Moreover, within the UCP1-positive islands, there was evidence of brown adipocyte precursors, implying that it may be possible to recruit BAT by promoting the proliferation and differentiation of these cells. Studies over the past 3 years have led to a paradigm shift in our understanding of BAT in adult humans, fueling a resurgence of interest in the potential of activating BAT to enhance energy expenditure, therefore assisting in controlling body weight and possibly preventing metabolic diseases. Limitations of PET/CT and Alternative Approaches for Assessing BAT According to studies using PET/CT techniques, the prevalence of BAT in adults varies between 2% and 100%. 34 Such variability may be attributed to most measurements being performed at thermoneutral conditions and in a fasting state when we expect BAT activity to be reduced or even minimal. For example, when the assessment of BAT was performed after a 2-hour mild cold exposure, a condition known to induce BAT activity, the prevalence of BAT increased to 96%, whereas no BAT was observed at thermoneutral conditions. 32 Alternatively, this variability may be due to the inability of PET/CT scans to discriminate between the presence of BAT and its activity (FDG uptake). This point is highlighted by several studies showing poor reproducibility from subjects having multiple PET/CT scans Lee et al 35 reported that the average probability of obtaining subsequent positive PET/CT scans among patients with BAT was 13%. Similarly, in 33 subjects who each had 5 PET/CT scans, Rousseau et al 37 showed that FDG uptake was identical on all 5 scans in only 5 patients, ie, a 15% reproducibility. No association was found between FDG uptake and outdoor temperature in that study. However, another study with poor BAT reproducibility from multiple PET/CT scans (16% reproducibility) 36 showed that FDG uptake was inversely associated with outdoor temperature. 27 To disentangle the discrepancies between PET/CT and histological methods of defining BAT, Lee et al 38 compared adipose tissue biopsies taken from 3 subjects in whom PET/CT was positive versus 14 subjects in whom PET/CT was negative. Regardless of PET/CT status, signature transcripts of BAT (ADRB1, ADRB2, ADRB3) were present in all patients, and UCP1 staining was seen in all but 1 PET/CT-negative patient. Together, these findings suggest that the prevalence of BAT reported in retrospective PET/CT studies may be largely underestimated, suggesting that BAT is highly prevalent in adult humans. 35,38 Currently, 18 F-FDG PET/CT scanning is the only noninvasive method for examining the presence of BAT. However, this technique is limited by poor reproducibility, 35 37,39 high cost, and the radiation exposure during the scan, which generally precludes the inclusion of healthy subjects in such studies. Therefore, there is a need to develop alternative noninvasive techniques for measuring the presence of BAT and, most important, BAT activity in humans. Recent advances in magnetic resonance imaging (MRI) have led to the development of techniques such as the iterative decomposition with echo asymmetry and least-squares estimation (IDEAL) MRI. This method distinguishes WAT and BAT on the basis of their unique lipid compositions because WAT contains greater lipid than BAT. Hu et al 40 demonstrated the feasibility of using IDEAL MRI to distinguish WAT from BAT in excised samples from mice carcasses. To date, there have been no published studies using IDEAL MRI to examine BAT in humans. Another approach is to use the negative Hounsfield units from computer-assisted tomography to differentiate between WAT and BAT. Hu et al 41 examined PET/CT scans from 101 pediatric and adolescent patients and identified that regions of metabolic active tissue (BAT) had more positive Hounsfield units than inactive (WAT) fat. The feasibility of infrared thermography as another noninvasive approach for measuring BAT in humans was recently published in a pilot study of 87 subjects. 42 Using infrared thermography in 1 subject, Lee et al 42 found that the highest skin temperature was localized to the supraclavicular fossae, which corresponds to the most common location of BAT. After cold exposure, supraclavicular skin temperature was decreased by only 0.9 C compared with a reference skin area where temperature was decreased by 2 C. Even without a change in core temperature in cold, energy expenditure increased by 14%, an increase similar to that reported in

