Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity Yihai Cao1,2,* 1Department

of Microbiology, Tumor and Cell Biology, Karolinska Institute, 171 77 Stockholm, Sweden of Medicine and Health Sciences, Linko¨ping University, 581 85 Linko¨ping, Sweden *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cmet.2013.08.008 2Department

White and brown adipose tissues are hypervascularized and the adipose vasculature displays phenotypic and functional plasticity to coordinate with metabolic demands of adipocytes. Blood vessels not only supply nutrients and oxygen to nourish adipocytes, they also serve as a cellular reservoir to provide adipose precursor and stem cells that control adipose tissue mass and function. Multiple signaling molecules modulate the complex interplay between the vascular system and the adipocytes. Understanding fundamental mechanisms by which angiogenesis and vasculatures modulate adipocyte functions may provide new therapeutic options for treatment of obesity and metabolic disorders by targeting the adipose vasculature. Introduction Forty-two years ago, Dr. Judah Folkman proposed the concept of antiangiogenic therapy based on his observations in experimental tumor models (Folkman, 1971). Today, Folkman’s concept has successfully been translated into the development of antiangiogenic drugs for the treatment of cancer and ophthalmological diseases in human patients (Folkman, 2007). Would the same antiangiogenic concept work for treatment of obesity and its related disorders? The definite answer to this question relies on understanding the fundamental mechanisms that underlie vascular modulation of adipose tissue functions. Similar to tumor tissues, adipose tissues are highly vascularized; brown adipose tissue (BAT) is one of the most vascularized tissues in the body (Bra˚kenhielm and Cao, 2008; Cao, 2007, 2010). Throughout adulthood, white adipose tissue (WAT) constantly undergoes expansion or shrinkage depending on energy consumption and the metabolic demand of the host. Alterations of adipose tissue mass and functions coordinately necessitate angiogenesis or vascular regression, which are controlled by various cell-type-derived growth factors, cytokines, and adipokines (Cao, 2007). Hypervascularization in BAT and WAT also implies the existence of an intimate interplay between vascular and adipose compartments. In support of this view, abundant recent data show that cells located in the vessel wall, a significant component in the adipose microenvironment, provide precursor cells that ultimately differentiate into preadipocytes and adipocytes (Gupta et al., 2012; Tang et al., 2008). Understanding vascular functions and the intimate interplay between vascular cells and adipocytes may provide an outstanding opportunity for therapeutic intervention for obesity and its related metabolic diseases by targeting the vascular compartment. However, targeting vasculaturebased therapeutic approaches for the treatment of obesity currently remain hypothetical and have not been tested in human subjects.

Cell-Type Switch between Vessel Walls and Adipocytes A recent study demonstrates that white-fat progenitors reside in the mural cell compartment of the adipose vessel wall (Tang et al., 2008) (Figure 1). Furthermore, these specialized pericytes and smooth-muscle-like cells express PPAR-g stem cell antigen 1(Scal1), CD34, a-smooth muscle actin (a-SMA), plateletderived growth factor receptor b (PDGFR-b), and NG2, which distinguishes this adipose-tissue-derived pericyte population from other cells. Similarly, genetic labeling of Zfp423, a multizinc-finger transcriptional regulator, demonstrates a perivascular origin of preadipocytes, and this transcription factor is both necessary and sufficient for the development of a common precursor of white and brown adipocytes (Gupta et al., 2012). Vascular endothelial cells also retain multipotent stem cell-like features and can, under the influence of transforming growth factor b2 (TGF-b2), differentiate into adipocytes, as well as other cell types (Medici et al., 2010). This view is supported by the fact that some mural cells and a very small subset of CD31+ capillary endothelial cells express Zfp423, a gene controlling preadipocyte determination, raising the possibility of the vascular origin of preadipocytes (Gupta et al., 2012) (Figure 1). The VE-cadherin promoter-driven lineage-tracing experiments provide an independent line of evidence that both pericytes and murine endothelial cells can differentiate into preadipocytes and adipocytes (Tran et al., 2012). Taken together, these findings support the concept of endothelial and perivascular origins of preadipocytes, suggesting that adipogenesis, angiogenesis, and vascular remodeling are tightly and coordinately regulated. Similar to vascular cells, mature white adipocytes also exhibit multilineage potential to differentiate into multiple mesenchymal lineages including bone, cartilage, cardiomyocytes, and skeletal muscles (Jumabay et al., 2010) (Figure 1). In response to bone morphogenetic proteins 4 and 9 (BMP4 and BMP9), these mature adipocytes with dedifferentiation potentials, also referred to as dedifferentiated fat cells (DFATCs), can spontaneously undergo mesenchymal-endothelial transition and differentiate into endothelial cells that actively participate in neovascularization Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc. 1

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Figure 1. Adipocyte and Endothelial Cell Differentiation and Transdifferentiation Reciprocal cellular switch between vascular cells and adipocytes (ACs). Under certain circumstances, cells located in the vessel wall including endothelial cells (ECs), pericytes (PCs), and vascular smooth muscle cells (VSMCs) can undergo differentiation to become BAT and WAT adipocytes. Inversely, mature white adipocytes can dedifferentiate into DFATCs that give rise to ECs and VSMCs, which actively participate in adipose angiogenesis. Mature adipocytes also exhibit multilineage potential to differentiate into multiple mesenchymal lineages including bone, cartilage, cardiomyocytes, and skeletal muscles.

(Jumabay et al., 2012). Additionally, DFATCs can also differentiate into smooth muscle cells (Sakuma et al., 2009). The interplay and cell-type switch between adipocytes and vascular cells are mediated by multiple signaling systems that determine cell proliferation and differentiation. For example, genetically constitutive activation of the PDGFR-b signaling system promotes pericyte proliferation but inhibits its differentiation toward WAT adipocytes (Olson and Soriano, 2011). PPAR-g has also been reported to coordinate preadipocyte differentiation and angiogenesis in PPAR-g-activator-treated mouse models (Gealekman et al., 2008). Vascular endothelial growth factor B (VEGF-B), originally described as an angiogenic factor, has recently been described as a factor that mediates fatty acid transport and insulin sensitivity (Hagberg et al., 2012). These studies show that multiple signaling systems are involved in the modulation of adipocyte-endothelial cell interaction and their functions. Angiogenic Driving Force and Reciprocal Interactions between Various Cell Types Both intrinsic adipocyte and constant microenvironmental alterations in various adipose tissues determine dynamic changes of the angiogenic driving force. For example, during the early stage of embryonic development, the formation of the initial adipose niche by stem cell-derived preadipocytes is spatiotemporally coupled to the switch of an angiogenic phenotype (Cao, 2007; Crandall et al., 1997). An early study showed that differentiation from preadipocytes to mature adipocytes was concomitantly linked to elevated production of angiogenic factors that induces robust angiogenic responses (Castellot et al., 1982). Adipose-derived stem cells produce high levels of a range of proangiogenic factors, including fibroblast growth factor 2 (FGF-2), VEGF, hepatocyte growth factor (HGF), and PDGFs (Lee et al., 2012), and an adipocyte differentiationinducer, PPARg activator, can further boost expression levels of VEGF, VEGF-B, and angiopoietin-like factor-4 (Gealekman et al., 2008) (Figure 2). Ingrowth of blood vessels in adipose tissues permits and facilitates reciprocal interaction between 2 Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc.

adipocytes and endothelial cells. Preadipocytes induce and guide endothelial cell migration (Borges et al., 2007) via FGF- and VEGF-dependent pathways, whereas the VEGF-triggered signaling pathway in endothelial cells promotes preadipocyte differentiation via a paracrine mechanism (Fukumura et al., 2003). The oxygen tension in the adipose is determined by several factors, including vascular density and function, metabolic status, and the composition of other cell types. However, it is unclear whether this is a cause or a consequence of obesityassociated metabolic disorders. Enlargement of adipocyte sizes would increase the intercapillary distance, resulting in relatively decreased blood perfusion to each adipocyte and a global decrease of adipose oxygen tension. Thus, the vascular density per adipocyte in a rigid field is relatively reduced. However, enlargement of adipocyte sizes may also result in decreased metabolic demands, and adipocytes may not necessarily suffer from tissue hypoxia. It is also highly plausible that the vessel growth rate and rapid expansion of adipocyte sizes in obese individuals occur in an uncoordinated manner, leading to insufficient angiogenesis and tissue hypoxia. Adipose tissue hypoxia has consistently been reported in various animal models of obesity, including genetically obese ob/ob and db/db mice, obese Zucker rats, and high-fat diet (HFD)-fed mice (Goossens and Blaak, 2012; Ye et al., 2007). Consequently, expression levels of certain hypoxia-inducible genes, such as hypoxiainducible factor-1 alpha (HIF-1a) and glucose transporter, are elevated in obese WAT relative to those in lean mice (Goossens and Blaak, 2012). In response to hypoxia, white and brown adipocytes may employ different mechanisms to induce targeted gene expression. For example, HIF-1a in white adipocytes is required for the induction of hypoxia-targeted gene expression, but it is not essential for hypoxia-induced targeted gene expression in brown adipocytes (Pino et al., 2012). VEGF is a direct transcriptional target of HIF-1a. Thus, the hypoxia-HIF-1a-VEGF axis may serve as an angiogenic driving force of rapidly expanding WAT tissues. Indeed, enforced expression of VEGF in adipose tissues produced a hypervascularized phenotype (Elias et al., 2012; Sun et al., 2012). Additionally, adipose tissue hypoxia may potentially upregulate expression levels of endothelial VEGFR2, as seen in other tissues (Brogi et al., 1996), which transduces the VEGF-triggered angiogenic signaling. Notably,

