3939

SummaryAdipose tissue is formed at stereotypic times and locations in adiverse array of organisms. Once formed, the tissue is dynamic,responding to homeostatic and external cues and capable of a15-fold expansion. The formation and maintenance of adiposetissue is essential to many biological processes and whenperturbed leads to significant diseases. Despite this basic andclinical significance, understanding of the developmental biologyof adipose tissue has languished. In this Review, we highlightrecent efforts to unveil adipose developmental cues, adiposestem cell biology and the regulators of adipose tissuehomeostasis and dynamism.

Key words: Adipose, Adipocyte, Development, Stem cells, Niche,Vascular niche

Introduction: fat factsAdipose tissue plays a myriad of roles; it serves as a central nexusof metabolic communication and control, an arbiter ofthermoregulation, a buffer against trauma and the cold, and aregulator of reproduction and satiety. Fat is also associated withemotionally charged issues, imparting various psychosocialimprints that have changed over the centuries: from a sign of wealthin the middle ages, to a Rubenesque celebration during theRenaissance, to fear and loathing in modern Hollywood (unless ofcourse it is injected into the lips or other ‘cosmetically desirable’locations). Despite the current social aversion to fat, the increase inthe percentage of people who are overweight has been marked andis observed in nearly all regions of the world. The global prevalenceof obesity is ~15% and in the United States more than two thirdsof the population is overweight (Hossain et al., 2007). The healthcare system is reeling not only from an obesity epidemic, but alsofrom a host of obesity-related negative sequelae including diabetes,cardiovascular disease, cirrhosis and even cancer. The reasons forthis increase are well known – increased food consumption anddecreased exercise – but this is only because the body ispredisposed to fat accumulation.

What is also becoming clear is that a variety of evolutionary anddevelopmental forces have laid the foundation for the developmentof obesity. Both invertebrates and vertebrates have fat-storingtissues and many aspects of the mechanistic underpinnings areconserved. These include the developmental programs,transcriptional cascades and basic proteins that regulate fatsynthesis, storage and lipolysis (McKay et al., 2003; Suh et al.,2006; Suh et al., 2008; Suh et al., 2007). It appears that fat tissuesevolved primarily as a safe harbor to store energy in times of plentyand to provide fuel when food sources become insufficient.However, in addition to serving as purely a storage depot, adipose

tissue is now recognized as the body’s largest endocrine organ,controlling many aspects of systemic physiology by secretinghormones (adipokines), lipids, cytokines and other factors (Gestaet al., 2007; Nawrocki and Scherer, 2004; Spiegelman and Flier,2001). Although many of the regulatory molecules are not yetidentified, they control a wide variety of biological actions,including appetite, glucose homeostasis, insulin sensitivity, aging,fertility and fecundity, and body temperature.

In mammals, adipose tissue forms in utero, in the peripartumperiod and throughout life. Notably, even in adult humans newadipocytes are generated continually and at substantial rates(Spalding et al., 2008). Adipose tissue is composed of adipose stemcells (the precursor cells that give rise to new adipocytes),adipocytes (the fat-storing cells) and various other cell types, whichinclude mural, endothelial and neuronal cells. Adipose tissuecomprises various discrete depots, such as inguinal, interscapular,perigonadal, retroperitoneal and mesenteric depots, which areplaced in defined positions throughout the body. These variousdepots develop at specific and distinguishable pre- and postnataltimes and they have discrete and distinct morphologies (Gregoireet al., 1998; Hossain et al., 2007; MacDougald and Mandrup,2002). These observations support the notion that developmentalcues are pivotal for the proper formation, migration andmorphology of adipose tissue. Yet relatively little is known aboutadipocyte development because adipocytes are spread throughoutthe body, because of inherent difficulties in handling the enormousand fragile adipose cells, and because of historic reliance on cellculture models.

However, a new horizon is in sight; developmental tools arebeginning to be utilized and interest in adipose tissue developmentis expanding. This recent tectonic shift towards a developmentalfield has begun to improve our understanding of adiposedevelopmental biology and has positioned the field for extensiveadvancement. This is timely, as the obesity crisis has increased thedemand for new therapeutic strategies for adipose-related illnesses.Thus, there is an urgent need to understand better the mechanismsand molecules that control the formation of adipocytes and theexpansion of adipose tissue due to caloric excess. Here, we reviewthe recent advancements in adipose tissue development, thediscovery and regulation of adipose stem cells, adipocyte turnover,and potential therapeutic interventions for those suffering fromobesity and diabetes.

Brown and white fat: flavors of the monthIn mammals, there are two main classes of adipose tissue that areseparated into major histological divisions, are molecularly distinctand are functionally distinguishable. White adipose tissue (WAT)is designed for energy storage whereas brown adipose tissue (BAT)dissipates energy and generates heat. WAT is far more prevalentand more biologically relevant, whereas BAT primarily functionsin the neonatal period, although recent reports have shown BAT tobe present and have functional roles in adults (Cypess et al., 2013;Lidell et al., 2013). Adipose tissue is also present in a variety oflocations, such as palms, soles, scalp, periarticular regions and

Development 140, 3939-3949 (2013) doi:10.1242/dev.080549© 2013. Published by The Company of Biologists Ltd

The developmental origins of adipose tissueDaniel C. Berry1, Drew Stenesen1, Daniel Zeve1 and Jonathan M. Graff1,2,*

1Department of Developmental Biology, University of Texas Southwestern MedicalCenter, Dallas, TX 75390, USA. 2Department of Molecular Biology, University ofTexas Southwestern Medical Center, Dallas, TX 75390, USA.

*Author for correspondence (jon.graff@utsouthwestern.edu)

REVIEW

Dev

elop

men

t

3940

orbits, that are thought to have so called mechanical roles (Gesta etal., 2007). In addition, adipocytes are positioned throughout theskin and bone marrow, and recent evidence indicates that in thoselocations adipocytes regulate stem cell biology, for example bycontrolling epidermal and hematopoietic lineage decisions.

White adipose tissue: the great white whaleWhite adipose tissues coordinate systemic metabolism. This keyrole is highlighted by the metabolic dysfunctions (e.g.hyperglycemia, hyperlipidemia, hypertension, diabetes, liverdisease, increased carcinogenesis, etc.) that are a consequence oftoo many (obesity) or too few (lipodystrophy) white adipocytes.There are two main divisions of white adipose tissue: subcutaneousand visceral. Subcutaneous and visceral depots are simplyseparated based upon gross anatomical location (Fig. 1A,B). Yetthis gross placement appears to define a variety of essentialfeatures, including developmental timing, microscopic appearance,molecular signature and biological function. In rodents,subcutaneous depots, such as interscapular and inguinal, appear toform prior to visceral depots, which include retroperitoneal,perigonadal and mesenteric WAT (Fig. 1C,D). Each adipose depothas a specific, distinct and reproducible morphology and texture.Histologically, subcutaneous fat is heterogeneous and contains

mature unilocular adipocytes intercalated with small multilocularadipocytes, whereas visceral fat is more uniform and appears toconsist primarily of large unilocular adipocytes (Fig. 2A,B)(Tchernof et al., 2006; Tchkonia et al., 2007). Of note,retroperitoneal adipose depots, which are placed within the visceralclass, have features that may be intermediate between subcutaneousand visceral depots. Current thinking indicates that subcutaneousand visceral depots are likely to have different contributions tometabolism, with inherent health ramifications. Increasedsubcutaneous fat deposition, sometimes called a pear-shaped orfemale pattern of distribution, might protect against certain aspectsof metabolic dysfunction (Snijder et al., 2003a; Snijder et al.,2003b). However, visceral depots, also known as an apple or malepattern, are thought to be associated with metabolic complicationsand appear to increase the risk of diabetes, hyperlipidemia andcardiovascular disease (Grauer et al., 1984). Because of these stillill-defined associations, it has become popular to termsubcutaneous adipose as ‘good fat’ and visceral as ‘bad fat’.

What might account for these associations? It is plausible thatthe aforementioned histological differences may be at play.Subcutaneous depots contain an abundance of small multilocularadipocytes, which might be protective against metabolicderangement (Salans et al., 1973; Weyer et al., 2000). In addition,

REVIEW Development 140 (19)

Visceral

PGW RPW MWAT

D

ISCW

IGW

SQC

A

P

SQ

Vis

Obese

Mou

se

SQ

Vis

A

P

Lean B

Vis

A

P

Lean A

P

Obese

Hum

an

A

Vis

SQSQ

Fig. 1. The anatomical distributions of fat.(A,B) Coronal MRI of a lean (left) and an obese(right) human (A) and mouse (B); yellow linesdelineate subcutaneous and visceral depots. A,anterior; P, posterior; SQ, subcutaneous; Vis,visceral. (C,D) Physical orientation, distributionand location of subcutaneous and visceraldepots. Subcutaneous fat (C) is located in theinguinal (IGW) and interscapular (ISCW) regions.Visceral fat (D) is positioned at perigonadal(PGW), retroperitoneal (RPW) and mesenteric(MWAT) regions of the body. Yellow arrowsindicate the denoted depot location within thebody of the mouse.

