Adipocytes, Innate Immunity and Obesity: A Mini-Review

The role of adipose tissue (AT) inflammation in obesity and its multiple related-complications is a rapidly expanding area of scientific interest. Within the last 30 years, the role of the adipocyte as an endocrine and immunologic cell has been progressively established. Like the macrophage, the adipocyte is capable of linking the innate and adaptive immune system through the secretion of adipokines and cytokines; exosome release of lipids, hormones, and microRNAs; and contact interaction with other immune cells. Key innate immune cells in AT include adipocytes, macrophages, neutrophils, and innate lymphoid cells type 2 (ILC2s). The role of the innate immune system in promoting adipose tissue inflammation in obesity will be highlighted in this review. T cells and B cells also play important roles in contributing to AT inflammation and are discussed in this series in the chapter on adaptive immunity.


INTRODUCTION
Obesity is a growing healthcare problem in the United States and globally. It is a leading cause of preventable death and currently impacts more than 35% of the US population (1). This number is estimated to rise with a projected 42% of the US adult population being obese by 2030 (2). Obesity adversely impacts the entire body leading to increased Type 2 diabetes and associated complications, Alzheimer's disease, vascular dementia, obstructive sleep-apnea, accelerated atherosclerosis, heart failure, fatty liver disease, nonalcoholic steatohepatitis, osteoarthritis, altered immune system, impaired response to vaccines, and increased susceptibility to cancer compared to aged-matched lean individuals (3). Obesity is characterized by an expansion of both visceral and subcutaneous adipose tissue (AT) in the setting of chronic over-nutrition. The ensuing chronic low-grade inflammation sets the stage for many of the extensive complications. Thus, understanding mechanisms that mediate the immunological changes in obesity may unlock new therapeutic strategies. This review places a special emphasis on the innate immune system and the adipocyte.

INNATE AND ADAPTIVE IMMUNITY
The role of the immune system is to identify self-versus non-self to eliminate potential toxins, allergens, and pathogens without destroying the host tissue. The immune system is composed of two key functional responses, innate and adaptive immunity. The innate immune response is the initial line of defense after the host's external barrier. This immune response is not antigen-specific, but rather recognizes molecular patterns that are inherent to the toxin, allergen, or pathogen. This allows for the rapid activation of an immune response which is followed by the development of an antigen-specific immune response. The adaptive immune system upon the first encounter of a toxin, allergen, or pathogen undergoes expansion to aid the innate immune response. Upon resolution, a subset of these adaptive immune cells persists and creates a distinct population of memory immune cells. Memory cells are faster to respond to future encounters with the same pathogen, allergen, or toxin. The cellular component of both immune responses arises from the hematopoietic stem cell. This pluripotent cell further differentiates within the bone marrow to generate either the common lymphoid progenitor or the common myeloid progenitor. The common myeloid progenitor gives rise to lineage-specific colony-forming cells which then further develop into a majority of the innate immune cells as well as megakaryocytes (platelets) and erythrocytes (red blood cells). The cellular components of the innate immune response include the granulocytes including monocytes, macrophages, neutrophils, basophils, eosinophils, mast cells, and dendritic cells, as well as adipocytes. While the adipocyte is not traditionally viewed as an immune cell, recent research has demonstrated that the adipocyte releases adipokines, microRNAs and lipids to influence the innate immune response (4)(5)(6)(7). The adipocyte also expresses MHCII molecules during high-fat diet feeding allowing the adipocyte to interact with naïve T cells resulting in T cell differentiation and activation (8,9). The common lymphoid progenitor gives rise to the key immune cells within the adaptive immune response including B cells and T cells as well, as more recently discovered innate immune cells including the NK cell and the innate lymphoid cell types 1, 2, and 3 (10,11). The key link between the innate and adaptive immune system is antigen presentation.

INFLAMMATION IN ADIPOSE TISSUE; CONTRIBUTION OF INNATE IMMUNITY
The AT immune cell microenvironment in the lean state is a well-balanced crosstalk between the adipocyte and the stromal vascular fraction (SVF) or the cellular compartment of AT. In lean mice, the SVF is comprised of mesenchymal stem cells (12), endothelial progenitor cells (13)(14)(15)(16) as well as numerous immune cells including anti-inflammatory immunoregulatory T cells, Tregs (17), innate lymphoid type 2 cells (ILC2) (18), alternatively activated macrophages (19,20), and eosinophils (21). These cells work in concert to ensure the maintenance of homeostasis within AT including maintaining systemic insulin sensitivity. However, upon high-fat diet (HFD) feeding, there is a disruption of the anti-inflammatory milieu with increased differentiation and recruitment of pro-inflammatory immune cells creating a chronic, low-grade inflammatory state. In murine obesity, AT is characterized by the early, transient infiltration of neutrophils (22) followed by the accumulation of pro-inflammatory CD8+ T cells (23), CD4 Th1 cells (8) and M1 macrophages (19,20) all of which surround the dying adipocyte forming a crown-like structure. Innate immunity is an early and key component in sustaining AT inflammation.