5 2786 Circulation June 5, 2012 studies using PET/CT. 32 This finding was confirmed in 87 subjects where they found that mean skin temperature was significantly higher in the supraclavicular fossae compared with the mediastinal region, an area known not to contain BAT. It is unclear whether these changes in skin temperatures are due to increased blood flow or a true thermogenic response in BAT. Further studies are needed to validate the accuracy of infrared thermography by PET/CT. 42 Recently, several new compounds have shown promising results for the detection and quantification of BAT. 43 Bartelt et al 43 demonstrated the feasibility of measuring active BAT by MRI and the incorporation of superparamagnetic iron oxide nanocrystals into the lipoprotein cores as tracers. However, this technique, although promising because of its low toxicity, may not be sensitive enough to detect inactive BAT. In contrast to 18 F-FDG, another innovative compound, 4-18 F-fluorobenzyltriphenyl phosphonium ( 18 F-FBnTP), may provide a unique opportunity to detect and quantify both inactive and active BAT depots. 44 By targeting the electrochemical gradients, 18 F-FBnTP accumulates in mitochondria as a function of their uncoupling state and is washed out in relationship to mitochondrial thermogenic activity. 44 This unique physical property of 18 F-FBnTP warrants a new direction in the noninvasive investigation of BAT in both humans and rodents. Although measures of BAT by 18 F-FDG glucose uptake and PET/CT have so far been considered the gold standard, will these novel approaches lead to a better understanding of the physiological role of BAT in humans? The ability to detect inactive BAT by IDEAL MRI or PET/CT with 18 F-FBnTP will certainly provide new insights into the true prevalence of BAT in humans and its metabolic role and allow us to start examining its plasticity in vivo. Associations Between the Presence of BAT and Clinical Variables Higher BAT Prevalence in Women Versus Men Most analyses from retrospective studies using PET/CT show that the prevalence of BAT is higher in women than in men. 31,35,45,46 Nedergaard et al 34 speculate that this is attributed to the fact that women experience cold sensations at higher temperatures than men. As a consequence, the chance of detecting BAT during a PET/CT scan may be increased in women. However, this explanation overlooks studies in rodents that strongly suggest that BAT metabolism is sex dependent with more BAT in female rats For instance, compared with male rats, female rats have higher UCP1 content, greater multilocular lipid arrangement, and both longer and denser cristae in their mitochondria, indicating higher thermogenic capacity and activity. 49 Furthermore, this is accompanied by a more sensitive 3 -adrenoceptor response to norepinephrine stimulation in female rats. 49 Testosteroneor progesterone-treated brown adipocytes have opposite actions on UCP1 expression, with testosterone-treated cells showing fewer and smaller lipid droplets with a dosedependent inhibition of UCP1 mrna expression under adrenergic stimulation by norepinephrine. 48 On the other hand, progesterone and 17- -estradiol-treated cells showed larger lipid droplets and progesterone-stimulated norepinephrine-induced UCP1 mrna expression. 48 These findings, albeit in rodents, suggest that BAT metabolism is sex dependent and influenced by the hormonal environment, giving potential insight into why women have greater reported BAT prevalence than men. BAT Is Inversely Associated With Aging BAT was traditionally believed to be present only in newborns and to decline with age owing to a decreased need for thermogenesis. As early as 1972, Heaton et al 7 observed that interscapular BAT was consistently present in the first decade of life, disappearing gradually up to 30 years of age with a sharp decline thereafter. We now know that BAT is present in adulthood, with numerous studies reporting a decline in BAT prevalence with age. 35,39,51 However, the exact timing of this age-related decline in BAT is unclear. When BAT-positive subjects were stratified into age tertiles, BAT was 3 times more likely to be detected in subjects 50 years of age compared with those 64 years. 