Dev

elop

men

t

3941REVIEWDevelopment 140 (19)

there is an increase in interstitial tissue in subcutaneous depots(Fig. 2A). In conjunction with histological differences,subcutaneous depots have increased rates of adipose turnover andnew adipocyte formation, and it is postulated that ‘younger’adipocytes do not confer metabolic dysfunction (Björntorp et al.,1971; Salans et al., 1973). Lipolytic rates also appear to be differentbetween the two depots (Fisher et al., 2002). Other factors thatmight account for the differences include blood flow and proximityto metabolically important visceral organs, such as the liver, whichproduces glucose, lipids and cholesterols, and the intestine wherefood is absorbed. It is also known that aspects of regional fatdeposition are inherited; for instance, it is widely appreciated thatthe propensity to accumulate subcutaneous versus visceral fat runswithin families (Shi and Clegg, 2009). That is, one can only controlwhether one becomes fat, not where one puts it.

Subcutaneous and visceral adipose depots also respond toexternal stimuli in different ways. For example, in humans andmice, steroid hormones confer depot-specific consequences (Shiand Clegg, 2009); estrogen is thought to increase subcutaneousdepots, whereas adipose tissues within the neck, or within theviscera, are more responsive to glucocorticoids. Furthermore,subcutaneous and visceral depots display unique anddistinguishable responses to thiazolidinediones (TZDs), a class ofdrugs widely prescribed for diabetes (Tontonoz and Spiegelman,2008). Additional support for complexity within the lineagesderives from the clinical observation that lipodystrophy(pathological absence of fat) preferentially affects different adiposelocations. For example, some patients with an inheritedlipodystrophy lack all metabolic fat but retain normal amounts ofmechanical fat (Herbst, 2012).

Transplantation methodologies have been employed in anattempt to probe whether the functional differences betweenvisceral and subcutaneous depots might be intrinsic (autonomous)

or extrinsic (non-autonomous). However, these studies have not yetproduced a definitive answer, with some indicating inherentdifferences and others suggesting that these functional attributesalter when locations change. For example, Kahn and colleaguesperformed an elegant series of murine reciprocal transplantscoupled with extensive metabolic analyses (Tran et al., 2008). Thestudies indicated that transplantation of subcutaneous depots intovisceral locations improved metabolic parameters, whereastransplantation of visceral depots into subcutaneous locations didnot (Tran et al., 2008). In addition to metabolic and morphologicalchanges, a striking rearrangement in gene expression towards asubcutaneous signature occurred within three weeks oftransplantation (Satoor et al., 2011). Thus, there appear to be bothintrinsic and extrinsic forces that regulate adipose depot biology.The factors and relative contributions of each are likely to havefundamental and clinical significance and are areas of futureexploration.

A variety of studies support the notion that distinct and definablemolecular programs might underlie the functional differencesbetween subcutaneous and visceral adipose tissues (Gesta et al.,2007). For example, gene expression profiling of adipose tissues inrodents and humans demonstrated divergent and discrete molecularsignatures between the various depot locations. Notably, the patternof gene expression, even within a depot, changes depending onadipocyte size (Jernås et al., 2006). Subcutaneous fat depots tendto express higher levels of leptin, angiotensinogen and glycogensynthase compared with visceral fat. Omental fat, part of thevisceral compartment, expresses increased levels of the insulinreceptor, 11β hydroxysteroid dehydrogenase (11β HSD) andinterleukin 6 (IL6) (Gesta et al., 2007; Masuzaki et al., 2001).Large-scale gene expression analyses comparing rodentsubcutaneous and visceral adipose progenitor signatures alsoidentified a host of differentially expressed transcripts (Gesta et al.,2006). These include a group of developmental regulators such ashomeobox (HOX) and forkhead box (FOX) family transcriptionfactors (Gesta et al., 2006; Macotela et al., 2012). Moreover, thegene trends that were detected in mice were also observed inhuman adipose samples (Gesta et al., 2006). Additional datasuggest that a subset of these genes correlate with body mass index(BMI) and by extension relate to metabolic dysfunction andcardiovascular disease (Gesta et al., 2006). The role of these genesin fat formation and regulation remains unclear. However, many ofthe identified genes are involved in developmental patterning, andit is possible that they might regulate aspects of adipose tissueformation. Studies on developmental signaling cascades have alsohighlighted the notion that developmental forces are important toadipose development and to regional distinctions. This can beillustrated by the WNT pathway, which regulates adipose tissueformation in a spatially significant manner (Longo et al., 2004;Zeve et al., 2012). In relation to gene changes, fat-storing cellsderived from subcutaneous progenitors accumulate more lipid andexpress higher levels of PPARγ and C/EBPα upon differentiationcompared with visceral progenitor cells (Baglioni et al., 2012;Tchkonia et al., 2002). Subcutaneous progenitor cells also haveimproved growth rate and different electrophysiological properties(Baglioni et al., 2012; Macotela et al., 2012). Collectively, thereappear to be intrinsic genetic depot-specific differences in adiposestem cells that result in different adipogenic potential, geneticexpression patterns, growth rates and biological properties. Thus,these studies support the notion that distinct developmental cuesare likely to determine the complexity of the adipose lineage andthat all fat stem cells and adipocytes are not created equally.

A B

SQ Visceral Brown adipose

C

Fig. 2. Heterogeneity of fat. (A-C) Adipose depots are positioned atstereotypic times and locations with defined sizes; shown aresubcutaneous (SQ) fat depots (A), perigonadal (visceral) fat depots (B) andintrascapular brown adipose tissue (C) from three unrelated mice.Histologically, white adipose depots differ in adipocyte size;subcutaneous depots are more heterogeneous, containing both uni- andmultilocular adipocytes and interstitial tissue, whereas visceral depots aremore uniform and contain mostly unilocular adipocytes. Morphologically,brown adipose tissue is very distinct from white adipose tissue withuniformity of multilocular lipid droplets.

Dev

elop

men

t

3942

Brown adipose tissue: burning down the houseBrown adipocytes convert nutrients into chemical energy in the formof heat (Kajimura et al., 2010). The primary role of BAT appears tobe in the neonatal period when temperature control is challenged(Gregoire et al., 1998; MacDougald and Mandrup, 2002; Spiegelmanand Flier, 2001). Brown fat cells express a unique thermogenic andmitochondrial genetic program that promotes mitochondrialbiogenesis, energy uncoupling and energy dissipation, whichprovides essential heat to the organism (Kajimura et al., 2010).Energy dissipation is accomplished by an abundance ofmitochondria, hence the eponymous color of BAT, as well as byspecialized proteins, including uncoupling protein 1 (UCP1). UCP1collapses the electron gradient to generate heat rather than ATP(Cannon and Nedergaard, 2004). Similar to WAT, BAT is present atseveral stereotypic places and can be increased or decreased byenvironmental cues (Fig. 2C). However, the brown adipocytes thatare present during neonatal life appear distinct from those that arepresent, or can be induced, in adults (Wu et al., 2012).

In the neonatal period, the most prominent BAT depot is locatedat the dorsal region between the scapulae (Cannon and Nedergaard,2004). The cells in the interscapular region of the body are derivedfrom the dermomytome, an unusual tripotent engrailed 1 (EN1)-positive cell lineage, which gives rise to brown fat bundles, thedermis and muscle cells (Atit et al., 2006). Studies also indicatethat this specialized region of the body descends from a myogenicfactor 5 (MYF5)-positive source, but it is clear that only a subsetof brown adipocytes do so. Interestingly, recent data indicate thata subset of white adipocytes also arises from MYF5-positive cells(Sanchez-Gurmaches et al., 2012), and additional studies show thatbrown and white fat progenitor cells share commonality.Granneman and colleagues demonstrated that a platelet-derivedgrowth factor receptor alpha (PDGFRα)-positive lineage traces tobrown and white adipose tissue and that PDGFRα is present inproliferating brown and white adipose progenitors (Lee et al.,2012).

Brown adipocytes were once thought to be absent in adulthumans; however, recent data indicate that some adults haveenergy-burning adipocytes with ‘brown-like’ characteristics, andpromotion of their formation or maintenance has potential as ananti-obesity avenue (Cypess et al., 2009; Cypess et al., 2013; Lidellet al., 2013; Whittle et al., 2011). In small-scale studies, it appearsthat obesity reduces the appearance of brown-like cells and,because of the therapeutic possibilities, a variety of studies havebeen directed towards elucidating the molecular cascades andfactors underlying brown fat formation, with the aim of triggeringcells to have BAT-like energy-consuming actions (Cannon andNedergaard, 2004; Kajimura et al., 2010; Seale et al., 2008; Sealeet al., 2009). Many initial studies examined possible conservationof function between the white and brown lineages, and revealedthat aspects of the overlapping core transcriptional machinery playrelated roles in the two lineages (Kajimura et al., 2010). In addition,various enzymes involved in lipogenesis and lipolysis areexpressed in both types of adipocytes. However, it also appears thatbrown fat formation has unique aspects. For instance, bonemorphogenetic protein 7 (BMP7) can induce cultured cells toundergo a brown adipocyte differentiation response (Tseng et al.,2008). In vivo studies also highlight the potential role of BMP7 inthe formation or maintenance of BAT; Bmp7 null embryos haveless brown fat compared with controls and virally overexpressingBMP7 increases brown but not white adipose tissue (Tseng et al.,2008). Thus, it appears that there are distinct signals that promotebrown fat formation over white adipocyte formation and vice versa.