Adipocyte
The adipocyte, unlike most traditional immune cells, links the innate and adaptive immune systems through adipokine, lipid and exosome release and through antigen presentation. While the adipocyte is the primary site of energy storage for the body and performs multiple metabolic activities, it can assume the role of a highly functional immune cell, releasing anti-and proinflammatory cytokines and hormones (adipokines), as well as lipids, which also act as signaling molecules (24). Since its discovery as an endocrine cell, the adipocyte has been identified to secrete more than 50 adipokines/cytokines including adiponectin (4,25), leptin (5), TNFa (26)(27)(28)(29), visfatin (30), and resistin (31) among many others (32) which impact local and systemic metabolism and inflammation.
The first adipokine described was leptin which revolutionized our understanding of the critical role that adipocytes play in whole-body energy homeostasis (33). Mutations in the leptin gene in the ob/ob mouse model led not only to hyperphagia and weight gain but also disruptions in fertility and body temperature regulation (34). Treatment of ob/ob mice with recombinant leptin, but not db/db (leptin receptor-deficient) mice, led to improved body weight and decreased food intake (35). However, contrary to what was initially hypothesized, leptin is found in higher levels in obese patients as compared to lean controls (36) suggesting the presence of leptin resistance (37). Increasing evidence supports an immunologic role of leptin. Leptin deficiency is associated with greater susceptibility to death after administration of LPS or TNFa which is partially corrected with leptin administration (38,39). Macrophages from leptindeficient mice have impaired phagocytosis and altered cytokine production (40,41). In neutrophils, leptin appears to increase ROS production (42), inhibit apoptosis (43), and affect neutrophil migration (44) suggesting that leptin impacts cells that mediate the innate immune response. More recently, Scherer and colleagues (45) demonstrated that hyperleptinemia is a driving force for metabolic disorders. Interestingly, a partial decrease of circulating leptin in obesity reestablishes hypothalamic leptin sensitivity and effectively reduces weight gain and enhances insulin sensitivity.
Unlike leptin, adiponectin, a key adipokine involved in energy homeostasis, is reduced in obese subjects and has antiinflammatory effects. Adiponectin is found in higher levels in AT and blood of lean subjects (46). Ob/ob mice with adiponectin overexpression have an increased ability to expand their subcutaneous AT associated with in a reduction of systemic and local AT inflammation. These mice also develop less ectopic lipid deposition in the liver and skeletal muscle leading to improvements in insulin sensitivity despite greater amounts of AT (47). Within the innate immune system, adiponectin acts primarily on macrophages resulting in a greater polarization of M2-like macrophages, decreased M1-like macrophages, and a reduction in ROS production (48). In neutrophils, adiponectin functions to decrease the production of the neutrophil chemokine CXCL8 (49) and ROS via modulation of NADPH oxidase (50). These observations highlight the yin-yang relationship of leptin and adiponectin, which functions as an anti-inflammatory regulator of the innate immune response. These hormones are the most well-known adipokines, but as discussed above, numerous other have been identified. Consistently obese versus lean humans and mice reveal increased proinflammatory and extracellular matrix gene expression, but the function of many adipokines remains unknown (51,52).
Lipid release is another important mechanism by which adipocytes can impact immune cells in the AT microenvironment. Lipids such palmitate and other unsaturated fatty acids can bind to toll-like receptors (TLRs) on the surface of immune cells, such as macrophages, and are converted to ceramides and diacylglycerols during states of lipid overabundance as occurs in obesity. These toxic lipids enhance proinflammatory signaling (7). More recently, branched-chain fatty acid esters of hydroxyl fatty acids (FAHFAs) produced by adipocytes were shown to bind to G-protein coupled receptors (GPRs) 40 and 120 to inhibit inflammation and improve insulin secretion and sensitivity (53). Within the blood and subcutaneous adipose tissue of insulin-resistant humans and mice, there is a reduction of several FAHFAs most notably palmitic acid esters of hydroxyl stearic acids. Supplementation of these via oral ingestion or subcutaneous administration improves glucose and insulin handling (53). These observations indicate adipocytes release both pro-and anti-inflammatory lipids as a mechanism to modulate the immune system.
Adipocytes are composed of large unilocular lipid droplets containing triacylglycerols and neutral free fatty acids (FFAs), which are a major mechanism for energy storage and release. Adipocytes not only release lipids and secrete adipokines, but they can also employ exosomes-extracellular vesicles (40-150 nm in size) of endosomal origin to participate in this process. Exosomes are increasingly recognized as a novel mechanism by which adipocytes communicate with other cells and target tissues. Their cargo contains adipokines, lipids, and microRNAs. Release is dependent on nutritional status and degrees of adiposity: increased release in obesity and decreased release with caloric restriction or lipodystrophy. Exosomes can be taken up by endocytosis, pinocytosis, or phagocytosis, and can be directed to target cells by adhesion molecules on the exosomal surface (6). Recently, Flaherty, et al. reported that adipocytes of mice release 1% of their lipid content daily ex vivo via exosomes, which was increased in obese mice. This release of exosomes contributed to macrophage foam cell formation, suggesting that exosomes contribute to the orchestration of AT immune cells (54). MicroRNAs, which regulate protein translation, are another important component of adipocyte-derived exosomes. Adipocytes are a major source of microRNAs in the circulation with greater than 55 differentially expressed microRNAs between lean and obese individuals. Adipocyte microRNAs contribute to the regulation of metabolism, inflammation, and multiple biologic processes locally and systemically (55). However, controversy exists as to whether adipocyte exosomes represent a minority or majority of circulating exosomes (54,56).
Finally, one of the most unique features of the adipocyte is its ability to function as an antigen-presenting cell, which is described in detail in Chapter XX of this series by Deng et al. The adipocyte can present antigen to promote differentiation and activation of interferon gamma-producing CD4 Th1 cells. This activity is increased early in obesity, after only 2 weeks HFD in mice, before the AT macrophage increase, suggesting adipocytes both instigate and maintain AT inflammation (8). Adipocytes provide a critical link between the innate and adaptive immune systems.

