Brown adipocyte and browning thermogenesis: metabolic crosstalk beyond mitochondrial limits and physiological impacts

ABSTRACT

Brown adipocytes were proposed to reverse metabolic conditions such as obesity and diabetes, which make them potential for therapeutic applications. Brown adipocytes and browning process are capable of thermogenesis, the uncoupling metabolism which allows them to promote balanced energy expenditure, a fundamental mechanism for improving metabolic disorders. Thermogenesis process is not only performed by the thermogenin UCPs within the mitochondria, but instead, is globally regulated within brown and browning adipose tissues, which induces signalling molecules that can be sent to nearby and distant tissues to generate systemic effects on metabolism. This review highlights thermogenesis and describes the crosstalk between different organelles within browning and brown adipocytes, as well as their interorgan axes to regulate whole body metabolism. Finally, browning and thermogenesis activation will also be discussed in terms of physiological conditions, in which, we propose that thermogenesis and functional activities of brown adipocytes should be considered individually in future clinical application.

KEYWORDS:

• Thermogenesis
• diabetes
• obesity
• triglyceride
• glucose tolerance

1. Introduction

The ability to generate heat is the key function of brown adipose tissue (BAT), which is required to maintain thermal stability in newborns as a crucial survival mechanism. When it was discovered that adult bodies retain a portion of brown fat, the role of BAT was found to extend into systemic metabolism and energy expenditure [1]. A human anatomical map of brown fat deposition has been identified, showing that BAT volume and the potential for heat generation are varied among different individuals, which is dependent on their physiological characteristics [1,2]. Specific characterization of different BAT depots has also been uncovered, suggesting that depending on the position, BAT has diverse capacities to stimulate a systemic effect in adult humans [3,4]. Regulators of brown fat activation have been shown in the literature, highlighting the involvement of multiple cellular pathways that, once activated, induce thermogenesis performed by uncoupling protein-1 (UCP1) in the mitochondria as well as other metabolic programs [4–6]. Thermogenesis is the key activity that leads to the modulation of systemic metabolism and improvement of energy expenditure, two fundamental mechanisms for therapeutic application against obesity complications and diabetes [7–9]. As UCP1 is the thermogenin where heat dissipation occurs, the question of how UCP1 is fuelled and how systemic effects are generated from local brown adipocytes mitochondria remains critical. Thermogenesis is regulated in a process that involves the activation of UCP1 machinery and the emission of signals from local BAT to distant tissues to initiate systemic regulation to maintain metabolic balance [10]. Other cell compartments have been shown to be involved in the regulation of BAT function in addition to mitochondria and UCP1 activity [11,12]. The integration between UCP1 inside the mitochondria and other organelles inside the cells raises the question of whether thermal signals could be interchanged between different cell compartments and thereby transported between different cells. This review will highlight the complex network and interaction between different cell compartments within brown adipocytes, with a discussion on thermal and metabolic signal transduction from thermogenic to systemic effects. Various physical and pathophysiological characteristics that affect BAT function will also be mentioned and discussed with regard to individual health conditions. From the available evidence, the current knowledge gap in understanding BAT will be mentioned with suggestions for future research.

2. Browning and activation of BAT

Adipocytes can be classified into two major types, white and brown adipocytes, which have distinct metabolic programs. There are also beige and pink adipocytes. Beige adipocytes are described as carrying both white and brown phenotypes, the origin of which is controversial; however, it is suggested that beige adipocytes generated from interscapular WAT (iWAT), retroperitoneal WAT (rWAT) are distinguished subset as these adipocytes carry distinct adipocyte markers, and is generated from distinguished precursor population but not from pre-existing adipocytes [13]. Pink adipocytes are mammary adipocytes, which is transdifferentiated from white adipocytes during pregnancy and have milk secretory potential [14].

The basic morphological differences between white and brown adipocytes are the contents and numbers of the mitochondria, which are associated with the expression of UCP1, and the sizes and numbers of lipid droplets. While white adipocytes are observed with few mitochondria located around the nucleus, and with a large and unilocular lipid droplet occupying the major area of the cells, brown adipocytes show numerous small lipid droplets randomly arranged within the cytoplasm with a high number of mitochondria [15]. The number of mitochondria and the sizes and numbers of lipid droplets are inversely correlated as the cells gain or lose thermogenic capacity, and have been discussed in relation to particular pathological development [16]. It must be noted that BAT physiology is varied among different species and is different from human. For examples, hibernating animals would store fat during summer for use in winter and BAT activation is seasonal dependent, which has been extensively reviewed elsewhere [17]. Recent study has shown that perhaps human BAT is more closely related to BAT in mice housed at thermoneutral condition (30°C) and energy-rich diet, which were physiologically modelled with human living condition [18]. Interestingly, the data showed that BAT in physiologically humanized mice was both morphologically and molecularly similar to human BAT, suggesting a potential translational opportunity in future research [18].

The browning process is described as the transdifferentiation of white adipocytes into brown or beige adipose morphology, which increases the cell thermogenic capacity with a rise in mitochondria number and upregulation of UCP1 expression, and has been proposed to have therapeutic potential [19,20]. Browning process also involves adipogenesis of brown and beige adipocytes from adipocyte precursors, and activation of thermogenesis of brown adipocytes in present BAT [21,22]. Thus, the browning process may also mark the more activated state of brown and beige adipocytes, where thermogenesis is increased under stimulation.

In reverse, the whitening process of brown adipocytes is described as the increase in intracellular fatty acid accumulation that could be associated with or without cell enlargement, decrease in UCP1 expression and loss of thermogenic capacity, and infiltration of macrophages surrounding adipocytes forming crown-like structures (CLS), which may involve tissue inflammation and cell apoptosis [23,24]. Physiological factors such as ageing and obesity has been shown to result in whitening of brown adipocytes and BAT inactivation, which has been thoroughly reviewed, indicating that maintaining BAT is important to improve general health [25–28].