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review

Figure 2. Reciprocal Interplay between Vascular and Adipocyte Compartment Cells in the vessel wall, i.e., endothelial cells (ECs), pericytes (PCs), and vascular smooth muscle cells (VSMCs), may produce various soluble and insoluble factors that crosstalk to adipocytes (ACs) via a paracrine regulatory mechanism. Conversely, adipocytes produce a range of angiogenic factors, cytokines, and adipokines that collectively modulate vascular growth, regression, vascular survival, vascular remodeling, and blood perfusion. In the adipose microenvironment, other cell types including inflammatory cells (IFCs) and mesenchymal stromal cells (MSCs) also actively participate in the complex regulation of vascular functions. In expanding obese adipose tissues, hypoxia may potentially facilitate the interaction between endothelial cells and adipocytes by elevating the production of hypoxia-regulated angiogenic factors such as VEGF.

not all data support the hypoxia-HIF-1a-angiogenesis pathway in the adipose tissue. For example, in a genetic mouse model, constitutive overexpression of an active form HIF-1a in adipose tissue fails to induce a proangiogenic response but unexpectedly triggers adipose tissue fibrosis and local inflammation (Halberg et al., 2009). Consistent with this notion, levels of VEGF, one of the classical targets of HIF-1a, were not elevated. Mechanistically, it is unclear why VEGF levels were not elevated in constitutive HIF-1a expression. One possibility would be that the deletion mutant HIF-1a was unable to induce and to sustain VEGF expression in adipocytes. Unlike other cell types, adipocytes may employ other HIF-1a-independent mechanisms to induce VEGF expression and angiogenesis. These speculations and discrepant findings warrant further mechanistic investigations. Obese WAT exhibits chronic inflammation by infiltrating macrophages, neutrophils, and immune cells, which, together with adipocytes and stromal vascular cells, produce various inflammatory cytokines (Donath and Shoelson, 2011). Noticeably, neovascularization in adipose tissues could further intensify the inflammatory process by increasing the infiltration of inflammatory cells. Adipose tissue hypoxia in obesity is not only the primary driving force of angiogenesis, but also an essential trigger for initiating chronic inflammation (Trayhurn, 2013). Specific expression of VEGF in adipose tissues improves tissue hypoxia and suppresses fibrosis and local inflammation (Sun et al., 2012). It is unclear why VEGF-enhanced adipose angiogenesis resulted in anti-inflammatory rather than proinflammatory responses. Analysis of the expression of inflammatory factors demonstrated that levels of interleukin 6 (IL-6) and tumor necrosis factor alpha (TNF-a) are significantly downregulated. Insufficient angiogenesis in response to adipose tissue hypoxia has also been suggested to activate the chronic inflammatory process (Ye,

2009). In support of this notion, an independent study shows that enforced expression of VEGF in the adipose tissue leads to the accumulation of M2 anti-inflammatory and fewer M1 proinflammatory macrophages (Elias et al., 2012). Inversely, deletion of VEGF in adipose tissues in mice also resulted in reduced inflammation with increased adipose hypoxia (Sung et al., 2013). Similar to inflammatory cells, adipose mesenchymal cells produce an array of angiogenic factors and cytokines, including VEGF, PDGF, FGF, HGF, and TGF-b (Cao, 2007) (Figure 2). In addition to angiogenic factors and proangiogenic cytokines, adipose-tissue-derived hormones, adipokines, play a crucial role in the modulation of adipose angiogenesis. It is recognized that adipose tissues produce more than 600 bioactive factors that are collectively termed adipokines. These adipokines positively and negatively modulate adipose angiogenesis, vascular homeostasis, and vascular integrity. For example, whereas leptin induces angiogenesis, vascular fenestration, and vascular remodeling (Bouloumie´ et al., 1998; Cao et al., 2001; Sierra-Honigmann et al., 1998), adiponectin may inhibit angiogenesis (Bra˚kenhielm et al., 2004b; Man et al., 2010). Adipose-derived resistin, chemerin, omentin, vaspin, and visfatin are all able to modulate angiogenesis (Bozaoglu et al., 2010; Kim et al., 2007b; Kukla et al., 2011; Mu et al., 2006; Northcott et al., 2012). It appears that some of these proadipokines can synergistically stimulate angiogenesis with known angiogenic factors; examples include leptin plus VEGF and leptin plus FGF-2 (Cao et al., 2001). Several known endogenous angiogenesis inhibitors, such as adiponectin, thrombospondin, and soluble VEGF receptors (sVEGFRs), are downregulated in angiogenic adipose tissues (Xue et al., 2009). Intriguingly, hypoxia potently downregulates Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc. 3

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Figure 3. Angiogenic Functions in Adipose Tissues Immunohistochemical CD31 staining of mouse inguinal WAT and interscapular BAT demonstrates that both adipose depots contain high microvascular density. In particular, BAT is the probably the most vascularized tissue in the body. Microvascular networks in each of these adipose tissues may display distinct functions. In WAT, angiogenesis promotes energy storage by enhancing lipid transport and deposition and may modulate adipocyte-related endocrine, paracrine, and autocrine functions. In contrast, the same angiogenic process augments lipolysis in metabolically active BAT. The angiogenic process is tightly regulated by various angiogenic and vascular remodeling factors, which coordinately control vascular growth. For example, adipocyte- and nonadipocyte-derived VEGF could build up an ‘‘angiogenic gradient’’ that guides vascular sprouting by formation of endothelial cell tips. This process is coordinately regulated by other signaling systems including the PDGF-BBPDGFR-b signaling in pericytes and the Dll4-Notch signaling to prevent excessive vascular sprouting and undirected growth.

expression levels of thrombospondin, pigment epitheliumderived growth factor (PEDF), and sVEGFR1 (Bienes-Martı´nez et al., 2012; Famulla et al., 2011; Shitaye et al., 2010), possibly leading to further acceleration of hypoxia-induced angiogenesis. Consistent with antiangiogenic function, genetic deletion of thrombospndin-2 in adipose tissue led to accelerated adipogenesis and hypersusceptibility of obesity development (Shitaye et al., 2010). These findings suggest that upregulation of proangiogenic factors and downregulation of endogenous angiogenesis inhibitors may be both necessary and sufficient for switching on an adipose angiogenic phenotype. 4 Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc.

Vascular Functions in White and Brown Adipose Tissues Adipose tissues, especially BAT, are probably the most vascularized tissues in the body, and the adipose vessel density is usually higher than those found in pathological tissues (Lim et al., 2012; Xue et al., 2010) (Figure 3). The adipose vasculature has multiple functions in the modulation of adipocyte functions (Cao, 2010). These include (1) providing nutrients and oxygen essential for the maintenance of adipocyte survival and functions; (2) removing metabolic products from adipose tissue; (3) transporting nonadipose-tissue-derived growth factors, cytokines, and hormones for modulating adipocyte functions and growth; (4) transporting adipose-tissue-derived growth factors, adipokines, and cytokines for removal of tissues globally regulating physiological functions via the endocrine mechanism; (5) paracrine and juxtacrine regulation of adipocyte functions through the production of various factors and cytokines from vascular cells; (6) supplying adipocyte vessel wall stem and precursor cells that can eventually differentiate into mature adipocytes; (7) supplying circulating stem cells from nonadipose tissues to adipose tissues; (8) preparation of adipose niche formation during embryonic development by the vasculature; (9) supplying other cell types such as inflammatory cells that secondarily affect adipocyte function; and (10) alteration of the adipose microenvironment such as hypoxia and acidosis, which control adipocyte function, preadipocyte differentiation, and adipose tissue mass.

