Adipose tissue is specialized connective tissue that plays an important part in controlling whole-body energy homeostasis. Two types of adipose tissues have been classified in mammals: white adipose tissue (WAT) and brown adipose tissue (BAT).

White adipose tissue is the predominant form of adipose tissue that is composed of single lipid droplet fat cells. WAT is found throughout the body in the subcutaneous and visceral regions, and also in loose connective tissue. WAT acts as an energy reserve by absorbing and storing fatty acids from the blood in lipid droplets, and synthesizing triglycerides de novo from glucose. This energy reserve can be tapped to meet the body's energy demands when energy expenditure exceeds energy intake. Additionally, WAT has an endocrine function because it secretes hormones, growth factors and cytokines such as leptin, TNF-a, and interleukin 6 that are involved in maintaining metabolic homeostasis.

Evolutionarily, WAT is important in times when food is scarce; however, modern day sedentary life style couples with increase in intake of high caloric food have perturbed this energy balance causing energy intake to be greater than energy expenditure. This energy imbalance causes an expansion of white adipocytes. These expanded white adipocytes are found in obese individuals and they affect the secretion of adipokines resulting in changes in metabolic homeostasis and associated pathologies such as insulin resistance. Obesity is thus a public health concern because it is a risk factor for many diseases, including type 2 diabetes, certain cancers, and cardiovascular diseases.

In contrast to WAT, the principal function of BAT is to burn energy to generate heat. BAT is composed of multilocular brown adipocytes and high volumes of blood vessels, nerves, and mitochondria. BAT produces and secretes hormones and cytokines but in lesser amounts than WAT. BAT is found in newborns at birth and its amount significantly decreases as humans mature. However, recent studies have detected depots of BAT in defined locations in adult humans {{95 Cypess,A.M. 2009; 11 Nedergaard,J. 2007; }}(4, Cypress and 10, Nedergaard).

Brown adipocytes are characterized by the densely packed mitochondria containing uncoupling protein-1 (UCP-1) in their inner mitochondrial membrane. UCP-1 is a member of the mitochondrial carrier protein family that is found exclusively in brown adipocytes. UCP-1 protein uncouples oxidative phosphorylation so that the energy generated by the proton motive force dissipates as heat. This type of heat production in BAT is known as nonshivering thermogenesis, and it is a crucial defense against hyperthermia in rodents and in human infants. UCP-1 knockout studies demonstrated that mice lacking UCP-1 were unable to generate heat via nonshivering thermogenesis, and were therefore profoundly intolerant to cold environment {{104 Enerback,S. 1997}}(Enerback).

BAT also depends on the extensive vascular bed for heat distribution and the abundance of nerves for its sympathetic control. Studies have shown that chronic cold exposure, which is sensed in the CNS, stimulates the release of catecholamines from brown adipocytes causing an expansion of these cells (Figure 1) {{12 Cannon,B. 2004 }}(7, Cannon and Nedergaard and more REFs). Furthermore, there is an increase in the expression of UCP-1 when catecholamines bind to b-adrenergic receptors in brown adipocytes {{12 Cannon,B. 2004; }} (Canon and Nedergaard).

In addition to its thermogenic property, BAT has also been recognized as having anti-obesity activity. Several BAT knockout studies using mice showed reduction in whole body energy expenditure and increase in the risk of obesity and other metabolic diseases {{103 Lowell,B.B. 1993; }}(Lowell 1993 and Hamann 1996). The heat producing capacity of BAT when it is working at maximal intensity can reached up to 300 W/kg, which is two orders of magnitude higher than the normal metabolic rate of a typical mammalian tissue (Cannon/Nedergaard) {{12 Cannon,B. 2004; }}. Moreover, brown adipocyte-deficient mice, but not UCP-1-ablated mice (C/N 200), became obese when fed a high-fat diet (C/N 462). This indicates that BAT as a unit modulates metabolic rate independent of UCP-1-mediated heat production (Cannon and Nedergaard) {{12 Cannon,B. 2004; }}. Other genetic studies demonstrated that mice gained less weight, were more insulin-sensitive, and had lower levels of free fatty acids when their BAT was stimulated to increase in both amount and function {{25 Cederberg,A. 2001; 27 Kopecky,J. 1995; 82 Xue,Y. 2008; }}(Cederberg 2001, Kopecky 1995, and Xue 2007). The knowledge of BAT as a thermogenic and anti-obesity modality couples with recent detection of BAT in human adults, has led to an increase in interest in BAT as a tool to combat obesity and other related metabolic diseases.

Brown Adipose Tissue: Thermogenic Property

Norepinephrine (NE) is the most important humoral factor influencing brown adipocytes. Several studies have demonstrated that NE binding to a b-adrenergic receptor (b-AR) stimulates thermogenesis {{12 Cannon,B. 2004; }}(REFs). Among the three b-AR subtypes, the b3-AR is the most significant adrenoceptor in mature brown adipocytes. b-ARs are Gs-protein coupled receptors. There are two forms of Gs proteins, GsaL and GsaS, found in brown adipocytes and other tissues. During brown preadipocyte differentiation into mature brown adipocytes, the GsaS variant increases while the expression of the GsaL subtype remains unchanged (C/N 72). Additionally, NE also stimulates a2-AR, which activates Gi inhibitory pathway in brown adipocytes {{12 Cannon,B. 2004; }} (Cannon and Nedergaard). This suggests that NE-stimulation is self-limiting because both the Gs and Gi can be stimulated depending on which adrenoceptors NE act (Figure 2) {{12 Cannon,B. 2004; }}(Cannon and Nedergaard).

The b-AR signaling cascade involves the activation of adenylyl cyclase (AC) by the G protein a-subunit. It is unclear as to which of the ten isoforms of AC is activated in brown adipocytes since multiple AC isoforms are expressed in BAT (C/N 125, 12). AC activation results in an increase in cAMP production, which leads to the activation of protein kinase A (PKA). PKA is a key player in the regulation of several downstream signals involved in brown adipocyte differentiation. PKA phosphorylates the transcription factor cAMP response element binding protein (CREB), which regulates the expression of UCP-1 (Figure 2) {{12 Cannon,B. 2004; }}(Cannon and Nedergaard). Additionally, PKA induces the activation of the p38 mitogen-activated protein kinase (p38 MAPK) pathway (C/N 107), which is also involved in UCP-1 gene expression.