Brown adipocytes activated by cold exposure increased metabolic activity and thermogenesis, the process of which includes the upregulation of glucose transporters GLUT1 and GLUT4, enhanced expression of glycolytic enzymes such as hexokinase (HK), lactate dehydrogenase (LDH), and phosphofructokinase (PFK) [29]. Thermogenic activation can occur in most of white and brown adipose tissues at different degrees in various regions, including inguinal, dorsolumbar, retroperitoneal, and subcutaneous WAT, and interscapular, cervical, supraclavicular, paravertebral, subcutaneous, and perirenal BAT [22,30–35]. Several stimulations such as cold exposure, beta-adrenergic application, exercise, and natural extracts have been shown to be effective in activating adipocytes towards gaining thermogenic capacity while simultaneously enhancing metabolic regulation [36,37]. The most commonly used and effective stimulations shown in research are cold exposure and beta-adrenegic administration, during which, β-adrenergic signalling is activated. In rodent, β3-adrenergic receptors (β3-AR) have been identified to activate brown adipocytes and browning of adipose tissues, which leads to the increase in UCP1 protein levels in different adipose depots, and lower free fatty acid (FFA) storage in WAT, and higher FFA oxidation and storage in BAT [38,39]. Even though recent evidence suggests that perhaps β3-AR is not important and browning can occur without the presence of β3-AR [40], the fact that β3-AR is involved in the activation cascade of adrenergic stimulation is widely confirmed [41–43]. However, in human, β2-AR but not β3-AR, was identified as the key receptors in activation of BAT thermogenesis, and was the dominant β-AR (up to 91%) expressed at RNA level in WAT [44], suggesting that human and rodents are different in β-AR in response to stimulation. The binding of norepinephrine to β-adrenergic receptors activate the production of cyclic adenosine monophosphate (cAMP) and protein kinase A (PKA) [45,46]. PKA activation phosphorylate CREB (cAMP response element-binding protein), and p38 mitogen-activated protein kinase (p38-MAPK), both of which regulate the transcription of PGC1α and upregulate the expression of UCP1 proteins [41,47] (Figure 1). PKA also phosphorylates mTOR, a key regulator of cell metabolism and response to nutrients in the microenvironment, which is required for the acquisition of brown morphology in adipose tissues [48]. The role of PKA in BAT stimulation and metabolic modulation has been shown to be correlated with the transcription factor PGC-1α, which is a key regulator of mitochondrial metabolism required for adaptive thermogenesis [49,50]. The involvement of PKA in browning was shown to be location-specific and dependent on individual physiological condition, as only the stimulation of subcutaneous white adipocytes increased lipolysis via mitochondrial uncoupling, and obesity condition reduced the cell ability to gain thermogenic capacity [46]. Apart from the activation through β-AR, a recent study has revealed a non-canonical pathway of UCP1 regulation and thermogenic activation via the fibroblast growth factor (FGF) 6 and 9 [51]. In this study, the authors found that FGF6 and FGF9 were induced in both exercise training and cold exposure, which activated the receptors FGFR3 and promoted biosynthesis of the hormone prostaglandin E2 to regulate UCP1 expression via the transcriptional co-activators FLII and the binding of oestrogen-related receptor alpha (ERRα) to UCP1 enhancer [51], suggesting a new novel way to activate thermogenesis.

Figure 1. Browning and thermogenesis activation. Acclimatization to cold-temperatures may trigger the sympathetic nervous system, resulting in norepinephrine secretion and subsequently activating thermogenesis in brown adipocytes through β-adrenergic receptors. In addition to norepinephrine, β-adrenergic receptor agonists such as CL-316,243 activate protein kinase a (PKA) and increase cAMP levels, leading to the acceleration of TGR breakdown and FFA production, a crucial substrate for the regulation of mitochondria and UCP-1. Thyroid hormone (TH) binds to TR on BAT to activate SIRT pathway via deacetylation of the downstream molecule FOXO1, resulting in adipogenesis of BAT. Leptin correlated with hypothalamic regulation to increase gene expression of Pgc-1a, Cidea and Ucp1 in BAT and improves energy expenditure in the body. Estrogen receptors are activated by oestradiol in the ventromedial hypothalamus to enhance sympathetic nervous system regulation via AMPK to upregulate Ucp1, Pgc1a and Pgc1b in BAT, which activate BAT thermogenesis. Activated brown adipocytes have enhanced lipolysis and triglyceride-FFA cycling, resulting in a reduction in accumulated triglycerides in the bloodstream and nearby organs and an increase in mitochondrial activity. Contact between small lipid droplets and mitochondria is required for thermogenic activity to occur. Furthermore, higher mitochondrial activity contributes to an increase in the expression of UCP1, a brown adipocyte-specific marker that plays an important role in energy-heat conversion in BAT. In addition, cold-temperature acclimation also leads to higher levels of leptin and thyroid hormones, leading to an increase in insulin sensitivity, thus boosting glucose transporter activity and subsequently decreasing glucose levels in the bloodstream. Conversely, exercise activates the secretion of Fndc5 from myocytes. Inside adipocytes, Fndc5 is cleaved to form the peptide irisin, which activates PPARγ to regulate the browning process.

Thermogenic genes can also be regulated by the activation of thyroid receptors a and b (TRα and β), which are expressed in both WAT and BAT [52]. The specific isoforms of TR in BAT may vary greatly depending on different species [53–55]. Thyroid hormone (TH) has two major isoforms. The less active isoform thyroxine (T4) can be converted to its active form 3,3′,5-triiodothyronine (T3) via the action of the type II deiodinase (DIO2) [56,57]. TH plays an important role in regulating metabolism in various tissues during growth and development [58]. It was found that in interscapular BAT (iBAT), T3 activates TRα, which promotes adipose progenitor cell (APC) population and the transition of APCs from the stem-like morphology towards adipogenic commitment, via which, expanding iBAT depots with Myc-mediated glycolysis [59]. In another study, it was shown that T3 induced deacetylation of forkhead box O1 (FOXO1), the downstream target of sirtuins 1 (SIRT1), the enzyme activity of which is dependent on cellular NAD+ [60], suggesting the activation of SIRT1 pathway [61]. SIRT1 activation was shown to improve BAT metabolism and function, and enhance glucose homoeostasis [62]. In addition, SIRT1 induced the inhibition of mTOR, leading to increased brown adipocyte autophagic and lysosomal gene expression, resulting mitochondrial turnover and biogenesis, and improved thermogenesis in BAT [61]. T3 also enhances the expression of UCP3, another thermogenin that is involved in thermogenesis and is correlated with UCP1 expression in BAT under cold challenge [63,64]. The role of UCP3 in thermogenesis has been shown in various tissue including skeletal muscle and heart muscle [65], and even though UCP3 is involved in BAT thermogenesis, it is indispensable for nonshivering thermogenesis at least in hamster, and may only be complementary to UCP1 thermogenic function in mice BAT [66,67], suggesting that UCP1 is still the main thermogenin in BAT under cold exposure.