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Although vasculatures in WAT and BAT share the abovementioned common functions, the ultimately functional consequence is dependent on the metabolic status of adipocytes. In WAT, switching to an angiogenic phenotype would facilitate to the process of energy deposition, resulting in adipose tissue expansion. In expanding fat pads at the early stages of a HFD challenge, specific expression of VEGF in WAT enhances angiogenesis, leading to a ‘‘browning phenotype’’ of WAT (Elias et al., 2012; Michailidou et al., 2012; Sun et al., 2012), which is beneficial for the improvement of metabolism. Additionally, vasculatures in healthy WAT play an essential role in controlling adipokine and other hormone release and transport. Similar to microvessels in various endocrine organs, the adipose vasculature contains vasculature fenestrations, which are essential for maintaining the physiological functions of endocrine tissues (Cao et al., 2001). In obese individuals, active adipose angiogenesis in WAT is likely to further promote obesity. Angiogenesis in WAT may also control energy metabolism by modulating adipocyte-associated endocrine, paracrine, and autocrine functions, given that WAT is considered one of the largest endocrine tissues in the body (Harwood, 2012). Alteration of vascular density and structure would result in changes in the composition and architecture of adipocytes in relation to other cellular components in WAT. Ultimately, angiogenesis would have a significantly functional impact of energy metabolism through modulation of adipocyte-related endocrine, paracrine, and autocrine mechanisms. In contrast to in WAT, activation of angiogenesis in BAT would increase the oxygen supply and thereby fuel the process of energy consumption, leading to a lean phenotype. Vascular Changes during Thermogenesis BAT located in the interscapular region plays an essential role for thermogenesis, which is crucial for the survival of small mammals in cold environments (Cannon and Nedergaard, 2004). In humans, the amount of BAT markedly deceases after the infant stages. A large cohort analysis in adult humans using (18)F-fluorodeoxyglucose positron-emission tomographic and computer tomographic scanning shows that substantial depots of BAT exist in the supraclavicular region, the anterior area in the neck and thorax (Cypess et al., 2009). More importantly, the amount BAT has been inversely correlated with body-mass index (BMI), implying the potential role of BAT in the modulation of adult human metabolism (Cypess et al., 2009). However, the identity of BAT in adult humans has recently become a disputed issue. Recent findings show that the previously described BAT depots in adult humans are actually brown-like, also called brown-in-white (brite)/beige, adipose tissues, which is different from the classical BAT (Wu et al., 2012). These findings raise an important question of the existence of BAT in humans. However, another recent study employing a combination of high-resolution imaging techniques, histology, and biochemical analyses has discovered that the classical thermogenic organ of bona fide BAT does exist in human infants (Lidell et al., 2013). Thus, humans have two distinct types of BAT, i.e., bona fide BAT and inducible brown-like brite/beige adipose tissues. Cold acclimation of mice can lead to the transition of WAT into a BAT-like adipocyte phenotype, also termed brite or beige adipocytes (Petrovic et al., 2010) (Figure 4). Brite/beige adipocytes enrich uncoupling protein 1 (UCP1)-positive mitochondrial con-

tents that are essentially required for catalyzing nonshivering thermogenesis (Heaton et al., 1978). Intriguingly, PGC-1a potently upregulates VEGF expression and angiogenesis in vitro and in vivo through a noncanonical pathway, i.e., the HIF-independent hypoxia-response pathway (Arany et al., 2008). Consistent with these findings, VEGF expression levels in the whole inguinal adipose tissue are elevated through a hypoxia-independent but possibly PGC-1a-dependent mechanism (Xue et al., 2009). The VEGF-VEGFR2 signaling system is primarily responsible for the initiation of adipose angiogenesis in the brite/beige adipose tissue given that VEGFR2 blockade, but not VEGFR1 blockade, could abolish the cold-induced angiogenesis (Xue et al., 2009). Endogenous angiogenesis inhibitors such as thrombospondin are simultaneously downregulated, tipping the balance toward angiogenic stimuli (Figure 4). It is noteworthy that the cold-induced angiogenic response occurs only transiently and reaches a higher steady-state level after persistent exposure to cold (Lim et al., 2012; Xue et al., 2009). Angiogenesis is switched off by decreasing VEGF levels and by increasing production of an sVEGFR1 that negatively regulates VEGF function (Xue et al., 2009). Another regulatory mechanism may involve sympathetic release of neuropeptide Y (NPY) during cold-induced chronic stress, which stimulates adipose angiogenesis via activation of NPY-Y2 receptors (Ekstrand et al., 2003; Kuo et al., 2007). Exposure of adult humans to low temperatures also leads to activation of BAT and thermogenesis-associated metabolism (van Marken Lichtenbelt et al., 2009). It is reasonably speculated, although not currently proved, that cold activation of BAT concomitantly induces angiogenesis in human BAT tissues. If so, vascular functions under this circumstance warrant future investigation. A brite/beige phenotype is also observed in several mouse genetic models. For example, specific overexpression of forkhead box protein C2 (FOXC2) in the adipose tissue produces a brown-like phenotype and switches on an adipose angiogenic phenotype (Cederberg et al., 2001; Xue et al., 2008). Mechanistically, FOXC2 directly targets the angiopoietin-2 (Ang-2) promoter to trigger an angiogenic response, and the elevated level of Ang-2 results in marked vascular remodeling by ablating perivascular cell coverage (Xue et al., 2008). The FOXC2-associated angiogenic phenotype can be completely prevented by specific inhibition of the Ang-2 signaling pathway. Thus, activation of BAT and the brite/beige transition under various experimental models may employ different angiogenesis mechanisms. Vascular Functions in Modulation of Insulin Sensitivity The initial evidence of adipose vasculatures in modulation of insulin sensitivity was obtained from genetic and HFD-fed obese mouse models that received treatment of a generic angiogenesis inhibitor, TNP-470, which is relatively specific for endothelial cells (Bra˚kenhielm et al., 2004a; Rupnick et al., 2002) (Figure 5). More recently, several published works from independent groups provide compelling evidence supporting the crucial roles of adipose angiogenesis in the modulation of insulin sensitivity (Elias et al., 2012; Lu et al., 2012; Michailidou et al., 2012; Sun et al., 2012; Sung et al., 2013). In TNP-470-treated obese mice, reduction of adipose vascular density was correlated with improved insulin sensitivity (Bra˚kenhielm et al., 2004b). However, the TNP-470-improved insulin sensitivity might be Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc. 5

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Figure 4. Cold- and Adrenergic-ActivationInduced Adipose and Vascular Plasticity (A) Exposure to cold- and drug-induced adrenergic activation can lead to activation of BAT that executes nonshivering thermogenesis function via an UCP1-dependent pathway. Intriguingly, activation of thermogenic metabolism in BAT triggers a robust and VEGF-dependent angiogenic response. Concomitant to upregulation of VEGF and other angiogenic factors, endogenous angiogenesis inhibitors including thrombospodin (TSP) and a sVEGFR are downregulated to further ensure robust activation of angiogenesis. In rodents and probably other hibernated animals, cold exposure and drug-induced adrenergic activation induces a BAT-like phenotype in WAT (brite/beige), which displays similar thermogenic functions as BAT. Again, active angiogenesis occurs during the brite/ beige transition. (B) As VEGF is primarily involved in stimulation of angiogenesis during BAT activation and brite/ beige transition, adrenergic activation of VEGF expression has been studied, and the underlying mechanism involves hypoxia-independent activation of the vegf promoter by PGC-1a. However, hypoxia-dependent angiogenesis cannot be completely excluded.

attributable to its nonantiangiogenic activity because TNP-470 has neurotoxicity effects (Bhargava et al., 1999; Kudelka et al., 1997). A recent study shows that TNP-470 significantly inhibited food intake and increased energy expenditure, leading to reduced body-weight gain (White et al., 2012). Thus, it is highly plausible that the previously observed TNP-470-induced amelioration of insulin resistance is due to the suppression of food 6 Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc.

intake rather than adipose angiogenesis. The mechanism underlying TNP-470modulated energy balance may involve a novel regulatory signaling pathway by which the adipose vasculature acts centrally to regulate energy balance, as seen with an adipose-endotheliumspecific proapototic peptide (Kim et al., 2010; Kim et al., 2007c). Despite significant reduction of body fat in HFD-fed obese mice, White et al. (2012) showed that TNP-470-treated mice exhibited glucose intolerance, raising the possibility that other tissues such as muscle tissues also significantly contribute to glucose intolerance. Subsequent studies of vascular functions in the modulation of insulin sensitivity were mostly focused on VEGF, a bona fide endothelial growth factor, owing to its relatively specific targets on vascular endothelial cells. VEGF modulates insulin activity by regulating its production in pancreatic islets and its sensitivity in peripheral tissues such as adipose and muscle tissues. An important clue for how VEGF maintains these vascular features was achieved by generating a genetic mouse strain that carries specific VEGF-A deletion in b cells (RIP-Cre:Vegffl/fl). As expected, vascular density and fenestrations were markedly reduced in RIP-Cre:Vegffl/fl mice, as well as in mice with pancreatic-specific deletion of VEGF (PDX1-Cre:Vegf fl/fl) (Brissova et al., 2006; Iwashita et al., 2007; Watada, 2010). Interestingly, despite reduced insulin secretion in the islets, these mice do not seem to show obvious

