In the past two decades, numerous experimental and clinical studies have established the importance of inflammation and immunity in the development of obesity and its metabolic complications, including insulin resistance and type 2 diabetes mellitus. In this context, T cells orchestrate inflammatory processes in metabolic organs, such as the adipose tissue (AT) and liver, thereby mediating obesity-related metabolic deterioration. Costimulatory molecules, which are present on antigen-presenting cells and naïve T cells in the AT, are known to mediate the crosstalk between the adaptive and innate immune system and to direct T-cell responses in inflammation. In this Perspectives in Diabetes article, we highlight the newest insights in immune cell interactions in obesity and discuss the role of costimulatory dyads in its pathogenesis. Moreover, the potential of therapeutic strategies that target costimulatory molecules in the metabolic syndrome is explored.

Over the past few decades, obesity-associated morbidity and mortality have reached endemic proportions, affecting >1 billion individuals worldwide. Obesity is a major risk factor for insulin resistance (IR), the metabolic syndrome, type 2 diabetes mellitus (T2DM), and cardiovascular diseases (CVDs) (1). Yet, the pathogenesis of obesity and its complications remain incompletely understood, and effective therapeutic strategies against obesity-associated metabolic complications are still lacking.

In 1876, Dr. Wilhelm Ebstein made the remarkable finding that the nonsteroidal anti-inflammatory drug sodium salicylate improved glycosuria in T2DM patients, suggesting a role of inflammation in the pathogenesis of obesity-associated metabolic disorders (2). However, it took until the 1990s to establish a clear link among obesity, T2DM, and inflammation. In 1993, Hotamisligil et al. (3) established that tumor necrosis factor-α (TNF-α) levels were increased in obese adipose tissue (AT) and that this directly induced IR. Today, numerous clinical and experimental studies have identified systemic, low-grade inflammation of the AT as a critical process underlying the development of obesity and its associated disorders (4,5).

AT is a complex endocrine tissue that contains multiple cell types, including adipocytes and adipocyte precursors, vascular cells, immune cells, and neuronal cells, which all contribute to the inflammatory response during obesity. Although the order of events that contribute to AT dysfunction and systemic inflammation in the course of obesity is incompletely understood, several key processes have been identified. Nutritional excess promotes adipocyte expansion, resulting in adipocyte dysfunction. Adipocytes subsequently secrete adipokines, cytokines, and chemokines, such as leptin, resistin, TNF-α, interleukin (IL)-6, and MCP-1, which induce the accumulation of immune cells in the AT, and the ongoing inflammation causes IR (5,6).

In the course of obesity, almost the entire spectrum of immune cells becomes apparent within the AT (5). Macrophages are abundantly present in AT. In obesity, the number of AT macrophages correlates with the extent of IR, most likely because of the secretion of TNF-α and IL-6, which directly interfere with insulin signaling (7). Resident macrophages in lean AT have an anti-inflammatory M2-like phenotype characterized by the surface expression of CD206 and macrophage galactose-type C-type lectin 1 (MGL-1), produce anti-inflammatory mediators such as IL-10, and play a critical role in the maintenance of AT insulin sensitivity (8). During obesity, the majority of macrophages recruited have or acquire a proinflammatory M1 profile characterized by the expression of CD11c, inducible nitric oxide synthase, TNF-α, and IL-6 and reside in crown-like structures that surround necrotic adipocytes (8). Depletion of these CD11c+ macrophages reduces AT inflammation and restores insulin sensitivity (9). Besides the classical M1 and M2 macrophages, the obese AT contains a mixed macrophage population, which expresses both CD11c and CD206 and has a proinflammatory phenotype that promotes AT fibrosis and IR (10).

Neutrophils are recruited to the AT within 1 week after the start of a high-fat diet (HFD), albeit in low numbers. Genetic deficiency and pharmacologic inhibition of neutrophil elastase improve glucose tolerance and insulin sensitivity by the reduced neutrophil elastase–mediated degradation of insulin receptor substrate-1 (IRS-1) and ameliorate AT inflammation due to decreased Toll-like receptor 4–dependent expression of proinflammatory mediators in AT macrophages (11). Of note, eosinophils reduce AT inflammation by promoting IL-4– and IL-13–dependent M2 polarization of AT macrophages (12). Meanwhile, mast cells secreting IL-6 and interferon γ (IFN-γ) appear to play a proinflammatory role in metabolic disease (13).