Furthermore, the activation of lipolysis triggers the initiation of thermogenesis in brown adipocytes (Cannon and Nedergaard) {{12 Cannon,B. 2004; }}. Like thermogenesis, lipolysis is also stimulated through b3-AR (Figure 2) (C/N 25). The stimulation of lipolysis consists of two processes: 1) activation of hormone-sensitive lipase (HSL) and 2) deactivation of perilipin. Perilipin covers triglyceride droplets and protects them from being digested by HSL (C/N 56, 485). Activated PKA phosphorylates perilipin (C/N 124) causing it to dissociate from triglyceride droplets, thus exposing these droplets to HSL. Lipolysis of triglyceride droplets results in the release of glycerol and free fatty acids. Free fatty acids are transferred into the mitochondria via the general carnitine shuttle system and are then b-oxidized (b-ox) to release acetyl CoA (AcCoA), which gets further oxidized in the citric acid cycle (CAC) (Figure 3) {{12 Cannon,B. 2004; }}. Similarly to other cell types, the incorporation of acetyl CoA into the citric acid cycle results in the transfer of electrons through the respiratory chain. Unlike other cell types, brown adipocytes possess UCP-1 proteins, which utilize the proton motive force created by the transfer of electrons through the respiratory chain to generate heat rather than to synthesize ATP (Figure 3).

BAT synthesizes triglyceride de novo from glucose cleared from the circulation. Fatty acids made in brown adipocytes consist mostly of saturated and monounsaturated forms (Cannon and Nedergaard). Stimulation of BAT by NE increases the synthesis and release of lipoprotein lipase (LPL) from brown adipocytes (Figure 4) (C/N 108). NE also activates lipoprotein lipase activity, which leads to increased levels of intracellular free fatty acids available to be transported and combusted in the mitochondria (Figure 4) (C/N 108, 642). This contributes overall to reductions in fatty acids uptake and storage, and an increase in fatty acids metabolism. In order to combust triglycerides, an abundant amount of oxygen is required. Thus during peak thermogenesis, BAT consumes all the extra oxygen in the body that is not going to the brain or large muscle groups {{12 Cannon,B. 2004; }}. (Cannon and Nedergaard).

Additionally, BAT is one of the most insulin-responsive tissues because it can stimulate glucose uptake (C/N 766). BAT promotes the uptake of glucose when there is an elevated plasma insulin level (C/N 750). Insulin-induced glucose uptake in brown adipocytes is mediated by glucose transporter type 4 (GLUT4). Insulin upregulates the expression of GLUT4 (C/N 799), and induces the translocation of GLUT4 from intracellular stores to the plasma membrane of brown adipocytes (Figure 5) (C/N 748, 734). The glucose taken up by brown adipocytes is metabolized to provide either glycerol phosphate for triglyceride synthesis or 2-carbon units for de novo fatty acid synthesis (Figure 5) (C/N 774). The requirement of glucose uptake for lipid accumulation in BAT is supported by studies using transgenic mice lacking the insulin receptor in BAT. These transgenic mice displayed reduction in lipid accumulation as evident in a decrease in the weight of BAT, fasting hyperglycemia, and impair glucose tolerance without insulin resistance (C/N 289). These results demonstrate a major role of BAT in glucose clearance in mice that could not be compensated by glucose uptake in other tissues {{12 Cannon,B. 2004; }} (Cannon and Nedergaard).

Detection of Brown Adipose Tissue in Adults

Positron-emission tomography (PET) and computed tomography (CT) can now be used to detect BAT that has taken up the glucose analogue F-fluorodeoxyglucose (18F-FDG). In adipose tissue, 18F-FDG is taken up by GLUT4, and gets phosphorylated by hexokinase to 18F-FDG-6-phosphate once inside the cell (Ref). Phosphorylated 18F-FDG cannot leave the fat cell and is detected by PET (Ref). PET scan allows for the mapping of the points of origin of the photons emitted from the annihilation of the positron by the electron to produce two and three-dimensional maps of the body (Figure 6A). Multiple factors can impede the uptake of 18F-FDG by BAT, which include increase in ambient temperature {{101 Cypess,A.M. 2009; }} (4, Cypress 2009), anxiolytic agents such as benzodiazepines (4, 25/Gelfand 2005), beta-blockers (4, Cypress and 26/Parysow 2007), and certain diets (4, 20/Williams 2008).

In principle, 18F-FDG PET only monitors the uptake of glucose by tissues. CT is thus required for adipose tissue identification because it can distinguish between tissues with different density up to 1% due to its high contrast-resolution (Figure 6B) {{11 Nedergaard,J. 2007; }} (10, Nedergaard 2007). Other methods can also be performed to confirm the identification of BAT, such as tissue resection and histochemical staining for UCP-1 {{101 Cypess,A.M. 2009; }} (4, Cypress 2009). The most common locations of BAT detected in adult humans by 18F-FDG PET-CT are the cervical-supraclavical, ventral neck, mediastinal (10, 67/Truong 2004), paravertebral, para-aortic, and pararenal regions {{101 Cypess,A.M. 2009; 11 Nedergaard,J. 2007; }} (4, Cypress 2009 and 10, Nedergaard 2007).

Origin of the Brown Adipocyte

Adipose tissues are currently thought to originate largely in the somites, which are segmented epithelioid masses of mesoderm (Farmer 2008) {{33 Farmer,S.R. 2008; }}. The somite gives rise to the sclerotome and the dermomyotome. The sclerotome is a progenitor cell of the WAT lineage {{33 Farmer,S.R. 2008; }} (Farmer 2008). The dermomyotome gives rise to the myotome, which is myf5-expressing progenitor cell that can differentiate into either brown adipoblasts or myoblasts (Figure 7) (Farmer 2008) {{33 Farmer,S.R. 2008; }}. It must be noted that progenitor cells differ from multipotent stem cells in that they have lost their potential to differentiate into other cell types and are committed to follow a limited path.

White and brown adipocytes have long been assumed to share a common developmental origin because they both express genes that are involved in triglyceride synthesis and metabolism. They also undergo a similar program of morphological differentiation controlled by peroxisome proliferating-activated receptor g (PPARg) (7, 6/Rosen and Spiegelman 2000, 8/Tontonoz 1994, 15/Hamm 201). However, recent works have suggested that BAT is derived from the same progenitor as skeletal muscle, and is distinct from WAT not just in morphology and function, but also in developmental lineage. Timmons and coworkers showed that brown preadipocytes express muscle-related genes {{10 Timmons,J.A. 2007; }} (14, Timmons 2007). Atit and coworkers performed lineage-tracing experiments and showed that embryonic BAT, skin, and muscle were derived from cells expressing engrailed-1 (En1) in the dermomyotome (1, 36/Atit){{73 Atit,R. 2006; }}. However, a direct connection between muscle and BAT cannot be drawn from this study because En1 is expressed early in mouse development and marks a variety of cell lineages (Ref).

