Brown adipose tissue (BAT) is profoundly involved in the regulation of energy balance and the control of body weight, which are mediated by non-shivering thermogenesis in mammals [1,2]. As a main mediator of adaptive thermogenesis, BAT is highly dependent on the activity of uncoupling protein 1 (UCP-1), which represents the mitochondrial activity [3]. Previous studies have demonstrated that BAT is also important for human metabolism [4,5].

Resveratrol (RSV), which has biological activities similar to those of estrogen [6], is a polyphenolic phytoalexin that is known to have several effects on metabolic diseases, cardiovascular disease, and malignancies.

With respect to energy metabolism, RSV increased heat production in mice, which was accompanied by a prolonged life span and improved insulin sensitivity [7]. Furthermore, RSV had inhibitory effects on adipokine expression and secretion in human white adipose tissue [8]. However, there have been few reports demonstrating the exact role of RSV on BAT. Although Miranda et al. [9] reported the apoptotic role of RSV on BAT, Andrade et al. [10] demonstrated that RSV promoted the biological function of BAT through increasing SIRT1 and energy expenditure.

Although white adipose tissue has major functions in the development of obesity, recent studies have demonstrated that BAT has an important role as a modulator of metabolic disease. However, there are few reports on agents which could modulate the activity of BAT, especially in patients with type 2 diabetes. With this background, we evaluated the role of RSV on the mitochondria of BAT, both in vivo and in vitro. The protein levels of UCP-1, which represents the mitochondrial activity of BAT, were evaluated in this study.

Adipose tissues, both white and brown, play pivotal roles in energy homeostasis. While white adipocytes store energy and cause obesity, BAT induces non-shivering thermogenesis and consumes excess calories [13]. Given the role of BAT in energy expenditure, several studies have suggested BAT as a new therapeutic target for obesity. For thermogenesis, BAT utilizes a unique protein, UCP-1, which is located in the inner mitochondrial membrane and uncouples mitochondrial respiration, leading to the release of energy as heat [14]. Together with UCP-1, peroxisome proliferator-activated receptor coactivator 1α (PGC-1α) has been show to express multiple physiologic roles for mitochondrial biogenesis and activity [15]. Several studies have demonstrated that PGC-1α were expressed in BAT and were dysregulated in obesity, diabetes and neurodegeneration.

There are several physiologic regulators of the differentiation and activity of BAT, including cold exposure, the sympathetic nervous system, the β-adrenergic system, catecholamines, thyroid hormones, nutrition, and sex hormones [13]. In this study, we confirmed that over-nutrition and RSV play important roles in the mitochondrial differentiation of brown adipocytes. The HFD itself augmented the number of brown adipocytes in OLETF rats. There have been debates about the changes in brown adipocytes during the development of obesity. In some studies, both impaired mitochondrial activity and a reduced amount of brown adipocytes were observed in obese animal models and human patients [4,16,17]. On the other hand, other studies have found that the differentiation of brown adipocytes correlated with increased supplement of nutrition [18]. In this study, we used OLETF rats to confirm the role of brown adipocytes in type 2 diabetes combined with obesity. OLETF rats were generated with inactive cholecystokinin-1 receptors on the background of Long-Evans Tokushima Otsuka rats, leading to the phenotype of type 2 diabetes with mild obesity. To eliminate the genetic influence of OLETF rats on brown adipocyte, the brown preadipocytes were isolated from Sprague Dawley rats for in vitro study. As described in this study, the HFD led to a significant increase in body weight and induced the differentiation of brown adipocytes. However, the HFD itself did not affect the mitochondrial activity of brown adipocytes. Although we could not find the exact mechanism, impaired biogenesis could have reduced mitochondrial activity, despite the increased differentiation of BAT after HFD intake [19].

Many studies have reported the effects of RSV on obesity [20,21,22]. Although most of them have focused on changes in white adipocyte, a few recent studies focused on the role of BAT or the 'browning' of white adipocyte [23,24]. They evaluated the development of brown-adipocyte-like characteristics in white adipocytes, through the main mechanism of SIRT1 activator combined with phytoestrogen activity [25]. Furthermore, Miranda et al. [9] reported that RSV leaded Akt/protein kinase B and foxo1 dephosphorylation/deacetylation and induced apoptosis of brown adipocyte. In the present study, RSV restored both the differentiation and activation of BAT and increased the expression of ER-α, especially in the HFD group. Considering that a sex-dependent difference in BAT expression was reported even in human subjects, and RSV treatment of postmenopausal women had favourable effects on estrogen-associated metabolism [26,27,28], RSV might modulate the activity of brown adipocytes through the estrogen pathway. However, RSV had no effects on white adipocytes, which were suggested by previous studies to be the target of the anti-obesity effects of RSV [29,30]. In our obesity-related inflammatory marker analysis, there were no differences in the levels of leptin and adiponectin following RSV treatment. Considering that the histologies and organ weights of the heart, liver, epididymal fat, retroperitoneal fat, gastrocnemius muscle, and soleus muscle did not differ with the RSV treatment, improved insulin sensitivity may have been achieved through the increased differentiation and activation of BAT. The differences in body weight were diminished after 19 weeks of age in the OLETF rats, possibly due to the relatively long study duration, the high dose of the HFD treatment, and differences in animal species for type 2 diabetes.

Andrade et al. [10] demonstrated that early treatment of RSV on mice fed with SD increased BAT thermogenesis markers by increasing SIRT1 and energy expenditure. However, Miranda et al. [9] reported that RSV induced apoptosis of apoptosis. The differences might be resulted from the time-point of RSV treatment. In our study, the effect of RSV on brown adipogenesis was confirmed at each step during in vitro study. Although RSV treatment was performed at different stages of adipogenesis, only the early treatment before the application of IM significantly affected the differentiation and activity of BAT. This suggests that RSV treatment might have a role in the early development of brown adipocytes with more prominent mitochondrial activities than those induced by obesity.

There are several limitations of current study. Although mitochondrial activities were evaluated by UCP-1, the role of RSV on mitochondrial biogenesis of BAT, such as PGC-1α, was not demonstrated. Furthermore, the direct mechanism of RSV involving mitochondrial activation of BAT was not clarified. Finally, the effect of RSV on BAT thermogenesis was not defined in in vivo condition using obesity animal model. However, we identified the relationship between RSV and mitochondrial activities of BAT in current study, both from in vitro and in vivo conditions.

In conclusion, we confirmed that RSV might have a role on promotion of mitochondrial activity in BAT, both in vitro and in vivo. In the in vivo study, RSV restored the impaired mitochondrial activity induced by the HFD, increased the expression of ER-α, and improved insulin sensitivity. Further studies evaluating the relationship between mitochondrial activity and ER-α could be helpful in understanding the anti-obesity effects of RSV.