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Figure 5. Antiobesity Therapy by Targeting Angiogenesis In WAT, inhibition of angiogenesis may provide a new therapeutic option for the treatment of obesity and diabetic retinopathy. In contrast, delivery of proangiogenic factors to metabolically active BAT may further accelerate metabolism, leading to a lean phenotype and improved insulin sensitivity. Stimulation of angiogenesis is also beneficial for the treatment of nonhealing diabetic ulcers.

glucose intolerance and insulin resistance (Toyofuku et al., 2009; Watada, 2010). VEGF has also been reported to acts as a survival factor for b cells through maintenance of vascular numbers and integrity (Xiao et al., 2013). The role of adipose VEGF in obesity and insulin sensitivity was also investigated using gain-of-function and loss-of-function genetic mouse models. In one study, temporal repression of systemic VEGF expression leads to a lean phenotype and resistance to diet-induced obesity owing to development of BAT-like adipocytes in WAT (Lu et al., 2012). In contrast, several other studies show that overexpression, but not repression, of VEGF in adipose tissues protects against diet-induced obesity and insulin resistance (Elias et al., 2012; Sun et al., 2012; Sung et al., 2013), although the activation of the BAT-like phenotype remains a significant part of the discrepancy. Sung et al. (2013) show that specific deletion or overexpression of VEGF in adipose tissues results in alterations of the adipose microenvironment and tissue hypoxia. In adipose VEGF-deleted mice, reduction of vascular density leads to hypoxia, apoptosis, inflammation, and metabolic defects in response to HFD. Inversely, high VEGF levels improve hypoxia and compromise obesity-related metabolic disorders. Although the discrepancy around VEGF in controlling metabolism remains unresolved, it highlights complex mechanisms that underlie the vasculature in the modulation of adipo-

cyte functions. A recent study showed that VEGF-B, a VEGF family member that exclusively binds to VEGFR1 (Cao, 2009), modulates endothelial fatty acid uptake and transport in muscular tissues (Hagberg et al., 2010). Moreover, genetic deletion of the vegfb gene and pharmacological inhibition of VEGF-B triggered signaling in diabetic mice that restored insulin sensitivity and glucose tolerance (Hagberg et al., 2012). In addition to the VEGF family, a genetically modified angiopoietin-1 variant, a more potent version of its native form, can improve insulin sensitivity by stimulation of adipose and muscle angiogenesis (Jung et al., 2012; Sung et al., 2009). Although these preclinical findings are seemingly contradictory in gain-of-function and loss-of-function experimental settings, careful analysis of all data by taking into consideration the different experimental conditions reveals consistent rather than contradictory conclusions regarding the VEGF paradox. It appears that the ultimate consequences of VEGF-modulated angiogenesis in controlling adipose tissue functions are context dependent (Sun et al., 2012). In expanding fat pads at the early stages of a HFD challenge, specific expression of VEGF in WAT enhances angiogenesis, leading to a ‘‘browning phenotype’’ of WAT (Sun et al., 2012). Interestingly, The VEGF-overexpressing adipose tissues show markedly elevated expression levels of UCP1 and PGC-1a, two browning adipose markers in brite/beige adipose tissues (Kajimura et al., 2010). Additionally, VEGF-induced browning adipose tissues exhibit a smaller average size of adipocytes, improvement of hypoxia, reduced fibrosis, and decreased inflammation (Sun et al., 2012; Sung et al., 2013). Importantly, overexpression of VEGF in adipose tissues significantly improves glucose tolerance, insulin sensitivity, and other metabolic abnormalities in HFD-fed mice. Consistent with these findings, genetic overexpression of aP2-driven VEGF in WAT and BAT markedly increased microvessel density that protects these tissues from hypoxia and HFD-induced obesity and metabolic abonormalities (Elias et al., 2012). The overall conclusions from this study (Elias et al., 2012) are in general agreement with two other studies (Sun et al., 2012; Sung et al., 2013) that adipose VEGF expression protects mice from HFD-induced obesity, glucose intolerance, insulin resistance, and adipose tissue inflammation. Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc. 7

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Conversely, genetic deletion of endogenous VEGF in mouse adipose tissues produced an opposing phenotype. Adipose tissues in the VEGF deletion mice show low vascular density, high degrees of tissue hypoxia, elevated apoptosis of adipocytes, inflammation, and metabolic defects on a HFD (Sung et al., 2013). These findings demonstrate that stimulation of angiogenesis in expanding adipose tissue is beneficially protective from development of metabolic disorders and insulin resistance (Sung et al., 2013). In another study, genetic repression of VEGF expression in various tissues results in a similar protective effect against HFD-induced obesity (Lu et al., 2012), as seen in adipose-tissue-specific VEGF overexpression expression models (Elias et al., 2012; Sun et al., 2012; Sung et al., 2013). Intriguingly, systemic VEGF repression also produced a browning adipose phenotype with upregulation of UCP1 and cell-death-inducing DNA fragmentation factor-a-like effector A (CIDEA) (Lu et al., 2012). Although molecular mechanisms underlying systemic repression of VEGF against HFD-induced obesity remain unknown, it is speculated that complex signaling systems might be involved in global regulation of metabolism, in association with vascular changes in various tissues and organs. Systemic VEGF repression may alter vascular numbers and structures in nonadipose tissues and organs, which subsequently modulate global metabolism in the body. For example, it is known that VEGF under physiological conditions is a maintenance factor for vascular homeostasis and systemic inhibition of the VEGFVEGFR2 signaling pathway in adult mice leads to marked regression of microvascular networks in various tissues (Kamba et al., 2006; Yang et al., 2013). In particular, microvasculatures in endocrine glands are exceptionally susceptible VEGF blockades. In the thyroid, pancreatic islets, and adrenal glands, for example, a more than 50% reduction of vascular density has been observed in these healthy tissues in response to anti-VEGF treatment (Kamba et al., 2006; Yang et al., 2013). Although functional impacts of VEGF inhibition on endocrine functions need to be proven in this pharmacological-intervention mouse model, sustained VEGF suppression as in the case of a genetically repressive mouse model would probably affect global metabolism and insulin sensitivity. Thus, nonadipose tissue effects of VEGFassociated vascular regression may significantly modulate metabolism and glucose tolerance. Taken together, these various findings from different laboratories may not necessarily contradict each other or the role of VEGF in the modulation of adipose tissue functions, metabolism, and insulin sensitivity manifested in a spatiotemporal and context-dependent manner. Several preliminary studies on human subjects suggest a possible correlation between adipose VEGF levels and insulin sensitivity. In one study, adipose levels of VEGF, VEGF-B, VEGF-C, and VEGF-D are negatively correlated with the development of insulin resistance in human obese patients (Tinahones et al., 2009). Other human patient data show that angiogenesis in subcutaneous adipose tissues is associated with the development of insulin resistance by enhancing lipid storage (Kim et al., 2007a). Analysis of a relatively small group of healthy young males implies the existence of a positive correlation between circulating VEGF levels and BMI, which is disassociated from insulin sensitivity (Loebig et al., 2010). It is noteworthy that unlike preclinical research, clinical studies with human sam8 Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc.