In fact, hyperthyroidism also led to the rise of hypothalamic TH levels, and the increase of TH in the ventromedial nucleus of the hypothalamus was shown to upregulate UCP1 expression in BAT at both 4°C and 23°C, which resulted in weight reduction [68], suggesting that TH can induce mediation on thermogenesis, systemic metabolism and energy expenditure. Hypothalamic regulation of thermogenesis and energy expenditure can also be performed by the secretion of the enzyme leptin, which regulates BAT thermogenesis possibly via increasing sympathetic nerve activity to BAT and involving BAT circuit in the sympathetic region, and enhances UCP1 expression in high-fat diet mice model [69–73]. In lipodystrophic mice model, leptin increased Pgc-1a, Cidea and Ucp1 expression in BAT, as well as improved body temperature even though metabolic benefit was unknown [74]. The molecular action of how leptin affects BAT thermogenesis and metabolism is limitedly explored, however, the increase in energy expenditure by leptin may be correlated with TH regulation in the hypothalamus [75]. These data suggest a systemic interaction that BAT is involved in the regulation of thermogenesis and energy expenditure. Limited is known about how individual physiological conditions affect BAT activity on systemic metabolism and future examination is warranted with aspects related to the physiological and metabolic status of each individual.

3. The mechanism of thermogenesis – from cells to systemic metabolic effects

3.1. Thermogenic regulation

3.1.1. The emerging role of mitochondrial metabolism

The mitochondria are the primary metabolic site in brown adipocytes, where also occurs thermogenesis. Therefore, thermogenic regulation is correlated to the characteristics of the mitochondria. While brown and white adipocytes share adipocytic characteristics, the abundant presence of mitochondria and UCP1 is the major morphological difference between the two. Thus, the question of how brown adipocytes can perform such ultimate functions in metabolism is focused on mitochondrial activity and UCP1 function. Not only mitochondrial metabolism and biogenesis is important for BAT function, but mitochondrial morphology has also been shown to be involved in the thermogenic capacity of beige adipocytes, as cells with positive UCP1 showed round-shaped mitochondria and increased mitochondrial fission [76].

Increased mitochondrial biogenesis and various metabolites production in brown adipocytes are most likely the thermogenic mechanism underlying systemic effects on metabolism and energy expenditure. It has been shown that lipolytic products are important for brown adipocytes in enhancing systemic metabolism [77–79]. In mice, increased lipolysis in brown adipocytes was found to be correlated with increased selective uptake of triglyceride into BAT and increased the clearance of triglyceride-rich lipoproteins (TRLs) after meals [77,78]. A thorough study into metabolites and their roles in BAT found that lipolytic products enhanced the permeability of the brown adipocyte membrane, which then allowed the internalization of plasma TRLs into brown adipocytes [78]. On the other hand, the internalization of TRLs and fatty acids into BAT is supported by lipoprotein lipase (LPL), which enhances local BAT lipolytic activity, hence, improving lipid metabolism [78]. The clearance of triglyceride remnants is finally driven by the liver, the total effect of which reduces lipidemia and prevents atherosclerosis [77].

In fact, the communication between different organs, especially between BAT and the livers during browning activation, has been shown to improve energy expenditure by changing the absorption of metabolites in these organs. As shown by Simcox and colleagues, acylcarnitines are the circulating metabolites produced by the livers during cold exposure and play a role in the thermogenic activity of BAT [80]. Acylcarnitine is an intermediate product of fatty acid oxidation that present with different length of carbon chain and have been shown to associated with high risk of diabetes [81]. Different sizes of acylcarnitines, including short, medium and long chains, have been shown to associated with increased insulin resistance in human and mice [81–84], the molecular mechanism of which remains unknown, but was suggested to be involved in downstream metabolite conversion, or activation of inflammation [85,86]. Cold exposure activates the secretion of long-chain fatty acids (FA) from WAT, which is converted into acylcarnitine in the liver. Once acylcarnitine enters the blood stream, it is preferentially taken by brown adipocytes to fuel thermogenesis, while FA and its transport into the liver are blocked [80]. This reduces the accumulation of FA in the WAT and liver, which hinders the development of metabolic diseases such as lipid disorders [80]. The increase in triglyceride uptake in brown adipocytes is also an effect of increased expression of CD36, a transporter that is specific for fatty acid imports [10,78]. At the same time, the increased expression of glucose transporters, such as GLUT1 and GLUT4, has been shown to improve the uptake of glucose into brown adipocytes, which improves glucose homoeostasis and insulin sensitivity [87]. These data indicate that lipolysis and thermogenesis metabolic activities in brown adipocytes can further enhance the cell function in regulating plasma concentrations of glucose and triglycerides. As the increase in metabolite uptake into the cells holds the keys to fuelling thermogenesis and enhancing energy expenditure, the interaction between the metabolite pool and cell function is important for the maintenance of BAT function. More research is needed to determine how metabolites are regulated within and outside the BAT area to optimize the thermogenic effect; however, it is suggested that brown adipocytes are subject to changes in metabolic programming and metabolites after activation stimulation.