31 Interestingly, body mass index (BMI) was not correlated with the presence of BAT, but when multivariate analysis was performed, BMI became a significant negative predictor with increasing age. 31 Similar findings were obtained by Pfannenberg et al, 45 who showed that age was the strongest negative determinant of BAT mass. Furthermore, their analysis suggested that the metabolic effect of BAT, as reflected by its relationship with BMI, declines with age in men but not women. Whether BAT has a protective role in age-related obesity remains to be established. The decline in BAT activity with aging appears to be related to the decrease in the amount of BAT depots rather than activity. Using immunohistochemistry 7,52 or PET/CT, 46 there is a notable decrease in BAT after adolescence that is maintained from 20 to 80 years of age before a steep decline after 80 years of age. It is also very likely that following the initial drop in BAT after adolescence, both the function and sensitivity of BAT diminish during adulthood. 53 Such observations may explain why the presence of cold-activated BAT is 50% in subjects in their 20s but 10% in subjects in their 50s or 60s. 54 Consistent with this hypothesis, Zingaretti et al 6 proposed that at a certain age, humans switch from a phenotype characterized by leanness, small white adipocytes and the presence of brown adipocytes to a phenotype characterized by increasing obesity, large adipocytes and the absence of brown adipocytes. BAT and Measures of Body Composition The inverse association between the amount of active BAT and BMI is well accepted ,45 However, only 4 studies have examined associations between BAT and more specific measures of body composition. 20,32,33,54 In 23 of 24 healthy men, van Marken Lichtenbelt et al 32 observed a strong negative association between cold-induced BAT and percent body fat measured by dual-energy x-ray absorptiometry. Interestingly, the 1 subject who did not have cold-induced BAT activation had the highest percent body fat (and BMI) of the cohort. 32 Similarly, Saito et al 33 reported strong negative correlations between cold-induced BAT and total and visceral

6 Tam et al Brown Adipose Tissue 2787 fat measured by CT. However, when subjects who were BAT positive were compared with those who were BAT negative, there were no significant differences in fat deposition, although this may have been due to the fact that the BATpositive subjects were younger. These findings suggest that the relationships between the presence of BAT, obesity, and age are innately intertwined because BAT is generally observed in younger, leaner subjects. Yoneshiro et al 54 attempted to disentangle these associations by testing the idea that decreased BAT was associated with body fat accumulation with age. In both sexes, significant positive correlations between age and all adiposity parameters (percent fat, visceral and subcutaneous fat) were observed only in BATnegative subjects, leading to the hypothesis that the presence of BAT may be protective of age-associated body fat accumulation. 54 Recently, Wang et al 20 measured BAT by 18 F- FDG PET/CT in 14 patients with pheochromocytoma and 14 age- and BMI-matched normal subjects. BAT was detected in 6 of 14 patients with pheochromocytoma in whom total metanephrine, a byproduct of epinephrine and norepinephrine degradation by catechol-o-methyl transferase and a surrogate measure of plasma catecholamine concentrations, was elevated. On the contrary, there was no detectable BAT in patients without elevated metanephrine and normal control subjects. Interestingly, there was a positive association between BAT activity and total metanephrine and an inverse association between BAT activity and visceral fat. Given that BAT activity was measured by PET/CT in thermoneutral conditions, these group differences may have been even more pronounced if BAT had been measured after cold exposure. Nevertheless, this recent finding in humans reinforces the data from the 1980s showing relationships between BAT activity and body composition in rodents. 