It was not until Seale and coworkers traced the fate of cells expressing the myoblast- associated Myf5 gene that the notion that BAT and skeletal muscle share a common progenitor was confirmed. Their study showed the detection of Myf5 expression in BAT and skeletal muscle, but not in WAT. Additionally, Crisan and coworkers demonstrated that multilineage progenitor cells expressing surface antigen CD34 from skeletal muscle of both fetal and adult tissues can differentiate in vitro into brown adipocytes expressing high levels of UCP-1; this differentiation was not observed in progenitors of WAT {{35 Crisan,M. 2008; }}(1, Crisan 2008).

Additionally, Seale and coworker showed that brown adipocytes arising from WAT in response to b-adrenergic stimulation were not derived from the Myf5-expressing progenitor cells. This suggests that WAT may contain brown fat-related genes, whose expression may be upregulated by an increase exposure to b-adrenergic stimulation (14, Farmer 2008) and also casts some doubt on the skeletal muscle lineage as the exclusive source of all normal BAT. It also indicates that brown adipocytes derived from WAT may have greater plasticity, i.e., the ability of differentiated cells to undergo transdifferentiation, than brown preadipocytes derived from the myogenic lineage.

Development of Brown Adipocytes from Mesenchymal Stem Cells

Pluripotent mesenchymal stem cells (MSCs) have the potential to develop into mesodermal cell types, including adipoblasts, myoblasts, osteoblasts, and chondroblasts. The enlargement of adipose tissue depots is a result of both an increase in the size and numbers of adipocytes. To increase the number of adipocytes, MSCs in the vascular stroma of adipose tissue are stimulated by a group of signals to become preadipocytes in a process called commitment differentiation (42, 3/Yu ZK 1997). During this stage, MSCs under go cell division resulting in one daughter cell becoming a preadipocyte while the other daughter cell remains a pluripotent stem cell, maintaining the stem cell population. NE can stimulate preadipocytes to undergo mitotic clonal expansion to further increase the number of cells committed to differentiate into mature adipocytes, (42, 4/Tang Q 2003 and C/N 77). The terminal differentiation from preadipocytes to mature adipocytes is activated by a second set of signals. This process illustrates how a single stem cell can produce large numbers of mature adipocytes (Figure 8). Parallel processes of capillaries and nerve terminals expansion also occur during the recruitment and proliferation of brown preadipocytes.

Transcriptional Control of Brown Adipocyte Differentiation


Bone morphogenic proteins (BMPs) are members of the transforming growth factor-b (TGF-b) superfamily. There are about 20 BMPs that have been discovered to date, and they all control multiple key steps of embryonic development and determination (27, Chen 2004). BMPs act on serine/threonine kinase receptors, which are composed of two subtypes, type I (BMPR1) and type II (BMPR2). Each subtype consists of three different receptors (27, Koenig 1994, Ten Dijke 1994, Rosenzweig 1995, Kawabata 1995). BMP binding causes the receptors to heterodimerize, forming a complex consisting of BMPR1 dimer and BMPR2 dimer (27, Moustakas 2002). The two major signaling pathways involving BMPs are SMAD (the vertebrate homologues of Drosophila mother against decapentaplegic ) pathway and the p38 MAPK pathway (3, 12/Canalis 2003). The SMAD pathway is stimulated by BMPR1. BMPR1 phosphorylates SMAD1, SMAD5, and SMAD8, causing them to form a complex with SMAD4, which then gets translocated into the nucleus and regulates gene expression (42, Huang). In contrast, BMPR2 phosphorylates and activates p38 MAPK (42, Huang). Both signaling pathways are important for the induction of MSC commitment toward adipocyte differentiation (42, Huang).

Among the members of the BMP family, BMP2 and BMP4 promote adipogenesis (40, Ahrens 1993 and Ji 2000; 41, Bowers). BMP2 acts on PPARg to induce MSCs differentiation into preadipocytes (40, Hata). Hata and coworkers demonstrated that BMP2 enhances the effect of PPARg and promotes adipocyte differentiation by treating MSCs with exogenous PPARg and BMP2. The mechanisms by which BMP2 affects PPARg are two-fold. First, BMP2 activates the SMAD1 pathway and induces PPARg expression (40, Hata). Second, BMP2 activates p38 MAPK and upregulates PPARg transcriptional activity (40, Hata). The steps following p38 MAPK activation have yet to be determined. It is unlikely that p38 MAPK acts by phosphorylating PPARg because the one MAPK phosphorylation site located on PPARg is at serine 112, whose phosphorylation has been shown to inhibit PPARg activity (Hu).

BMP4 is a secretory protein that remains associated with the cell extracellular surface and the extracellular matrix (41, Bowers). Its mRNA level was observed to increase in committed preadipocytes during proliferation (41, Bowers). Bowers and coworkers revealed that the presence of BMP4 is necessary for MSCs to achieve complete adipocyte differentiation. They treated proliferating preadipocytes with noggin, a naturally occurring BMP4 antagonist, and observed a complete blockade of terminal differentiation as indicated by the absence of triacylglycerol accumulation and PPARg2 expression (41, Bowers). Bowers and coworkers also discovered that BMP4 promotes adipogenesis by activating the SMAD signaling pathway (41, Bowers).

Previous studies have reported that BMP4 contains two promoters, 1A and 1B, that are differentially regulated during development (41, 25/26). Bowers and coworkers showed that 1A promoter is used to drive the expression of BMP4 in preadipocytes by day 3, and its usage stopped by day 7. In contrast, uncommitted MSCs utilized 1B promoter in the BMP4 gene at both day 3 and day 7 (41, Bowers). These findings suggest that by preferentially activating the 1A promoter region on the BMP4 gene using 5-azacytidine, a potent DNA methylation inhibitor, MSCs can be induced to develop into preadipocytes.

In addition to the function of BMPs as signals that stimulate MSCs to commit to become preadipocytes, BMPs also play a role in brown adipocyte differentiation. BMP2, BMP6, and BMP7 significantly inhibit the expression of Runx2, a key transcription factor associated with osteoblast differentiation (3, Tseng). This suggests that BMPs may promote adipogenesis by inhibiting osteogenic differentiation (3, Tseng 2008). BMP7 has specific functions in brown adipocyte differentiation because it upregulates the expression of UCP-1, PGC-1a, other brown fat selective genes, and mitochondrial genes (Figure 9) (3, Tseng 2008). Additionally, BMP7 significantly suppresses the expression of several adipogenic inhibitors, including necdin, PREF-1 and WNT10a (Figure 9) (3, Tseng 2008). This suppression results in an increase in the expression of PRDM16, a key transcription factor in brown adipocyte differentiation (Figure 9) (3, Tseng 2008).