T cells constitute ∼10% of the stromal vascular fraction of lean AT, with CD4+ T cells outnumbering CD8+ T cells. Approximately 50% of these CD4+ cells are anti-inflammatory regulatory T cells (Tregs), whereas T helper (Th) 1 CD4+ T cells and Th2 CD4+ T cells are present in equal numbers. During the development of diet-induced obesity (DIO), the number of AT T cells increases as does the CD8+-to-CD4+ T-cell ratio, whereas the percentage of Tregs decreases dramatically (1416). This change in T-cell subsets is mediated by the expression of Stat3 (17). CD8+ cells seem to precede macrophage infiltration and promote the recruitment of AT macrophages by secreting MCP-1, MCP-3, and RANTES (regulated on activation, normal T cell expressed and secreted) (18). In later stages of obesity, both CD4+ and CD8+ T cells are crucial in the recruitment and M1 polarization of macrophages through IFN-γ (14,18).

Genetic and antibody-mediated depletion of CD8+ T cells limits AT inflammation and ameliorates IR (18). Besides CD8+ cells, CD4+ cells also have been shown to contribute to obesity-related metabolic dysfunction. Within the obese AT, CD4+ Th1 cells are found in abundance, and depletion of these IFN-γ–secreting Th1 cells ameliorates AT inflammation by reducing T-cell and macrophage influx and IR (14). The CD4+ Treg population also plays a major role in preventing obesity-associated IR. The visceral AT contains relatively high numbers of Tregs, and by depleting them, AT inflammation and IR occur. Of note, the visceral AT Tregs are of a special phenotype, with peroxisome proliferator–activated receptor γ as the major orchestrator of Treg accumulation, phenotype, and function (19). The other T-cell subset proven to promote IR is Th17, which is increased 3- to 10-fold in obese subjects and which reduces glucose uptake in skeletal muscle (20).

Natural killer T (NKT) cells are a special subset of T cells and have characteristics of both adaptive and innate immune cells. Type 1 or invariant NKT cells express a T-cell receptor (TCR) with an invariant α-chain (Vα14Jα18), whereas type 2 or variant NKT cells express a more diverse TCR repertoire. NKT cells react to lipid antigens presented on CD1d, which results in the secretion of various cytokines (IL-4, IL-10, IFN-γ, and TNF-α) that may elicit Th1, Th2, and Treg responses. Experiments with mice with loss- and gain-of-function in NKT activity revealed a gamut of outcomes ranging from beneficial (2125), to null (26), to detrimental (2729) effects of NKT cells on metabolism. These divergent effects may result from the various strategies applied (CD1d−/− that affects all NKT cells vs. Jα18−/− that affects only type 1 NKT cells), various diets, various durations, or other local, yet not identified factors. However, these data indicate that NKT cells are tightly controlled during the course of obesity.

B cells are recruited to obese AT, and increased B-cell activation is observed in obese subjects. Experimental studies have demonstrated that B cells from obese mice secrete more proinflammatory (IFN-γ, IL-6, and IL-8) and less anti-inflammatory (IL-5 and IL-10) cytokines (30). Additionally, during obesity, B cells are directly or indirectly activated through lipid-induced Toll-like receptor signaling or through T-cell–dependent mechanisms, respectively, to produce IgG2c (auto)antibodies that promote AT inflammation and IR. In accordance, CD20-mediated depletion of B cells ameliorates metabolic parameters in obese mice (30).

Thus, accumulating evidence suggests that both the adaptive and the innate immune systems are active in the early and later phases of obesity. Hence, it is likely that complex immunological mechanisms with delicate interactions between the innate and adaptive immune systems are involved.

T cells are present in the AT of lean and obese subjects, and AT T-cell content positively correlates with waist circumference in subjects with T2DM. T cells infiltrate the AT in the early phase of DIO through chemokines, such as SDF1, which attract them to the AT (31). To execute their effector functions, newly recruited T cells need to be activated. CD8+ T cells are activated through the interaction with antigen-loaded MHC class I (MHCI) molecules, which are expressed on all nucleated cells. In contrast, CD4+ T-cell activation requires MHC class II (MHCII)-dependent antigen presentation on antigen-presenting cells (APCs) as well as costimulatory molecules to strengthen the interaction between APCs and T cells, thereby preventing T-cell anergy and warranting a proper immune response (32).