The involvement of lipid metabolism in browning capacity and thermogenesis has been indicated in both transcriptional and lipidomic studies [4]. Lipolysis process in adipocytes involves the breakdown of triglycerides into glycerol and FFA, which is regulated by the action of adipose triglyceride lipase (ATGL), and the feeding of FFAs and glycerol to fuel ATP production in other tissues or thermogenesis marked the differences in white and brown adipocytes [88]. The inhibition of ATGL resulted in decreased brown adipocyte gene expression and reduced FA oxidation via PPARα and PPARδ pathway, suggesting that lipolytic products are important for thermogenic program [89]. In a recent study, it has been shown that lipolysis regulated and was required for the expression of majority of genes in β-adrenergic program in brown adipocytes, independently of PPAR pathway and FA oxidation [90], suggesting the role of lipolysis in connecting browning and metabolic programming. Among those metabolites that were shown to increase in brown adipocytes, the production of cardiolipin (CL), a component that comprises the inner mitochondrial membrane, has been shown to be involved in the regulation of systemic metabolism homoeostasis [79]. Accordingly, CL, which is formed under the synthetic activity of the enzyme CL synthase 1, was found to be the dominant product of lipid metabolism during browning [79]. The synthesis of CLs is required for normal mitochondrial respiration and thermogenesis and is involved in mitochondrial-nucleus communication via the endoplasmic reticulum stress response [79]. The production of reactive oxygen species (ROS), as a major product of metabolism, increased during the cold exposure, which is associated with the increase in metabolic rate and oxygen consumption of brown adipocytes [91]. In vivo study showed that mitochondria isolated from BAT in cold-exposed mice were more oxidized when compared to mice at room temperature, which was measured via the decreased ratio of glutathione (GSH) and its oxidized form glutathione disulphide (GSSH) (GSH/GSSH) [92]. Increased ROS production in BAT during cold exposure oxidized a range of cellular and mitochondrial thiols, including cysteine residues on the UCP1 protein, which activate UCP1 function [91]. This has been far the only identified mechanism of ROS in regulating BAT thermogenesis. In reverse, it has been shown that cytosolic ROS accumulation induced by glucose-6-phosphate dehydrogenase (G6PD) deficiency activated ERK, which in turn inhibited thermogenic gene expression such as Ucp1, Ppargc1a, Cidea and Dio2 [93]. Furthermore, a proteomic study of lipid droplets (LDs) has shown that protein content in LDs changes and increases during thermogenic activation by cold exposure [12]. These LDs were shown to be attached to activated mitochondria in close association, suggesting a requirement for LDs and mitochondrial interaction during cold challenges [12]. Perhaps, lipid metabolism is the key activity in the initiation of thermogenesis. Another study with proteomic screening of BAT in comparison to adjacent WAT has revealed that BAT increases mitochondrial metabolic proteins and exclusive expression of creatine kinases, which was then further confirmed by another study about the role of creatine metabolism, suggesting the involvement of local metabolism in BAT function [94,95]. Even though the silencing of creatine kinase B in mice brown adipocytes did not alter the expression of UCP1, and was shown to regulate ATP turnover rather than uncoupling respiration, it significantly decreased oxygen consumption and energy expenditure [96]. This evidence indicated that metabolites are important driving factors in the function of BAT and that their production is crucial to adipocyte activity. The question of how these metabolites crosstalk with other molecules to form the regulatory network would warrant further investigation.

3.1.2. UCP1 as a thermogenic machinery

The thermogenic ability of BAT is performed via the uncoupling activity of UCP1, and is associated with high metabolic profiles and increased nutrient and substrate uptake [4]. In general, UCP1 is characterized as an uncoupling complex that dissipates proton H⁺ into heat but not ATP, which is fundamental for thermogenesis [17]. In fact, this property of UCP1 has been suggested to underlie brown adipocyte homoeostasis, as cells with deficient UCP1 resulted in a significant increase in ROS production, calcium overload, and mitochondrial dysfunction [97]. Mitochondrial ROS production during cold induction has been shown to activate UCP1-dependent thermogenesis via thiol oxidation of cysteine residues on UCP1 proteins [91]. Different models of proton leaks have been proposed, which further suggested that fatty acids (FA) may play an important role in regulating UCP1 activity and the transport of protons through UCP1 [98].

In terms of regulation, the expression of UCP1 is known to be dependent on a number of cellular pathways, which include the activation of cyclic adenosine monophosphate (cAMP) and protein kinase A (PKA) signalling [99]. Accordingly, the stimulation of β-adrenergic receptors leads to the induction of cAMP in brown adipocytes, which phosphorylates Salt-inducible kinase 2 (Sik2), leading to the activation of histone deacetylase 4 (Hdac4) and inhibition of Ucp1 gene expression but not thermogenic capacity or mitochondrial biogenesis [99]. Thermogenesis is a specific characteristic of UCP1. However, thermogenesis can occur independently of UCP1, such as the creatine cycle, FA oxidation (mentioned in the section above), and Ca2+/SERCA (sarco/endoplasmic reticulum calcium ATPase) cycle, which involved the transport of Ca2+ secreted from the sarcoplasmic reticulum (SR) into the endoplasmic reticulum (ER). In detail, endoplasmic reticulum (ER) is the main site of Ca2+ storage, and energy derived from ATP hydrolysis by ATPase is used for SERCA to pump Ca2+ across the membrane, and a part of this energy can be converted into heat [100]. While SERCA2 was shown to be the ubiquitous, in rats, SERCA1 was identified to be the primary isoform involved in thermogenesis of brown fat, the mechanism of which involved SERCA1 to transport Ca2+ from the ER to the mitochondria and cytosol [100,101]. In a recent study, the generation of heat was shown to be compensated in the absence of UCP1 via the action of SERCA2b in beige, but not brown, adipocytes under chronic cold challenge [102]. Specifically, in mice lacking UCP1, during cold exposure, increased SERCA expression was only observed in inguinal WAT (iWAT) but not in iBAT, in which SERCA was colocalised with both ER and SR [102]. Similarly, norepinephrine treatment induced ER releasing Ca2+ in the cells, which was associated with enhanced thermogenesis [102]. In addition, the ablation of SERCA2 was also shown to cause depletion of Ca2+ storage in the ER, resulting in ER stress and mitochondrial dysfunction, which led to altered WAT and BAT function and systemic metabolism [103]. These findings suggested that even though SERCA is involved in UCP1-independent thermogenesis, it still plays an important role in the function of brown and beige adipose tissues and is required for thermogenic program. Thus, browning is an adaptive process relying on the acquisition of thermogenesis to enhance metabolism and energy expenditure, which highlights the role of heat generation in browning. The consideration of different thermogenic mechanisms and understanding how thermogenesis is used by the body will enhance the success of using browning in clinical applications.