12,18 Whether BAT can be activated by -adrenergic agonists or catecholamines in classic depots of white fat remains to be investigated. However, these brown adipocytes are impossible to detect by imaging techniques. BAT and Outdoor Temperature One of the major indications that BAT is present in adulthood and activated by cold exposure was provided by Huttunen et al, 55 who found BAT-positive depots in neck necropsies obtained from Finnish outdoor workers in the cold. In recent years, the association between BAT and outdoor temperature has been well described by many groups, 31,33,36,37,46,51 although only 1 study has specifically examined seasonal differences in the detection of BAT. Saito et al 33 performed FDG PET/CT scans in the same 8 subjects in summer and winter. The presence of BAT was observed in only 2 of 8 subjects in summer but increased to 6 of 8 subjects in winter. Yoneshiro et al 56 showed that short-term cold exposure (2-hour) resulted in a significant increase in energy expenditure only in BAT-positive, not BAT-negative, male subjects. Whether this finding suggests that BAT is actively recruited during cold exposure in humans is uncertain. Mice and rats that undergo long-term cold exposure have hyperplasia of BAT and increased 2-deoxyglucose uptake. 57,58 Moreover, cold acclimation in rodents has been shown to induce a transdifferentiation of WAT to BAT. 59,60 It is unknown whether long-term cold exposure in humans will induce changes in the amount of BAT and metabolic activity. Is There a Relevant Role for BAT in Obesity Therapy? As described above, the last few years have heralded a renaissance in the study of BAT and consequently (re)ignited the hypothesis that BAT thermogenic potential could be targeted to increase energy expenditure and thus have an antiobesity effect. 53,61,62 However, numerous areas still require thorough investigation before the exploitation of BAT for obesity therapy is considered. In a broad context, both total abundance and activity need to be stimulated, resulting in increased energy expenditure. Approaches that may lead to increased BAT activity (and possibly abundance) include cold exposure, stimulation of the sympathetic nervous system, and/or increasing the activity of the thyroid axis. Other more invasive approaches include BAT transplantation and/or the stimulation of non-bat progenitor cells to differentiate into brown adipocytes. Transcriptional regulators of brown adipocyte differentiation include bone morphometric protein 7, PRDM16, Mir193b-365, orexin, forkhead Box C2, Plac8, and RIP140, all of which have been described in rodents (see Figure 4). 9,63 68 Transcriptional regulators of brown adipocyte differentiation in humans are largely unknown and may depend on whether BAT is in a diffuse or a discrete location. Indeed, given that BAT and WAT are both present in various adipose tissue depots in humans, some authors suggest that under certain conditions WAT could transdifferentiate to BAT and vice versa. Accordingly, these transdifferentiating cells have been described as brite or beige. 69 Regardless of the method by which upregulation of BAT thermogenesis occurs, a fundamental question remains: How would an overactive BAT operate at the whole-body level, and would upregulation be sufficient to induce negative energy balance and thus weight loss? In a simplified integrated model, BAT thermogenesis occurs mainly through the interaction of 2 components: short-term (ie, minutes or hours) modulation of the activity of the sympathetic nervous system in response to meals, physical activity, or variations in temperature and long-term changes in thyroid hormones in response to changes in energy balance (Figure 5). These interactions are likely to regulate both BAT thermogenesis and the fuel supply to BAT. Indeed, the sympathetic and thyroid axes are involved in peripheral (WAT) and local (BAT) lipolysis and in the regulation of plasma glucose by the liver. Therefore, BAT becomes important to dissipate excess energy intake by potentially increasing energy expenditure to match energy intake. In response to food intake, insulin and gut hormones are transiently increased, thereby promoting substrate disposal and norepinephrine-induced BAT thermogenesis, whereas other hormones involved in the long-term regulation of energy balance (thyroid hormones and leptin) would hardly be affected. As a consequence, short-term excess energy intake (meals) may be only partially buffered by activation of BAT. However, it is assumed that for weight control (which requires long-term regulation) one needs long-term stimulation of BAT (increased amount and optimal activity). This simplified model