The binding of BMP7 to BMPR activates the SMAD pathway in both brown and white preadipocytes (3, Tseng 2008). However, the activation of p38 MAPK and its downstream activating transcription factor-2 (ATF-2) only occur in brown preadipocytes (3, Tseng 2008). The p38 MAPK transduction pathway is important in the differentiation of brown adipoblasts because this signaling cascade induces the expression of PGC-1a and upregulates the expression of UCP-1 (21, Cao 2004). These results taken together confirm that BMP7 stimulates brown adipocyte differentiation by suppressing adipogenic inhibitors to induce the expression of PPARg, C/EBPa, and PRDM16, and by activating p38 MAPK to increase the level of PGC-1a and UCP-1 mRNAs.


Extensive studies in cell culture and in vivo have elucidated the pathway that controls the process of brown fat differentiation. The dominant regulator of both white and brown adipocytes formation is PPARg, a member of the peroxisome proliferator-activated receptor family (7, Tontonoz 1994, Rosen 1999, Nedergaard 2005). PPAR derives its name from the ability of the first-described member of the family, PPARa, to respond to various ligands and induce peroxisome proliferation. However, the other PPAR isoforms do not share this ability with PPARa. PPARg is a ligand-activated transcription factor that regulates gene expression, and it is highly expressed in both WAT and BAT during differentiation of preadipocytes to adipocytes (Sears 1996). Moreover, brown adipocytes have been demonstrated to increase in both number and size by stimulating PPARg with exogenous ligands, such as thiazolidinediones (17, 48/Berthiaume 2004).

PPARg appearance in BAT is well established in the rat embryo by day 18.5, after which it is found mostly in BAT and not in other tissues (17, 17/Braissant 1998). The PPARg gene gives rise to two different mRNAs, PPARg1 and PPARg2 (17, Nedergaard). PPARg2 comprises of at least 95% of all PPARg mRNAs in brown adipocytes, and it contains the full template including the translation initiation sequence for PPARg1 (7, 16/Lindgren 2004). Lindgren and coworkers detected the same levels of PPARg1 and PPARg2 proteins in brown adipocytes. They concluded that PPARg1 must derive from the PPARg2 template and translation must occur with equal probability at both PPARg1 and PPARg2 initiation sites (17, 16/Lindgren 2004). Even though PPARg1 and PPARg2 are induced during adipogenesis, it must be noted that PPARg1 is also found in other cell types such as colonic epithelium and macrophage (20, Rosen 2006).

Little is known about the functional differences of the two isoforms. However, several studies have demonstrated that PPARg2 plays a more important role in adipocyte differentiation. Ren and coworkers suppressed the expression of PPARg1 and PPARg2 using engineered zinc finger repressor proteins and found that overexpression of exogenous PPARg2, but not PPARg1, rescued adipogenesis (Ren 2002). Mueller and coworkers performed a genetic analysis using PPARg knockout fibroblasts and determined that both PPARg1 and PPARg2 have intrinsic ability to stimulate adipogenesis. However, at a lower concentration of ligand, PPARg2 was more effective at stimulating adipogenesis than PPARg1 (Mueller 2002). This is because PARg2 can bind more effectively to the TRAP coactivator complex than PPARg1 (Mueller 2002). The TRAP coactivator complex contains a subunit called TRAP220, which is a PPARg2 coactivator (28, Ge 2002), and the interaction between TRAP220 and PPARg2 is essential in PPARg2-mediated adipogenesis (28, Ge 2002).

PPARg protein has domains that are common to all nuclear hormone receptors. The C-terminal region of PPARg is responsible for dimerization with the retinoic-acid receptor (RXR), and it contains the major transcriptional activation region called the AF2 domain. Its N-terminal domain has transcriptional activity when linked to a DNA-binding domain, and it serves as the regulator of the receptor's biological function. Deletion of the N-terminal region leads to a greater transcriptional activity and an increased adipogenic activity (13, 11/Tontonoz 1994). In addition, phosphorylation of PPARg serine 112 residue by Erk 1 and 2, members of MAPK family, reduces PPARg ligand-binding affinity and transcriptional activity (13, 13/Huand 14/Shao 1998).

PPARg forms a heterodimer with RXR (13, 20/Kliewer 1992). Heterodimerization is required for the binding of PPARg to PPAR-response element (PPRE), which is located in the distal enhancer region of the UCP-1 gene (Figure 10) (Sears 1996). Nedergaard and coworkers demonstrated that a chronic exposure to PPARg agonists increases the level of UCP-1 gene expression. However, a PPARg knockout study showed no change in the level of UCP-1 gene expression (17, 51/He 2003). This suggests that certain genes, once activated, may not require the continued expression of PPARg and/or that other factors are also involved in regulating UCP-1 gene expression (17, 51/He 2003).


Coactivators and corepressors are accessory proteins that mediate gene activation or repression by acting on binding sites distant from the target gene. Most of these accessory proteins have very little specificity as to which nuclear receptors they interact with, and when and where they are expressed. Unlike most other accessory proteins, a coactivator discovered by Puigserver and coworkers called PPARg coactivator-1a (PGC-1a) showed tissue selectivity and physiological state specificity. PGC-1a is expressed in BAT but not in WAT. Its expression is upregulated by cold temperature (16, Puigserver 1998). Moreover, PGC-1a increases the expression of UCP-1 and mitochondrial genes (16, Puigserver 1998). These findings illustrate that PGC-1a plays an important role in cold-induced thermogenesis.

PGC-1a is a protein with 795 amino acids. Like most other coactivators and corepressors, PGC-1a contains a RNA-binding motif, two serine and arginine-rich domains, and an LXXLL motif. Whereas almost all known coactivators and corepressors utilize LXXLL sequence to mediate nuclear receptor-protein interactions (Heery 1997 and Torchia 1997), PGC-1a uses a proline-rich domain to bind to a region that overlaps with the DNA binding domain and the hinge region of PPARg (Ref: mostly from 16, Puigserver 1998). Another unique feature of PGC-1a is the three different sites for phosphorylation by protein kinase A (PKA).

PGC-1a is highly enriched in brown adipocytes, and its signaling cascade is mediated by p38 MAPK, as depicted in Figure 11 (21, Cao 2004 and Puigserver 1998). Stimulation of a b-AR by either exposure to cold temperature or bAR agonists increases cAMP production. Cyclic AMP activates PKA and PKA triggers the p38 MAPK signaling cascade. Activated p38 MAPK phosphorylates both ATF-2 and PGC-1a. Phosphorylated ATF-2 stimulates PGC-1a transcription and directly regulates UCP-1 gene expression by binding to one of the enhancer domains located in the UCP-1 gene, called cAMP-responsive element 2 (CRE2) (21, Cao 2004). Phosphorylated PGC-1a is recruited to PPRE on the UCP-1 promoter, and subsequently upregulates UCP-1 expression (21, Cao 2004). In addition, PKA acts independently of p38 MAPK to enhance UCP-1 transcriptional activity by phosphorylating and thus activating CRE binding protein (CREB) (21, Cao 2004). CREB in turn binds to the CRE4 region located proximal to the promoter of UCP-1 gene (21, Cao 2004).