3.2. The involvement of other cell compartments

Apart from the mitochondria, the involvement of different cell compartments such as lysosomes and ER was found to be essential in BAT thermogenesis, suggesting that BAT function in energy expenditure is not only dependent on mitochondrial activity, but that other organelles also play an important role in the regulation of thermogenic capacity (Figure 2). Lysosomes are known to act as cell clearance organelles that perform the degradation and recycling of cellular macromolecules via autophagosomes and endocytosis, which is crucial for cell metabolism and homoeostasis [104]. Lysosomes have recently been discovered to play a role in cellular pathway regulation and cell fate decision, making them a promising therapeutic target for a variety of diseases such as cancer, immunity, obesity, and other pathological conditions [104–106]. The role of lysosomes in thermogenesis has also been investigated, showing that lysosomes play an important role in regulating the browning process. This is supported by the evidence that the expression of lysosomal acid lipase (LAL) was shown to regulate UCP1 expression, which directly controls thermogenesis in BAT [11]. Specifically, mice deficient in LAL expression showed decreased UCP1 expression in BAT and decreased energy expenditure even at room temperature [11]. Furthermore, the depletion of LAL was found to alter BAT morphology and metabolism, and significantly reduce the temperature of the body, and despite enhanced glucose uptake to compensate for energy deficit, the mice were intolerant to cold exposure with severe hypothermia [11]. Lysosomes also take part in regulating the mitophagy of brown adipocytes, which has been shown to be involved in mitochondrial homoeostasis and thermogenic maintenance under cold challenges [107–109]. The role of LAL in processing and transporting triglyceride-rich lipoproteins (TRL) in endothelial cells (ECs) adjacent to BAT was shown to regulate the recruitment and proliferation of ECs via HIF-alpha (hypoxic induced factor) activation, which enhances WAT browning and differentiation of brown adipocyte precursors [110]. These data suggested that the crosstalk between lysosomal and mitochondrial metabolism is important for signal transduction not only within the cell environment, but also between different cell types to initiate thermogenesis. This is particularly essential in sustaining the prolonged effect of BAT in therapeutic application.

Figure 2. Thermogenic activation forms a signal transduction network to generate systemic effects. Thermogenic mechanisms in brown adipocytes and signal transduction to distant tissues. 1. Long-chain FAs are transported from the WAT to the liver, where they are converted to acylcarnitine (AC). AC is preferably used by BAT to fuel thermogenesis in the mitochondria. UCP-1 in the mitochondria dissipates protons into heat instead of producing ATP. 2. Lipoprotein lipase (LPL) supports the transport of TG and FA across the blood stream into BAT. 3. in turn, lipolytic products from the mitochondria further regulate the activity of LPL to enhance membrane permeability and the import of substances into BAT. 4. Cardiolipins (CLs) are synthesized and exported from the mitochondria to the ER to support communication with the nucleus via the stress response and Nrf1, which results in the regulation of thermogenic and transporter genes such as UCP1, CD36, and glucose transporters. 5. the involvement of lysosomes occurs via the activity of lysosomal acid lipase (LAL), which regulates UCP1 expression and mitophagy and remodels BAT morphology and function. 6. the endoplasmic reticulum (ER) protein Calsyntenin 3β binds to S100 proteins and mediates their secretion as a neurotrophic factor to recruit innervation. 7. Activated BAs produce a secretome in the blood stream to regulate distance to support their maintenance and enhance systemic metabolism.

Similarly, as an important organelle for protein stabilization, ER also plays an important role in regulating cellular activity. Following Ca2+/SERCA cycle in regulating thermogenic activation, it was shown that the fusion of ER to mitochondrial was evidenced of the crosstalk between mitochondrial metabolism with other organelles in BAT [111]. It was shown that ER homoeostasis, which is regulated by the nuclear factor erythroid-2, like-1 (Nrf1), is required for BAT to increase thermogenesis during cold exposure by enhancing proteasomal activity [112]. A recent study has discovered a new role for calsyntenin 3β, which is a mammal-specific protein that is located in the ER of brown adipocytes [113]. According to the data, calsyntenin 3β strongly binds to an unusual protein with no signal peptide, S100b, and mediates the secretion of S100b as a neurotrophic factor, which has a function to recruit sympathetic innervation of adipose tissue [113]. These data show that regulatory factors from different cell organelles may perform distinct mechanisms on the homoeostasis of BAT, suggesting that these thermogenic regulators play an important role in controlling BAT function. Deepened research into the role of lysosomes and ERs components and function are beneficial to understand their roles and the crosstalk with the mitochondria, by which the mechanism of thermogenesis in BAT is controlled and activated. For instance, the questions of how signals are sent from different organelles, and how they are received, and under which pathways they perform regulation on browning, thermogenesis and energy expenditure will add knowledge in future applications when developing potential targets for obesity and diabetes.

3.3. BAT as a secretory organ

The ability of BAT to secrete cytokines has been demonstrated with numerous pieces of evidence, implying the involvement of various regulatory factors in distant tissues [87,114]. Regulatory factors secreted by BAT, which include growth factors, cytokines, and other peptides, can generate either autocrine, endocrine, or paracrine effects on neighbouring and distant tissues such as the brain, liver, and immune system, to control their biological and physiological processes [115,116]. Evidence suggests that the communication between BAT and other tissues in the body during browning activation can extend to the kidneys, the heart, and even the gut [117–119], indicating that BAT systemic effects are multiorgan oriented. An example of a secretory factor is the production of the pro-inflammatory cytokine, interleukin 6 (IL-6), which plays the main role in the communication between BAT and immune cells [120]. IL-6 has also been shown to regulate the uptake of glucose and increase insulin sensitivity in BAT [87]. Systemic regulation of BAT also involves the regulation of satiety as BAT expressed high levels of the gut hormone secretin. After meal, the rise in plasma secretin levels activated thermogenesis in BAT via lipolysis, and promoted satiety to reduce food intake, which may act via heat sensing in the brain to stimulate the feeling of satiety at least in mice [121]. In human, prandial BAT thermogenesis was also observed in human together with increased glucose uptake, and negatively correlated with caudate glucose uptake and inhibitory response, which led to increased satiety in human, suggesting a BAT-brain axis [122,123]. Future research is warranted to explore the molecular mechanisms underlying BAT-brain axis as well as their effects on whole body energy expenditure.