ples, particularly in small-cohort populations, are complex, and data can be influenced by other factors such as lifestyle, smoking, alcohol intake, and physical exercise. Therapeutic Challenges of Targeting Adipose Vasculature As discussed above, angiogenesis in BAT and WAT may produce opposing effects in controlling energy expenditure and lipid deposition. Could the proangiogenic or antiangiogenic concept be translated for the treatment of obesity and its related metabolic disorders? At the time of writing, therapeutic angiogenesis for treatment of obesity and metabolic diseases remains a paradoxically disputed issue, and there are limited data in the literature to conclusively support either option (Figure 5). Early studies using genetic and HFD-induced obese models in mice showed that delivery of generic angiogenesis inhibitors, including TNP-470 and angiostatin, markedly prevents obesity development (Bra˚kenhielm et al., 2004a; Rupnick et al., 2002). These findings have been validated by several subsequent studies (Kim et al., 2007c; White et al., 2012). In these low-metabolic obese mice, systemic treatment with broad-spectrum angiogenesis inhibitors suppresses adipose angiogenesis in various WAT depots, which is, in part, responsible for the antiobese effect. TNP-470 may target a broad-spectrum of cell types other than endothelial cells (Kudelka et al., 1997), and these non-vascular-related effects might also contribute to its antiobese activity. One of the intriguing findings of these studies is that TNP-470-treated obese mice demonstrate improved insulin sensitivity and blood lipid profiles (Bra˚kenhielm et al., 2004a). In clinical scenarios, the reduction of existing adipose tissue mass in obese individuals should be considered, and this may employ a rather different vasculature-related mechanism such as selective regression of adipose blood vessels. To achieve this goal, molecular players that control vascular survival signals in WAT should be identified and defined as potential therapeutic targets. Owing to their crucial roles in the regulation of angiogenesis, targeting VEGF-induced angiogenic signaling pathways for antiobese implications has become a focused theme in most studies. Despite the availability of various VEGF inhibitors, there is only limited published data to show the impact of anti-VEGF agents on adipose tissue growth. Specific inhibition of VEGFR2, but not VEGFR1, results in restriction of adipose tissue expansion, largely due to antiangiogenesis, in HFD-fed mice (Fukumura et al., 2003; Tam et al., 2009). Treatment with VEGFR2-specific blockade in HFD-fed mice did not result in significant body-weight changes during the first 5–6 weeks of diet-induced obesity (Tam et al., 2009). However, prolonged treatment with VEGFR blockade significantly decreased the body-weight gain compared with the HFD-fed control group. The VEGFR2-blockade-treated group had significantly lower food intake relative to the control group, and the VEGFR2blockade-induced body-weight loss and food-intake reduction was reversible after cessation of treatment (Tam et al., 2009). Most other data were obtained from VEGF genetic models using the gain-of-function and loss-of-function approaches. As discussed in the section of insulin sensitivity, the findings around VEGF-related obesity and metabolism exhibit spatiotemporal and context-dependent manner. Although some genetic studies are consistent with the pharmacological intervention of the

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review VEGF-VEGFR2 signaling system, demonstrating a lean phenotype in VEGF-deleted adipose tissues (Lu et al., 2012; Sung et al., 2013), others show that overexpression, but not repression, of VEGF protects mice from obese development (Elias et al., 2012; Sun et al., 2012). In both adipose VEGF overexpression and repression genetic models, activation of a BAT-like phenotype has been reported (Lu et al., 2012; Sun et al., 2012). Although the discrepancy of these findings warrants future clarification, these data, at least, demonstrate that targeting VEGF for treatment of obesity is not a straightforward approach. Careful examinations of these various findings show that antiVEGF treatment of established adipose tissues should be separated from those of expanding adipose tissues at the early stage of obesity. Systemic anti-VEGF treatment of expanding adipose tissues at the early stage of obesity results in increased adipocyte sizes and glucose intolerance (Sun et al., 2012). Paradoxically, delivery of the same VEGF blockade to established ob/ ob mice resulted in increased adipocyte death, weight loss, and improvement of glucose tolerance (Sun et al., 2012). It should be emphasized that systemic treatment with VEGF blockade may be likely to affect the pre-existing vasculatures in various tissues and organs as reported in nonobese mice (Kamba et al., 2006; Yang et al., 2013). Thus, the ultimate metabolic alterations of systemic pharmacological intervention reflect both adipose and nonadipose vascular changes. Interestingly, systemic delivery of a VEGF blockade to mice at the early stage of obesity only results in vascular changes in the adipose tissue, but not other tissues (Sun et al., 2012). Perhaps this is a dose-related phenomenon, given that angiogenic vessels in the adipose tissue would be more sensitive to anti-VEGF treatment than quiescent vasculatures in other tissues. Increases of anti-VEGF dosages may be likely to affect other vasculatures, owing to the fact that VEGF is a bona fide maintenance factor for vascular homeostasis in various tissues and organs (Kamba et al., 2006; Yang et al., 2013). In contrast to the systemic model, local deletion of the Vegf gene in the adipose tissue increases tissue hypoxia, inflammation, and metabolic defects including glucose intolerance in association with reduction of vascularity (Sung et al., 2013). The systemic impact of anti-VEGF agents on modulation of body weight, BMI, metabolism, glucose tolerance, and insulin sensitivity can be learnt from various cancer patients who receive anti-VEGF drugs, such as bevacizumab (Hurwitz et al., 2004). However, it seems to be difficult to obtain any conclusive results from human patient trials using bevacizumab and other VEGF inhibitor owing to several reasons: 1) given that antiangiogenic monotherapy does not, in most human cancer types, produce beneficial effects, anti-VEGF drugs are, in most cases, codelivered to patients together with chemotherapeutics, which significantly suppress appetite, vomiting, hepatotoxicity, gastrointestinal (GI) functions including stomatitis, mucositis, and diarrhea, and hematological toxicity (Perry, 1992). These chemotherapy-induced broad and GI-associated side effects significantly reduce body weight; (2) in almost all FDA-approved antiangiogenic trials, cancer patients with advanced and metastatic diseases are the primarily subjects for clinical trials. Patients with advanced malignant diseases often suffer from cachexia as a waste syndrome and their food-intake is minimal; and (3) anti-VEGF monotherapy can also significantly affects GI

tract functions (Tol et al., 2008). Thus, it is difficult to obtain conclusive data from these patients who receive antiangiogenic therapy. Unless chemotherapy is removed, food intake is accurately measured, metabolism is standardized, and systemic disease is well controlled, it would be difficult to speculate the body-weight outcome of anti-VEGF drugs in cancer patients. HIF-1a in adipose tissue is another attractive target given that HIF-1a-induced fibrotic response has been linked to hypoxiaassociated metabolic dysfunction in adipose tissues (Halberg et al., 2009). In support of this view, treatment of HFD-fed mice with a HIF-1a selective inhibitor, PX-478, effectively inhibits HFD-induced HIF-1a activation (Sun et al., 2013). Interestingly, inhibition of HIF-1a selectively reduces body-weight gain in HFD-fed mice, but not in normal-chow-fed mice. The anti-HIF1a-treated mice show increased energy expenditure and improvement of HFD-induced metabolic abnormalities, which are well correlated with reduced adipose fibrosis and inflammation. Genetic inhibition of endogenous HIF-1a in adipose tissues also produced similar metabolic improvements in HFD-fed mice (Sun et al., 2013). Thus, pharmacological targeting HIF-1a in the adipose tissue provides an attractive approach for treatment of obesity and metabolic disorders through possible conceivable mechanisms of inhibition of adipose fibrosis and inflammation. A challenging issue with this approach is how anti-HIF-1a agents are specifically delivered to the adipose tissues without interference with its physiological functions in other tissues. Given differential functions of the adipose vasculature in BAT and WAT, it is reasonably to speculate that local delivery of proangiogenic factors in metabolically active BAT would be beneficial as an antiobese therapy. Conversely, selective inhibition of angiogenesis in WAT would restrict energy storage and lead to a similar antiobese effect (Figure 5). The paradoxical view of angiogenesis-based therapy can also be extended to obesity- and diabetes-associated complications. For example, the type 2 diabetes-associated nonhealing foot ulcers often seen in obese patients lack sufficient proangiogenic signals to induce neovascularization, resulting in delayed wound healing. In this case, local delivery of proangiogenic factors such as PDGF-BB to the ulcer bed considerably improves wound healing (Mulder et al., 2009). On the contrary, inhibition of retinal angiogenesis and vascular leakiness in patients with diabetic macular edema has also been demonstrated as an effective approach for the improvement of visual acuity and for decreasing the risk of diabetic retinopathy progression (Smiddy, 2012). Thus, opposing angiogenesis principles, depending on the mechanisms of pathological angiogenesis, may be applied for the treatment of these obesity- and diabetes-related clinical complications. Conclusion Accumulating preclinical data show the intimate interplay between adipocyte and vascular compartments, which are reciprocally dependent on each other for determining energy expenditure or storage. Blood vessels interact with adipocytes at different levels to modulate the adipose microenvironment that is constantly altered throughout the entire lifetime. In response to metabolic changes, blood vessels undergo remodeling, growth, or regression to justify optimal levels of oxygen perfusion, adipocyte survival signals, supply of essential Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc. 9

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review nutrients, and removal of metabolic products, which collectively control adipose tissue functions. At the cellular level, adipocytes and cells in the vessel wall possess stem cell features that allow them to reciprocally switch to opposite cell types. Differentiation from vascular cells into adipocytes and vice versa ensures a prompt response to metabolic changes in adipose tissues. At the molecular level, adipocytes and vascular cells produce a range of soluble or insoluble factors and cytokines to reciprocally control their functions via a paracrine mechanism. Although adipocyte-derived factors in the regulation of adipose angiogenesis are relatively well characterized, the paracrine mechanisms by which endothelial cells modulate adipocyte function remain poorly understood. Additionally, several other cell types, including inflammatory cells and mesenchymal stromal cells, also actively participate in the regulation of adipose angiogenesis and modulate adipocyte functions. Apparently, angiogenesis in BAT provides sufficient fuels for energy expenditure, whereas the same angiogenic process in WAT can also promote energy storage. Owing to these opposing functions and the complex mechanisms that underlie interplay between vascular cells and adipocytes, targeting angiogenesis for treatment of obesity and its related metabolic disorders may not be a straightforward approach. Future therapeutic success is dependent on in-depth mechanistic understanding the role of angiogenesis in the modulation of adipose tissue expansion and regression. Despite currently limited experimental data and mechanistic understanding of adipose angiogenesis, targeting the adipose vasculature for the treatment of obesity and metabolic disease is an undisputed and exciting therapeutic approach that may be beneficial for millions of obese and diabetic patients. In fact, clinical studies have already validated the beneficial outcomes of diabetic nonhealing ulcers and macular edema by therapeutically targeting pathological diseases. ACKNOWLEDGMENTS The author thanks Patrik Andersson and Sharon Lim for the artistic work. The author’s laboratory is supported by research grants from the Swedish Research Council, the Swedish Cancer Foundation, the Karolinska Institute Foundation, the Karolinska Institute distinguished professor award, the Torsten So¨derbergs Foundation, the European Union Integrated Project of Metoxia (project no. 222741), and the European Research Council (ERC) advanced grant ANGIOFAT (project no. 250021).

