Macrophage ontogeny in the control of adipose tissue biology

Abstract

Macrophages are found in large numbers in the adipose tissue where they closely associate with the adipocytes and the vasculature. Adipose tissue macrophages are a heterogenous population of cells with ‘hard wired’ diversity brought upon by distinct developmental lineages. The purpose of this review is to provide a brief history of macrophages in control of adipose tissue metabolism with the emphasis on the importance of macrophage ontogeny.

Introduction

Adipose tissue and organismal metabolism

Daily and seasonal variations in caloric intake dictate a need for cycles of energy storage and expenditurê). To accommodate these changes in nutritional availability, all eukaryotes from yeast to man have evolved mechanisms to store calories in the form of lipid droplets(2). Amongst all eukaryotes only vertebrates including all mammals, birds, reptiles, amphibians and many fish have cells that are readily identifiable as adipocytes, although the anatomical location of these cells varies greatly between species(2, 3). At the molecular level, orthologous lipid storage genes performing similar functions can be found in worms, flies, and mammals(4). Adipocytes store large amounts of energy in the form of esterified lipids in a manner that is not toxic to the cell or the organism as a whole, a function that is indispensable to the survival of the organism(1, 5, 6). Adipocytes release fatty acids into the circulation when glucose is limiting. These fatty acids are generated by breaking down triacylglycerols that contain more energy per unit mass than do carbohydrates and can essentially be stored anhydrously(1, 5–7).

In addition to the primary function of adipocytes and the adipose tissue in energy storage, there seem to exist adipose depot-specific functions. Transplantation studies have demonstrated that transferring subcutaneous fat to the visceral compartment leads to reduced adiposity and improvement in glucose homeostasis in the recipient tissue(8). Additionally, many diseases that affect adipose tissue show depot-specific outcomes. For example, glucocorticoid excess is associated with redistribution of fat only to visceral stores with relative wasting of subcutaneous fat. A similar pattern is seen in the acquired lipodystrophy associated with certain HIV treatment regimens(9). Congenital lipodystrophy can also preferentially affect specific depots, with different patterns of fat loss associated with distinct genetic lesions (For a comprehensive review see 9).

Macrophages in organismal and adipose tissue metabolism

Macrophages were first hypothesized to play a role in energy metabolism via their function in inflammation, a phenomenon that is associated with insulin resistance, hyperglycemia, and altered fatty acid synthesis(10–12). However, it was not until the early 1980s that it was experimentally demonstrated that macrophage-derived factors, when added to human adipocyte cells, could decrease insulin response(10), reduce lipoprotein lipase activity(11), and decrease fatty acid synthesis(12). In the following decades, It was shown that obesity is associated with a heightened inflammatory condition(13, 14) that involves activation and recruitment of bone-marrow derived monocyte/macrophages into the adipose tissue of obese animals(15, 16), a phenomenon that causes inflammation, insulin resistance, and type 2 diabetes(10, 13–15, 17–28). It is now known that all the adipose tissue depots of healthy or obese animals contain a large population of innate and adaptive immune cells, numerically dominated by macrophages that surround the adipocytes and vasculature(19, 20, 29).

Macrophage heterogeneity in the adipose tissue

Adipose tissue macrophages are a diverse population of cells that express distinct surface markers and possess unique anatomical locations(29–33). This diversity in macrophages is in part proposed to result from reprogramming of one macrophage subset into another(34–41). In this model, adipose tissue macrophages alter their transcriptional programs to ‘polarize’ from an homeostatic state into an inflammatory state or vice versa depending on the environmental cues. For example, macrophages are proposed to ‘polarize’ into an inflammatory state in the adipose tissue of obese animals and result in obesity associated complications such as insulin resistance(7, 18, 42–47). However, recent studies(34–41) indicate that there exist layers of ‘hard wired’ macrophage diversity that result from developmental processes that take place in the embryo and the adult bone marrow to define myeloid cell function in vivo. For example, erythro-myeloid progenitors(EMP) that reside in the yolk-sac differentiate during embryogenesis to give rise to long lived, self-maintaining, tissue resident macrophages such as microglia and Kupffer cells(19, 21). In contrast, bone-marrow hematopoietic stem cells (HSC) differentiate at steady state along several lineages to generate a number of distinct cells types, including several monocyte subsets that infiltrate tissues to give rise to macrophage-like cells with short lifespans that are continuously renewed from the bone-marrow HSCs(19, 21, 33). There now exists experimental evidence indicating the presence of such ‘hard wired’ developmental heterogeneity in the adipose tissue macrophages. In amphibians, the adipose tissue contains both self-renewing macrophages that populate the adipose tissue before the establishment of bone marrow hematopoiesis and macrophages that originate from the bone-marrow(33). Similarly in mice, there exist a population of adipose tissue macrophages that are long lived(29), self-renewing(48), and molecularly distinct from the other macrophage subsets(32) suggesting a potential developmental heterogeneity in the adipose tissue macrophages of mice. Indeed, recent studies on the heterogeneity of macrophages using fate mapping models indicate that in the adipose tissue, yolk-sac derived tissue resident macrophages and bone-marrow derived monocytes/macrophages coexist (33)(manuscript under review)(Figure 1). Of note, a population of adipose tissue macrophages that express markers reminiscent of yolk-sac derived macrophages in mice are also found in humans, suggesting a potential developmental heterogeneity comparable to mice and amphibians(30).