As BAT needs substrates to fulfill its high metabolic profile, the formation of the vascular and nervous systems is critical to maintaining its supply to BAT and the communication between BAT and other organs. Data evidence suggests that during the browning process, the vascular endothelial growth factors (VEGF) are secreted by activated brown adipocytes during the browning process, resulting in an autocrine effect that is important for the recruitment and formation of new vessels [124,125]. The secretion of nerve growth factor (NGF) and neuregulin 4 (NRG4), which, respectively promote the development of sympathetic neurons and neurites, has been shown to be important for the innervation of BAT to support the activation of the browning process [126,127]. In addition, activated BAT also produce several factors and signalling molecules such as bone morphology protein 8 (BMP8), SLIT protein 2 (SLIT2), G protein regulator 20 (GPR20) or insulin-like growth factor (IGF-1), which can target distant tissues and directly regulate their activity to induce systemic effects on energy expenditure [127–130]. However, the precise mechanism of how these molecules coordinate to generate the total effect on systemic metabolism requires more investigation.

The application of transcription and protein screening technology such as expression arrays, RNA sequencing, and mass spectrometry has been used to improve our understanding of the BAT secretome [4,6,95,127,131]. The development of big data analysis has made it possible to thoroughly dissect and characterize BAT along with its thermogenic capacity. For example, a gene expression microarray study by Rosell and colleagues has suggested a list of upregulated signalling molecules that were induced by cold exposure [127]. Recent emerging use of the metabolomic technique in research has improved the ability to examine metabolic pathways and specific metabolites to understand the role of BAT metabolism in the regulation of total energy expenditure. Metabolomic study by Whitehead and colleagues has shown that BAT metabolism produces a number of metabolites, including 3-methyl-2-oxovaleric acid (MOVA), 5-oxoproline (5OP), and β-hydroxyisobutyric acid (BHIBA), which can signal activation and upregulation of brown-adipocyte-like phenotype, including UCP1, CIDEA, and PGC1α in both human and mice primary adipocytes, at the same time with regulating skeletal muscle metabolism in mice and improve BMI in human, suggesting an interorgan axis between beige-brown adipose-muscle axis [132]. Furthermore, the authors showed that MOVA and 5OP function involved the activation of cAMP-PKA and phosphorylation of p38-MAPK pathway, while extracellular activation of mTOR was shown to underly the mechanism of BHIBA in regulating gene expression of adipocytes and myocytes, and these signals were enriched with cold conditioning and decreased in obese mice. This evidence suggested that BAT and browning adipocytes can regulate metabolism of adjacent tissues and directly regulate metabolic program in these tissues during their metabolic homoeostasis regulation, and emphasized the importance of metabolites that are specifically catalysed by BAT and browning adipocytes. A breakthrough in data integration methods that involve metabolomic and genomic data, followed by functional studies, will be useful to solve the puzzle of how BAT affects systemic metabolism and energy expenditure and what signals are required for the spreading of thermogenesis in the body. Understanding the role of BAT as a secretory organ will be beneficial to enhance the control of using BAT in different subjects with various physiological conditions, supporting the use of BAT as a personalized treatment.

3.4. BAT capacity and pathogenesis varied with specific individual physiological characteristics

Lifestyles, daily habits, and individual health conditions have particular impacts on BAT activities and may affect BAT function and pathogenesis (summarized in Table 1). For instance, exercise has been shown to enhance BAT activation and capacity by regulating UCP1 expression and BAT morphology in in vitro and animal models. An in vitro study showed that adipocytes differentiated from fibroblast source enhanced UCP1 expression in co-cultured with contracting muscle cells, which were induced by electrical pulse stimulation [142]. In vivo, animal studies showed that exercise at different density level increased the expression of browning master gene PGC-1α and UCP1 in WAT, which could be associated with improved brown adipogenesis from preadipocytes [143–145]. In an attempt to understand the mechanism of browning in exercise, a study used transgenic mice with increased PGC-1α expression in myocytes, which partially recapitulated the exercise program, and showed that these myocytes increased Fndc5 secretion [146]. When white adipocytes were treated with Fndc5, Fndc5 was cleaved to form the peptide irisin to trigger the PPARα pathway, which is involved in the browning process of WAT in vivo and primary subcutaneous white adipocytes in vitro [146]. In addition, a microarray study in mice proposed that exercise might promote thermogenesis in BAT via the upregulation of COX2 and prostaglandin to activate UCP1 expression under VEGF signalling pathway [147]. These data suggested a relationship between active exercise and the browning effect on white adipocytes at least in rodents. However, the effects of exercise on browning in human remained controversial either due to experimental design or physiological effects [148]. There are evidences that exercise was effective in promoting adipose tissue browning in people of both normal- and overweight in humans, which were evaluated through the measurements of anthropometry and biochemical characterizations of the individuals, and gene expression of beige and brown adipocyte markers [149]. Tanaka and his colleagues showed that men, but not women, increased BAT volume at specific adipose regions after vigorous intensity exercises [150]. However, no indication about the role of BAT in systemic metabolism in these subjects were mentioned, and the involvement of ambient temperature and sex hormones might require further clarification [150]. In fact, despite the evidence of increased brown adipose gene expression in adipose depots, exercises were shown to be less effective in promoting a general metabolic improvement, and varied greatly with individuals, especially in those of different body weight and genders [149]. In a recent randomized control study, the authors showed that despite the reduction in adiposity, improvement in muscular fitness and cardiorespiration, no evidence in body weight reduction, glucose metabolism or BAT volume was found after 24 weeks of supervised moderate or vigorous exercise intensity in young participants [151], suggesting that exercise may not be an effective method to increase brown adiposity and thermogenesis in human. Research into wider range of participants with various technique measuring brown adipocyte capacity may give stronger conclusions on the role of exercise on browning, thermogenesis, and systemic metabolism.

Table 1. BAT and browning capacity are varied with individual health conditions.

Health condition	Model	Effect on BAT function
Lean body	Human	More BAT content [2].
Obese body	Human	More browning capacity in WAT [2]. Compromised browning and lipolysis [133].
Aging	Human	Beige progenitors cannot acquire full brown characteristics [134]. Reduced cold-induced thermogenesis [135].
Naturally, skinny build	Human	Sera can induce browning commitment in MSCs [136].
Genders	Human, Mice	Females have more BAT content and are more sensitive to cold-induced thermogenesis than males [137–139].
High fat diets	Mice	Developed hypertrophic BAT, infilled with T cells and macrophages [120].
Diabetes	Rat	Increased PGC-1α and UCP1 expression without changes in glucose uptake and diabetic hyperglycaemia [140]
Cachexia in cancers	Mice	Increased UCP1 expression, high thermogenesis, and metabolism [141].