7 2788 Circulation June 5, 2012 A Recruitment and differentiation of BAT precursors into discrete BAT depots Discrete BAT Transcrip onal regulators & BMPs Skeletal Muscle Transcrip onal regulators & BMPs Dermomyotomal precursors (Myf5) Brown Preadipocyte Brown Adipose B Recruitment and differentiation of stem cells into diffuse BAT Brown adipose-like cell in WAT Diffuse BAT WAT-resident brown precursor Transcrip onal regulators & BMPs Mesodermal stem cells White Preadipocyte White Adipose fits with observations that in subjects with active BAT (as assessed by PET/CT), BAT mass and activity are associated with BMI. 32,33,54,70 However, a chronically active BAT would likely affect glucose and adipose tissue regulation, triggering counterregulatory mechanisms that may ultimately downregulate its activity (Figure 5C). The model of human pheochromocytoma provides an example of the potential of long-term BAT thermogenesis to decrease body weight. This idea is also supported by rodent models in which increased BAT activity and uncoupling lead to a reduction in body weight However, in such a model, the constant increase in BAT metabolic activity with extra stimulation in response to meals, cold exposure, or stress may trigger compensatory mechanisms in response to the lowering of plasma glucose and/or free fatty acids. These compensatory mechanisms may include hyperphagia, fatigue, or somnolence, and stress that will favor positive energy balance. Furthermore, weight loss with a sustained decrease in adiposity will cause a decrease in circulating leptin, which in turn leads to a decrease in sympathetic nervous system activity and activity of the thyroid axis. Although theoretical, this example using integrated physiology highlights some of the White Adipose tissue Figure 4. Regulation of brown adipogenesis in rodents. In rodents, it has been largely believed that Myf5 cells are early brown fat or muscle cell precursors differentiating into discrete BAT (A). A range of transcriptional factors regulating brown adipocyte differentiation have been identified, including bone morphometric protein 7, PR domain containing 16, Mir193b-365, orexin, forkhead Box C2, Plac8, and RIP140. These transcriptional regulators act in combination with other proteins such as peroxisome proliferator activated receptor g coactivator 1 to suppress myocyte differentiation and promote a brown fat cell phenotype. Diffuse BAT (B) originates from white adipose lineage in response to cold or beta-adrenergic stimulation and may be derived from brown precursors, white preadipocytes of mature white adipocytes. BAT indicates brown adipose tissue; WAT, white adipose tissue. possible counterregulatory processes that may oppose BATmediated weight loss (Figure 5). Long-term cold exposure may represent a relatively easy method for stimulating BAT thermogenic capacity in humans, even if there is clear interindividual variability in the response to cold-induced energy expenditure. 15,74 Furthermore, a significant lowering in ambient temperature may not be well tolerated by many individuals, who will compensate by wearing extra layers of clothes. Whether daily short-term (ie, several minutes) exposure to cold (a cold pill ) may be sufficient to stimulate BAT mass and responsiveness remains to be tested. Apart from the reports of high BAT mass and likely activity in lumberjacks working long term in extreme cold temperatures, 55 recent studies reported seasonal differences in BAT activity, 31,39,46 with higher BAT prevalence in winter versus summer. In Western countries, occupations are mostly performed indoors and thus in thermoneutral conditions. Only short exposures to cold are experienced during the winter months. It is therefore likely that the reported increase in BAT activity is induced by only daily short cold exposures. However, some data indicate that the seasonal changes in the photoperiod may be more important than the outdoor tem-