Figure 1). Macrophages of distinct developmental origin coexist in the adipose tissue of lean and obese animals.

Upon high fat diet feeding, adipocytes and by extension the adipose tissue undergo dramatic expansion. Adipocyte hypertrophy is associated with endoplasmic reticulum stress which results in cell death and recruitment of monocytes into the adipose tissue. The recruited cells are primarily found in association with dead/dying adipocytes in “crown-like structures”. The present body of data indicate that macrophages dynamically adapt their population sizes (the proportion and the absolute number of different macrophage subsets) to accommodate changes associated with dramatic alterations in the size of the adipocytes and the adipose tissue.

The presence of developmentally distinct macrophage populations in the adipose tissue suggests that ‘reprogramming’ of macrophages proposed in context of obesity and inflammation could reflect a change in the proportion and the absolute number of different macrophage subsets resulting from recruitment of bone-marrow derived monocytes into the adipose tissue of obese animals(Figure 1).

Homeostatic roles of macrophages in the adipose tissue

Macrophages regulate fat storage in the adipose tissue

Macrophages are present in the adipose tissue of lean as well as obese animals(7, 18, 42–47, 49–55) (Figure 1). The function of these macrophages is well characterized in the context of obesity and metabolic syndrome(described below), however, their function at steady state and their contribution to lipid storage in the adipose tissue remains somewhat enigmatic. Depletion of all macrophages in mice using CSF1 blocking antibodies postnatally provided perhaps the first glimpse into the function of macrophages. These experiments revealed that in the absence of all macrophages adipocytes are smaller in size(51). Later studies revealed that macrophages are necessary for the formation of adipocytes in tissue engineering chambers(50, 56). However, it was not until the generation of Csf1r mutant rats that it became apparent that macrophages are required for proper development of the adipose tissue(57). Interestingly, the defective adipose tissue formation does not seem to require the absence of all macrophages as experiments in Trib1 deficient mice that lack specifically mannose receptor+ macrophages in the adipose tissue can recapture the same phenotype(32). This unique subset of macrophages is found directly associated with the adipocytes and is long lived, self-maintaining, and most likely of embryonic origin(29, 32) suggesting a role for yolk-sac derived macrophages in proper development of the adipose tissue. Our studies in mice that specifically lack tissue resident macrophages but still maintain HSC-derived monocytes/macrophages demonstrate that yolk-sac derived macrophages are required for expansion of adipocytes postnatally and adipocyte hypertrophy upon high fat diet feeding(manuscript under review). This function is in part mediated through the production of PDGFcc, a PDGF/VEGF family growth factor, by yolk-sac derived macrophages(manuscript under review). It is noteworthy that other macrophage-derived factors such TNF have also been implicated in adipocyte differentiation as well as modulation of adipocyte size(53, 58). These observations on the role of macrophages in control of adipocyte expansion is of particular interest as too much fat(obesity) and too little fat(lipodystrophy) are both associated with metabolic complications such as insulin resistance and hyperglycemia(1).

Metabolic sensors in macrophages

In response to a lipid rich diet, adipose tissue macrophages of embryonic origin increase PDGFcc transcription to promote adipocyte hypertrophy(manuscript under review). This observation requires that macrophages act as metabolic sensors in the adipose tissue. In agreement with this hypothesis, macrophages respond to dietary changes in Drosophila and mice(59, 60) and in particular, some adipose tissue macrophages are observed in close proximity to the vasculature where they may sample the circulation(29). This sampling of the blood would allow for the sensing of circulating metabolites in addition to removal of potentially noxious material.

