I. Introduction

Adipose tissue plays a central role in the interplay between nutrition, energy balance, and human health. There are 2 types of adipose tissue, white and brown. White adipose tissue (WAT) stores energy, whereas brown adipose tissue (BAT) dissipates it. Overnutrition and/or physical inactivity result in an excess of WAT, the hallmark of obesity. In contrast, BAT is thermogenic, a property conferred by the presence of a unique protein, uncoupling protein 1 (UCP1). Located in the inner mitochondrial membrane, UCP1 uncouples mitochondrial respiration, releasing energy as heat. This unique property protects animals from hypothermia (1).

The traditional belief that BAT exists only in infants but not in adults has resulted in a paucity of research in humans. However, the discovery of fat with high metabolic activity in adults by functional imaging using positron emission tomography (PET) brought about a resurgence in research interest on BAT identity, abundance, prevalence, regulation, and significance in humans (2–9).

This review will cover 1) the characteristics and ontogeny of BAT, 2) its prevalence and regulation, 3) metabolic relevance, 4) the potential roles of BAT in health and diseases, and 5) the avenues for therapeutic targeting of BAT in obesity. These questions will be discussed on the background of known biology of BAT in rodents.

II. A Historical Perspective

Understanding the biology of a body organ requires knowledge of location, morphology, regulation, and function. In contrast to other classic metabolic organs, such as skeletal muscle, WAT, and liver, the biology of BAT in humans has remained elusive since Gessner first described its presence in hibernators in 1551. The 1960s heralded a golden era of human BAT research that withered in the late 1980s with the view that significant deposits of BAT did not persist beyond childhood. In the last decade, unequivocal evidence of BAT in adult humans has led to resurgence in global research interest. Table 1 summarizes major developments in human BAT research.

Data table     

Table 1. Timeline and Summary of Major Developments in Human BAT Research

At the beginning of the last century, anatomists described similarities between fat masses located in the dorsal and cervical region of human fetuses and fat depots in the interscapular area of hibernating mammals (10–12). It was, however, not until the 1960s that BAT was ascribed a regulatory role in thermogenenesis (13–18). It was proposed that BAT was a heat-producing tissue in small mammals and human infants, defending newborns from hypothermia. BAT is histologically and functionally distinct from WAT, and the presence of the facultative proton transporter UCP1 confers upon it the unique ability to generate heat through respiratory uncoupling (19).

The thermogenic properties of BAT originally were of interest only to a few scientists studying hibernating animals. Serial publications revealed a 6-fold increase in heat production from BAT after cold acclimatization in rodents, dissipating heat to the body via dense vascularization juxtaposing deep viscera (13, 20–22). The striking thermogenic capacity of BAT led some researchers to regard it as an electric blanket for animals in the cold (23). Because temperature changes are cues to food availability in nature, BAT studies were extended to investigating response to nutrient variations. In the 1970s, Rothwell and Stock (24) observed near identical morphological changes in BAT between cold-exposed and high-fat diet-fed animals. Heat production in BAT during cold exposure corresponded closely to that after high-fat feeding. The strong association between diet-induced thermogenesis (DIT) and cold-induced thermogenesis (CIT) in animals led to the proposal that BAT played a major role in both. Meanwhile, from human cadaveric studies, Heaton (25) found that BAT persisted up to the eighth decade of life. These findings led to the hypothesis that BAT failure could contribute to development of obesity in adult humans (26–29), resulting in a tripling of BAT publications between 1980 and 1982.

During this early phase of human BAT research, investigations were restricted to examining depots around the adrenal bed, a location accessible during elective abdominal surgery. This approach overlooked BAT in extra-abdominal locations, underestimating its abundance and distribution in adults. The scientific consensus at the time did not support a definite metabolic role in energy homeostasis in adult humans (30, 31), with Rothwell and Stock raising the doubt of “Whither brown fat?” (32). It was recognized there were major difficulties identifying BAT depots in humans and that the view would “continue to be controversial until a method for quantitative noninvasive measurement of total BAT thermogenesis is developed” (30, 33).

It took another 2 decades for such noninvasive methods to become available. PET scanning technology has ushered in a new era of metabolic imaging, catalyzing the resurgence in BAT research. The rebirth of human BAT research interest has been viewed as a renaissance in metabolic medicine (34). The research questions in human BAT are the same as those posed in the 1980s; however, the field has been enriched by advances and insights from animal studies in the intervening years. This review will integrate new knowledge from animal studies into an appraisal of its physiological significance in humans.