It must be noted that while PGC-1a is important in regulating adaptive thermogenesis, its presence is not required for adipogenesis. The ability of brown adipocytes to respond to cold-induced thermogenesis is reduced in PGC-1a knockout cells (27, Uldry 2006). This is evident in the significant reduction of UCP-1 gene expression, which cannot be rescued by dibutryl-cAMP treatment (27, Uldry 2006). However, PGC-1a knockout preadipocytes are capable of differentiating into cells containing multilocular fat droplets (27, Uldry). The levels of UCP-1 and cidea mRNAs, which are markers of the brown fat phenotype, are similar in both wild-type cells and PGC-1a knockout cells (27, Uldry 2006).

A novel PGC-1 related transcription coactivator called PGC-1b was discovered in 2002 (Lin, 2002). PGC-1b mRNA is abundantly found in BAT, and its level is elevated during brown adipocyte differentiation (Lin, 2002). In contrast to PGC-1a, the expression of PGC-1b is not affected by cold exposure or treatment of cells with forskolin (Lin 2002). While little is known about the function of PGC-1b, Uldry and coworkers have established an important role of PGC-1b in brown adipocyte differentiation. The expression of the UCP-1 gene decreased by 90% during differentiation in cells lacking both PGC-1a and PGC-1b (27, Uldry 2006). Cells deficient in both PGC-1a and PGC-1b also showed a significant reduction in the expression of mitochondrial genes, such as cox4i and cox5b, and a complete loss in cytochrome c expression during cell differentiation (27, Uldry 2006). These findings confirmed that the presence of either PGC-1a or PGC-1b is essential in the development of normal brown adipocyte phenotype in differentiating cells.


FOXC2 is a member of the forkhead/winged helix transcription factor family. The highly conserved monomeric DNA-binding domain called the winged helix gives the forkhead/winged helix family its unique characteristic (30, 2/Gajiwala 2000). FOXC2 is restricted to the adipocytes of human adults, and it has been shown to promote the development of the brown adipocyte (29, Cederberg). FOXC2 transgenic mice have enlarged bilobed interscapular BAT depots and reduced intraabdominal WAT (29, Cederber). Transgenic expression of FOXC2 in WAT of mice stimulates the expression of UCP-1, mitochondrial, and PCG-1 mRNAs (29, Cederberg). This illustrates that FOXC2 has the ability to convert WAT to an adipose tissue expressing important brown adipocyte-selective genes (29, Cederberg).

FOXC2 transgenic mice exhibit lean, insulin-sensitive, and resistant to diet-induced obesity phenotypes (29, Cederberg). These phenotypes resulted from the upregulation of bARs, which causes an increase in cells sensitivity to bAR stimulation (29, Cederberg). There are two forms of PKA holoenzyme, type I and type II (30, Dostmann 1990). PKA type I binds to cAMP with a higher affinity than PKA type II, thus resulting in a faster activation of the cell signaling pathway (30, Dostmann 1990). Normally, PKA type II is expressed in WAT and BAT (30, Cummings 1996, and Dostmann 1990). However, the adipose tissue of FOXC2-expressing cells shows high levels of PKA type I expression (29, Cederberg and 30, Dahle). This finding demonstrates that FOXC2 increases cells' sensitivity to cAMP by upregulating the expression of PKA type I (Figure 11).

Additionally, FOXC2 stimulates vascularization in adipose tissue (32, Xue 2008). Adipose tissue requires continuous vascular remodeling to distribute heat and transport hormones and metabolites in order to accommodate the constant regression and expansion of its adipocytes (32, Xue 2008). Both WAT and BAT of FOXC2 transgenic mice contain a high volume of vascular networks consisting mostly of dense vascular plexuses (32, Xue 2008). FOXC2 stimulates the expression of many angiogenic factors, including VEGF, PDGF, and Ang. These angiogenic factors have all been reported to be involved in regulating vascular maturation, remodeling, and stabilization (32, 35/Cao 2002, 37/Vajkoczy 2002). It must be noted that the highest level of angiogenic factor detected in FOXC2 transgenic mice is Ang-2 (32, Xue 2008). Xue and coworkers showed that FOXC2 regulates Ang-2 expression by acting on the five forkhead-binding sites located in the promoter region of Ang-2 (32, Xue 2008).


While PGC-1a is critical for cAMP-dependent thermogenic programming in brown adipocytes, several genetic studies have demonstrated that PGC-1a does not give adipocytes their brown adipocyte identity (Lin 2004 and Uldry 2006). Seale and coworkers utilized global expression analysis of murine transcriptional components and found PRDM16, a PR-(PRD1-BF-1-RIZ1 homologous) domain containing protein, to be selectively expressed in brown adipocytes. They observed a 20-fold increase in PRDM16 expression during brown adipocyte differentiation (26, Seale). Furthermore, PRDM16 expression was neither affected by cAMP treatment nor induced by exposure to cold temperature (26, Seale 2007). These results suggest that PRDM16 expression in brown adipocytes may be associated with cell differentiation and not with adaptive thermogenesis.

PRDM16-expressing adipocytes have high levels of UCP-1 and PGC-1a mRNAs, and have a significant elevation in mitochondrial density (26, Seale 2007). PRDM16 also stimulates the expression of mitochondrial genes that are enriched in BAT, such as cytochrome c, cox5b, and cox8b (26, Seale 2007). Moreover, PRDM16 significantly represses the expression of several genes that are preferentially expressed in white adipocytes such as resistin, a WAT-secreted hormone that promotes insulin resistance (Steppan 2001 and 26, Seale 2007). Seale and coworkers demonstrated that the expression of PRDM16 in stromal-vascular cells prior to differentiation resulted in the activation of brown adipocyte differentiation, as evident in 200-fold upregulation of UCP-1 gene expression. These findings confirm the role of PRDM16 as a factor controlling the expression of brown adipocyte phenotypes.

The results from PRDM16 knockout studies showing reductions in expression of brown fat selective genes and mitochondrial genes further confirmed the requirement of PRDM16 in brown adipocyte determination and differentiation (26, Seale 2007). Even more interesting is the finding that PRDM16 knockout cells exhibit a 5-25% increase in the expression of myogenic genes, including MyoD, myogenin, myosin light chain, muscle creatine kinase and Myhc (5, Seale 2008). Seale and coworkers demonstrated that ectopic expression of PRDM16 in myoblasts completely inhibited myogenic differentiation (5, Seale 2008). This suggests that PRDM16 acts at the level of Myf5 precursor cells to inhibit skeletal muscle gene expression concomitant with activation of brown adipocyte differentiation.