8 Tam et al Brown Adipose Tissue 2789 Figure 5. Integrated physiological control of brown adipose tissue (BAT) thermogenesis. Hypothetical illustration of an integrated physiological model of BAT thermogenesis accompanied by some of the physiological implications and counterregulatory mechanisms. In the basal postabsorptive state, the BAT present in the body is controlled by the sympathetic nervous system and thyroid axis (A). In response to short-term cold exposure (and possibly food intake), sympathetic activity is transiently increased, leading to a short-term stimulation of BAT (cold-induced thermogenesis [CIT] or thermic effect of food [TEG]; B). This transient increase participates in the regulation of substrate supply to BAT and therefore short-term energy balance. Long-term signals such as leptin and thyroid hormones are unaffected. When BAT amount and activity are increased by long-term cold exposure (C), high substrate disposal into BAT will induce a progressive depletion of body fat stores and weight loss. Ultimately, the adipose tissue loss will lead to a decrease in circulating leptin and other signals defending energy stores. These adaptations will trigger compensatory mechanisms such as an increase in food intake and a decrease in sympathetic nervous system and thyroid axis activities, both predisposing to weight regain. perature in the stimulation of BAT activity. 46 Nevertheless, to what extent such a regimen would be able to induce BAT and to help increase heat dissipation (and therefore caloric excess) in response to food intake is unknown. Although controversial, 11 a link between BAT activity and diet-induced thermogenesis in humans has recently been reported recently. 74 Furthermore, other studies indicate a role for BAT in facultative thermogenesis and body weight regulation. Wijers et al 75 examined 10 lean and 10 obese subjects exposed to 48 hours of mild cold (16 C) while living in a respiratory chamber. On cold exposure, lean subjects had significantly higher mean daytime energy expenditure, a greater change in distal temperature, and a smaller decrease in proximal skin temperature compared with the obese subjects. In this study, cold-induced thermogenesis and diet-induced thermogenesis were correlated in lean subjects. Similar increases in energy expenditure after exposure to mild cold (22 C) for 30 hours were observed in lean women. 76 These findings, albeit in a small number of subjects, indicate that lean individuals are able to counteract changes to cold by increasing energy expenditure, whereas obese subjects experience a change in temperature distribution and therefore an increase in insulation. Recent data point to the role of thyroid hormones in the regulation of energy expenditure and, more specifically, BAT thermogenesis through the modulation of hypothalamic fat metabolism. 77 Besides the new insights into the central effects of hyperthyroidism, this newly identified pathway by which metabolic inefficiency can be manipulated may provide a new target for obesity. 78 Novel approaches aimed at increasing BAT mass have been proposed. The discovery of transcription factors and microrna implicated in the differentiation of progenitor cells into BAT bring us 1 step closer to the induction of BAT tissue in vivo 63 and/or its transplantation 9,79,80 to increase thermogenesis to limit weight gain. 80,81 In addition to the limitations and caveats already pointed out by Nedergaard and Cannon 81 and Kozak, 11 the prevalence of BAT appears to be higher than reported in retrospective studies in 90% of the subjects investigated. 