Neutrophils
Neutrophils are one of the initial inflammatory cells recruited to sites of host injury. As a component of the innate immune system, neutrophils have four primary activities including, phagocytosis, degranulation, reactive oxygen species (ROS) production, and neutrophil extracellular trap (NET) formation (57). Within mouse peripheral blood, neutrophils comprise 10-25% of the circulating immune cells (58), whereas in humans they compromise 50-70% of circulating immune cells. Furthermore, unlike humans, mouse neutrophils do not have defensins (anti-microbial peptides). While there are several differences between mouse and human neutrophils, mouse models are still routinely used for genetic manipulation (59).
Early studies in mice observed the transient infiltration of neutrophils into the AT after the start of HFD resulting in maximal levels by day 3 and undetectable levels by day 28 (22). However, others have suggested a more prolonged presence, contributing to about 2% of the SVF (60). Despite the early and relatively small contribution of the neutrophil in obese mice, loss of neutrophil elastase (60) or myeloperoxidase (61) leads to a decrease in AT inflammation and macrophage recruitment and promotes resolution of insulin resistance. However, loss of neutrophil NET formation does not impact obesity-related inflammation or insulin resistance in HFD-fed mice (62).
Within human AT, there have been fewer studies. One identified the presence of neutrophils within subcutaneous AT and reported that neutrophils were contained within the vasculature with no or limited infiltration into the tissue similar to vascular pools of neutrophils found in the liver (63). Another study showed limited infiltration in obesity in both visceral and subcutaneous AT (64); however, the quantity, cause of recruitment, and function of these cells within human AT remains unknown. Despite this, inflammatory lipids such as leukotrienes are known to attract and activate neutrophils. Under inflammatory conditions adipocyte and macrophages produce increased IL-8, a powerful neutrophil chemoattractant (65,66). Additionally neutrophils can self-recruit via increased production of CXCL2, another known neutrophil chemoattractant (67). Using a mouse peritonitis model, Tynan et al. demonstrated that lipids extracted from human adipocytes promoted migration and accumulation of neutrophils and macrophages, and activated these cells to produce cytokines (68). These effects were similar whether the adipocytes were obtained from lean or obese subjects, as fatty acid profiles, analyzed by gas chromatography, were not different. Oleic acid was also shown to recruit neutrophils in a similar mouse model (69). Additionally, adipocyte lipolysis has been shown to attract neutrophils and enhance their production of IL-1b leading to the activation of adipocytes and other immune cells (70). Further studies will be useful to determine the types of adipocyte lipids and other factors that attract and activate neutrophils into AT.