What could then be the nutritional sensor in macrophages that monitors changes in circulating metabolites? The nuclear receptors (NRs) are promising candidates for this function. NRs are a large family of ligand-activated transcription factors that regulate several important processes, including development, reproduction, and metabolism(61, 62). NRs respond to lipophilic hormones and vitamins, but more importantly dietary lipids(63). NRs include members such as the Peroxisome proliferator activated receptors(PPAR) and liver X receptors(LXR)(61–63).

Several studies have shown that the administration of PPARy agonists inhibits the development of atherosclerosis in low-density lipoprotein(LDL) receptor–deficient(LDLR^−/−)(64, 65) and apolipoprotein E-deficient(apoE^−/−) mice(66). It is proposed that these effects of PPARy agonists are in part mediated through PPARy activation in macrophages therefore leading to upregulation of scavenger receptor CD36(75) and genes responsible for cholesterol efflux(67). These transcriptional changes promote removal of lipids and fatty acids from the environment and stimulate efflux and loading of such molecules into lipoproteins for transport to the liver where they are processed(67), thus shifting the balance from lipid loading to lipid efflux in the atherosclerotic lesions. PPARy may also exert anti-inflammatory effects in macrophages directly(68) or indirectly through LXR(69–73). Disruption of PPARy in myeloid cells impairs insulin response, glucose homeostasis, and downregulates the expression of genes involved in oxidative phosphorylation and mitochondrial homeostasis in the skeletal muscle and liver(74). This leads to decreased insulin sensitivity in these tissues(74). PPARy reduces monocyte CCR2 expression(76, 77) therefore contributing to altered recruitment of monocytes into different tissues including adipose tissue of obese animals. Other members of the peroxisome proliferator activated receptors are also implicated in macrophage biology. For example, macrophages sense very-low-density lipoprotein (VLDL) via the activation of the nuclear receptor PPAR5 (78) to upregulate genes involved in lipid handling such as Perilipin. Altogether, these studies indicate that PPAR and LXR activity in macrophages may control lipid metabolism, glucose metabolism, and monocyte recruitment into the adipose tissue of obese animals therefore providing a link between circulating fatty acids, lipid sensing in macrophages, inflammation in the adipose tissue, and complications associated with metabolic syndrome.

Pathogenic roles of macrophages in obesity

Macrophages in adipose tissue inflammation

Obesity promotes adipocyte hypertrophy and endoplasmic reticulum stress(82) which may be the initial signals to trigger inflammation(31, 83) in the adipose tissue. Upon adipose tissue inflammation in obese animals, monocytes are recruited to the adipose tissue under the influence of secreted proteins such as MCP-1(CCL2)(19, 20). These recruited monocytes/macrophages are often found surrounding the dead/dying adipocytes in so-called ‘crown-like structures’ where they scavenge cell debris and free lipid droplets(84)(Figure 1). It is proposed that adipose tissue inflammation and monocyte recruitment are responsible for insulin resistance. Accordingly, targeted ablation of either MCP-1 or its receptor (CCR2) reduces monocyte infiltration of the adipose tissue and improves insulin sensitivity(21). Interestingly, these mice do not present with changes in body weight, suggesting that bone-marrow derived monocytes/macrophages may have little to no role in adipocyte hypertrophy or weight gain in animals on lipid rich diet(21). Overexpression of MCP-1 in the adipose tissue causes monocyte recruitment and insulin resistance(27, 28). Decrease in monocyte infiltration into the adipose tissue is also associated with a reduction in crown-like structures, TNF, and iNOS in the adipose tissue of obese animals(19, 21, 24), further emphasizing the function of bone-marrow derived monocytes/macrophages in the obesity associated inflammation and the subsequent metabolic syndrome.

Of note, cultured adipocytes in vitro made hypertrophic with oleic acid display insulin resistance without triggering an inflammatory response(85) suggesting monocyte infiltration may not be the only contributing factor to induction of insulin resistance in the adipose tissue.