Bozaoglu, K., Curran, J.E., Stocker, C.J., Zaibi, M.S., Segal, D., Konstantopoulos, N., Morrison, S., Carless, M., Dyer, T.D., Cole, S.A., et al. (2010). Chemerin, a novel adipokine in the regulation of angiogenesis. J. Clin. Endocrinol. Metab. 95, 2476–2485. Bra˚kenhielm, E., and Cao, Y. (2008). Angiogenesis in adipose tissue. Methods Mol. Biol. 456, 65–81. Bra˚kenhielm, E., Cao, R., Gao, B., Angelin, B., Cannon, B., Parini, P., and Cao, Y. (2004a). Angiogenesis inhibitor, TNP-470, prevents diet-induced and genetic obesity in mice. Circ. Res. 94, 1579–1588. Bra˚kenhielm, E., Veitonma¨ki, N., Cao, R., Kihara, S., Matsuzawa, Y., Zhivotovsky, B., Funahashi, T., and Cao, Y. (2004b). Adiponectin-induced antiangiogenesis and antitumor activity involve caspase-mediated endothelial cell apoptosis. Proc. Natl. Acad. Sci. USA 101, 2476–2481. Brissova, M., Shostak, A., Shiota, M., Wiebe, P.O., Poffenberger, G., Kantz, J., Chen, Z., Carr, C., Jerome, W.G., Chen, J., et al. (2006). Pancreatic islet production of vascular endothelial growth factor—a is essential for islet vascularization, revascularization, and function. Diabetes 55, 2974–2985. Brogi, E., Schatteman, G., Wu, T., Kim, E.A., Varticovski, L., Keyt, B., and Isner, J.M. (1996). Hypoxia-induced paracrine regulation of vascular endothelial growth factor receptor expression. J. Clin. Invest. 97, 469–476. Cannon, B., and Nedergaard, J. (2004). Brown adipose tissue: function and physiological significance. Physiol. Rev. 84, 277–359. Cao, Y. (2007). Angiogenesis modulates adipogenesis and obesity. J. Clin. Invest. 117, 2362–2368. Cao, Y. (2009). Positive and negative modulation of angiogenesis by VEGFR1 ligands. Sci. Signal. 2, re1. Cao, Y. (2010). Adipose tissue angiogenesis as a therapeutic target for obesity and metabolic diseases. Nat. Rev. Drug Discov. 9, 107–115. Cao, R., Bra˚kenhielm, E., Wahlestedt, C., Thyberg, J., and Cao, Y. (2001). Leptin induces vascular permeability and synergistically stimulates angiogenesis with FGF-2 and VEGF. Proc. Natl. Acad. Sci. USA 98, 6390–6395. Castellot, J.J., Jr., Karnovsky, M.J., and Spiegelman, B.M. (1982). Differentiation-dependent stimulation of neovascularization and endothelial cell chemotaxis by 3T3 adipocytes. Proc. Natl. Acad. Sci. USA 79, 5597–5601. Cederberg, A., Grønning, L.M., Ahre´n, B., Taske´n, K., Carlsson, P., and Enerba¨ck, S. (2001). FOXC2 is a winged helix gene that counteracts obesity, hypertriglyceridemia, and diet-induced insulin resistance. Cell 106, 563–573. Crandall, D.L., Hausman, G.J., and Kral, J.G. (1997). A review of the microcirculation of adipose tissue: anatomic, metabolic, and angiogenic perspectives. Microcirculation 4, 211–232. 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 and importance of brown adipose tissue in adult humans. N. Engl. J. Med. 360, 1509–1517.

REFERENCES Arany, Z., Foo, S.Y., Ma, Y., Ruas, J.L., Bommi-Reddy, A., Girnun, G., Cooper, M., Laznik, D., Chinsomboon, J., Rangwala, S.M., et al. (2008). HIF-independent regulation of VEGF and angiogenesis by the transcriptional coactivator PGC-1alpha. Nature 451, 1008–1012. Bhargava, P., Marshall, J.L., Rizvi, N., Dahut, W., Yoe, J., Figuera, M., Phipps, K., Ong, V.S., Kato, A., and Hawkins, M.J. (1999). A Phase I and pharmacokinetic study of TNP-470 administered weekly to patients with advanced cancer. Clin. Cancer Res. 5, 1989–1995. Bienes-Martı´nez, R., Ordo´n˜ez, A., Feijoo-Cuaresma, M., Corral-Escariz, M., Mateo, G., Stenina, O., Jime´nez, B., and Calzada, M.J. (2012). Autocrine stimulation of clear-cell renal carcinoma cell migration in hypoxia via HIF-independent suppression of thrombospondin-1. Sci. Rep. 2, 788.

Donath, M.Y., and Shoelson, S.E. (2011). Type 2 diabetes as an inflammatory disease. Nat. Rev. Immunol. 11, 98–107. Ekstrand, A.J., Cao, R., Bjorndahl, M., Nystrom, S., Jonsson-Rylander, A.C., Hassani, H., Hallberg, B., Nordlander, M., and Cao, Y. (2003). Deletion of neuropeptide Y (NPY) 2 receptor in mice results in blockage of NPY-induced angiogenesis and delayed wound healing. Proc. Natl. Acad. Sci. USA 100, 6033–6038. Elias, I., Franckhauser, S., Ferre´, T., Vila`, L., Tafuro, S., Mun˜oz, S., Roca, C., Ramos, D., Pujol, A., Riu, E., et al. (2012). Adipose tissue overexpression of vascular endothelial growth factor protects against diet-induced obesity and insulin resistance. Diabetes 61, 1801–1813.

Borges, J., Mu¨ller, M.C., Momeni, A., Stark, G.B., and Torio-Padron, N. (2007). In vitro analysis of the interactions between preadipocytes and endothelial cells in a 3D fibrin matrix. Minim. Invasive Ther. Allied Technol. 16, 141–148.

Famulla, S., Lamers, D., Hartwig, S., Passlack, W., Horrighs, A., Cramer, A., Lehr, S., Sell, H., and Eckel, J. (2011). Pigment epithelium-derived factor (PEDF) is one of the most abundant proteins secreted by human adipocytes and induces insulin resistance and inflammatory signaling in muscle and fat cells. Int J Obes (Lond) 35, 762–772.

Bouloumie´, A., Drexler, H.C., Lafontan, M., and Busse, R. (1998). Leptin, the product of Ob gene, promotes angiogenesis. Circ. Res. 83, 1059–1066.

Folkman, J. (1971). Tumor angiogenesis: therapeutic implications. N. Engl. J. Med. 285, 1182–1186.

10 Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc.