A previous study by Nishikata and coworkers showed that PRDM16 binds directly to DNA at the two DNA binding regions consisting of seven C2H2 repeats at the N-terminus and three C2H2 repeats at the C-terminus. However, by altering DNA binding sequences, Seale and coworkers demonstrated that PRDM16 direct binding to the DNA is not required to induce brown fat differentiation. Therefore, PRDM16 must induce brown fat differentiation by way of protein-protein interactions. PRDM16 was found to bind directly to PPARg at two different regions, zinc-finger 1 and zinc-finger 2 (Figure 12) (5, Seale 2008). PRDM16 and PPARg interaction is independent of ligand binding (5, Seale 2008). Even though both PRDM16 and PPARg are capable of inducing the expression of genes that are common to both WAT and BAT, such as Ap2 and adiponectin, only PRDM16-expressing adipocytes contain numerous small lipid droplets, which are characteristics of BAT (5, Seale 2008). However, PRDM16 cannot promote adipogenesis in the absence of PPARg (5, Seale 2008). Therefore, it can be concluded that while PPARg is essential for adipose tissue differentiation, PRDM16 is required for brown adipocyte determination.


CCAAT/enhancer-binding protein (C/EBP) is a family of transcription factors belonging to a class of basic zipper proteins. These proteins consist of two domains: 1) basic DNA-binding domain and 2) C-terminal dimerization domain called leucine zipper. There are six different members belonging to the C/EBP family, which are C/EBPa, C/EBPb, C/EBP g, C/EBPd, C/EBPe, and CHOP that have been characterized to date (24, Machado). These proteins can both homodimerize and heterodimerize and bind to the same C/EBP consensus sequence (25, Tanaka). Among the six members, C/EBPb mRNA is found in nuclear extracts in rat BAT, and is detected in both fetal and adult rat BAT (24, Machado). C/EBPb mRNA expression is also induced by cold exposure (24, Machado). Moreover, an increased in C/EBPb mRNA is detected by day 18 of fetal rat development, before the rise in UCP-1 mRNA, which is not detected until day 20 (24, Machado). This suggests that the expression of C/EBPb is required early in development for the regulation of brown adipocyte differentiation.

Brown adipocytes express three isoforms of C/EBPb, which all come from a single mRNA. There are the two active forms called LAP (liver-enriched transcriptional activatory protein), and a dominant transcriptional inhibitor form called LIP (liver-enriched transcriptional inhibitory protein) (23, 8/Descombes). Kajimura and coworkers discovered that PRDM16 preferentially bound to LAP, but not to LIP, via the interaction between the two zinc finger domains (6, Kijimura). The complex formed by PRDM16 and C/EBPb increases the transcriptional activity of PGC-1a and PPARg2, because the latter two genes contain C/EBP binding sites in their promoters (Figure 12) (6, Kijimura and 25, Tanaka). Moreover, there are significant decreases in the levels of PPARg mRNA and other brown fat selective genes, which include cidea, elovl3, cox7al, and cox8b, in C/EBPb-knockout mice (6, Kijimura). This confirms the importance of C/EBPb in the expression of normal brown adipocyte phenotypes.

BAT of C/EBPb knockout mice embryos have significantly reduced numbers of lipid droplets and UCP-1 expression as compared to wild-type mice embryos (25, Tanaka). BAT from both C/EBPb knockout mice and PRDM16 knockout mice show a reduction in the expression of brown adipocyte-selective genes, but an increase in the expression of skeletal muscle genes (6, Kijimura). These results demonstrate the role of PRDM16 and C/EBPb transcriptional complex in the initiation of the myoblast to brown adipocyte switch. PRDM16 in complex with C/EBPb acts in Myf5-myoblastic precursor cells to induce the expression of PPARg and PGC-1a and consequently determines the fate of precursor cells to differentiate into brown adipocytes.


RIP140 is a corepressor of many nuclear receptors that plays a role in controlling the balance between fat accumulation and energy dissipation. RIP140-null mice are leaner and exhibit resistance to obesity when fed a high-fat diet (33, Leonardsson). In comparison to wild-type mice, white adipocytes of RIP140-null mice are smaller, whereas the size and appearance of brown adipocytes remain unchanged (33, LEonardsson). The absence of RIP140 leads to an increase in UCP-1 gene expression in white adipocytes causing WAT to appear brown adipocyte-like. However, adipogenesis is not affected by the absence of RIP140 as observed in mice (33, Leonardsson).

RIP140 regulates the transcription of several target genes that are also regulated by PGC-1a (34, Hallberg). RIP140 represses the transcription of important mitochondrial genes such as cox8b and cytochrome c, whereas PGC-1a stimulates their expressions (34, Hallberg). RIP140 was found to directly interact with both PGC-1a and PGC-1b as demonstrated by a GST pull-down assay (34, Hallberg). There are several regions necessary for the interaction between PGC-1a and RIP140, but the most important ones are the LXXLL motifs in RIP140 (34, Hallberg). In conclusion, RIP140 regulates the expression of mitochondrial genes important in brown adipocytes by modulating the activity of PGC-1a and PGC-1b.


Steroid receptor coactivators (SRCs), consisting of SRC-1/NcoA-1, SRC-2/TIF-2/GRIP-1, and SRC-3/p/CIP, are members of the p160 SRC family. SRCs are recruited to the nuclear receptor ligand binding domain (NR-LBD) upon stimulation by a ligand. There are three a-helical LXXLL motifs located in the SRC N-terminal region that facilitate the binding of SRCs to NR-LBD (37, Aranda and Pascual 2001). SRCs also contain conserved leucine-rich motifs in their C-terminal region that mediate their interactions with other coregulators (37, Aranda and Pascual 2001). Furthermore, the expression of SRC-1 and SRC-2 mRNAs was found to increase during the differentiation of preadipocytes into adipocytes (37, Picard).

SRC-1-knockout mice have significantly higher body weight than wild-type mice due mainly to an increase in WAT accumulation (37, Picard). Reductions in mRNA levels of UCP-1, PGC-1a, and acetyl-CoA oxidase are also observed in BAT of SRC-1-knockout mice (37, Picard). These results suggest that mice lacking SRC-1 are at risk of becoming obese because they have reduced capacity to metabolize fatty acids and expend energy. SRC-1 directly binds to PGC-1a to activate PGC-1a transcriptional activity (37, Puigserver 1999). This interaction is mediated by four highly conserved leucine-rich motifs in the C-terminal region of SRC-1 (37, Picard).

The functions of SRC-2 are reciprocal to that of SRC-1. SRC-2-knockout mice are protected against obesity induced by high-fat diet (37, Picard). SRC-2-knockout mice also exhibit significantly lower total body fat content and their white adipocytes are significantly reduced in size as compared to wild-type mice (37, Picard). Reductions in the levels of perilipin A and B and other proteins involved in fatty acid uptake and storage are observed in SRC-2-knockout mice (37, Picard).

