The epidemic of obesity is developing faster than the scientific understanding of an efficient way to overcome it, as reflected by the low success rate of short- and long-term weight loss interventions. From a clinical standpoint, this suggests that the body tends to defend a set point of possible genetic origin in the context of a weight-reducing program. As described in this paper, this limited therapeutic success may depend on adaptive thermogenesis, which represents in this case the decrease in energy expenditure (EE) beyond what could be predicted from the changes in fat mass or fat-free mass under conditions of standardized physical activity in response to a decrease in energy intake. This issue has been documented in recent studies that have shown in obese individuals adhering to a weight reduction program a greater than predicted decrease in EE, which in some cases was quantitatively sufficient to overcome the prescribed energy restriction, suggesting a role for adaptive thermogenesis in unsuccessful weight loss interventions and reduced body weight maintenance. As also discussed in this paper, this ‘adaptive thermogenesis’ can be influenced by environmental factors, which have not been frequently considered up to now. This is potentially the case for plasma organochlorine concentration and oxygen desaturation in obstructive sleep apnea syndrome. It is concluded that health professionals should be aware that in some vulnerable individuals, adaptive thermogenesis can be multi-causal, and has the capacity to compensate, at least partly, for the prescribed energy deficit, possibly going beyond any good compliance of some patients.
The accumulating evidence for the existence of mechanisms that regulate body energy reserves and their responsiveness to stressors of different nature is bringing the nutrition specialists up to a point where they might need to revisit obesity treatment. Indeed, the regulatory systems that allow the maintenance of a relatively stable body weight throughout life could also be the most important threat to the capacity to overcome obesity, once it is established. Currently, higher energy intake and/or lower physical activity-induced energy expenditure (EE) are both considered to be responsible for the state of positive energy balance that characterizes a large proportion of Westerners, directly contributing to the epidemic of obesity.1 Despite the fact that weight loss programs are centered on modification of both sides of the energy balance equation, that is energy intake and physical activity EE, long-term success is low.2 This is probably explained at least in part by the failure to maintain a regular exercise regimen and to follow nutritional advice on what constitute a healthy dietary regimen.2 However, as presented herein, adaptive thermogenesis, which is described as the decrease in EE beyond what could be predicted from the changes in fat mass or fat-free mass under conditions of standardized physical activity in response to a decreased energy intake, could represent in some individuals another factor that impedes weight loss and compromises the maintenance of a reduced body weight. Indeed, it has been well demonstrated that the decrease in EE during energy restriction and weight loss programs can be greater than expected from the decrease in fat mass and fat-free mass,3, 4, 5 despite these two variables accounting for over 82% of the variance in EE.6, 7, 8, 9, 10 This paper thus specifically focuses on an adaptive reduction in EE in response to energy restriction that can be observed in any component of total EE,3, 4, 11, 12 traditionally divided into resting and non-resting EEs (including physical activity and non-exercise activity EEs), and thermic effect of food.
Adaptive thermogenesis: a historical perspective
The lack of consideration of adaptive thermogenesis in the clinical context comes from the fact that during a period of more than 80 years, research has failed to clearly identify its significant role in the regulation of human body weight. Few decades have passed now, as the first studies to test experimental starvation in human reported a decrease in EE above that explained by fat mass and fat-free mass losses,7, 13, 14, 15, 16, 17 suggesting the presence of an adaptive component of thermogenesis in response to decreased energy intake. However, controversy arose over the years with regards to demonstrating this phenomenon in opposite circumstances, that is an increase in EE beyond expectation during overfeeding. At the beginning of the 20th Century, Neumman,18 followed by Gulick,19 demonstrated weight stability despite significant quantitative variations in the diet, and proposed the term luxusconsumption to described component of EE, which would explain the alterations of metabolism during overfeeding and would be responsible for the discrepancy between energy intake and body weight changes. In 1931, Neumman and Gulick's hypothesis was contradicted,20 and 50 years later their studies were re-examined by Forbes et al.21 who also had a contradictory interpretation, suggesting (in opposition to the luxusconsumption theory) a predominant role for energy intake in body weight maintenance.