Other browning pathways have been employed in different strategies to perform sustainable browning methods in vitro and in vivo [152]. However, those browning procedures have not been tested in association with specific physical conditions. Perhaps combining thermogenic activation with physical characteristics may support the regulation of gene expression during the browning process, which is essential for BAT function and capacity. Furthermore, recent studies have found that BAT content and deposition are varied in different people, depending on their physiological and metabolic characteristics [2]. People of different ages, body types, and diets can be affected differently by BAT activating factors and possess distinct BAT formation capacities [2,134,136]. It was shown that even though lean people have more BAT content than obese individuals, it was found that obese people have more browning capacity in their white depots [2]. Adipocytes from obese people were also less responsive to sympathetic-induced browning signals and showed lower levels of lipolysis, even though total metabolic activity and thermogenic capacity might remain unchanged [133,153].

Aging also affects the browning capacity of adipose tissues because aged beige progenitor cells lose the ability to acquire a full brown characteristic [134]. Aged people also tended to have lower cold-induced thermogenic capacity compared to young adults [135]. Body build may also affect browning. In an in vitro study, which demonstrated that sera from naturally skinny people, those with BMI ranged from 18 to 25 but failed to increase weight gain despite calories intake far bypassing the body requirement, was potential in driving the differentiation of brown adipocytes [136]. The mechanism underlying naturally skinny people and their metabolism with energy expenditure is unknown, however, limited evidence suggested the genetic copy numbers of salivary amylase gene (AMY1), which inversely correlated with BMI, and ventromedial hypothalamus (VMH) regulation of skeletal muscle and BAT thermogenesis, as well as energy balance upon melanocortin receptor activation [154,155]. However, it was found that sera from naturally skinny people significantly induce brown commitment of MSCs with high mitochondrial content, reduced size of lipid droplets, and increased expression of UCP1 [136]. Such findings suggested that there were factors effective in driving BAT formation in the sera of skinny people; thus, it raised questions about the identification of those factors and whether they can be utilized for therapeutic treatment. Gender and sex hormones also have profound effects on thermogenesis and BAT function [137,138,156]. In both animal and human studies, female individuals have been shown to be more sensitive to cold-induced thermogenesis and have more BAT content when compared to males [139,156]. Specifically, the binding of oestradiol (E2), a form of female oestrogen hormone, to oestrogen receptor alpha activated the VMH, and increased activation of the sympathetic nervous system via the modulation of AMPK signalling in the VMH, which consequently, resulting BAT activation and upregulation of Ucp1, Pgc1α and Pgc1β [157]. The levels of 17β-oestradiol were also found to be correlated with BAT thermogenesis in human [158]. These data provide evidence about the correlation of BAT activation with sex hormones in different individuals.

BAT abnormal development follows the changes in physiological conditions such as diet and metabolic diseases, which can lead to altered thermogenesis, inhibiting browning, and enhancing whitening of BAs [120,134,159]. It has been shown that mice fed with a high-fat diet (HFD) developed hypertrophic BAT and unilocular adipocytes, which were infiltrated with macrophages and T cells [120]. Even though the changes in BAT morphology are associated with the increase in ER stress and ROS production, HFD BAT obtained maximal metabolic capacity, and no sign of fibrosis, apoptosis, or dysfunctional thermogenesis was detected, suggesting the stability of BAT function in systemic energy expenditure [120]. When mice fed with HFD were treated with bone morphology protein 7 (BMP7), a browning-inducing factor, BAT enhanced energy expenditure and reduced fat mass and body weight [160]. Only when prolonged HFD is maintained, it can compromise thermogenic capacity. These findings suggested that BAT could be a promising treatment for obesity due to its resistance to changes in environmental conditions, and the abilities to promote improvement in systemic metabolism. In addition, during HFD, BAT with deficient HIF2α showed increased inflammation, enlarged adipocytes, enhanced lipogenesis, and decreased angiogenesis, which resulted from upregulation of lipogenic genes and reduced expression of VEGFα, however, the loss of angiogenesis can be reversed by treating BAT with VEGFα [161], suggesting that angiogenesis may be one of the key regulators of BAT thermogenesis [125], and may be the prone site for the pathogenesis of BAT. This is in line with a recent study showing the involvement of endothelial cell lysosomal metabolism in regulating thermogenic capacity, which supported cell proliferation and brown adipocyte remodelling via the processing of TRLs inside lysosomes [110]. On the other hand, in uncontrolled diabetic STZ-rats, the action of TH even though could upregulate the expression of enzyme odothyronine deiodinases (D2), PGC-1α and UCP1 in BAT, there was no changes in glucose uptake, which could not reverse diabetic hyperglycaemia [140], suggesting that metabolic disorder results in BAT dysfunction and healthy perfusion and regulation is important for BAT to regulate energy expenditure. In fact, proper BAT function also requires brown adipocytes to regulate their own mitochondrial homoeostasis via proper mitophagy [107]. Increased mitophagy is observed in BAT activated by both cold exposure and β3-agonist stimulation, and has been shown to be critical for BAT to maintain optimal UCP1 activity and the removal of damaged mitochondria during thermogenesis [107]. In conditions such as cachexia in cancer, the alteration in systemic metabolism results in the changes in the expression of lipolytic enzymes and thermogenic genes, which are marked by increased UCP1 expression while mitochondrial protein does not change [141]. As a result, BAT exhibit high thermogenesis, resulting in elevated temperatures and hypermetabolic in cachexia, as well as worsening the medical condition of mice models [141]. These data indicated that the activation of BAT and the browning process can be expanded and manipulated in different individuals with varied physiological and metabolic characteristics and the pathological condition may reduce BAT metabolic effect systemically.