In vitro adipocyte differentiation: artificialsweetenersPioneering studies on in vivo formation of adipose tissue wereconducted in the first half of the 20th century but these werescattered and primarily observational (Clark and Clark, 1940;Napolitano, 1963; Napolitano and Gagne, 1963). Over the last fewdecades, the major focus of the field has been cell culture modelingprimarily studying in vitro adipogenesis. The term adipogenesisfirst described the transition of cultured and confluent 3T3-L1fibroblast cells to individual lipid-laden cells upon incubation in apowerful cocktail composed of artificial ‘inducers’: cAMPinducers, glucocorticoid agonists and insulin or insulin-like growthfactor (IGF) (Green and Kehinde, 1975; Gregoire et al., 1998).These fat-storing cells appeared to have several characteristics ofadipocytes and have served as the primary model. This well-studied system has led to the elucidation of the cellular andbiological events that occur and the transcriptional hierarchy thatexists during the conversion of these cells from fibroblasts toadipocyte-like cells (Rosen and Spiegelman, 2000; Tontonoz andSpiegelman, 2008). It appears that aspects of the 3T3-L1adipogenic transition also occur in other in vitro systems, such asmouse embryonic fibroblasts (MEFs) and stromal vascular (SV)cells (see below), when induced with the adipogenic induction mix(Rosen and Spiegelman, 2000).

A significant advance derived from the 3T3-L1 cellular modelwas the identification of relevant molecules expressed during fataccumulation, most notably a transcriptional hierarchy. PPARγ, anuclear hormone receptor, is the centerpiece of this transcriptionalcascade and is necessary and sufficient for in vitro adipogenesis(Chawla et al., 1994; Tontonoz et al., 1994). PPARγ controls genesinvolved in lipid storage, lipid synthesis and glucose sensing(Barak et al., 1999; He et al., 2003; Miles et al., 2000). Multiplestudies using conditional and hypomorphic alleles, all solidify therole of PPARγ in differentiation of mature adipocytes. However, avariety of data indicate that PPARγ might have significant roles inaddition to those in adipocytes and indeed may play central rolesin early adipose lineage development. For example, PPARγ isexpressed by embryonic day (E)14.5, much earlier than adipocytetissue development begins, and appears to mark the location ofsubcutaneous fat depots that are present perinatally (Barak et al.,1999; Brun et al., 1996). Furthermore, PPARγ is expressed inadipose stem cells and appears to have key functions in the adiposestem cell compartment, including roles in stem cell proliferation,self-renewal, and cell determination (Tang et al., 2011; Tang et al.,2008). Collectively, a variety of studies suggest that PPARγ iscrucial both for adipocyte terminal differentiation and as adeterminant that controls adipose tissue lineage development andlocation.

Elegant in vitro studies have also identified many additional genesthat may be expressed in adipocytes or might regulate adipocytedifferentiation and function. Multiple signaling molecules can initiatethe cell model adipogenic cascade, including insulin, thyroidhormone, glucocorticoids and TGFβ superfamily members(BMP2/BMP4/GDF3) (Chapman et al., 1985; Flores-Delgado et al.,1987; Zamani and Brown, 2011). During the initiation ofdifferentiation, a transcriptional cascade occurs with activation ofmultiple factors, such as KLF4, KLF5, SMAD1/5/8, CREB,KROX20 (EGR2), glucocorticoid receptor, STAT5A, STAT5B,C/EBPβ and C/EBPδ. In turn, these factors can alter PPARγexpression (Cristancho and Lazar, 2011; Siersbæk et al., 2012). Othertranscription factors, such as thyroid receptor, C/EBPα, SREBP-1c(SREBF1), KLF15 and LXR (NR1H), are involved later in

REVIEW Development 140 (19)

Dev

elop

men

t

3943REVIEWDevelopment 140 (19)

adipocyte maturation and regulate the expression of PPARγ as wellas that of other genes involved in fat storage (Siersbæk et al., 2012).Terminal differentiation is characterized by the appearance of lipiddroplets and the expression of lipid/carbohydrate-storage proteins[e.g. AP2 (FABP4), CD36, LPL, perilipin and GLUT4 (SLC2A4)]and adipokines (leptin) (Cawthorn et al., 2012). Multipledevelopmental signaling pathways, hormones and environmentalcues also inhibit in vitro adipogenesis, such as canonical WNTsignaling, retinoic acid, TNFα, Hedgehog (Hh) signaling,SMAD/TGFβ and HIF1α (Berry et al., 2012; Cawthorn et al., 2007;Ross et al., 2000; Suh et al., 2006; Sul, 2009; Wang et al., 2006). Insummary, studies in cell culture models indicate that the formationof fat-storing cells is controlled by complex interactions ofstimulatory and inhibitory signals. A key challenge for the field,however, is to determine whether these molecules have roles inadipose tissue formation in vivo.

Adipose stem cells: getting to the root of the stemAlthough some aspects of in vitro models reflect in vivo events, ithas become increasingly clear that in vivo adipocyte differentiationis distinct. Therefore, the identification and characterization of theadipose stem cell is essential to understanding adipose tissuedevelopment, formation and maintenance. However, ourunderstanding of the developmental biology of adipose tissue haslagged and our understanding of adipose stem cells is primitive.Stem cells are crucial components for tissue development andmaintenance, and the notion of adipose progenitors stems from the1940s. In the 1940s, it was noted that fibroblast-like cells changemorphology and acquire unilocular lipid droplets (Clark and Clark,1940). The developmental or embryonic origin of adipose stemcells remains largely unexplored and a variety of basic insights,such as germ layer derivation, remain murky. Conventionalwisdom holds that WAT originates from the mesoderm germ celllayer despite a lack of direct evidence supporting this hypothesis(Duan et al., 2007); however, not all adipose tissue originates fromthe mesoderm (Monteiro et al., 2009). For instance, a variety ofstudies, such as those using chick-quail chimeras and modern Cre-dependent cell marking studies in the mouse, indicate thatcraniofacial adipose depots originate from neural crest cells (Billonet al., 2007; Le Lièvre and Le Douarin, 1975).

Identifying and characterizing adipose stem cellsRecent efforts have begun to tease apart the endogenous adiposestem/progenitor compartment. Tissue dissociation studies have

implicated the adipose stromal vascular fraction (SVF) as apossible site of origin for adipose stem cells. The SVF is aheterogeneous mixture of cells that can be isolated from adiposetissue after dissociation and centrifugation, in which adipocytesfloat and the remaining cells pellet (Hollenberg and Vost, 1968).The SVF is composed of a myriad of cells, which includesfibroblasts, endothelial cells, hematopoietic cells and neural cells.Yet within this multiplicity of cell types the SVF appears to containbone fide adipose stem cells that proliferate, self-renew anddifferentiate (Rodeheffer et al., 2008; Tang et al., 2008; Yamamotoet al., 2007). The isolation, identification and characterization ofsuch adipose stem cells has only recently been performed usingtwo different approaches: tissue dissociation coupled withfluorescence-activated cell sorting (FACS) and lineage tracing.Utilizing cell surface markers, Friedman and colleagues identifieda CD24+ cell population within the SVF that had high adipogenicpotential in vitro (Rodeheffer et al., 2008). When this cellpopulation was injected into wild-type mice it did not form afunctional fat pad; however, when transplanted into A-Zip mice (anengineered lipodystrophic model that lacks fat) the cells were ableto form adipose-like tissue. Thus, it appears that the A-Zip mice,and possibly lipodystrophic mice in general, harbor a pro-adipogenic environment presumably containing various signals thatinduce adipocyte differentiation (Rodeheffer et al., 2008).

Contemporaneously, Tang et al. also developed a flow-basedmethod for isolating adipose stem cells that could form adiposetissue after transplantation, even in wild-type mice (Tang et al.,2008). More importantly, by exploiting the observation that PPARγis expressed at low levels in adipocyte stem cells in order to labelgenetically and lineage trace adipocyte stem cells, they reportedthat PPARγ-expressing cells are present prior to birth and longbefore adipocytes are formed (Tang et al., 2008). Moreover, thesecells proliferate, self-renew, differentiate and are capable ofrepopulating the stem cell pool and contributing to the formationof all adipocytes (Tang et al., 2011; Tang et al., 2008; Zeve et al.,2009).

Remarkably, a new notion arose from these studies, that theseadipose stem cells are present in a vascular niche, residing as asubset of perivascular mural cells within adipose depots thatexpress a variety of mural cell markers, such as PDGFRβ, α-smooth muscle actin and NG2 (CSPG4) (Fig. 3). Transplantationof PDGFRβ mural cells, isolated from adipose depots, reconstitutedadipose depots in the recipients; yet equivalent PDGFRβ-positivemural cells from other organs did not give rise to adipocytes or a

H&E

GFP

/PE

CA

M/S

MA

Tissue section

BA

SVP

C D

Fig. 3. From the vascular niche to adipocyte. (A) Illustration conceptualizing the adipose vascular niche depicting blood vessels (red), adipose stemcells (green with green nuclei), mural cells (green with black nuclei) and mature adipocytes (yellow with green nuclei). (B) Mouse subcutaneous adiposetissue stained with Hematoxylin and Eosin (H&E) stain. (C) Immunofluorescence image of mouse subcutaneous adipose tissue section showing theexpression of GFP (marking adipose stem cells, green), PECAM (an endothelial marker, red) and smooth muscle actin (SMA) (a mural cell marker, blue).(D) Stromal vascular particulates (SVPs) were isolated from mouse subcutaneous fat depots and immunostained for GFP (adipose stem cell, green),PECAM (endothelial marker, red) and SMA (mural cell marker, blue). Yellow arrows indicate adipose stem cells (green) residing at the vascular interface. D

evel

opm

ent

3944

fat pad (Tang et al., 2008). These data support the notion thatadipose stem cells reside at the vascular interface and resemblemural cells that can proliferate and differentiate into matureadipocytes.