These opposing conclusions well illustrate the important controversy that has animated obesity research, namely to either consider a defect in thermogenesis or an excess of ingested calories as the leading cause for a positive energy balance and long-term development of obesity.22 Despite convincing evidence for a role for brown adipose tissue thermogenesis in the development of obesity in experiments from various types of obese rodents and on rodents exhibiting voluntary overfeeding through the provision of a cafeteria diet,23, 24 the same does not seem to be true in humans.25 Therefore, this paper will argue that adaptive thermogenesis is an adaptive component of EE, which can be significant during food deprivation. Indeed, it is suggested that adaptive thermogenesis could be quantitatively more important than what is generally perceived by health professionals and nutrition specialists. We therefore propose to revisit the concept of adaptive thermogenesis from a clinical point of view, as a factor that could be of a quantitative significance in relation to its capacity to modify the outcome of a weight loss intervention.
Characterization of adaptive thermogenesis in the clinical context
The Minnesota study16 figures among the early demonstration of an adaptive decrease in thermogenesis in response to energy restriction. In this study, EE, body mass, and body composition were measured in 32 men subjected to semi-starvation (half of their energy needs) during 24 weeks.7 Comparison with baseline control period showed that fat-free mass and fat mass loss-adjusted basal metabolic rate was reduced by ∼20 and ∼25% after 12 and 24 weeks of starvation, respectively, representing a ∼1490 and ∼1700 kJ/day greater than predicted decrease in basal metabolic rate. More recently, the study of Leibel et al.4 contributed substantially to the quantification of adaptive thermogenesis, also in circumstances of severe energy restriction and body weight changes. This study described the changes in EE in non-obese and obese individuals subjected to a 10% body weight gain and to a 10 and 20% diet-induced (800 kcal/day) body weight loss.4 Total and resting EEs were measured by indirect calorimetry using a respiratory chamber and a ventilated hood, respectively, at the beginning of the study and at each weight plateau after at least 14 days of weight stability. Predicted values of EE were also determined at each weight plateau with regression equations relating initial measures of EE to fat-free mass and fat mass at the initial body weight. The difference between measured and predicted EE was then calculated for each weight plateau and was found to be significant. Indeed, the adaptive thermogenesis termed ‘observed-minus-predicted total EE’ in this study was −244 and −301 kcal/day (∼1025 and 1264 kJ/day) following a 10 and 20% body weight loss, respectively, in the obese individuals.4
A significant adaptive decrease in thermogenesis has also been reported in less drastic conditions of energy restriction. Studies conducted in the clinical context of weight reduction programs in obese individuals have demonstrated that a decrease in energy intake averaging 2000–3000 kJ is sufficient to induce a greater than predicted decrease in EE. Indeed, in concordance with the observations reported above, an adaptive reduction in EE in the context of a weight loss intervention was reported in the study of Doucet et al.3 In that study, resting EE was measured by indirect calorimetry, and predicted resting EE was obtained based on a reference group regression. Similar to the method used in the study of Leibel et al., a regression equation relating, this time, the resting EE of control participants to their fat mass and fat-free mass in a context of weight stability, was used to assess the predicted resting EE of participants at baseline and at weeks 2 and 8 of the intervention. Comparison with the predicted values showed, in men, a decrease in resting EE, which was ∼469 and ∼953 kJ/day greater than what was expected based on the reference regression at weeks 2 and 8, respectively.3 In obese women, a similar pattern was observed, and the decrease in resting EE was ∼635 and ∼614 kJ/day greater than what was predicted from the reference regression.3 The greater than predicted decrease in resting EE after a 10% weight loss in men in this study (∼622 kJ/day)3 was of the same magnitude as that reported by Leibelet al.4 (137 kcal/day, ∼573 kJ/day) for a similar decrease in body weight. These studies thus showed that a significantly greater than predicted decrease in EE in humans is present during periods of energy restriction before any major changes in body weight3 as well as during short-term maintenance of a significant body weight loss.4
The existence of adaptive thermogenesis in circumstances of long-term maintenance of reduced energy intake and body weight was also observed in the Biosphere 2 study.26 The Biosphere 2 study provided the possibility of describing the variations in EE of five participants after a 2-year period of energy restriction, in comparison with a group of control individuals matched for height and body weight.26 In that study, EE was measured by indirect calorimetry with a whole-body respiratory chamber, 18 months after participants had lost 15% of their initial body weight and maintained it. Participants were thus in condition of stable body weight at the time measurements were taken. It was found that participants' sleeping metabolic rate (defined as the average EE from all 15-min periods between 2330 and 0500 hours during which the spontaneous physical activity was <1.5%) was ∼ 400 kJ/day (∼ 7%) lower than the control group sleeping metabolic rate after correction for age, sex, fat mass and fat-free mass.26 Therefore, this study showed that an adaptive reduction in EE persists during maintenance of a reduced body weight. Although this study was conducted in normal weight individuals, other studies in weight-reduced, formerly obese individuals have also shown that these individuals have a lower resting metabolic rate for their body size and composition than never-obese controls.27 This suggests that the adaptive decrease in thermogenesis could be a factor that complicates long-term weight loss maintenance.