38,82 If true, is it necessary to increase BAT mass? Instead, increasing the responsiveness of the BAT already present in the organism by acute sympathetic stimulation will likely be more efficacious. The development and validation of measurement techniques such as IDEAL MRI and 18 F-FBnTP with PET/CT in humans will be of importance in the study of human BAT physiology in the years to come. Conclusions Targeting BAT for obesity therapy is still at a preliminary stage. As early as 1982, Rothwell and Stock 14 urged that we need to exercise caution in extrapolating indirect histological and thermographic evidence to a major role for BAT in human energy metabolism. Despite significant advances in BAT research, multiple questions remain: What is the exact amount of BAT in humans? What is its exact role in the energy balance? Do the correlations between BAT mass and BMI indicate a role for BAT in the regulation of body weight? The advances in imaging techniques and cold activation of BAT before PET/CT scans indicate that BAT prevalence is likely to be high. Therefore, other important questions to be answered: Do most people keep a hibernating BAT that needs to be activated to regulate body weight by increased thermogenesis and thus buffer excess energy intake? What physiological counterregulations would be engaged if BAT could be transformed into an energy store

9 2790 Circulation June 5, 2012 burner? Will humans exhibit increased appetite such as that encountered with exercise training? Will the main drivers for thermogenesis, namely the sympathetic and thyroid axes, be turned off? All these questions must be answered before BAT thermogenesis can be used efficiently and safely as an antiobesity strategy. Sources of Funding Dr Lecoultre is supported by a grant from the Swiss National Science Foundation PBLAP None. Disclosures References 1. Bouchard C. The biological predisposition to obesity: beyond the thrifty genotype scenario. Int J Obes (Lond). 2007;31: Schulz LO, Bennett PH, Ravussin E, Kidd JR, Kidd KK, Esparza J, Valencia ME. Effects of traditional and western environments on prevalence of type 2 diabetes in Pima Indians in Mexico and the U.S. Diabetes Care. 2006;29: Swinburn B, Sacks G, Ravussin E. Increased food energy supply is more than sufficient to explain the US epidemic of obesity. Am J Clin Nutr. 2009;90: Keith SW, Redden DT, Katzmarzyk PT, Boggiano MM, Hanlon EC, Benca RM, Ruden D, Pietrobelli A, Barger JL, Fontaine KR, Wang C, Aronne LJ, Wright SM, Baskin M, Dhurandhar NV, Lijoi MC, Grilo CM, DeLuca M, Westfall AO, Allison DB. Putative contributors to the secular increase in obesity: exploring the roads less traveled. Int J Obes (Lond). 2006;30: Clarke JR, Brglevska S, Lau EW, Ramdave S, Hicks RJ. Atypical brown fat distribution in young males demonstrated on PET/CT. Clin Nucl Med. 2007;32: Zingaretti MC, Crosta F, Vitali A, Guerrieri M, Frontini A, Cannon B, Nedergaard J, Cinti S. The presence of UCP1 demonstrates that metabolically active adipose tissue in the neck of adult humans truly represents brown adipose tissue. FASEB J. 2009;23: Heaton JM. The distribution of brown adipose tissue in the human. J Anat. 1972;112: Virtanen KA, Lidell ME, Orava J, Heglind M, Westergren R, Niemi T, Taittonen M, Laine J, Savisto NJ, Enerback S, Nuutila P. Functional brown adipose tissue in healthy adults. N Engl J Med. 2009;360: Seale P, Bjork B, Yang W, Kajimura S, Chin S, Kuang S, Scime A, Devarakonda S, Conroe HM, Erdjument-Bromage H, Tempst P, Rudnicki MA, Beier DR, Spiegelman BM. PRDM16 controls a brown fat/skeletal muscle switch. Nature. 2008;454: Himms-Hagen J. Obesity may be due to a malfunctioning of brown fat. Can Med Assoc J. 1979;121: Kozak LP. Brown fat and the myth of diet-induced thermogenesis. Cell Metab. 2010;11: Rothwell NJ, Stock MJ. A role for brown adipose tissue in diet-induced thermogenesis. Nature. 1979;281: Cannon B, Nedergaard J. Brown adipose tissue: function and physiological significance. Physiol Rev. 2004;84: Rothwell NJ, Stock MJ. Luxuskonsumption, diet-induced thermogenesis and brown fat: the case in favour. Clin Sci (Lond). 1983;64: van Marken Lichtenbelt WD, Schrauwen P. Implications of nonshivering thermogenesis for energy balance regulation in humans. Am J Physiol Regul Integr Comp Physiol. 2011;301:R285 R Trayhurn P, Thurlby PL, James WP. Thermogenic defect in pre-obese ob/ob mice. Nature. 