Macrophages
One of the most well-studied immune cells in AT and a key component of the innate immune response is the macrophage. Within mouse models of obesity, macrophages comprise up to 40% of the SVF (71) and are shown to be involved in the development of insulin resistance (72), atherosclerosis progression (73), and other obesity-related complications. In mice, macrophages have been classified into the relatively simplistic M1 and M2 phenotypes with obesity increasing the prevalence of pro-inflammatory M1-like macrophages (19). Ablation of these CD11c+ proinflammatory macrophages decreases AT inflammation and interferes with the development of insulin resistance, suggesting that macrophages are key mediators of insulin sensitivity (72). Furthermore inhibiting macrophage recruitment through the genetic depletion of CCR2 (74) or MCP-1 (75) also leads to the repression of AT inflammation and insulin resistance during HFD feeding. In contrast, accumulation of anti-inflammatory PPARg positive macrophages (M2 macrophages) leads to improvements in AT inflammation and insulin sensitivity (76), while loss of macrophage PPARg increases AT inflammation and insulin resistance (77). The balance between the pro and antiinflammatory macrophage subtypes is much less defined within human AT with most AT macrophages expressing both common M1 and M2 markers (78,79). Although there is still some debate on this topic with others suggesting a greater abundance of CD206 macrophages after weight loss (80). Further research on human adipose tissue macrophage subsets still needs to be done. The interaction between adipocytes and ATMs begins with the formation of crown-like structures characterized by macrophage accumulation surrounding dying adipocytes. This process is mediated by the adipocyte secretion of MCP-1 resulting in macrophage accumulation and activation (81). Once accumulated, these macrophages release TNFa which increases adipocyte release of FFAs (82). FFAs are capable of binding TLR4 on both the adipocyte and the macrophage resulting in NFkB activation and release of IL1b by macrophages (83). Adipocyte turnover occurs with approximately 10% of adipocytes undergoing apoptosis annually (84). These dying adipocytes are removed via trogocytosis by ATMs (85). Adipokine secretion by adipocytes also alters macrophage function. Increased leptin in obesity increases the phagocytic function of ATMs and is associated with an increase in circulating C-reactive protein (41,86). Adiponectin secretion is thought to be inhibited by TNFalpha secretion which is increased in obesity. In the lean state, FIGURE 1 | The earliest changes in adipose tissue inflammation include adipocyte hypertrophy with direct secretion/exosome release of key adipokines (increased leptin/decreased adiponectin) which in combination with increased adipocyte MHCII expression result in the differentiation of pro-inflammatory Thelper type 1 cells (Th1) resulting in decreased regulatory T cells (Tregs). Adipocytes during this time frame also increase IL-8 production resulting in neutrophil accumulation. Later in the time course (around 8 weeks), macrophage infiltration is stimulated by toxic lipid production and increased MCP-1 secretion which inhibits innate lymphoid cell type 2 (ILC2) differentiation.
adiponectin is known to inhibit the development of foam cells from macrophages and decreases endothelial cell activation and monocyte adhesion (87). The interaction between adipocytes and ATMs is a key contributor to the chronic low-grade inflammation of obesity (88).

Innate Lymphoid Cells Type 2
Another key innate immune cell that is not well-defined within AT is the innate lymphoid cell type 2 (ILC2). ILC2 cells are important in the maintenance of insulin sensitivity and are decreased in the setting of HFD. Innate lymphoid cells express CD4+ related cytokines, mirror Th1, Th2, and Th3 expression profiles (89), but differ in that they do not express B or T cell receptors despite arising from the common lymphoid progenitor cell (90). The ILC2 cell is similar to Th2 cells in that it contains the transcription factor GATA3 (91) and secretes IL5 and IL13. Within mouse models of obesity, the administration of IL25 leads to improvements in glucose tolerance and weight loss and is associated with the infiltration of ILC2 cells, alternatively activated macrophages, and eosinophils. Depletion of ILC2 cells in obese Rag1-/-mice leads to worsening insulin sensitivity and weight gain, while repletion of ILC2 cells reverses these negative metabolic consequences (18). Furthermore, in murine models, ILC2 cells appear to be the primary source of IL-5 and IL-13 and are necessary for the maintenance of alternatively activated macrophages and eosinophils, two key cells implicated in the anti-inflammatory state of lean AT (92). These cells are thought to contribute to an improved metabolic phenotype through the beiging of WAT characterized by increased expression of Ucp1 leading to an increase in caloric expenditure and attenuation of weight (93,94). These researchers also confirmed that ILC2 cells are markedly decreased in the SAT of obese compared to lean humans (93); thus highlighting a key role for ILC2 cells in AT (95).

CONCLUSIONS
Multiple changes in the innate immune system are key contributors to inflammation in obese AT resulting in the development of obesity-related diseases as summarized in Figure 1. Despite the greater understanding of the immunologic role of AT, further investigation should seek to answer the following questions: (1) what are the causes of AT immune cell infiltration and activation, (2) how does the adipocyte contribute to these changes and interact with AT immune cells, (3) is there a role of the gut microbiota in alteration of AT inflammation and (4) how do AT immune cells change during weight-loss. Through the elucidation of the answers to these key questions, immunologic therapies, potentially targeting the adipocyte, for the treatment of obesity and its inflammatory complications can be developed.

AUTHOR CONTRIBUTIONS
AB, AJ, and WH were involved in all aspects of writing, editing, and decision to publish. All authors contributed to the article and approved the submitted version.