The variation in EE on an individual basis remains a relevant issue in the clinical context of obesity treatment, as changes in daily EE in response to weight gain or loss varies substantially among individuals.28 That some experience relatively large overcompensatory responses and other low or absent responses28 might be partly explained by genetic factors29, 30 and suggest that adaptive thermogenesis might be a threat to weight loss success for some individuals more than others. In this regard, the deviation from the predicted change in resting EE in both the highest male and female responder from the study of Doucet et al.3 (unpublished data) are presented in Table 1. In these two participants, the greater than predicted decrease in EE after 8 weeks of weight reduction program was nearly as important as the prescribed daily reduction in energy intake (−2900 kJ/day). These observations suggest that the adaptive decrease in thermogenesis could in some cases be sufficiently important to overcome the prescribed energy restriction, probably leading to incapacity to further lose weight. On the other hand, to be able to further lose weight, one strategy could be to adopt a very low calorie diet. However, this type of diet has been associated with weight regain in a 1-year period upon return to a low calorie diet,31 thus again suggesting that adaptive thermogenesis could limit the efficiency of weight reduction programs and also complicate weight loss maintenance.
Adaptive thermogenesis in physical activity EE
Physical activity, through enhancement of physical activity-related EE, elevation of post-exercise basal metabolism and preservation of lean mass can compensate, at least partly, for the decrease in EE in response to low energy intake.32 It could therefore be assumed that physical activity would also prevent the greater than predicted decrease in EE in this situation. However, some studies did not support this notion. Lazzer et al.33 reported the outcome of a weight reduction program that included aerobic and anaerobic exercises in addition to a moderate energy restriction diet in obese adolescents. Although this 9-month program was successful in decreasing body weight, indirect calorimetry measurements revealed a significant decrease in basal (−6.3%), sedentary (−12.6%) and sleeping EE (−11.7%) adjusted for fat-free mass and/or fat mass (P<0.001) compared to initial values (pre-weight loss).33 Moreover, it was found at the end of the program that sleeping EE and sedentary EE adjusted for fat-free mass were significantly lower in obese adolescents compared to non-obese adolescents tested under similar activity program, despite the former being still characterized by a greater body weight and fat-free mass. Based on these results, it seems that energy restriction combined with exercise is also accompanied by a greater than predicted decrease in EE.33
In their study, Lazzer et al.33 measured the change in physical activity-induced EE and found, in concordance with other studies,11, 33, 34 that it was significant and in proportions unaccounted for by the change in body mass. Indeed, the energy cost of physical activity (walking) adjusted for body weight and fat-free mass was significantly decreased at the end of the intervention in boys (−17.6%).33 This resulted in a significantly lower daily EE in adolescents for the same physical activity program than before weight loss (11.84 vs 14.76 MJ/day).33 Doucet et al.3, 11 also measured the exercise EE in men before and after they had successfully completed a 15-week energy restrictive weight loss intervention. EE was measured by indirect calorimetry, whereas subjects were exercising (walk on a treadmill) and predicted EE was assessed with a regression equation based on age, fat mass and fat-free mass as determinants of EE of a weight stable control group tested in a similar physical activity protocol. The 11% decrease in body weight at the end of this intervention was accompanied by a significant decrease in net exercise EE, so that measured exercise EE was significantly lower than predicted (15.5 vs 17.3 kJ/min, respectively).11 Concordant results were obtained in a study measuring the impact of a long-term energy deficit induced by endurance training on variations in EE. In that study, participants were submitted to a 93-day training period during which energy intake was maintained at the pre-training level in order to induce a 14.2MJ/day energy deficit. However, despite a negative energy balance being held theoretically constant by the exercise program, the body weight loss decreased, corresponding only to 65% of the net energy deficit by the second half of the training period.34 This discrepancy between energy deficit induced by the diet and the energy equivalent from change in body mass suggest the presence of an adaptive decrease in thermogenesis. Therefore, these studies suggest that interventions combining exercise and energy restriction in order to amplify the energy deficit might not observe the expected decrease in body weight, as in that condition, an adaptive decrease in thermogenesis occurs both in the active and resting states.