With increased interest in the search for an effective browning method in application, it is important to understand how browning is regulated and the capacity of browning in different conditions. Perhaps, depending on the personal health profile, BAT storage and activation can be different between individuals, suggesting that BAT application may need to be associated with personalization setup. In conclusion, brown adipocytes are subject to the changes in different physiological and metabolic conditions. Thus, to achieve the optimal potential of BAT, it is important to understand the mechanism of browning and brown adipocyte recruitment with respect to different health conditions and individuals. The identification of regulatory factors essential for BAT activation and differentiation is crucial for maintaining BAT function and ensuring the success of using BAT in clinical applications. Moreover, the activation and presence of brown adipocytes are important for systemic metabolism, as BAT metabolism and thermogenesis provide factors that function similarly to cytokines to recruit more brown adipocytes and accelerate energy expenditure in distant tissues [116].

4. Conclusion

The ability to generate heat instead of ATP in the form of energy and to regulate metabolic homoeostasis marks brown adipocytes potentially for cell therapy that can be safely applied in metabolic disorders such as obesity and diabetes. In this paper, we have mentioned our points that thermogenesis in BAT is not solely specific to UCP1 and the mitochondria. Brown adipocytes have multiple regulatory repertoires to sustain heat generation, creating a complex thermogenic machinery that can adapt to specific conditions to enhance thermogenesis. Thus, it is essential to understand the precise mechanism of how BAT activation is controlled and induced, and how the function of BAT can be managed and optimized. It appears that BAT thermogenesis and metabolic regulatory functions are synchronized in a network of communication with the signals generated locally from within the cell organelles to systemically transferring to nearby tissues, and further propagating to the body system [80,162]. Because BAT are composed of adipocytes with various thermogenic capacities [4,163], the question of how to identify and recruit high-performance brown adipocytes, or how to improve adipocyte browning, may open up a new research avenue, laying the groundwork for advanced BAT applications.

Despite our lengthy discussion about the machinery of thermogenesis, the specific characteristics underlying the parameters of BAT activation, as well as quantitation for how much BAT is needed to maintain thermogenesis and systemic effects, remain a critical question for generating a more clinically related outlook. In other words, there are no definite morphologic/phenotypic parameters that can translate BAT activity into metabolic regulation. Browning is, thus, more a concept of activation to improve energy expenditure by adipose tissue than the necessity to increase UCP1 expression and thermogenesis in local BAT. The activation of browning should be directed towards enhancing the ability of BAT in integration with other tissues to generate a systemic effect on energy expenditure. Furthermore, the ability to prolong and sustain the thermogenic and metabolic homoeostasis of BAT is also a considerable aspect in clinical application. It is known that the effectiveness of BAT relies on personal physiology, as BAT activation is impaired in metabolic diseases and ageing conditions [134,164]. Preserving BAT function and browning potential is important on healthy ageing and metabolic outcomes. Therefore, an understanding of how metabolic regulating capacity is regulated in different physiological and pathological conditions will be essential for the control of BAT in clinical applications, which we suggest to be developed as a personalized treatment rather than a general therapy. Recently, cold-induced thermogenesis involved in BAT has been shown to enhance total glucose metabolism, which competed with tumours for glucose availability, and consequently changed tumour metabolism and inhibited its growth and metastasis [165]. These data suggested a new application for BAT in therapeutic application as an energy regulator of tumour cell growth, which, if utilized in a targeted manner, may propose a promising supporting treatment for cancers. Additionally, it should be clarified how the use of exogenous BAT would be most beneficial in different physiological conditions, and how thermogenesis can be safely exploited. Despite these obstacles, using BAT and brown adipocytes over drugs and other chemical treatments would be a safe and specific treatment for patients with metabolic conditions, and we hope to see future work to advance the research.

Abbreviations

T3	=	3,3′,5-TRIIODOTHYRONINE
AC	=	ACYLCARTINIES
APC	=	ADIPOSE PROGENITOR CELL
ATGL	=	ADIPOSE TRIGLYCERIDE LIPASE
BMP	=	BONE MORPHOLOGY PROTEIN
BAT	=	BROWN ADIPOSE TISSUE
SERCA2b	=	CA2+ ATPASE LOCALIZED INTO THE SARCOPLASMIC RETICULUM
CL	=	CARDIOLIPIN
cAMP	=	CYCLIC ADENOSINE MONOPHOSPHATE
ER	=	ENDOPLASMIC RETICULUM
EC	=	ENDOTHELIAL CELLS
ERR	=	ESTROGEN-RELATED RECEPTOR
FA	=	FATTY ACIDS
FGF	=	FIBROBLAST GROWTH FACTOR
FOXO	=	FORKHEAD BOX O
FFA	=	FREE FATTY ACID
GPR20	=	G PROTEIN REGULATOR 20
HFD	=	HIGH-FAT DIET
Hdac4	=	HISTONE DEACETYLASE 4
IGF-1	=	INSULIN-LIKE GROWTH FACTOR
IL-6	=	INTERLEUKIN 6
LD	=	LIPID DROPLETS
LPL	=	LIPOPROTEIN LIPASE
LAL	=	LYSOSOMAL ACID LIPASE
mTOR	=	MAMMALIAN TARGET OF RAPAMYCIN
MSC	=	MESENCHYMAL STEM CELL
Nrf1	=	NUCLEAR FACTOR ERYTHROID-2, LIKE-1
PPAR	=	PEROXISOME PROLIFERATOR-ACTIVATED RECEPTORS
PKA	=	PROTEIN KINASE A
SR	=	SACOPLASMIC RETICULUM
AMY	=	SALIVARY AMYLASE GENE
SIRT	=	SIRTUINS
SNS	=	SYMPATHETIC NERVOUS SYSTEM
TH	=	THYROID HORMONE
TR	=	THYROID RECEPTORS
T4	=	THYROXINE
TRL	=	TRIGLYCERIDE-RICH LIPOPROTEIN
DIO2	=	TYPE II DEIODINASE (DIO2)
UCP1	=	UNCOUPLING PROTEIN 1
VEGF	=	VASCULAR ENDOTHELIAL GROWTH FACTOR
VMH	=	VENTROMEDIAL HYPOTHALAMUS
WAT	=	WHITE ADIPOSE TISSUE

Author’s contributions

Van Thi Tuong Nguyen planned and constructed the paper, and did the major work for the review. Vuong Van Vu contributed to writing defined sections and the major drawing of the diagrams. Phuc Van Pham suggested the ideas, and correct figures, edit the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work was supported by the Stem Cell Institute, University of Science Ho Chi Minh City, Viet Nam [04-2023].