Some recent observations raise the possibility that endothelialcells and hematopoietic stem cells might give rise to a subset ofadipocytes by changing fate into mural cells and then convertingto a progenitor population (Crossno et al., 2006; Gupta et al., 2012;Medici et al., 2010; Sera et al., 2009; Tran et al., 2012). Suchlineage changes do not have significant precedence in otherendothelial compartments but may turn out to be important inadipose biology. However, endothelial cell deletion of PPARγ didnot alter fat formation, under normal chow conditions, as would beexpected if endothelial cells are major components of the adiposefate (Kanda et al., 2009). Recently, data indicate that neitherendothelial nor hematopoietic lineage markers label adipocyteseven under high fat diet conditions (Berry and Rodeheffer, 2013).

Factors that regulate adipose stem cellsThe observation that PPARγ marks adipocyte stem cells that residein the perivascular niche raises several important questions aboutthe functional role of PPARγ within these cells. TZDs, a group ofglucose-lowering drugs prescribed for people with type 2 diabetes,are PPARγ agonists that promote insulin sensitivity and lowerblood glucose. However, a common and distressing side effect forpatients treated with TZDs is significant weight gain and increasedfat mass (de Souza et al., 2001; Hiragun et al., 1988; Sandouk etal., 1993a; Sandouk et al., 1993b). This process is multifaceted inthat TZDs stimulate new adipocyte formation and apoptosis oflarge adipocytes (Okuno et al., 1998), cause fat depositionremodeling from visceral to subcutaneous (Akazawa et al., 2000;Kelly et al., 1999), increase fluid retention (Muto et al., 2001) andaugment appetite (Shimizu et al., 1998). This clinical observationcoupled with the finding that PPARγ marks the adipose stem cellraised the possibility that TZDs might alter adipocyte stem cells,for example by inducing them to differentiate thereby increasingthe number of adipocytes and leading to weight gain (Yki-Järvinen,2004). Consistent with this hypothesis, in vivo administration ofTZDs to mice enhanced differentiation of the adipose stem cellpopulation and increased formation of additional mature adipocytes(Tang et al., 2011). Furthermore, rosiglitazone (a TZD andclinically prescribed medicine) enhanced adipose stem cellproliferation and reduced the fraction of adipose stem cells in thequiescent state. Intriguingly, chronic rosiglitazone administrationreduced the number of adipose stem cells, reduced the adipogenicpotential of the extant stem cell compartment and appeared toexhaust the adipocyte stem population (Tang et al., 2011). Thesestudies indirectly implicated PPARγ in adipose stem cellproliferation, self-renewal, lineage decisions, stem and nicheidentity and adipocyte differentiation (Tang et al., 2011). Studiesexamining these notions are a major area of future experimentation.

The adipose stem cell niche: blood brothersThe niche, the microenvironment in which stem cells reside, ishighly specialized and controls stem cell behaviors, such asproliferation, quiescence and differentiation. In the 1940s, Clarkand Clark examined fat formation in rabbit ears using a chamberedsystem to displace the tissue and watch capillary growth and fatformation into the chamber (Clark and Clark, 1940). The chambermethodology allowed ‘real-time’ visualization of fat formation ina quasi in vivo setting. These innovative experiments also indicatedthat numerous minute lipid droplets appeared first and coalesced to

form a large unilocular droplet as the cell morphology changed(Clark and Clark, 1940). Electron micrographs from the 1960sprovided further evidence that adipocyte progenitor cells reside atthe blood vessel and ‘peel’ away as they begin the transition tomature adipocytes (Napolitano, 1963; Napolitano and Gagne,1963). This notion was further supported by Hausman andcolleagues who observed the appearance of vascular structures thatslightly protrude before the formation of adipocytes (Crandall etal., 1997). Their microscopic studies also indicated thatangiogenesis recruits adipose stem cells and stimulates these stemcells to differentiate (Crandall et al., 1997). These observationswere supported by studies of Han et al., whose data suggested thatangiogenesis precedes visceral epididymal adipose tissue depotdevelopment (Han et al., 2011). This relationship between vesselsand adipose lineage cells appears to be reciprocal, as adipocytestem cells stimulate blood vessel formation. For instance, in vivomixing of adipose stem cells with endothelial cells showed thatadipose stem cells stimulate vasculogenesis and indicate thatadipocyte differentiation and angiogenesis occur in concert(Traktuev et al., 2009). Furthermore, adipose stem cells secretemultiple angiogenic regulators, some of which are transcriptionallyregulated by PPARγ, including vascular endothelial growth factor(VEGF), angiopoietin-like 4, fibroblast growth factor 2 and matrixmetalloproteinases (MMPs) (Cao, 2007; Fukumura et al., 2003;Gealekman et al., 2008; Kersten et al., 2000). Moreover, treatingadipose tissue fragments with TZDs increased angiogenic sproutformation and similar results were obtained in vivo in mouse(Gealekman et al., 2008). Inhibiting or activating angiogenicfactors controls adiposity; in the epididymal fat depot, bluntingVEGF/VEGFR2 signaling inhibits fat pad formation (Han et al.,2011), and transgenically overexpressing VEGF in matureadipocytes enhances vasculogenesis and reduces adipocyte size(Nishimura et al., 2007; Sun et al., 2012), demonstrating theorchestration between blood vessels and adipocytes.

Although the niche is known to be a major element of stem cellcontrol, examination of the niche has been hindered by difficultiesin assessing it as an intact structural unit. The ability to study theadipose stem cell vascular niche has been aided by procedures thatmaintain the native structure of the microenvironment, thusallowing the isolation of stromal vascular particulates (SVPs) as anorganotypic culture system (Tang et al., 2011; Tang et al., 2008).Isolation of SVPs showed that these vessels do not contain lipid-filled adipocytes but rather adipose stem cells that wrap around theblood vessel and express a bevy of mural cell markers (Fig. 3C,D).Upon adipogenic signals, such as TZDs, the stem cells on the SVPtransition to lipid-filled adipocytes loosely associated with theparticulate (Tang et al., 2008). Taken together, it seems likely thatthe vascular network is the niche for adipocyte stem cells, and thatthis microenvironment provides adipogenic cues to inhibit orinitiate adipocyte differentiation. The identification of such cues isan area that may attract significant attention.

Adipocyte turnover: churning the fatStem cells play important roles in tissue development as well as inhomeostatic maintenance and response to environmental stimuli.There are two distinguishable responses to adipose tissueexpansion: hypertrophy and hyperplasia. However, the level ofcontributions of the two responses varies depending upon geneticbackground, modifier effects, diet, biological and hormonal milieu,and depot preference for fat storage. Adipose tissue can expandfrom 2-3% to 60-70% of body weight in response to positiveenergy balance (Fig. 1A,B; Fig. 4D) (Hossain et al., 2007). A

REVIEW Development 140 (19)

Dev

elop

men

t

3945REVIEWDevelopment 140 (19)

hypertrophic response is characterized by pre-existing adipocytesenhancing triglyceride storage; estimates indicate that a singleadipocyte can increase its volume two- to threefold (Hirsch andBatchelor, 1976; Salans et al., 1973). A prolonged hypertrophicresponse appears to be a causal link between adipose expansionand metabolic dysfunction such as reducing adipocyte insulinsensitivity (Hossain et al., 2007; MacDougald and Mandrup, 2002).During obesogenesis, there is significantly higher adipocyteturnover, shortened adipocyte lifespan and a markedly increasedrate of apoptosis (Strissel et al., 2007). Hypertrophy also promoteslocal inflammation, for example by recruiting ‘unhealthy’macrophages to adipose tissue thereby facilitating adipocyte celldeath and other events that trigger maladaptive metabolicconsequences (Osborn and Olefsky, 2012). Furthermore,hypertrophy provokes increases in several adipocyte extracellularsignals, such as IGF1, IGFBP and TNFα, that are linked to negativeoutcomes. Of note, these factors appear able to alter various aspectsof adipose stem cell biology, such as proliferation, quiescence anddifferentiation (Cao, 2007).

Adipose tissue can also expand by forming new adipocytes,and newly formed, small adipocytes might be protective againstmetabolic dysregulation (de Souza et al., 2001; Okuno et al.,1998; Strissel et al., 2007). In the 1960s, it was shown in rodentsthat cells within adipose depots proliferate, that adipocytes arerenewed and that adipocytes have a finite lifespan (Hellman andHellerstrom, 1961; Hollenberg and Vost, 1968). Indeed, caloricexcess stimulates the adipose stem cell compartment to proliferateand differentiate. For example, Joe et al. reported a significantincrease in bromodeoxyuridine (BrdU) incorporation in the SVFof subcutaneous adipose depots in response to high-fat diet,whereas visceral fat increased as a result of hypertrophy (Joe etal., 2009). Other studies have begun to examine the source(s) ofhigh-fat diet-induced increases in proliferation observed in theadipose lineage. Several studies have since aimed to gain a betterunderstanding of adipocyte turnover but have yielded mixedfindings. In humans, it was found that 8.4% of adipocytes

turnover every year (Spalding et al., 2008). However, estimatesindicate that young adult mice generate ~15% of adipocytes eachmonth. The stem cell compartment also appears to be active withsome estimates suggesting that 4.8% of mouse adipose stem cellsare replicating at a given time (Rigamonti et al., 2011). Thesefindings suggest that adipose stem cell proliferation not onlycontributes to adipocyte formation but also replenishes the stemcell pool. Support for this notion comes from the findings thatduring postnatal development [specifically, the first 30 days oflife (P0-P30 in mice)], there is rapid turnover and replenishingevents in the adipose depot (Fig. 4A,B) (Tang et al., 2008).Furthermore, adipose tissue maintains its ability to expand as theorganism ages, and aging is associated with increased percentageof body fat, which cannot only be attributed to hypertrophy(Fig. 4C). Further studies to examine adipose stem cellproliferation are needed to define the mechanisms that controlthese events, how the stem cell divisions regulate adipose biologyand whether interventions directed towards cycle cellmanipulation are therapeutically tenable.