At least four cell types may act as APCs to activate CD4+ T cells in the AT. Dendritic cells (DCs) are professional APCs present in the AT (33). The number of DCs positively correlates with BMI in men (33). Although AT DCs induce Th17 differentiation of naïve T cells, detailed knowledge of the underlying mechanisms is lacking. AT macrophages also process and present antigen on MHCII molecules and express costimulatory molecules (34,35). Macrophage–T-cell interactions result in the polarization of naïve T cells toward IFN-γ–producing Th1 cells (34,35). Furthermore, AT B cells may act as APCs that induce MHCII–dependent T-cell activation as well as T-cell–dependent antigen production, which is further promoted by the expression of multiple costimulatory molecules on B cells (36). Additionally, adipocytes may function as APCs during obesity because DIO increases the expression of MHCII and costimulatory molecules on the adipocyte, which induces a Th1 response (35). However, whether the adipocyte as APC activates the T cell or whether the activated T cell that has encountered a more classical APC activates the adipocyte and how exactly CD8+ T cells are activated (through MHCI) still needs to be elucidated.

Of note, AT-mediated T-cell activation is observed in the early stages of DIO before the infiltration of monocytes/macrophages, which suggests that MHCII-expressing adipocytes may trigger CD4+ cell activation in early obesity, whereas macrophages regulate T-cell activation in the later stages of the disease (31,35). The source of the antigens in the AT is unknown; however, several mechanisms may promote the development of novel antigens, including palmitoylation and oxidation of AT proteins. Furthermore, stress-induced protease activity and endoplasmic reticulum stress-induced protein misfolding may produce antigenic peptides in the AT (35). The importance of MHCII-mediated antigen presentation in CD4+ T-cell activation is emphasized by the observations in MHCII-deficient mice, which show impaired Th1 differentiation and reduced AT inflammation and metabolic deterioration (35).

After binding of the TCR to the MHCII molecule, costimulatory molecules provide additional stimuli that license the T cell and APC to initiate an immune response (Fig. 1A) (32). Besides the classical role in costimulation, costimulatory molecules are expressed on a variety of other immune cells, such as neutrophils and mast cells, and nonimmune cells, including platelets, endothelial cells, smooth muscle cells, adipocytes, hepatocytes, and pancreatic cells (32,37). Interactions between costimulatory molecules result in the activation of these cells and promote inflammation (Fig. 1B) (32,3739). Hence, costimulatory interactions are likely to mediate a rather broad crosstalk between innate and adaptive immunity during the pathogenesis of obesity and its related complications. Multiple costimulatory dyads of the B7 family and the tumor necrosis factor superfamilies are expressed on cell types involved in obesity-associated inflammation, and some have been identified as critical contributors to the pathogenesis of obesity (Table 1) (32,4051).

CD80 (B7.1) and CD86 (B7.2) are expressed by DCs, macrophages, B cells, and T cells. Both proteins interact with the costimulatory receptor CD28, which is constitutively expressed on CD4+ and CD8+ T cells (32,40). CD80/86-CD28 interactions promote activation, priming, and proliferation of naïve T cells by inducing the production of growth factors (IL-2) and antiapoptotic proteins (BCL-X) (32,40). Binding of CD80/86 to the coinhibitory factor cytotoxic T-lymphocyte–associated protein 4 (CTLA-4) on T cells dampens CD28-mediated T-cell activation (32,40). Although less well characterized, it has been reported that CD80 and CD86 also bind to other costimulatory receptors, including inducible costimulator (ICOS) and programmed cell death 1 (PD-1) (32,40).

Although the mRNA levels of CD80/86 are upregulated in the AT and the AT-derived stromal vascular fraction of obese mice (52), the expression of CD80/86 on AT macrophages is reduced in obese subjects compared with lean individuals and negatively correlates with HOMA-IR (53). We and others recently showed that obese CD80−/−86−/− mice exhibit increased obesity-related metabolic deterioration with increased AT macrophage infiltration (52,53). Analysis of AT T cells revealed a decrease in CD4+ T cells, especially in the CD4+CD25+FoxP3+ Tregs in CD80−/−86−/− mice (52,53). We also described that a similar reduction in the Treg population was taking place in the liver of CD80−/−86−/− mice, thereby exacerbating the development of nonalcoholic steatohepatitis under HFD conditions (52). However, CD80−/−86−/− mice have an inborn deficiency in the development of Tregs, which explains the unexpected aggravation of DIO in these mice (54). We therefore believed it of interest to investigate whether antibody-mediated blockage of CD80/86 in a DIO mouse that contains Tregs would improve IR and AT inflammation and hepatosteatosis because it has been shown that CD80/86-mediated activation of effector T cells results in their proliferation and induces the secretion of proinflammatory cytokines, including TNF-α and IL-6 (32,40). Indeed, antibody inhibition of both CD80 and CD86 costimulatory molecules but not of each of them alone protected mice from the development of obesity-related anomalies, pathologies, and especially nonalcoholic steatohepatitis and IR (52).