In the above studies, the adaptive thermogenesis was perceived as being a lower than predicted decrease in total 24-h,4, 26 resting,3, 4, 35 and exercise EEs11, 33, 34, 36 compared to the subjects' initial values or compared to a control group, in short- and long-term energy restriction studies (6 weeks to 2 years) involving obese and lean individuals. Although differences in study settings and methodologies make it difficult to quantify precisely the adaptive decrease in EE, the interpretation of these results nevertheless suggests that the adaptive decrease in thermogenesis reported was quantitatively important. Therefore, it suggests that energy restrictive diet, whether combined or not with an exercise program, represents a situation where a significantly greater than predicted decrease in EE occurs. This raises the sensitive issue that in a clinical context of obesity treatment, adaptive thermogenesis could be, in some individuals, a significant threat to the successful outcome of a weight loss intervention on a short- or long-term basis.
Determinant factors associated with the adaptive thermogenesis
The adaptive component of thermogenesis that has been documented under conditions of negative energy balance is under the influence of hormones and sympathetic nervous system activity that have been shown to explain variations in EE beyond what could be explained by changes in body weight and composition.9, 37, 38, 39, 40, 41, 42, 43, 44 Indeed, leptin,11, 39, 45 insulin,46, 47, 48 thyroid hormones9, 38, 41, 43, 45 as well as sympathetic activity37, 41, 42, 45 have been shown in several studies to be associated with a greater than predicted variation in EE. Moreover, fat depletion per se has also been considered as a determinant factor for adaptive thermogenesis.7, 49 It is, however, not the aim of this paper to cover extensively EE regulatory systems, and readers can refer to several papers that have been written on the topic.5, 42, 48, 49, 50, 51, 52, 53, 54, 55 Similarly, cellular mechanisms on the basis of human body thermogenesis, such as mitochondrial adenosine triphosphate synthesis efficiency,56, 57 uncoupling proteins53, 58 and their related increased muscle efficiency59 will not be covered in this paper, as they have been directly addressed in their respective cited references. This conceptual paper will rather describe three conditions that have been associated to adaptive thermogenesis and that may thus represent limiting factors in weight loss, and confer a significant susceptibility to weight gain/regain in some individuals: body weight loss and regain cycles, organochlorine plasma concentration and hypoxia in severe form of obstructive sleep apnea. These last two factors have not been traditionally associated with energy metabolism and are therefore more frequently than otherwise not considered in the clinical context of obesity treatment, despite demonstration in some studies of their potential to influence EE.12, 60, 61, 62
Body weight loss and regain cycles
Studies that have been conducted in animals to assess the effect of body weight loss–regain cycles on energy metabolism have reported a decrease in EE above that explained by body weight loss,63, 64 an increased caloric efficiency (weight gain/kcal food intake),65, 66, 67 an increased and a decreased time required losing and gaining body weight, respectively,65, 66 and a reduced resting rates of O2 consumption.68 This suggests the presence of an adaptive response in EE to body weight instability. In order to further investigate this issue and to better characterize changes in EE in response to exceptional variations in body weight, two men leaving for extreme expeditions were tested in the whole-body indirect calorimeter (metabolic chamber) in Laval University. Subject 1 was tested before departure for a 65-day cross-country ski expedition in Antarctica during which he lost 13.2 kg, and 18 weeks after the expedition once he had return to his initial body weight and composition. Thus, both pre- and post-expedition anthropometrical and EE measurements were taken under conditions of standardized diet, body weight and composition. Nevertheless, the subject showed a 10% lower 24-h EE compared to pre-expedition values (Table 2). Similar results were obtained with subject 2 who lost 8 kg during a 3-week expedition through Greenland.69 Indeed, despite recovery of his initial body weight and composition 12 days after the expedition, EE was also reduced compared to pre-expedition levels, under standardized conditions of food intake and activity (Table 2). Moreover, EE measured 3 and 16 weeks after the expedition remained below the pre-expedition values (results not shown). These two case studies demonstrate that in lean and healthy man, one (short) cycle of body weight loss–regain can induce a substantial greater than predicted from body composition decrease in daily energy needs, suggesting that body weight variations per se seem to be a determinant of adaptive thermogenesis. Interestingly, the extreme variations in body weight experienced by the two explorer men (∼14 and 12% from initial body weight, respectively) share similarities with the targeted weight loss proposed to obese individuals undergoing energy restriction programs. These results thus suggest in complement to the study of Wadden et al.,70 that the adaptive decrease in thermogenesis might occur during the first weight loss–regain cycle. Therefore, unsuccessful weight loss maintenance could be associated, upon return to initial body weight, with a lower than predicted EE as observed in animals,66 which could predispose to further weight gain, or difficulty trying to lose weight again. This suggests that success in the first attempt to lose weight might be primordial, enhancing the relevance of considering adaptive thermogenesis in the clinical context of obesity treatment.