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Folkman, J. (2007). Angiogenesis: an organizing principle for drug discovery? Nat. Rev. Drug Discov. 6, 273–286. Fukumura, D., Ushiyama, A., Duda, D.G., Xu, L., Tam, J., Krishna, V., Chatterjee, K., Garkavtsev, I., and Jain, R.K. (2003). Paracrine regulation of angiogenesis 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 to PPARgamma activators through adipocyte VEGF and ANGPTL4 production. Am. J. Physiol. Endocrinol. Metab. 295, E1056–E1064. Goossens, G.H., and Blaak, E.E. (2012). Adipose tissue oxygen tension: implications for chronic metabolic and inflammatory diseases. Curr. Opin. Clin. Nutr. Metab. Care 15, 539–546. 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., and Spiegelman, B.M. (2012). Zfp423 expression identifies committed preadipocytes and localizes to adipose endothelial and perivascular cells. Cell Metab. 15, 230–239. Hagberg, C.E., Falkevall, A., Wang, X., Larsson, E., Huusko, J., Nilsson, I., van Meeteren, L.A., Samen, E., Lu, L., Vanwildemeersch, M., et al. (2010). Vascular endothelial growth factor B controls endothelial fatty acid uptake. Nature 464, 917–921. Hagberg, C.E., Mehlem, A., Falkevall, A., Muhl, L., Fam, B.C., Ortsa¨ter, H., Scotney, P., Nyqvist, D., Same´n, E., Lu, L., et al. (2012). Targeting VEGF-B as a novel treatment for insulin resistance and type 2 diabetes. Nature 490, 426–430. Halberg, N., Khan, T., Trujillo, M.E., Wernstedt-Asterholm, I., Attie, A.D., Sherwani, S., Wang, Z.V., Landskroner-Eiger, S., Dineen, S., Magalang, U.J., et al. (2009). Hypoxia-inducible factor 1alpha induces fibrosis and insulin resistance in white adipose tissue. Mol. Cell. Biol. 29, 4467–4483. Harwood, H.J., Jr. (2012). The adipocyte as an endocrine organ in the regulation of metabolic homeostasis. Neuropharmacology 63, 57–75. Heaton, G.M., Wagenvoord, R.J., Kemp, A., Jr., and Nicholls, D.G. (1978). Brown-adipose-tissue mitochondria: photoaffinity labelling of the regulatory site of energy dissipation. Eur. J. Biochem. 82, 515–521. Hurwitz, H., Fehrenbacher, L., Novotny, W., Cartwright, T., Hainsworth, J., Heim, W., Berlin, J., Baron, A., Griffing, S., Holmgren, E., et al. (2004). Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N. Engl. J. Med. 350, 2335–2342. Iwashita, N., Uchida, T., Choi, J.B., Azuma, K., Ogihara, T., Ferrara, N., Gerber, H., Kawamori, R., Inoue, M., and Watada, H. (2007). Impaired insulin secretion in vivo but enhanced insulin secretion from isolated islets in pancreatic beta cell-specific vascular endothelial growth factor-A knock-out mice. Diabetologia 50, 380–389. Jumabay, M., Zhang, R., Yao, Y., Goldhaber, J.I., and Bostro¨m, K.I. (2010). Spontaneously beating cardiomyocytes derived from white mature adipocytes. Cardiovasc. Res. 85, 17–27. Jumabay, M., Abdmaulen, R., Urs, S., Heydarkhan-Hagvall, S., Chazenbalk, G.D., Jordan, M.C., Roos, K.P., Yao, Y., and Bostro¨m, K.I. (2012). Endothelial differentiation in multipotent cells derived from mouse and human white mature adipocytes. J. Mol. Cell. Cardiol. 53, 790–800. Jung, Y.J., Choi, H.J., Lee, J.E., Lee, A.S., Kang, K.P., Lee, S., Park, S.K., Park, T.S., Jin, H.Y., Lee, S.Y., et al. (2012). The effects of designed angiopoietin-1 variant on lipid droplet diameter, vascular endothelial cell density and metabolic parameters in diabetic db/db mice. Biochem. Biophys. Res. Commun. 420, 498–504. Kajimura, S., Seale, P., and Spiegelman, B.M. (2010). Transcriptional control of brown fat development. Cell Metab. 11, 257–262. Kamba, T., Tam, B.Y., Hashizume, H., Haskell, A., Sennino, B., Mancuso, M.R., Norberg, S.M., O’Brien, S.M., Davis, R.B., Gowen, L.C., et al. (2006). VEGF-dependent plasticity of fenestrated capillaries in the normal adult microvasculature. Am. J. Physiol. Heart Circ. Physiol. 290, H560–H576. Kim, D.H., Woods, S.C., and Seeley, R.J. (2010). Peptide designed to elicit apoptosis in adipose tissue endothelium reduces food intake and body weight. Diabetes 59, 907–915.

Kim, J.Y., van de Wall, E., Laplante, M., Azzara, A., Trujillo, M.E., Hofmann, S.M., Schraw, T., Durand, J.L., Li, H., Li, G., et al. (2007a). Obesity-associated improvements in metabolic profile through expansion of adipose tissue. J. Clin. Invest. 117, 2621–2637. Kim, S.R., Bae, S.K., Choi, K.S., Park, S.Y., Jun, H.O., Lee, J.Y., Jang, H.O., Yun, I., Yoon, K.H., Kim, Y.J., et al. (2007b). Visfatin promotes angiogenesis by activation of extracellular signal-regulated kinase 1/2. Biochem. Biophys. Res. Commun. 357, 150–156. Kim, Y.M., An, J.J., Jin, Y.J., Rhee, Y., Cha, B.S., Lee, H.C., and Lim, S.K. (2007c). Assessment of the anti-obesity effects of the TNP-470 analog, CKD-732. J. Mol. Endocrinol. 38, 455–465. Kudelka, A.P., Levy, T., Verschraegen, C.F., Edwards, C.L., Piamsomboon, S., Termrungruanglert, W., Freedman, R.S., Kaplan, A.L., Kieback, D.G., Meyers, C.A., et al. (1997). A phase I study of TNP-470 administered to patients with advanced squamous cell cancer of the cervix. Clin. Cancer Res. 3, 1501–1505. Kukla, M., Berdowska, A., Gabriel, A., Sawczyn, T., Mazur, W., Sobala-Szczygie1, B., Grzonka, D., Zaje˛cki, W., Tomaszek, K., Bu1dak, R.J., and ZwirskaKorczala, K. (2011). Association between hepatic angiogenesis and serum adipokine profile in non-obese chronic hepatitis C patients. Pol. J. Pathol. 62, 218–228. Kuo, L.E., Kitlinska, J.B., Tilan, J.U., Li, L., Baker, S.B., Johnson, M.D., Lee, E.W., Burnett, M.S., Fricke, S.T., Kvetnansky, R., et al. (2007). Neuropeptide Y acts directly in the periphery on fat tissue and mediates stress-induced obesity and metabolic syndrome. Nat. Med. 13, 803–811. Lee, T.J., Bhang, S.H., Yang, H.S., La, W.G., Yoon, H.H., Shin, J.Y., Seong, J.Y., Shin, H., and Kim, B.S. (2012). Enhancement of long-term angiogenic efficacy of adipose stem cells by delivery of FGF2. Microvasc. Res. 84, 1–8. 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. Lim, S., Honek, J., Xue, Y., Seki, T., Cao, Z., Andersson, P., Yang, X., Hosaka, K., and Cao, Y. (2012). Cold-induced activation of brown adipose tissue and adipose angiogenesis in mice. Nat. Protoc. 7, 606–615. Loebig, M., Klement, J., Schmoller, A., Betz, S., Heuck, N., Schweiger, U., Peters, A., Schultes, B., and Oltmanns, K.M. (2010). Evidence for a relationship between VEGF and BMI independent of insulin sensitivity by glucose clamp procedure in a homogenous group healthy young men. PLoS ONE 5, e12610. Lu, X., Ji, Y., Zhang, L., Zhang, Y., Zhang, S., An, Y., Liu, P., and Zheng, Y. (2012). Resistance to obesity by repression of VEGF gene expression through induction of brown-like adipocyte differentiation. Endocrinology 153, 3123– 3132. Man, K., Ng, K.T., Xu, A., Cheng, Q., Lo, C.M., Xiao, J.W., Sun, B.S., Lim, Z.X., Cheung, J.S., Wu, E.X., et al. (2010). Suppression of liver tumor growth and metastasis by adiponectin in nude mice through inhibition of tumor angiogenesis and downregulation of Rho kinase/IFN-inducible protein 10/matrix metalloproteinase 9 signaling. Clin. Cancer Res. 16, 967–977. 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-like cells. Nat. Med. 16, 1400–1406. Michailidou, Z., Turban, S., Miller, E., Zou, X., Schrader, J., Ratcliffe, P.J., Hadoke, P.W., Walker, B.R., Iredale, J.P., Morton, N.M., and Seckl, J.R. (2012). Increased angiogenesis protects against adipose hypoxia and fibrosis in metabolic disease-resistant 11b-hydroxysteroid dehydrogenase type 1 (HSD1)-deficient mice. J. Biol. Chem. 287, 4188–4197. Mu, H., Ohashi, R., Yan, S., Chai, H., Yang, H., Lin, P., Yao, Q., and Chen, C. (2006). Adipokine resistin promotes in vitro angiogenesis of human endothelial cells. Cardiovasc. Res. 70, 146–157. Mulder, G., Tallis, A.J., Marshall, V.T., Mozingo, D., Phillips, L., Pierce, G.F., Chandler, L.A., and Sosnowski, B.K. (2009). Treatment of nonhealing diabetic foot ulcers with a platelet-derived growth factor gene-activated matrix (GAM501): results of a phase 1/2 trial. Wound Repair Regen. 17, 772–779. Northcott, J.M., Yeganeh, A., Taylor, C.G., Zahradka, P., and Wigle, J.T. (2012). Adipokines and the cardiovascular system: mechanisms mediating health and disease. Can. J. Physiol. Pharmacol. 90, 1029–1059. Olson, L.E., and Soriano, P. (2011). PDGFRb signaling regulates mural cell plasticity and inhibits fat development. Dev. Cell 20, 815–826.

Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc. 11

Please cite this article in press as: Cao, Angiogenesis and Vascular Functions in Modulation of Obesity, Adipose Metabolism, and Insulin Sensitivity, Cell Metabolism (2013), http://dx.doi.org/10.1016/j.cmet.2013.08.008

Cell Metabolism

Review Perry, M.C. (1992). Chemotherapeutic agents and hepatotoxicity. Semin. Oncol. 19, 551–565.

correlative study with hypoxic induced factor and cyclooxygenase-2. PLoS ONE 4, e8213.

Petrovic, N., Walden, T.B., Shabalina, I.G., Timmons, J.A., Cannon, B., and Nedergaard, J. (2010). Chronic peroxisome proliferator-activated receptor gamma (PPARgamma) activation of epididymally derived white adipocyte cultures reveals a population of thermogenically competent, UCP1-containing adipocytes molecularly distinct from classic brown adipocytes. J. Biol. Chem. 285, 7153–7164.

Tol, J., Cats, A., Mol, L., Koopman, M., Bos, M.M., van der Hoeven, J.J., Antonini, N.F., van Krieken, J.H., and Punt, C.J. (2008). Gastrointestinal ulceration as a possible side effect of bevacizumab which may herald perforation. Invest. New Drugs 26, 393–397.

Pino, E., Wang, H., McDonald, M.E., Qiang, L., and Farmer, S.R. (2012). Roles for peroxisome proliferator-activated receptor g (PPARg) and PPARg coactivators 1a and 1b in regulating response of white and brown adipocytes to hypoxia. J. Biol. Chem. 287, 18351–18358. Rupnick, M.A., Panigrahy, D., Zhang, C.Y., Dallabrida, S.M., Lowell, B.B., Langer, R., and Folkman, M.J. (2002). Adipose tissue mass can be regulated through the vasculature. Proc. Natl. Acad. Sci. USA 99, 10730–10735. Sakuma, T., Matsumoto, T., Kano, K., Fukuda, N., Obinata, D., Yamaguchi, K., Yoshida, T., Takahashi, S., and Mugishima, H. (2009). Mature, adipocyte derived, dedifferentiated fat cells can differentiate into smooth muscle-like cells and contribute to bladder tissue regeneration. J. Urol. 182, 355–365. Shitaye, H.S., Terkhorn, S.P., Combs, J.A., and Hankenson, K.D. (2010). Thrombospondin-2 is an endogenous adipocyte inhibitor. Matrix Biol. 29, 549–556. Sierra-Honigmann, M.R., Nath, A.K., Murakami, C., Garcı´a-Carden˜a, G., Papapetropoulos, A., Sessa, W.C., Madge, L.A., Schechner, J.S., Schwabb, M.B., Polverini, P.J., and Flores-Riveros, J.R. (1998). Biological action of leptin as an angiogenic factor. Science 281, 1683–1686. Smiddy, W.E. (2012). Clinical applications of cost analysis of diabetic macular edema treatments. Ophthalmology 119, 2558–2562. Sun, K., Halberg, N., Khan, M., Magalang, U.J., and Scherer, P.E. (2013). Selective inhibition of hypoxia-inducible factor 1a ameliorates adipose tissue dysfunction. Mol. Cell. Biol. 33, 904–917. 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 effects of VEGF-A on adipose tissue dysfunction. Proc. Natl. Acad. Sci. USA 109, 5874–5879. Sung, H.K., Kim, Y.W., Choi, S.J., Kim, J.Y., Jeune, K.H., Won, K.C., Kim, J.K., Koh, G.Y., and Park, S.Y. (2009). COMP-angiopoietin-1 enhances skeletal muscle blood flow and insulin sensitivity in mice. Am. J. Physiol. Endocrinol. Metab. 297, E402–E409. Sung, H.K., Doh, K.O., Son, J.E., Park, J.G., Bae, Y., Choi, S., Nelson, S.M., Cowling, R., Nagy, K., Michael, I.P., et al. (2013). Adipose vascular endothelial growth factor regulates metabolic homeostasis through angiogenesis. Cell Metab. 17, 61–72. Tam, J., Duda, D.G., Perentes, J.Y., Quadri, R.S., Fukumura, D., and Jain, R.K. (2009). Blockade of VEGFR2 and not VEGFR1 can limit diet-induced fat tissue expansion: role of local versus bone marrow-derived endothelial cells. PLoS ONE 4, e4974.

Toyofuku, Y., Uchida, T., Nakayama, S., Hirose, T., Kawamori, R., Fujitani, Y., Inoue, M., and Watada, H. (2009). Normal islet vascularization is dispensable for expansion of beta-cell mass in response to high-fat diet induced insulin resistance. Biochem. Biophys. Res. Commun. 383, 303–307. 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 white and brown fat cells. Cell Metab. 15, 222–229. Trayhurn, P. (2013). Hypoxia and adipose tissue function and dysfunction in obesity. Physiol. Rev. 93, 1–21. van Marken Lichtenbelt, W.D., Vanhommerig, J.W., Smulders, N.M., Drossaerts, J.M., Kemerink, G.J., Bouvy, N.D., Schrauwen, P., and Teule, G.J. (2009). Cold-activated brown adipose tissue in healthy men. N. Engl. J. Med. 360, 1500–1508. Watada, H. (2010). Role of VEGF-A in pancreatic beta cells. Endocr. J. 57, 185–191. White, H.M., Acton, A.J., and Considine, R.V. (2012). The angiogenic inhibitor TNP-470 decreases caloric intake and weight gain in high-fat fed mice. Obesity (Silver Spring) 20, 2003–2009. Wu, J., Bostro¨m, 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 are a distinct type of thermogenic fat cell in mouse and human. Cell 150, 366–376. Xiao, X., Guo, P., Chen, Z., El Gohary, Y., Wiersch, J., Gaffar, I., Prasadan, K., Shiota, C., and Gittes, G.K. (2013). Hypoglycemia reduces vascular endothelial growth factor A production by pancreatic beta cells as a regulator of beta cell mass. J. Biol. Chem. 288, 8636–8646. Xue, Y., Cao, R., Nilsson, D., Chen, S., Westergren, R., Hedlund, E.M., Martijn, C., Rondahl, L., Krauli, P., Walum, E., et al. (2008). FOXC2 controls Ang-2 expression and modulates angiogenesis, vascular patterning, remodeling, and functions in adipose tissue. Proc. Natl. Acad. Sci. USA 105, 10167–10172. Xue, Y., Petrovic, N., Cao, R., Larsson, O., Lim, S., Chen, S., Feldmann, H.M., Liang, Z., Zhu, Z., Nedergaard, J., et al. (2009). Hypoxia-independent angiogenesis in adipose tissues during cold acclimation. Cell Metab. 9, 99–109. Xue, Y., Lim, S., Bra˚kenhielm, E., and Cao, Y. (2010). Adipose angiogenesis: quantitative methods to study microvessel growth, regression and remodeling in vivo. Nat. Protoc. 5, 912–920. Yang, Y., Zhang, Y., Cao, Z., Ji, H., Yang, X., Iwamoto, H., Wahlberg, E., La¨nne, T., Sun, B., and Cao, Y. (2013). Anti-VEGF- and anti-VEGF receptor- induced vascular alteration in mouse healthy tissues. Proc. Natl. Acad. Sci. USA 110, 10218–10223.

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 the adipose vasculature. Science 322, 583–586.

Ye, J. (2009). Emerging role of adipose tissue hypoxia in obesity and insulin resistance. Int J Obes (Lond) 33, 54–66.

Tinahones, F., Salas, J., Mayas, M.D., Ruiz-Villalba, A., Macias-Gonzalez, M., Garrido-Sanchez, L., DeMora, M., Moreno-Santos, I., Bernal, R., Cardona, F., and El Bekay, R. (2009). VEGF gene expression in adult human thymus fat: a

Ye, J., Gao, Z., Yin, J., and He, Q. (2007). Hypoxia is a potential risk factor for chronic inflammation and adiponectin reduction in adipose tissue of ob/ob and dietary obese mice. Am. J. Physiol. Endocrinol. Metab. 293, E1118–E1128.

12 Cell Metabolism 18, October 1, 2013 ª2013 Elsevier Inc.