1977;266: Hadi M, Chen CC, Whatley M, Pacak K, Carrasquillo JA. Brown fat imaging with (18)F-6-fluorodopamine PET/CT, (18)F-FDG PET/CT, and (123)I-MIBG SPECT: a study of patients being evaluated for pheochromocytoma. J Nucl Med. 2007;48: Lean ME, James WP, Jennings G, Trayhurn P. Brown adipose tissue in patients with phaeochromocytoma. Int J Obes. 1986;10: Lean ME. Brown adipose tissue in humans. Proc Nutr Soc. 1989;48: Wang Q, Zhang M, Ning G, Gu W, Su T, Xu M, Li B, Wang W. Brown adipose tissue in humans is activated by elevated plasma catecholamines levels and is inversely related to central obesity. PLoS ONE. 2011;6: e Kuji I, Imabayashi E, Minagawa A, Matsuda H, Miyauchi T. Brown adipose tissue demonstrating intense FDG uptake in a patient with mediastinal pheochromocytoma. Ann Nucl Med. 2008;22: Weyer C, Tataranni PA, Snitker S, Danforth E Jr, Ravussin E. Increase in insulin action and fat oxidation after treatment with CL 316,243, a highly selective beta3-adrenoceptor agonist in humans. Diabetes. 1998;47: Redman LM, de JL, Fang X, Gamlin B, Recker D, Greenway FL, Smith SR, Ravussin E. Lack of an effect of a novel beta3-adrenoceptor agonist, TAK-677, on energy metabolism in obese individuals: a double-blind, placebo-controlled randomized study. J Clin Endocrinol Metab. 2007;92: Connacher AA, Lakie M, Powers N, Elton RA, Walsh EG, Jung RT. Tremor and the anti-obesity drug BRL 26830A. Br J Clin Pharmacol. 1990;30: Wheeldon NM, McDevitt DG, Lipworth BJ. Cardiac effects of the beta 3-adrenoceptor agonist BRL35135 in man. Br J Clin Pharmacol. 1994; 37: Acheson KJ, Ravussin E, Schoeller DA, Christin L, Bourquin L, Baertschi P, Danforth E Jr, Jequier E. Two-week stimulation or blockade of the sympathetic nervous system in man: influence on body weight, body composition, and twenty four-hour energy expenditure. Metabolism. 1988;37: Cohade C, Mourtzikos KA, Wahl RL. USA-Fat : prevalence is related to ambient outdoor temperature-evaluation with 18F-FDG PET/CT. J Nucl Med. 2003;44: Tatsumi M, Engles JM, Ishimori T, Nicely O, Cohade C, Wahl RL. Intense (18)F-FDG uptake in brown fat can be reduced pharmacologically. J Nucl Med. 2004;45: Parysow O, Mollerach AM, Jager V, Racioppi S, San RJ, Gerbaudo VH. Low-dose oral propranolol could reduce brown adipose tissue F-18 FDG uptake in patients undergoing PET scans. Clin Nucl Med. 2007;32: Soderlund V, Larsson SA, Jacobsson H. Reduction of FDG uptake in brown adipose tissue in clinical patients by a single dose of propranolol. Eur J Nucl Med Mol Imaging. 2007;34: Cypess AM, Lehman S, Williams G, Tal I, Rodman D, Goldfine AB, Kuo FC, Palmer EL, Tseng YH, Doria A, Kolodny GM, Kahn CR. Identification and importance of brown adipose tissue in adult humans. N Engl J Med. 2009;360: van Marken Lichtenbelt WD, Vanhommerig JW, Smulders NM, Drossaerts JM, Kemerink GJ, Bouvy ND, Schrauwen P, Teule GJ. Cold-activated brown adipose tissue in healthy men. N Engl J Med. 2009;360: Saito M, Okamatsu-Ogura Y, Matsushita M, Watanabe K, Yoneshiro T, Nio-Kobayashi J, Iwanaga T, Miyagawa M, Kameya T, Nakada K, Kawai Y, Tsujisaki M. High incidence of metabolically active brown adipose tissue in healthy adult humans: effects of cold exposure and adiposity. Diabetes. 2009;58: Nedergaard J, Bengtsson T, Cannon B. Three years with adult human brown adipose tissue. Ann N Y Acad Sci. 2010;1212:E20 E Lee P, Greenfield JR, Ho KK, Fulham MJ. A critical appraisal of the prevalence and metabolic significance of brown adipose tissue in adult humans. Am J Physiol Endocrinol Metab. 2010;299:E601 E Cohade C, Osman M, Pannu HK, Wahl RL. Uptake in supraclavicular area fat ( USA-Fat ): description on 18F-FDG PET/CT. J Nucl Med. 2003;44: Rousseau C, Bourbouloux E, Campion L, Fleury N, Bridji B, Chatal JF, Resche I, Campone M. Brown fat in breast cancer patients: analysis of serial (18)F-FDG PET/CT scans. Eur J Nucl Med Mol Imaging. 2006; 33: Lee P, Zhao JT, Swarbrick MM, Gracie G, Bova R, Greenfield JR, Freund J, Ho KK. High prevalence of brown adipose tissue in adult humans. J Clin Endocrinol Metab. 2011;96: Pace L, Nicolai E, D Amico D, Ibello F, Della Morte AM, Salvatore B, Micol PL, Salvatore M, Soricelli A. Determinants of physiologic (18)F-FDG uptake in brown adipose tissue in sequential PET/CT examinations. Mol Imaging Biol. 2011;13: Hu HH, Smith DL Jr, Nayak KS, Goran MI, Nagy TR. Identification of brown adipose tissue in mice with fat-water IDEAL-MRI. J Magn Reson Imaging. 2010;31: Hu HH, Chung SA, Nayak KS, Jackson HA, Gilsanz V. Differential computed tomographic attenuation of metabolically active and inactive