Organochlorines: obesogen pollutants?
Organochlorines are chemical products that accumulate in the fat of organisms due to their lipophilic properties,71 and humans are particularly susceptible to accumulate organochlorine compounds via consumption of animal products such as fish, meat and dairy products.72 In humans, body mass index and fat mass have been positively associated with plasma organochlorine concentration,73 obese individuals being characterized by higher plasma organochlorine concentration than lean persons.62, 73 Moreover, lipolysis is positively associated with an increase in plasma organochlorine concentration74, 75 leading to an hyperconcentration of these pollutants in plasma and tissues during body weight and fat mass loss.61, 74, 76, 77, 78 Indeed, it was found that men engaged in a weigh reduction program had a significant increase in plasma and in abdominal and femoral subcutaneous adipose tissue organochlorine concentration after a significant body weigh loss.61, 77 Similarly, (1, 1-dichloro-2, 2-bis (p-chlorophenyl) ethene) DDE concentration, an organochlorine compound, was found to be increased by 61 and 55% in subcutaneous and in abdominal adipose tissue, respectively, following a significant body weight loss after an intestinal bypass operation in obese individuals.76
What is more of a concern regarding adaptive thermogenesis is the fact that release of these organochlorine compounds into the circulation during weight loss has been associated with decreased EE. Indeed, in the context of a 15-week weight reduction program, an increase in plasma organochlorine concentration has successively been shown to be the best predictor of the significant decrease in resting metabolic rate61 and of the significant greater than predicted decrease in sleeping EE,12 explaining 32 and 47% of these variables, respectively (see Table 3). Interestingly, in this last study,12 this corresponded to a greater percentage than what change in plasma leptin concentration (20%) and plasma leptin concentration after weight loss (13%) did explain. These relationships are surprising but could nevertheless be explained, at least partly, through the relation between plasma organochlorine concentration and lowered plasma-free thyroxine and total thyroxine levels that have been reported in several human studies.79, 80, 81 In this regard, an increase in plasma organochlorine concentration during body weight loss was shown to be inversely correlated with changes in tri-iodothyronine (T3) serum concentration61 and skeletal muscles enzymes activity.82 Therefore, it suggests that an increase in plasma organochlorine concentration could partly account for the adaptive decrease in thermogenesis during body weight loss, possibly through a deleterious action on thyroid hormone status. Based on studies underlining the role of pollutants in the development of obesity,83 the potential of organochlorine as other obesogen substances certainly remain an area of interest.