Adipose stem cells presumably confer the ability of adiposedepots to turnover, and they may also underlie the ability ofadipose tissue to regenerate. In humans, lipectomy and liposuctionare commonly performed medical procedures that remove excessadipose tissue. However, removal of adipose tissue causes other fatdepots to remodel with both hypertrophic and hyperplasticresponses (Mauer et al., 2001; Reyne et al., 1983), and fullregeneration of the removed fat pad also occurs leading to repeatoperations. Experiments using rodent model systems have aimedto characterize this regenerative phenomenon. When inguinal fat(subcutaneous) depots were removed from rats the fat pad wasregenerated within ~13 weeks (Larson and Anderson, 1978).Lipectomies in obese mice stimulated the rate of local and globallipogenesis and stimulated adipocyte differentiation (Bueno et al.,2011). This indicates that stem cells are potential sources of therenewal, respond to environmental cues and provide experimentalparadigms for further exploration.

SQ SQVisceral VisceralB

GFP

–Dox (P2-P30) +Dox (P2-P30)

A

P30 P2

SQ

P30 P90 P180

SQ

P30 P90 P180

Visceral SQ Visceral

Chow HFD Chow HFD

DC

P180

Fig. 4. Fat expansion. (A) Isolated postnatalday (P)2 and P30 subcutaneous (inguinal) fatdepots, demonstrating the extensiveexpansion that occurs within this shortwindow of time. (B) Adipo-trak mice (Tanget al., 2008) were treated with placebo(–Dox) or doxycycline (+Dox) at denotedtimes. In the absence of doxycycline the GFPmarker is expressed and the adipose depotsare green. In the presence of doxycycline, allGFP signal is lost, indicating that there was amassive recruitment of new stem cells andadipocyte replacement. (C) White adiposedepots enlarge with age; subcutaneous andvisceral fat depots were isolated from 30-,90- and 180-day-old mice. Depot weightnearly quintupled between 30 and 180 days.(D) Subcutaneous inguinal (SQ) and visceralperigonadal depots obtained from mice ona standard diet (chow) or a high fat diet(HFD), demonstrating the dramaticexpansion in response to HFD.

Dev

elop

men

t

3946

The therapeutic adipose stem cellThe ability to isolate and track the responses of adipocyte stemcells has provided a platform for the characterization andmanipulation of these cells, with the aim of providing somebeneficial therapeutic outcomes. Recent studies show that TZDsimprove symptoms of the metabolic syndrome in part bymobilizing adipocyte progenitors to differentiate into new insulin-sensitive adipocytes (Tang et al., 2011). This indicates that adiposestem cells are pharmacologically accessible, that increasedformation of new adipocytes may be beneficial and that adiposestem cells may be subject to additional types of manipulation.Efforts to control differentiation or proliferation might thus serveas weight loss intervention for overweight or pre-obese individuals.Alternatively, other methods directed at inducing adipocytedifferentiation in order to maintain new small insulin-sensitiveadipocytes (healthy adipocytes) might provide metabolic relief tothose suffering with obesity and metabolic dysfunction. In additionto pharmacological intervention, adipose stem cells are uniquelypoised for regenerative and genetic therapy. WAT providesminimally invasive access to large numbers of progenitor cellsnecessary for cell-based therapeutics. Furthermore, rejection is nota concern because these cells can be harvested from andadministered to the same person. In principle, applications couldinclude reprogramming cells into other lineages or inducing cellsinto the adipose lineage, for example to improve wound healing orto use as a ‘filler’ to repair defects or other cosmetic concerns.Plastic and reconstructive surgeons have long been plagued withthe inconsistencies of adipose grafting in wound healing and softtissue augmentation (Sterodimas et al., 2012). However, newtechnologies enriching graft material for adipose progenitor cellsare being developed and show initial promise (Gir et al., 2012). Anissue in surgical applications is that transplants of adipose depotstend to fail and necrose because of the lack of vascular supply.However, transplanted adipose stem cells recruit vessel formationand the formed adipose depots are well vascularized overcoming

the cause of traditional transplant failure (Han et al., 2011; Satooret al., 2011; Tran et al., 2008).

The ability to induce adipose stem cells into a variety oflineages, such as skeletal myocytes for muscle degeneration,cardiac myocytes for a failing heart, or neural fates for Alzheimer’sdisease, also has significant potential. Confirmation andidentification of the conditions necessary to promote lineagevariation might lead to production of cells or tissues amenable forautologous transplant of modified tissue to treat a host ofdegenerative diseases. Yet the full range of lineagecapacity/plasticity of isolated adipose progenitor cells remainsunknown. For instance, altering canonical WNT signaling (via aconditional allele of β-catenin) in adipose stem cells disruptedadipose development in mice (Fig. 5) (Zeve et al., 2009). Thesemutant mice had a marked paucity of adipocytes, and alipodystrophic phenotype, with ensuing hypertriglyceridemia.Notably, these mutants displayed essentially a complete loss ofsubcutaneous adipose tissue accompanied by a lineage change inthe subcutaneous adipose stem cells. Of note, the secretome ofWNT mutant stem cells was altered and the cells produced highlevels of a glucose-lowering hormone, glucodyne, which had manyactions similar to insulin but functioned through distinctmechanisms (Zeve et al., 2012).

ConclusionsIn recent decades, attention to adipose tissue has increased inparallel with a rising epidemic of obesity and its negative effectson whole body metabolism and increased incidence of variousillnesses and conditions. Only recently have research efforts shiftedto understanding the developmental biology of this tissue (Han etal., 2011; Rodeheffer et al., 2008; Tang et al., 2011; Tang et al.,2008). The complexity of the various adipose lineages (white,brown, induced brown, subcutaneous, visceral, etc.), the difficultyin working with such a fragile cell type, and the non-contiguousnature of this tissue has made it difficult to understand its

REVIEW Development 140 (19)

A

Wild-type adipose tissue

SQ

B

WNT-activated fibrotic tissue

WNT activationWWWNTNTNT acttiiivattttiiion

SQ

Fig. 5. Manipulating the adipose stem cell: kill the stem, kill the tree. (A) Illustration conceptualizing the importance of adipose stem cells to theformation of adipose tissue. This illustration demonstrates the extensive branching of the vascular network and shows that adipose stem cells (greennuclei) and mural cells (black nuclei) reside on the vasculature. Adipose stem cells then transition to a multilocular progenitor giving rise to theunilocular adipocyte. Inset: Histological image of subcutaneous fat from a wild-type mouse. (B) Disrupting adipose stem cells by activating WNTsignaling results in a fate change that promotes the formation of fibrotic tissue (purple). Inset: Histological image of subcutaneous fat from a mousewith activated WNT signaling. D

evel

opm

ent

3947REVIEWDevelopment 140 (19)

developmental origin, and multiple origins might exist. However,with the generation of developmental tools, such as lineage tracing,researchers are now poised to understand the developmental cuesand origin of adipose tissue and to answer a slew of interesting andundefined questions. For instance, what are the cell types of originof the adipose lineage? What is the timing of adipose lineagedetermination and specification? Do adipose stem cells arise in situon the blood vessel or do they migrate and arrive from elsewhere?What are the signals derived from the blood vessel niche thatstimulate these stem cells to proliferate and differentiate or thathold them in the quiescent state? What is the importance of adiposestem cells to the homeostasis and maintenance of the fat pad underboth normal energy intake and excess nutrient load? Do other anti-diabetes drugs alter adipose stem cell biology, similar to TZDtreatment? Do growth factors and development signaling pathwaysalter stem cell behavior and adipocyte formation? In the midst ofthe obesity epidemic, the recent discoveries and the answers tothese open-ended questions would provide hope that there is lightat the end of the tunnel.

AcknowledgementsWe thank all current and former members of the Graff lab for their helpfulcomments and discussion points leading to refinement of our concepts. Weespecially thank Dr Ayoung Jo and Kia Huang for providing data images forspecific figures, and Dr Todd Soesbe for MRI imaging.We apologize for theeditorial constraints that limit the ability to cite a vast number of papers thatpioneered the way for the above studies and notions for future advances.

FundingJ.M.G. is supported by the National Institute of Diabetes and Digestive andKidney Diseases and the National Institutes of Health [R01-DK066556, R01-DK064261 and R01-DK088220]. D.C.B. is supported by the National Heart,Lung and Blood Institute [5T32HL007360-34]. Deposited in PMC for releaseafter 12 months.

Competing interests statementJ.M.G. is a co-founder and shareholder of Reata Pharmaceuticals.

ReferencesAkazawa, S., Sun, F., Ito, M., Kawasaki, E. and Eguchi, K. (2000). Efficacy of

troglitazone on body fat distribution in type 2 diabetes. Diabetes Care 23,1067-1071.

Atit, R., Sgaier, S. K., Mohamed, O. A., Taketo, M. M., Dufort, D., Joyner, A. L.,Niswander, L. and Conlon, R. A. (2006). Beta-catenin activation is necessaryand sufficient to specify the dorsal dermal fate in the mouse. Dev. Biol. 296,164-176.