Brown adipocytes of SRC-2-knockout mice are smaller and contain enlarged mitochondria with more cristae (37, Picard). SRC-2-knockout mice consume more oxygen, a finding that is consistent with their lower body weight resulting from an increase in metabolic rate (37, Picard). Enhanced fatty acid oxidation and thermogenesis in SRC-2-knockout mice are evident in the increase expression of UCP-1, PGC-1a, and fatty acyl oxidase (37, Picard). Picard and coworkers demonstrated that SRC-1 and SRC-2 compete for the binding to PGC-1a. In the absence of SRC-2, the more active and stable SRC-1/PGC-1a complex is favored, and this accounts for the increase in thermogenic properties observed in SRC-2-knockout mice (37, Picard). On the other hand, the absence of SRC-1 leads to the formation of the less active SRC-2/PGC-1a complex resulting in a reduction in the total body energy expenditure (37, Picard). Therefore, the ratio of SRC-2 to SRC-1 is important in determining the transcriptional activity of genes involved in fat storage and adaptive thermogenesis (37, Picard)

SRC-3 has also been shown to play a role in adipocyte differentiation. There is a significant increase in the level SRC-3 expressed in the nucleus during the early steps of cells commitment toward adipocyte lineage (38, Louet and O'Malley). SRC-3-knockout mice have lower body weight compared to wild-type mice, and are protected against diet-induced obesity (38, Louet and 39, Coste). The lower body weight is due to a reduction in fat content, specifically a reduction in WAT volume (38, Louet and O'Malley). Moreover, BAT of SRC-3-knockout mice is smaller in size but has increased numbers of brown adipocytes with higher mitochondrial concentration (39, Coste). This finding is consistent with the observation that SRC-3-knockout mice consumed more oxygen and expended more energy than wild-type mice (39, Coste). The effect of SRC-3 on energy expenditure is mediated by the ability of SRC-3 to facilitate acetylation of PGC-1a (39, Coste). A recent study reported that PGC-1a is acetylated by GCN5, an endogenous acetyltransferase that is specific to PGC-1a (40, Lerin 2006). The acetylation of PGC-1a causes PGC-1a to be translocated from the promoter region to the subnuclear domain resulting in the repression of PGC-1a transcriptional activity (40, Lerin 2006). Coste and coworkers demonstrated that SRC-3 facilitates PGC-1a acetylation by inducing GCN5 expression. In summary, the absence of SRC-3 reduces the expression of GCN5, thereby promoting PGC-1a transcriptional activity.


Retinoblastoma gene encodes for the family of retinoblastoma (RB) pocket proteins that includes pRB, p107, and p130. Members of RB family have overlapping functions, but their primary role is in the regulation of mammalian cell cycle (35, 7/Harbour JW). They are expressed in almost all tissues, and are important for the development and differentiation of those tissues. RB knockout mice die in utero due to abnormalities in neural, muscle, and erythroid development (34, Clarke, Jacks and Lee all 1992). RB family, specifically pRB and p107, have been shown to regulate adipogenic determination and differentiation (35, Hansen and 36 Scime).

Rb-knockout mouse embryonic fibroblasts (MEFS) and embryonic stem cells differentiate into fat cells with brown adipocyte characteristics when treated with PPARg activator, such as rosiglitzone (35, Hansen and 36, Scime). pRB-null cells express UCP-1 and induce the expression of PGC-1a and PGC-1b during adipocyte differentiation (35, Hansen and 36, Scime). RB-knockout adipocytes also contain more mitochondria than wild-type adipocytes, and this is reflected by an increase in the expression of mitochondrial transcription factor A and enzymes of the respiratory chain (35, Hansen). Furthermore, pRB expression is present in the nuclei of white adipocyte precursors but not in brown adipocyte precursors (35, Hansen). Scime and coworkers confirmed that pRB directs preadipocyte differentiation into brown adipocyte by demonstrating that in the absence of pRB, preadipocytes differentiate into brown adipocytes.

Similar results were also found in p170-knockout mice. Dissection of neonatal P170-knockout mice reveals deficient amounts of WAT (36, Scime). This deficiency persists into adulthood; however, BAT of p170-knockout mice do not differ from BAT of wild-type mice (36, Scime). Scime and coworkers showed that the reduction in WAT was caused by the conversion of WAT into adipose tissue with brown adipocyte characteristics, as evident in an increase in the expression of PGC-1a and UCP-1. Furthermore, p170-knockout preadipocytes display a reduction in the level of pRB (36, Scime). This explains why there is a reduction in WAT in p170-deficient cells because of a reduction in the level of RB.

There are several mechanisms by which pRB regulates the switch between white versus brown adipocyte differentiation. Hansen and coworkers presented the first mechanism, which was formulated after they detected an increase in the expression of FoxC2 in Rb-knockout fibroblasts. They proposed that brown fat differentiation occurs as a result of pRB inactivation, which leads to an increase in FOXC2 expression during cell differentiation. This increases the level of type I PKA, and thus causes cells to be sensitive to cAMP (35. Hansen). Additionally, Scime and coworkers showed that pRB directly binds to the PGC-1a promoter region during differentiation. In mice, the binding of pRB inhibits PGC-1a expression in a dose-dependent manner (36, Scime). This finding demonstrates that pRb functions as a switch between white and brown adipocyte differentiation by regulating the expression of PGC-1a. As mentioned above, PGC-1a is a transcriptional coregulator that controls the expression of UCP-1, which is a marker of brown adipocyte differentiation.

Brown Adipose Tissue: Clinical Implications

There are reasons to believe that our knowledge of BAT as obtained from studies using small mammals is also applicable to humans. First, newborns have relatively large deposits of BAT containing high levels of UCP-1. Second, recent developments in imaging technology have allowed doctors and researchers to observe small depots of BAT distributed throughout adult humans. One of the explanations behind the reduction of BAT in adults compared to newborns is the higher ratio between heat production from basal metabolism and the smaller surface area exposed to the cold environment, since adult humans have a lower surface-to-volume ration than infants, are clothed and live indoors (Cannon and Nedergaard). Third, adult humans with pheochromocytoma, a pathological condition where patients have very high adrenergic activation, express UCP-1 containing BAT. But why has there been a recent surge of interest in BAT? What are its clinical implications that warrant this increase in research? The answer lies in the potential of BAT to counteract obesity and related diseases, such as type 2 diabetes.