Obstructive sleep apnea syndrome: the energetic consequences of severe oxygen desaturations
Obstructive sleep apnea syndrome (OSAS) is a highly prevalent disease in overweight and obese individuals.84 The relationship between OSAS and propensity to weight gain and obesity85, 86 seems therefore paradoxical based on the fact that this disease is also characterized by several energy consuming factors such as sleep deprivation, sleep fragmentation, increase in breathing effort and motor activity,87, 88, 89 as well as with activation of sympathetic activity90, 91 and increased EE.92, 93 In order to better investigate this issue, eight obese individuals with positive diagnosis for OSAS (determined by continuous nocturnal home oxymetry) were tested in the whole-body indirect calorimetric chamber in Laval University.60 EE was measured as described previously94 and the severity of OSAS was evaluated by the percentage of total recording time spent below 90% arterial oxygen saturation (% TRT <90% SaO2).60 Predicted daily and sleeping EE were calculated with an equation obtained from multiple regression analysis in a control group of a priori non-apneic individuals. No difference between predicted and measured total and sleeping EEs were observed. However, the severity of the hypoxic stress (% TRT <90% SaO2) in the apneic participants was negatively correlated with the difference between mean measured and predicted daily EE (Figure 1). A similar trend was observed with sleeping EE, indicating that the greater the time spent in hypoxia during the night, the greater the adaptive decrease in thermogenesis.60 Following these observations, another study was conducted in the same laboratory with a larger group of men in whom OSAS was diagnosed with polysomnography. In agreement with Hins et al.,60 a negative correlation between sleeping EE expressed on a per kg body weight basis (kcal/kg) and the % TRT <90% SaO2 was observed in men classified in the upper quartile for this variable (r=−0.93, P=0.02) (Major et al., abstract, Can Resp J). These results thus show that despite displaying an apparently similar or increased EE compared to healthy individuals, apneic patients are characterized by lower relative EE with increasing severity of their disease. In other words, patients who spend most of the time in severe oxygen desaturation during sleeping time are also the one characterized by the lower than predicted level in EE, indicating that repetitive intermittent hypoxia could contribute to adaptive thermogenesis. This suggests that even in the absence of energy restriction and its associated decreased EE, patients with severe form of OSAS might nevertheless display a decreased thermogenesis.
OSAS is characterized by repetitive apnea and hypopnea, which are the consequence of recurrent episodes of complete or partial upper-airway collapses/occlusion.95 These episodes create a decrease in oxygen saturation,96 which put the body in a hypercapnic and hypoxic state. This results in an increased sympathetic activity during sleeping and awake state,91, 97, 98 which has been shown to be independent of the obesity that often accompanied OSAS.98 In this regard, it has been shown that OSAS is associated with a reduced vascular response to α- and β-1 and β-2 receptor stimulation,99 which suggests that a continuous exposure to sympathetic flux results in the development of a compensatory resistance to sympathetic nervous system stimulus. As β-1 and β-2 adrenoceptors are involved in the sympathetically mediated variations in thermogenesis,100 the compensatory resistance to sympathetic stimulation that might be developed as a result of the repetitive episode of desaturation could explain, at least partly, the adaptive decrease in thermogenesis in severe apneic individuals. This is a clinically relevant issue, as it could in turn complicate the management of this disease for which weight loss happens to be an efficient therapy.101
In conclusion, based on studies that have shown a greater than predicted decrease in EE under energy restriction circumstances, this review presented arguments in support of the potential of adaptive thermogenesis to impede obesity treatment on a short- and long-term basis, at least in some individuals. In some cases, the adaptive decrease in thermogenesis was shown to be significantly related to a single cycle of body weight loss and regain, an increase in plasma organochlorine concentration following weight loss, and a lower than predicted EE was also shown to be associated with severe nocturnal oxygen desaturations in OSAS. This suggests that energy metabolism might be sensitive to stimuli of different physiological nature and that adaptive thermogenesis could be quantitatively more important than what is generally perceived by health professionals and nutrition specialists. However, from a clinical point of view, several issues remain to be investigated in order to more clearly identify adaptive thermogenesis determining factors and to develop strategies to cope with them. Along these lines, it is concluded that unsuccessful weight loss interventions and reduced body weight maintenance could be partly due, in some vulnerable individuals, to the adaptive thermogenesis, which is multicausal, quantitatively significant, and has the capacity to compensate for a given prescribed energy deficit, possibly going beyond any good compliance of some patients.
Geneviève C Major is a recipient of a studentship from the Heart and Stroke Foundation of Canada in collaboration with the Canadian Institute of Nutrition, Metabolism and Diabetes and of Cancer Research, and the Canadian Diabetes Association. Angelo Tremblay is partly funded by the Canada Research Chair on Physical Activity, Nutrition and Energy Balance. Paul Trayhurn is funded by grants from the BBSRC (UK) and the European Union (Ob-Age: OLK6-CT-2002-02288). Eric Doucet is a recipient of a CIHR/Merck-Frosst New Investigator Award and CFI/OIT New Opportunities Award.
About this article
Fluoxetine induces lean phenotype in rat by increasing the brown/white adipose tissue ratio and UCP1 expression
Journal of Bioenergetics and Biomembranes (2015)