Baglioni, S., Cantini, G., Poli, G., Francalanci, M., Squecco, R., Di Franco, A.,Borgogni, E., Frontera, S., Nesi, G., Liotta, F. et al. (2012). Functionaldifferences in visceral and subcutaneous fat pads originate from differences inthe adipose stem cell. PLoS ONE 7, e36569.

Barak, Y., Nelson, M. C., Ong, E. S., Jones, Y. Z., Ruiz-Lozano, P., Chien, K. R.,Koder, A. and Evans, R. M. (1999). PPAR gamma is required for placental,cardiac, and adipose tissue development. Mol. Cell 4, 585-595.

Berry, R. and Rodeheffer, M. S. (2013). Characterization of the adipocytecellular lineage in vivo. Nat. Cell Biol. 15, 302-308.

Berry, D. C., DeSantis, D., Soltanian, H., Croniger, C. M. and Noy, N. (2012).Retinoic acid upregulates preadipocyte genes to block adipogenesis andsuppress diet-induced obesity. Diabetes 61, 1112-1121.

Billon, N., Iannarelli, P., Monteiro, M. C., Glavieux-Pardanaud, C.,Richardson, W. D., Kessaris, N., Dani, C. and Dupin, E. (2007). Thegeneration of adipocytes by the neural crest. Development 134, 2283-2292.

Björntorp, P., Bengtsson, C., Blohmé, G., Jonsson, A., Sjöström, L., Tibblin,E., Tibblin, G. and Wilhelmsen, L. (1971). Adipose tissue fat cell size andnumber in relation to metabolism in randomly selected middle-aged menand women. Metabolism 20, 927-935.

Brun, R. P., Tontonoz, P., Forman, B. M., Ellis, R., Chen, J., Evans, R. M. andSpiegelman, B. M. (1996). Differential activation of adipogenesis by multiplePPAR isoforms. Genes Dev. 10, 974-984.

Bueno, A. A., Habitante, C. A., Oyama, L. M., Estadella, D., Ribeiro, E. B. andOller do Nascimento, C. M. (2011). White adipose tissue re-growth afterpartial lipectomy in high fat diet induced obese wistar rats. J. Physiol. Sci. 61,55-63.

Cannon, B. and Nedergaard, J. (2004). Brown adipose tissue: function andphysiological significance. Physiol. Rev. 84, 277-359.

Cao, Y. (2007). Angiogenesis modulates adipogenesis and obesity. J. Clin. Invest.117, 2362-2368.

Cawthorn, W. P., Heyd, F., Hegyi, K. and Sethi, J. K. (2007). Tumour necrosisfactor-alpha inhibits adipogenesis via a beta-catenin/TCF4(TCF7L2)-dependent pathway. Cell Death Differ. 14, 1361-1373.

Cawthorn, W. P., Scheller, E. L. and MacDougald, O. A. (2012). Adipose tissuestem cells meet preadipocyte commitment: going back to the future. J. LipidRes. 53, 227-246.

Chapman, A. B., Knight, D. M. and Ringold, G. M. (1985). Glucocorticoidregulation of adipocyte differentiation: hormonal triggering of thedevelopmental program and induction of a differentiation-dependent gene. J.Cell Biol. 101, 1227-1235.

Chawla, A., Schwarz, E. J., Dimaculangan, D. D. and Lazar, M. A. (1994).Peroxisome proliferator-activated receptor (PPAR) gamma: adipose-predominant expression and induction early in adipocyte differentiation.Endocrinology 135, 798-800.

Clark, E. R. and Clark., E. L. (1940). Microscopic studies of the new formation offat in living adult rabbits. Am. J. Anat. 67, 255-285.

Crandall, D. L., Hausman, G. J. and Kral, J. G. (1997). A review of themicrocirculation of adipose tissue: anatomic, metabolic, and angiogenicperspectives. Microcirculation 4, 211-232.

Cristancho, A. G. and Lazar, M. A. (2011). Forming functional fat: a growingunderstanding of adipocyte differentiation. Nat. Rev. Mol. Cell Biol. 12, 722-734.

Crossno, J. T., Jr, Majka, S. M., Grazia, T., Gill, R. G. and Klemm, D. J. (2006).Rosiglitazone promotes development of a novel adipocyte population frombone marrow-derived circulating progenitor cells. J. Clin. Invest. 116, 3220-3228.

Cypess, A. M., Lehman, S., Williams, G., Tal, I., Rodman, D., Goldfine, A. B.,Kuo, F. C., Palmer, E. L., Tseng, Y. H., Doria, A. et al. (2009). Identification andimportance of brown adipose tissue in adult humans. N. Engl. J. Med. 360,1509-1517.

Cypess, A. M., White, A. P., Vernochet, C., Schulz, T. J., Xue, R., Sass, C. A.,Huang, T. L., Roberts-Toler, C., Weiner, L. S., Sze, C. et al. (2013). Anatomicallocalization, gene expression profiling and functional characterization of adulthuman neck brown fat. Nat. Med. 19, 635-639.

de Souza, C. J., Eckhardt, M., Gagen, K., Dong, M., Chen, W., Laurent, D. andBurkey, B. F. (2001). Effects of pioglitazone on adipose tissue remodelingwithin the setting of obesity and insulin resistance. Diabetes 50, 1863-1871.

Duan, S. Z., Ivashchenko, C. Y., Whitesall, S. E., D’Alecy, L. G., Duquaine, D.C., Brosius, F. C., 3rd, Gonzalez, F. J., Vinson, C., Pierre, M. A., Milstone, D.S. et al. (2007). Hypotension, lipodystrophy, and insulin resistance ingeneralized PPARgamma-deficient mice rescued from embryonic lethality. J.Clin. Invest. 117, 812-822.

Fisher, R. M., Thörne, A., Hamsten, A. and Arner, P. (2002). Fatty acid bindingprotein expression in different human adipose tissue depots in relation torates of lipolysis and insulin concentration in obese individuals. Mol. Cell.Biochem. 239, 95-100.

Flores-Delgado, G., Marsch-Moreno, M. and Kuri-Harcuch, W. (1987). Thyroidhormone stimulates adipocyte differentiation of 3T3 cells. Mol. Cell. Biochem.76, 35-43.

Fukumura, D., Ushiyama, A., Duda, D. G., Xu, L., Tam, J., Krishna, V.,Chatterjee, K., Garkavtsev, I. and Jain, R. K. (2003). Paracrine regulation ofangiogenesis and adipocyte differentiation during in vivo adipogenesis. Circ.Res. 93, e88-e97.

Gealekman, O., Burkart, A., Chouinard, M., Nicoloro, S. M., Straubhaar, J.and Corvera, S. (2008). Enhanced angiogenesis in obesity and in response toPPARgamma activators through adipocyte VEGF and ANGPTL4 production.Am. J. Physiol. 295, E1056-E1064.

Gesta, S., Blüher, M., Yamamoto, Y., Norris, A. W., Berndt, J., Kralisch, S.,Boucher, J., Lewis, C. and Kahn, C. R. (2006). Evidence for a role ofdevelopmental genes in the origin of obesity and body fat distribution. Proc.Natl. Acad. Sci. USA 103, 6676-6681.

Gesta, S., Tseng, Y. H. and Kahn, C. R. (2007). Developmental origin of fat:tracking obesity to its source. Cell 131, 242-256.

Gir, P., Oni, G., Brown, S. A., Mojallal, A. and Rohrich, R. J. (2012). Humanadipose stem cells: current clinical applications. Plast. Reconstr. Surg. 129, 1277-1290.

Grauer, W. O., Moss, A. A., Cann, C. E. and Goldberg, H. I. (1984).Quantification of body fat distribution in the abdomen using computedtomography. Am. J. Clin. Nutr. 39, 631-637.

Green, H. and Kehinde, O. (1975). An established preadipose cell line and itsdifferentiation in culture. II. Factors affecting the adipose conversion. Cell 5, 19-27.

Gregoire, F. M., Smas, C. M. and Sul, H. S. (1998). Understanding adipocytedifferentiation. Physiol. Rev. 78, 783-809.

Gupta, R. K., Mepani, R. J., Kleiner, S., Lo, J. C., Khandekar, M. J., Cohen, P.,Frontini, A., Bhowmick, D. C., Ye, L., Cinti, S. et al. (2012). Zfp423 expressionidentifies committed preadipocytes and localizes to adipose endothelial andperivascular cells. Cell Metab. 15, 230-239. D

evel

opm

ent

3948

Han, J., Lee, J. E., Jin, J., Lim, J. S., Oh, N., Kim, K., Chang, S. I., Shibuya, M.,Kim, H. and Koh, G. Y. (2011). The spatiotemporal development of adiposetissue. Development 138, 5027-5037.

He, W., Barak, Y., Hevener, A., Olson, P., Liao, D., Le, J., Nelson, M., Ong, E.,Olefsky, J. M. and Evans, R. M. (2003). Adipose-specific peroxisomeproliferator-activated receptor gamma knockout causes insulin resistance infat and liver but not in muscle. Proc. Natl. Acad. Sci. USA 100, 15712-15717.

Hellman, B. and Hellerstrom, C. (1961). Cell renewal in the white and brown fattissue of the rat. Acta Pathol. Microbiol. Scand. 51, 347-353.

Herbst, K. L. (2012). Rare adipose disorders (RADs) masquerading as obesity.Acta Pharmacol. Sin. 33, 155-172.