Macrophages in ectopic fat deposition

Excess energy intake in lieu of limiting adipocyte hypertrophy results in the overflow of lipids that need to be stored in cells not specialized for long term fat storage in a process termed ectopic fat deposition(86). Ectopic deposition of lipids is often accompanied with inflammation in the affected organs such as the liver and skeletal muscle(15, 16, 21, 24–28). Interestingly, the development of insulin resistance in tissues such as the liver and skeletal muscle are often preceded by local deposition of lipids, suggesting that this process may directly or indirectly contribute to insulin resistance(87). Studies in mice deficient for MCP-1 or its receptor (CCR2) suggest that ectopic fat deposition requires the infiltration of monocytes into the affected organs(27)(manuscript under review). It is not clear how bone marrow-derived monocytes/macrophages promote ectopic fat deposition. There are indications that induction of inflammation in the liver may be sufficient to induce fat deposition in hepatocytes, however, the role of monocytes was not formally ruled out in these experiments(88). The proposed role of bone marrow-derived monocytes/macrophages in ectopic fat deposition demonstrates a functional dichotomy between these cells and EMP-derived macrophages that are required for fat storage in specialized fat storing cells i.e adipocytes. This further emphasizes the necessity to consider the developmental origin of macrophages when studying their functions in the adipose tissue and organismal metabolism.

Development of therapeutic targeting macrophages for treatment of obesity and accompanied morbidities

Elegant studies on the origin and function of macrophages in the adipose tissue of lean and obese animals have provided new and exciting opportunities for development of therapies that may improve the life of individuals suffering from morbid obesity. Targeted ablation of either MCP-1 or its receptor(CCR2) improves insulin sensitivity(19, 21, 24, 26, 89). Similarly, Ccr2 silencing or antagonism in the adipose tissue of wildtype mice or those suffering from genetic causes of obesity, such as leptin receptor deficient mice, results in reduced ectopic fat deposition and inflammation leading to improved glucose and insulin resistance(19, 21,24, 26, 89). These results suggest that targeting the CCR2/CCL2 axis may prove a viable therapeutic option in obese individuals suffering from metabolic syndrome. Indeed, a clinical trial has indicated the beneficial effects of the CCR2 antagonist CCX140-B on glycemic parameters in human subjects(90). However, it is important to note that targeted ablation of CCR2/CCL2 axis does not affect body weight(21). An approach that may prove successful in correcting the weight gain due to over feeding is through targeting of PDGF-receptor signaling. These receptors play complex roles in control of adipocyte size and therefore lipid storage(91–95). Our study on the role of adipose tissue macrophages in lean and obese mice demonstrate that blockade of macrophage-derived PDGFcc may be beneficial in preventing further weight gain in obese individuals(91–95)(manuscript under review). This treatment, however, does not prevent complications associated with obesity and therefore may be best used in combination with CCR2 antagonists.

An alternative approach is the use of CSF1R inhibitors or blocking antibodies, the benefit of such an approach is that it depletes all macrophages including yolk-sac derived macrophages that control adipocyte hypertrophy(96)(manuscript under review) and bone-marrow derived monocytes/macrophages that contribute to insulin resistance and ectopic fat deposition(15, 16, 21,24–28). However, systemic complications of macrophages depletion by CSF1R inhibitors and antibodies may preclude such dramatic measures.

Summary and outlook

In the last few decades, many studies using new genetic mouse models have provided insights into the roles of macrophages in the adipose tissue. Albeit many questions are still open, it is now demonstrated that yolk-sac derived macrophages promote the postnatal expansion of adipocytes during early development and control adipocyte hypertrophy in obese mice. In contrast, bone-marrow derived monocytes/macrophages are primarily involved in adipose tissue inflammation, ectopic fat deposition, and insulin resistance. Many challenges lie in identification of mechanisms by which macrophages sense the nutritional status of the organism and then inform adipocytes of the ever-changing environment. Mounting evidence suggests a role for yolk-sac derived macrophages in formation and expansion of adipocytes in adult mice, but it not clear if these cells may be involved in adipose tissue wasting or lipodystrophy. In regard to bone-marrow derived monocytes/macrophages, it is not apparent how and why macrophages promote ectopic lipid uptake by cells not specialized for fat storage such as hepatocytes. It is also not clear for what purpose bone-marrow derived monocytes/macrophage may be recruited to the adipose tissue of obese animals. One proposed reason is to act as buffers for the dramatic increase in lipid flux(97). Future work investigating the respective contributions of lineage-specific functions of macrophages will greatly benefit our understanding of adipose tissue biology and lipid storage in specialized and non-specialized fat storing cells.

Highlights.

• Adipose tissue is populated with developmentally distinct populations of macrophages

• Yolk-sac derived adipose tissue macrophages control adipose tissue development and adipocyte hypertrophy

• Bone-marrow derived monocytes/macrophages contribute to insulin resistance and lipid storage in cells not specialized for long term fat storage such as hepatocytes