III. Characteristics and Ontogeny

A. Tissue characteristics

White adipocytes are typified by their round appearance, containing a large single lipid droplet, displacing the nucleus to the periphery. Morphologically, brown adipocytes differ from white adipocytes by their polygonal shape, smaller size, central nucleus, and numerous small lipid droplets, lending them a multilocular appearance. Brown adipocytes contain well-developed mitochondria filling most of the cytoplasm. The high density of mitochondria is comparable to that of cardiomyocytes. In contrast, the mitochondrial content of white adipocytes is low. BAT is also endowed with a rich blood and nerve supply and a dense niche of perivascular mesenchymal cells, which are a nursery of preadipocytes. The abundance of mitochondria containing respiratory chain cytochrome enzymes with iron as a cofactor, as well as the vasculature within BAT, gives rise to a darker red (brown) color, compared with the paler hue exhibited by WAT.

B. Ontogeny

Because both WAT and BAT accumulate lipid droplets, the traditional view is that they share a common developmental origin. Recent evidence has revealed the existence of brown adipocyte-like cells within WAT, which harbor transcriptional regulators for differentiation into brown adipocytes. This section will outline common factors determining brown and white adipogenesis and developmental differences between BAT and WAT. The implications of these findings in human BAT are explored.

1. Transcriptional control of adipogenesis

Information on the developmental origin of BAT is drawn from rodent studies. BAT and WAT originate from mesenchymal stem cells, which can differentiate into adipocytes, osteoblasts, chondrocytes, and myoblasts (35). Adipogenesis comprises the 2 stages of commitment and differentiation. With appropriate stimulation, mesenchymal stem cells undergo a process of commitment in which they are recruited to an adipocyte lineage (preadipocytes), subject to clonal expansion, followed by terminal differentiation into mature adipocytes.

Brown and white adipocyte differentiation begins with commitment to adipogenesis, a process culminating in the accumulation of intracellular lipid. Adipogenesis is controlled by a cascade of interactions between several transcription factors, chiefly the CCAAT/enhancer-binding proteins (C/EBPs) α, β, and δ; peroxisome proliferator-activated receptor γ (PPAR-γ); and steroid response element-binding protein 1c (SREBP1c) (36, 37) (Figure 1).

figure

Figure 1. Origins and gene expression signatures of white, beige/brite, and brown adipocytes. Mesenchymal stem cells are multipotent, giving rise to the precursors of different adipocyte populations in addition to chondrocytes, osteoblasts, and myocytes. In these precursor cells, those expressing Myf5, originally thought to be a myogenic marker, can be differentiated to produce either brown adipocytes or myocytes (58). The transcriptional regulator PRDM16 controls the fate of these cells. Alternatively, a pool of Myf5-negative precursor cells can be induced to form white adipocytes as well as a recently identified population of beige/brite adipocytes in depots of white adipose tissue (84). Beige adipocyte precursors can be separated from other (white) preadipocytes by sorting using the markers CD137 or TMEM26 (97). Terminal differentiation in all 3 types of adipocytes is controlled by a series of adipogenic transcription factors, including PPARγ and C/EBPα, -β, and -δ. Brown and beige adipocytes are also enriched for the transcriptional coactivator PGC1-α, which directs the expression of key regulatory molecules responsible for mitochondrial biogenesis in response to external stimuli, such as cold exposure (50). Once differentiated, these different types of adipocytes (and the depots enriched in these cell types) can be identified by their unique gene expression signatures (68).

PPAR-γ is the most critical of these factors. It belongs to a superfamily of hormone nuclear receptors orchestrating the recruitment of adipogenic elements during preadipocyte differentiation. PPAR-γ is essential for both brown and white adipogenesis. Both mature brown and white adipocytes express high levels of PPAR-γ (38–40). In mice, deletion of PPAR-γ selectively in adipose tissue diminishes fat mass, whereas ectopic overexpression promotes adipogenic differentiation of nonadipose cells such as myocytes and fibroblasts, attesting to its importance as a master regulator of adipogenesis (41–43). Although many activators and repressors of adipogenesis have been identified, most modulate the expression/activity