Hiragun, A., Sato, M. and Mitsui, H. (1988). Preadipocyte differentiation in vitro:identification of a highly active adipogenic agent. J. Cell. Physiol. 134, 124-130.

Hirsch, J. and Batchelor, B. (1976). Adipose tissue cellularity in human obesity.Clin. Endocrinol. Metab. 5, 299-311.

Hollenberg, C. H. and Vost, A. (1968). Regulation of DNA synthesis in fat cellsand stromal elements from rat adipose tissue. J. Clin. Invest. 47, 2485-2498.

Hossain, P., Kawar, B. and El Nahas, M. (2007). Obesity and diabetes in thedeveloping world—a growing challenge. N. Engl. J. Med. 356, 213-215.

Jernås, M., Palming, J., Sjöholm, K., Jennische, E., Svensson, P. A.,Gabrielsson, B. G., Levin, M., Sjögren, A., Rudemo, M., Lystig, T. C. et al.(2006). Separation of human adipocytes by size: hypertrophic fat cells displaydistinct gene expression. FASEB J. 20, 1540-1542.

Joe, A. W., Yi, L., Even, Y., Vogl, A. W. and Rossi, F. M. (2009). Depot-specificdifferences in adipogenic progenitor abundance and proliferative response tohigh-fat diet. Stem Cells 27, 2563-2570.

Kajimura, S., Seale, P. and Spiegelman, B. M. (2010). Transcriptional control ofbrown fat development. Cell Metab. 11, 257-262.

Kanda, T., Brown, J. D., Orasanu, G., Vogel, S., Gonzalez, F. J., Sartoretto, J.,Michel, T. and Plutzky, J. (2009). PPARgamma in the endothelium regulatesmetabolic responses to high-fat diet in mice. J. Clin. Invest. 119, 110-124.

Kelly, I. E., Han, T. S., Walsh, K. and Lean, M. E. (1999). Effects of athiazolidinedione compound on body fat and fat distribution of patients withtype 2 diabetes. Diabetes Care 22, 288-293.

Kersten, S., Mandard, S., Tan, N. S., Escher, P., Metzger, D., Chambon, P.,Gonzalez, F. J., Desvergne, B. and Wahli, W. (2000). Characterization of thefasting-induced adipose factor FIAF, a novel peroxisome proliferator-activatedreceptor target gene. J. Biol. Chem. 275, 28488-28493.

Larson, K. A. and Anderson, D. B. (1978). The effects of lipectomy on remainingadipose tissue depots in the Sprague Dawley rat. Growth 42, 469-477.

Le Lièvre, C. S. and Le Douarin, N. M. (1975). Mesenchymal derivatives of theneural crest: analysis of chimaeric quail and chick embryos. J. Embryol. Exp.Morphol. 34, 125-154.

Lee, Y. H., Petkova, A. P., Mottillo, E. P. and Granneman, J. G. (2012). In vivoidentification of bipotential adipocyte progenitors recruited by β3-adrenoceptor activation and high-fat feeding. Cell Metab. 15, 480-491.

Lidell, M. E., Betz, M. J., Dahlqvist Leinhard, O., Heglind, M., Elander, L.,Slawik, M., Mussack, T., Nilsson, D., Romu, T., Nuutila, P. et al. (2013).Evidence for two types of brown adipose tissue in humans. Nat. Med. 19, 631-634.

Longo, K. A., Wright, W. S., Kang, S., Gerin, I., Chiang, S. H., Lucas, P. C., Opp,M. R. and MacDougald, O. A. (2004). Wnt10b inhibits development of whiteand brown adipose tissues. J. Biol. Chem. 279, 35503-35509.

MacDougald, O. A. and Mandrup, S. (2002). Adipogenesis: forces that tip thescales. Trends Endocrinol. Metab. 13, 5-11.

Macotela, Y., Emanuelli, B., Mori, M. A., Gesta, S., Schulz, T. J., Tseng, Y. H.and Kahn, C. R. (2012). Intrinsic differences in adipocyte precursor cells fromdifferent white fat depots. Diabetes 61, 1691-1699.

Masuzaki, H., Paterson, J., Shinyama, H., Morton, N. M., Mullins, J. J., Seckl,J. R. and Flier, J. S. (2001). A transgenic model of visceral obesity and themetabolic syndrome. Science 294, 2166-2170.

Mauer, M. M., Harris, R. B. and Bartness, T. J. (2001). The regulation of totalbody fat: lessons learned from lipectomy studies. Neurosci. Biobehav. Rev. 25,15-28.

McKay, R. M., McKay, J. P., Avery, L. and Graff, J. M. (2003). C elegans: a modelfor exploring the genetics of fat storage. Dev. Cell 4, 131-142.

Medici, D., Shore, E. M., Lounev, V. Y., Kaplan, F. S., Kalluri, R. and Olsen, B.R. (2010). Conversion of vascular endothelial cells into multipotent stem-likecells. Nat. Med. 16, 1400-1406.

Miles, P. D., Barak, Y., He, W., Evans, R. M. and Olefsky, J. M. (2000). Improvedinsulin-sensitivity in mice heterozygous for PPAR-gamma deficiency. J. Clin.Invest. 105, 287-292.

Monteiro, M. C., Wdziekonski, B., Villageois, P., Vernochet, C., Iehle, C.,Billon, N. and Dani, C. (2009). Commitment of mouse embryonic stem cellsto the adipocyte lineage requires retinoic acid receptor beta and active GSK3.Stem Cells Dev. 18, 457-463.

Muto, S., Miyata, Y., Imai, M. and Asano, Y. (2001). Troglitazone stimulatesbasolateral rheogenic Na+/HCO3- cotransport activity in rabbit proximalstraight tubules. Exp. Nephrol. 9, 191-197.

Napolitano, L. (1963). The Differentiation of White Adipose Cells. An ElectronMicroscope Study. J. Cell Biol. 18, 663-679.

Napolitano, L. and Gagne, H. T. (1963). Lipid-depleted white adipose cells: anelectronmicroscope study. Anat. Rec. 147, 273-293.

Nawrocki, A. R. and Scherer, P. E. (2004). The delicate balance between fat andmuscle: adipokines in metabolic disease and musculoskeletal inflammation.Curr. Opin. Pharmacol. 4, 281-289.

Nishimura, S., Manabe, I., Nagasaki, M., Hosoya, Y., Yamashita, H., Fujita, H.,Ohsugi, M., Tobe, K., Kadowaki, T., Nagai, R. et al. (2007). Adipogenesis inobesity requires close interplay between differentiating adipocytes, stromalcells, and blood vessels. Diabetes 56, 1517-1526.

Okuno, A., Tamemoto, H., Tobe, K., Ueki, K., Mori, Y., Iwamoto, K., Umesono,K., Akanuma, Y., Fujiwara, T., Horikoshi, H. et al. (1998). Troglitazoneincreases the number of small adipocytes without the change of whiteadipose tissue mass in obese Zucker rats. J. Clin. Invest. 101, 1354-1361.

Osborn, O. and Olefsky, J. M. (2012). The cellular and signaling networkslinking the immune system and metabolism in disease. Nat. Med. 18, 363-374.

Reyne, Y., Nougues, J. and Vezinhet, A. (1983). Adipose tissue regeneration in6-month-old and adult rabbits following lipectomy. Proc. Soc. Exp. Biol. Med.174, 258-264.

Rigamonti, A., Brennand, K., Lau, F. and Cowan, C. A. (2011). Rapid cellularturnover in adipose tissue. PLoS ONE 6, e17637.

Rodeheffer, M. S., Birsoy, K. and Friedman, J. M. (2008). Identification of whiteadipocyte progenitor cells in vivo. Cell 135, 240-249.

Rosen, E. D. and Spiegelman, B. M. (2000). Molecular regulation ofadipogenesis. Annu. Rev. Cell Dev. Biol. 16, 145-171.

Ross, S. E., Hemati, N., Longo, K. A., Bennett, C. N., Lucas, P. C., Erickson, R.L. and MacDougald, O. A. (2000). Inhibition of adipogenesis by Wntsignaling. Science 289, 950-953.

Salans, L. B., Cushman, S. W. and Weismann, R. E. (1973). Studies of humanadipose tissue. Adipose cell size and number in nonobese and obese patients.J. Clin. Invest. 52, 929-941.

Sanchez-Gurmaches, J., Hung, C. M., Sparks, C. A., Tang, Y., Li, H. andGuertin, D. A. (2012). PTEN loss in the Myf5 lineage redistributes body fat andreveals subsets of white adipocytes that arise from Myf5 precursors. Cell Metab.16, 348-362.

Sandouk, T., Reda, D. and Hofmann, C. (1993a). Antidiabetic agentpioglitazone enhances adipocyte differentiation of 3T3-F442A cells. Am. J.Physiol. 264, C1600-C1608.

Sandouk, T., Reda, D. and Hofmann, C. (1993b). The antidiabetic agentpioglitazone increases expression of glucose transporters in 3T3-F442A cellsby increasing messenger ribonucleic acid transcript stability. Endocrinology133, 352-359.

Satoor, S. N., Puranik, A. S., Kumar, S., Williams, M. D., Ghale, M., Rahalkar,A., Karandikar, M. S., Shouche, Y., Patole, M., Bhonde, R. et al. (2011).Location, location, location: Beneficial effects of autologous fat transplantation.Sci. Rep. 1, 81.

Seale, P., Bjork, B., Yang, W., Kajimura, S., Chin, S., Kuang, S., Scimè, A.,Devarakonda, S., Conroe, H. M., Erdjument-Bromage, H. et al. (2008).PRDM16 controls a brown fat/skeletal muscle switch. Nature 454, 961-967.