CTLA-4-immunoglobulin–mediated inhibition of CD80/86 in DIO mice reduced AT inflammation and IR. This could be partly due to the decrease in body weight, liver weight, and AT weight. However, the AT showed decreased levels of TNF-α and IL-6, MCP-1, and MCP-3, and increased numbers of AT M2 macrophages (55), which, besides the decrease in weight, is likely responsible for the reduced IR observed after CTLA-4 treatment. These data indicate that the CD80/86-CD28/CTLA-4 dyad may play a dual role in the development of DIO by inducing a protective Treg response and eliciting proinflammatory functions of effector T cells.

The CD40-CD40L dyad is a prominent member of the tumor necrosis factor receptor (TNFR) family (32). The costimulatory molecule CD40 and its ligand CD40L (CD154) are expressed by immune cells, including B cells, T cells, DCs and monocytes, and nonimmune cells, including platelets, endothelial cells, adipocytes, fibroblasts, hepatocytes, smooth muscle cells, pancreatic islet β-cells, and pancreatic ductal cells (32,38,39,56). CD40-CD40L interactions induce T-cell and APC activation, cytokine production, and B-cell isotype class switching as well as endothelial cell activation and monocyte migration (32,38,39,56).

The expression of CD40 on AT cells, including adipocytes, stromal AT cells, and immune cells, positively correlates with BMI (57). Although the expression of membrane-bound CD40L is restricted to immune cells and stromal cells within the AT, soluble CD40L (sCD40L), mainly derived from activated platelets and T cells, may also activate CD40+ adipocytes (38,39,57). Clinical studies have demonstrated that sCD40L plasma levels are increased in obese patients and positively correlate with BMI, waist circumference, fasting glucose, and leukocyte counts and that sCD40L levels decrease after bariatric surgery (58). CD40 ligation on adipocytes results in activation of classical proinflammatory signal transduction pathways, including extracellular signal–related kinase, p38, Jun NH2-terminal kinase, mitogen-activated protein kinase, and nuclear factor-κB (NF-κB), which results in the expression of cytokines and chemokines, including TNF, IL-6, and MCP-1, as well as the prothrombotic mediator plasminogen activator inhibitor 1 (57,59,60). These proinflammatory mediators subsequently activate endothelial cells and immune cells, which promote a generalized proinflammatory AT status. Additionally, CD40L stimulation directly affects lipid metabolism in adipocytes because it enhances lipid droplet accumulation in 3T3-L1 cells and promotes adipogenesis by inducing C-EPBα and peroxisome proliferator–activated receptor γ expression, both master regulators of adipogenesis. Moreover, CD40L directly targets glucose metabolism by reducing the expression of IRS-1 and GLUT-4, which decreases insulin-mediated glucose uptake in adipocytes and promotes AT IR (57).

Thus, in vitro studies have identified a proinflammatory role of CD40-CD40L interactions in AT inflammation and metabolic deregulation. Of note, in vivo studies revealed a more complex dual function for the CD40-CD40L dyad. CD40L−/− mice subjected to DIO were protected against weight gain, AT inflammation, and hepatosteatosis (45). The AT tissue of obese CD40L−/− mice contained lower numbers of T cells and macrophages, whereas Treg and M2 macrophage numbers were increased (45). Furthermore, MCP-1, IL-6, IFN-γ, TNF-α, and IL-12 expression was reduced in the AT of these mice, whereas adiponectin was increased (45,61). These results were mimicked in wild-type DIO mice treated with antagonistic CD40L antibodies that had similar weight gain, suggesting that the phenotype of the CD40L−/− mouse is independent of weight gain (45).