The best evidence that humans may be able to reactivate BAT come from observation made in newborns. Unclothed newborns double their resting energy expenditure without shivering when transferred from a thermoneutral temperature (34°C) to a mildly cold environment (28°C) (47, 20). Approximately half of their energy intake is used for cold-induced nonshivering thermogenesis under these conditions. However, evidence for adaptive nonshivering thermogenesis in adult humans is limited. There have been two reports, one occupational and the other experimental, that positively demonstrate the recruitment of an adaptive adrenergic nonshivering thermogenesis in adult humans (Cannon and Nedergaard). The results of these studies were weak because the level of nonshivering thermogenesis was only raised to 15% above basal level (Cannon and Nedergaard). It could be argued that the level of NE used may not have been sufficient to elicit a maximal response because of potential negative side effects involved in the use of high systemic NE concentrations (Cannon and Nedergaard). However, if there is a possibility of demonstrating nonshivering thermogenesis in adult humans, it may be through the expression of UCP-1 in other tissues or the recruitment of brown adipocytes.

One of these groups of investigations examined the promotion of UCP-1 expression in other organs, such as WAT (C/N 408) and skeletal muscle. Chemical uncouplers have been used to induce weight reduction, but this protocol affects the level of ATP production (Cannon and Nedergaard). Even though muscle and heart could sustain the level of ATP produced when UCP-1 is ectopically expressed, the muscle mass and composition were altered (C/N 440). An alternative to ectopic expression of UCP-1 is the promotion of the recruitment and activation of BAT. There are also limitations to this option. First, proliferating brown preadipocytes depend on b1-AR, whose agonists directly stimulate the heart, thus posing life-threatening side effects. Second, even if an increased amount of brown adipocytes could be achieved, BAT must still be continuously stimulated (Cannon and Nedergaard). Therefore, b3-AR agonists have been the focus of thermogenic anti-obesity agents because b3-AR agonists do not act on the heart like b1-AR agonists or cause tremor, marked vasodilation and hypokalemia like b2-AR agonists (Ref). Future development of drugs that can safely act on b3-ARs in brown adipocyte precursor cells located within WAT might enable the possibility of using brown adipocytes to combat obesity.

Despite the potential of b3-AR agonists in combating obesity, it must be noted that b3-AR agonists may not be as useful in promoting weight loss in obese individuals. Adipose tissue contains 7 kcal/g of energy and lean tissue contains 1 kcal/g (48 Arch). b3-AR agonists cannot be expected to raise the basal metabolic rate by more than 10%. Simple math demonstrates that a 10% metabolic rate of 2,500 kcal/day corresponds to about 36 g of adipose tissue (48, Arch). A weekly weight loss would be expected to be around 252 g, which means that it would take at least 20 weeks for a 100 kg person to lose 5% of body weight due to increased metabolic rate (48, Arch). Even though b3-AR agonists may have a small effect on weight loss, their role in stimulating fatty acid oxidation in BAT may still prove to be beneficial for the metabolism of obese individuals.

In addition to anti-obesity activity mediated by b3-AR agonists, b3-AR agonists produce a more rapid insulin sensitizing effect and therefore mediate an anti-diabetic activity. Several studies have shown that b3-AR agonists improve insulin sensitivity in rodents at dose levels that did not affect their body weight (48, 66/67). A similar result was also observed in humans (48, 70). Insulin resistance may be a consequence of elevated intracellular levels of long chain fatty acyl CoA, metabolites that activate certain isoforms of protein kinase C (PKC). Activated PKC in turn phosphorylates and reduces the activity of the insulin receptor, more specifically IRS-1, and some enzymes of carbohydrate metabolism, such as glycogen synthase (48, 71). b3-AR agonists stimulate fatty acyl CoA oxidation in BAT thereby reducing the levels of metabolites that affect the activity of insulin receptor (48, Arch).

Drugs targeting b3-ARs have not been proven successful because of the lack of selectivity and poor efficacy. A number of pharmaceutical companies have used human cloned b-adrenoceptors to identify agonists with increased efficacy and potency at the b3-ARs. However, b3-AR is expressed in such low numbers relative to b1-ARs and b2-ARs that even a highly selective b3-AR agonist may stimulate either b1 or b2 adrenoceptors. Additionally, it is extremely difficult to combine appropriate pharmacology with good oral bioavailability and other properties required in a drug (48, 18). Because of these challenges relating to b3-AR agonists, no b3-AR agonists have advanced beyond Phase II clinical studies.

An additional activity observed in BAT that warrants future research because of its potential in combating obesity is the production of triiodothyronine (T3) by BAT. Proliferation and differentiation of brown preadipocytes is accompanied by a large sympathetic-mediated increase in thyroxine 5'-deiodinase (type II) activity that increases local production of T3 from thyroxine (47, 21). T3 produced in BAT enhances the expression of UCP-1, but T3 can also be exported out of BAT. In this case, it mostly targets skeletal muscle (47, 22). The export of T3 by BAT may explain the lack of obesity observed in UCP-1-deficient mice. One study showed that UCP-1-deficient mice increase energy expenditure in the muscle via an increase in proton leakage in the mitochondria (47, 25). The proton leak in muscle mitochondria is regulated by thyroid hormone (47, 26). Furthermore, the increased expression of thyroxine 5'-deiodinase and T3 may explain the continual development of new brown adipocytes in UCP-1-deficient mice (47, Hagen). However, it is uncertain whether such a process occurs in humans. It must be noted that while thyroxine 5'-deiodinase is present in BAT of newborn humans (47, 28), deiodinase has only been identified in the heart and skeletal muscle of adult humans (47, 29/30).


Drawing from past research, it is clear that b3-AR in adipose tissue, specifically in BAT, plays an essential role in regulating fatty acid oxidation and thermogenesis in humans. Because of the potential of b3-AR agonists in treating obesity and Type 2 diabetes, b3-AR agonists have remained a target of study for researchers and pharmaceutical companies alike. Some pharmaceutical companies, such as Merck and Eli Lilly, have produced b3-AR agonists, but challenges still remain because these drugs lack selectivity and have poor efficacy and low oral bioavailability. Furthermore, rodent and human b3-adrenoceptors display different pharmacologies, and thus results drawn from human clinical studies and animal model studies have been inconsistent.

Additionally, the knowledge that BAT is the main site of fatty acid oxidation and thermogenesis, both of which are stimulated by b3-AR agonists, in rodents has increased interested in brown adipocytes. However, only small and localized BAT depots have been detected in human adults and may therefore not be sufficient to promote fatty acid oxidation in obese individuals and improve insulin sensitivity in Type 2 diabetic patients. Future research aimed at 1) inducing precursor cells to differentiate into brown adipocytes, 2) increasing the numbers of proliferating brown preadipocytes, and 3) stimulating the transdifferentiation of WAT into brown adipocyte-like tissue, will reveal whether BAT can influence energy balance and serve as a therapeutic use in treating obesity and Type 2 diabetes.

Please be aware that the free essay that you were just reading was not written by us. This essay, and all of the others available to view on the website, were provided to us by students in exchange for services that we offer. This relationship helps our students to get an even better deal while also contributing to the biggest free essay resource in the UK!