Seale, P., Kajimura, S. and Spiegelman, B. M. (2009). Transcriptional control ofbrown adipocyte development and physiological function—of mice and men.Genes Dev. 23, 788-797.

Sera, Y., LaRue, A. C., Moussa, O., Mehrotra, M., Duncan, J. D., Williams, C.R., Nishimoto, E., Schulte, B. A., Watson, P. M., Watson, D. K. et al. (2009).Hematopoietic stem cell origin of adipocytes. Exp Hematol 37, 1108-1120, e1-e4.

Shi, H. and Clegg, D. J. (2009). Sex differences in the regulation of body weight.Physiol. Behav. 97, 199-204.

Shimizu, H., Tsuchiya, T., Sato, N., Shimomura, Y., Kobayashi, I. and Mori, M.(1998). Troglitazone reduces plasma leptin concentration but increaseshunger in NIDDM patients. Diabetes Care 21, 1470-1474.

Siersbæk, R., Nielsen, R. and Mandrup, S. (2012). Transcriptional networks andchromatin remodeling controlling adipogenesis. Trends Endocrinol. Metab. 23,56-64.

Snijder, M. B., Dekker, J. M., Visser, M., Bouter, L. M., Stehouwer, C. D.,Kostense, P. J., Yudkin, J. S., Heine, R. J., Nijpels, G. and Seidell, J. C.(2003a). Associations of hip and thigh circumferences independent of waistcircumference with the incidence of type 2 diabetes: the Hoorn Study. Am. J.Clin. Nutr. 77, 1192-1197.

Snijder, M. B., Dekker, J. M., Visser, M., Yudkin, J. S., Stehouwer, C. D.,Bouter, L. M., Heine, R. J., Nijpels, G. and Seidell, J. C. (2003b). Larger thighand hip circumferences are associated with better glucose tolerance: theHoorn study. Obes. Res. 11, 104-111.

Spalding, K. L., Arner, E., Westermark, P. O., Bernard, S., Buchholz, B. A.,Bergmann, O., Blomqvist, L., Hoffstedt, J., Näslund, E., Britton, T. et al.(2008). Dynamics of fat cell turnover in humans. Nature 453, 783-787.

Spiegelman, B. M. and Flier, J. S. (2001). Obesity and the regulation of energybalance. Cell 104, 531-543.

REVIEW Development 140 (19)

Dev

elop

men

t

3949REVIEWDevelopment 140 (19)

Sterodimas, A., Boriani, F., Magarakis, E., Nicaretta, B., Pereira, L. H. andIllouz, Y. G. (2012). Thirtyfour years of liposuction: past, present and future. Eur.Rev. Med. Pharmacol. Sci. 16, 393-406.

Strissel, K. J., Stancheva, Z., Miyoshi, H., Perfield, J. W., 2nd, DeFuria, J., Jick,Z., Greenberg, A. S. and Obin, M. S. (2007). Adipocyte death, adipose tissueremodeling, and obesity complications. Diabetes 56, 2910-2918.

Suh, J. M., Gao, X., McKay, J., McKay, R., Salo, Z. and Graff, J. M. (2006).Hedgehog signaling plays a conserved role in inhibiting fat formation. CellMetab. 3, 25-34.

Suh, J. M., Zeve, D., McKay, R., Seo, J., Salo, Z., Li, R., Wang, M. and Graff, J.M. (2007). Adipose is a conserved dosage-sensitive antiobesity gene. CellMetab. 6, 195-207.

Suh, J. M., Stenesen, D., Peters, J. M., Inoue, A., Cade, A. and Graff, J. M.(2008). An RGS-containing sorting nexin controls Drosophila lifespan. PLoSONE 3, e2152.

Sul, H. S. (2009). Minireview: Pref-1: role in adipogenesis and mesenchymal cellfate. Mol. Endocrinol. 23, 1717-1725.

Sun, K., Wernstedt Asterholm, I., Kusminski, C. M., Bueno, A. C., Wang, Z. V.,Pollard, J. W., Brekken, R. A. and Scherer, P. E. (2012). Dichotomous effectsof VEGF-A on adipose tissue dysfunction. Proc. Natl. Acad. Sci. USA 109, 5874-5879.

Tang, W., Zeve, D., Suh, J. M., Bosnakovski, D., Kyba, M., Hammer, R. E.,Tallquist, M. D. and Graff, J. M. (2008). White fat progenitor cells reside in theadipose vasculature. Science 322, 583-586.

Tang, W., Zeve, D., Seo, J., Jo, A. Y. and Graff, J. M. (2011). Thiazolidinedionesregulate adipose lineage dynamics. Cell Metab. 14, 116-122.

Tchernof, A., Bélanger, C., Morisset, A. S., Richard, C., Mailloux, J., Laberge,P. and Dupont, P. (2006). Regional differences in adipose tissue metabolism inwomen: minor effect of obesity and body fat distribution. Diabetes 55, 1353-1360.

Tchkonia, T., Giorgadze, N., Pirtskhalava, T., Tchoukalova, Y., Karagiannides,I., Forse, R. A., DePonte, M., Stevenson, M., Guo, W., Han, J. et al. (2002).Fat depot origin affects adipogenesis in primary cultured and cloned humanpreadipocytes. Am. J. Physiol. 282, R1286-R1296.

Tchkonia, T., Lenburg, M., Thomou, T., Giorgadze, N., Frampton, G.,Pirtskhalava, T., Cartwright, A., Cartwright, M., Flanagan, J.,Karagiannides, I. et al. (2007). Identification of depot-specific human fat cellprogenitors through distinct expression profiles and developmental genepatterns. Am. J. Physiol. 292, E298-E307.

Tontonoz, P. and Spiegelman, B. M. (2008). Fat and beyond: the diversebiology of PPARgamma. Annu. Rev. Biochem. 77, 289-312.

Tontonoz, P., Hu, E., Graves, R. A., Budavari, A. I. and Spiegelman, B. M.(1994). mPPAR gamma 2: tissue-specific regulator of an adipocyte enhancer.Genes Dev. 8, 1224-1234.

Traktuev, D. O., Prater, D. N., Merfeld-Clauss, S., Sanjeevaiah, A. R.,Saadatzadeh, M. R., Murphy, M., Johnstone, B. H., Ingram, D. A. andMarch, K. L. (2009). Robust functional vascular network formation in vivo bycooperation of adipose progenitor and endothelial cells. Circ. Res. 104, 1410-1420.

Tran, T. T., Yamamoto, Y., Gesta, S. and Kahn, C. R. (2008). Beneficial effects ofsubcutaneous fat transplantation on metabolism. Cell Metab. 7, 410-420.

Tran, K. V., Gealekman, O., Frontini, A., Zingaretti, M. C., Morroni, M.,Giordano, A., Smorlesi, A., Perugini, J., De Matteis, R., Sbarbati, A. et al.(2012). The vascular endothelium of the adipose tissue gives rise to both whiteand brown fat cells. Cell Metab. 15, 222-229.

Tseng, Y. H., Kokkotou, E., Schulz, T. J., Huang, T. L., Winnay, J. N., Taniguchi,C. M., Tran, T. T., Suzuki, R., Espinoza, D. O., Yamamoto, Y. et al. (2008).New role of bone morphogenetic protein 7 in brown adipogenesis andenergy expenditure. Nature 454, 1000-1004.

Wang, Y., Kim, K. A., Kim, J. H. and Sul, H. S. (2006). Pref-1, a preadipocytesecreted factor that inhibits adipogenesis. J. Nutr. 136, 2953-2956.

Weyer, C., Foley, J. E., Bogardus, C., Tataranni, P. A. and Pratley, R. E. (2000).Enlarged subcutaneous abdominal adipocyte size, but not obesity itself, predictstype II diabetes independent of insulin resistance. Diabetologia 43, 1498-1506.

Whittle, A. J., López, M. and Vidal-Puig, A. (2011). Using brown adipose tissueto treat obesity - the central issue. Trends Mol. Med. 17, 405-411.

Wu, J., Boström, P., Sparks, L. M., Ye, L., Choi, J. H., Giang, A. H., Khandekar,M., Virtanen, K. A., Nuutila, P., Schaart, G. et al. (2012). Beige adipocytes area distinct type of thermogenic fat cell in mouse and human. Cell 150, 366-376.

Yamamoto, N., Akamatsu, H., Hasegawa, S., Yamada, T., Nakata, S.,Ohkuma, M., Miyachi, E., Marunouchi, T. and Matsunaga, K. (2007).Isolation of multipotent stem cells from mouse adipose tissue. J. Dermatol. Sci.48, 43-52.

Yki-Järvinen, H. (2004). Thiazolidinediones. N. Engl. J. Med. 351, 1106-1118. Zamani, N. and Brown, C. W. (2011). Emerging roles for the transforming

growth factor-beta superfamily in regulating adiposity and energyexpenditure. Endocr. Rev. 32, 387-403.

Zeve, D., Tang, W. and Graff, J. (2009). Fighting fat with fat: the expanding fieldof adipose stem cells. Cell Stem Cell 5, 472-481.

Zeve, D., Seo, J., Suh, J. M., Stenesen, D., Tang, W., Berglund, E. D., Wan, Y.,Williams, L. J., Lim, A., Martinez, M. J. et al. (2012). Wnt signaling activationin adipose progenitors promotes insulin-independent muscle glucose uptake.Cell Metab. 15, 492-504.

Dev

elop

men

t