In contrast to CD40L−/− mice, increased IR and hepatosteatosis notably characterize CD40−/− mice (46,62,63). CD40 deficiency in male mice on an HFD resulted in impaired insulin sensitivity despite no difference in diet-induced body weight gain. IR related to CD40 deficiency was associated with increased hepatosteatosis and enhanced AT inflammation. In particular, AT of CD40-deficient mice revealed elevated CD8+ T cells in the obese AT of CD40−/− mice accompanied by an M1-biased inflammatory response, as indicated by higher numbers of M1 macrophages and increased expression of TNF, IL-6, and IL-12 (46). To understand these contradictory results, we recently explored the involvement of various signaling pathways induced upon CD40 activation. After binding of CD40L, CD40 needs to recruit adaptor proteins, the TNFR-associated factors (TRAFs), to elicit intracellular signaling (39). CD40 contains a distal binding site for TRAF2, 3, and 5 and a proximal binding site for TRAF6. When obesity was induced in mice with a deficiency in the CD40-TRAF2/3/5 interaction in MHCII+ cells, a similar phenotype as in obese CD40−/− mice was observed (46). CD40-TRAF2/3/5−/− mice exhibited an aggravation of AT inflammation, IR, and hepatosteatosis (46). Of note, deficiency of CD40-TRAF6 interactions in MHCII+ cells instead protected mice against these obesity-induced aberrations (46). To exploit this potential therapeutic target, we developed a small molecule inhibitor of the CD40-TRAF6 interaction, which reduced AT inflammation, IR, and hepatosteatosis in DIO mice (46). Thus, the contradictory results on the role of CD40L and CD40 may be explained by the differential involvement of the TRAF proteins; although CD40−/− mice mimicked CD40-TRAF2/3/5–deficient mice, deficiency of CD40-TRAF6 signaling protected mice against the complications of DIO. Additionally, novel ligands for CD40 may explain the opposing phenotype of CD40L−/− and CD40−/− mice subjected to DIO (64). CD40-CD40L interactions promote the expression of proinflammatory mediators in other tissues besides the AT, including the vasculature and pancreas, which results in a state of continuous low-grade systemic inflammation that promotes the development of obesity-associated complications, including atherosclerosis and IR/T2DM (Fig. 2) (38).

In conclusion, CD40L and CD40 have opposing roles in the development of obesity, which is explained by the differential involvement of downstream signaling proteins. Small molecule–mediated inhibition of the CD40-TRAF6 interaction is a promising therapeutic strategy for obesity-related metabolic complications.

The constitutive expression of the TNFR family member CD137 (4-1BB, TNFRSF9) is low, but increases upon TCR-mediated activation of CD8+ T cells and, to a lesser extent of CD4+ T cells (32,65). Additionally, CD137 is expressed on DCs, NK cells, granulocytes, eosinophils, mast cells, and possibly inflammatory monocytes (32). CD137 binds to CD137L on professional APC. The expression of the CD137-CD137L dyad has also been described on nonimmune cells, including endothelial cells, smooth muscle cells, fibroblasts, and cardiomyocytes (65). CD137-CD137L interactions augment T-cell proliferation, cytotoxic T-cell activities, and the secretion of IFN-γ, TNF-α, IL-2, and IL-4 by T cells and APCs (65).

Obesity is associated with an increased expression of CD137 and soluble CD137 in the AT of men and mice (66). CD137-CD137L–mediated interactions between adipocytes and macrophages induce cytokine expression (e.g., TNF-α, IL-6, MCP-1) and promote monocyte and T-cell recruitment to the AT (48). Accordingly, CD137−/− mice exhibited reduced body weight gain and adiposity when subjected to DIO. Moreover, AT inflammation and hepatosteatosis were decreased, whereas glucose tolerance was improved (49). Counterintuitively, antibody-mediated stimulation of CD137 also reduced body weight, adiposity, and IR (67). The AT of antibody-treated mice contained more CD4+ and CD8+ T cells, whereas the number of macrophages was decreased. Furthermore, hepatic immune cells were increased in these mice, as were IL-6 and MCP-1 levels; however, hepatosteatosis was decreased. Of note, CD137 stimulation increased glucose and lipid metabolism and increased energy expenditure, which was attributed to increased expansion and activation of CD8+ T cells. Indeed, the size of secondary lymphoid organs (e.g., spleen, lymph nodes) was increased in antibody-treated mice. However, these mice also exhibited increased locomotor activity (67).

Together these studies suggest that inhibition of the CD137-CD137L improves DIO by inhibiting the underlying inflammatory processes, whereas stimulation of CD137 also ameliorates some metabolic and inflammatory parameters of DIO by increasing energy expenditure due to massive immune cell expansion. Because antibody-mediated stimulation of CD137 increases circulating leukocyte counts and results in lymphadenopathy and splenomegaly, this strategy is not feasible for the (long-term) treatment of obesity. Therefore, future studies should focus on therapeutic strategies that inhibit the CD137-CD137L dyad in obesity.

LIGHT (lymphotoxin-like inducible protein that competes with glycoprotein D for herpesvirus entry on T cells), a member of the TNF family, is expressed on activated T cells, monocytes, granulocytes, and immature DCs. LIGHT binds to HVEM (herpes simplex virus glycoprotein D for herpesvirus entry mediator, TNFRSF14), which is expressed on activated and resting immune cells, adipocytes, and endothelial cells. LIGHT-HVEM interactions induce a strong activation of NF-κB that results in T-cell activation and expansion and the secretion of proinflammatory mediators such as adhesion molecules, chemokines, and matrix metalloproteinases (17,37).

Soluble LIGHT (sLIGHT) concentrations are increased in morbidly obese and T2DM patients and correlate with BMI, triglycerides, fat mass, and glycated hemoglobin (HbA1c) (68). The expression of LIGHT and HVEM on T cells, monocytes, and adipocytes is increased in DIO mice. sLIGHT in AT is produced by activated T cells, monocytes, granulocytes, and immature DCs but not by adipocytes. Ligation of HVEM on adipocytes results in the NF-κB–mediated secretion of inflammatory mediators that promote macrophage and T-cell recruitment (67,69). Moreover, LIGHT-expressing T cells inhibited lipase expression in hepatocytes, thereby increasing plasma lipoprotein levels, which promote the development of obesity-associated complications (70). HVEM-deficient mice on an HFD were protected against metabolic deterioration. Moreover, AT inflammation was reduced as a result of decreased cytokine secretion by T cells. Energy expenditure was increased in these mice, possibly as a result of increased thermogenesis, as reflected by the increased expression of UCP1 (51). The phenotype of HVEM-deficient mice was mimicked by the use of HVEM blocking antibodies, emphasizing the great therapeutic potential of the LIGHT-HVEM dyad in obesity.

Experimental studies have demonstrated that both genetic deficiency and antibody-mediated inhibition of costimulatory molecules either improve or deteriorate obesity-associated AT inflammation and metabolic complications. Because systemic modulation of costimulation may result in severe immune suppression or activation, this strategy has not yet been applied for long-term treatment of chronic inflammatory diseases such as obesity and CVD (37,39). However, clinical studies that evaluated the efficacy of short-term antibody treatment against costimulatory molecules in other diseases have shown a great potential for these agents.

Administration of antagonistic CD40L antibodies, such as ruplizumab, ABI793, and IDEC131, was well tolerated in patients and improved disease severity in systemic lupus erythematosus, Crohn’s disease, and renal allograft rejection (71). However, clinical trials that evaluated the efficacy of anti-CD40L antibodies were stopped because of the occurrence of thromboembolic events resulting from disrupted CD40L-αIIbβ3 interactions in arterial thrombi (72). An alternative approach, blockage of CD40, was well tolerated and able to slightly improve disease severity in Crohn’s disease (73). The agonistic CD40 antibodies dacetuzumab and lucatumumab are also safe for patients and are currently being tested for treating malignancies such as chronic lymphatic leukemia, multiple myeloma, and non-Hodgkin’s lymphoma (74). Targeting the CD80/86-CD28/CTLA-4 pathway has also been proven successful in various diseases. Abatacept, a CTLA-4-Ig, has been proven efficient in several diseases, especially in rheumatoid arthritis (75), whereas belatacept, a CD80/86 inhibitor, is a very promising immunosuppressant after kidney transplantation (76). As previously mentioned, we recently showed that obese mice injected with a combination of anti-CD80 and anti-CD86 were less insulin tolerant and displayed ameliorated hepatosteatosis and reduced inflammation in the liver and AT, which make CD80/86 inhibitors a promising therapy for the metabolic syndrome (52).

Also as previously mentioned, clinical application of these strategies for long-term treatment of obesity may be complicated by immunosuppressive side effects, such as infectious and neoplastic complications. For example, long-term abatacept-mediated inhibition of CD80/86 to prevent allograft rejection is associated with increased occurrence of Epstein-Barr viral infection and lymphoproliferative disorders (77). To develop safe therapies that preserve the role of costimulatory molecules in immunity while selectively targeting their deleterious effects in obesity, future research should focus on the selective targeting of the signal transduction cascades that mediate the proinflammatory effects of costimulatory molecules during obesity. We recently developed such a compound that specifically inhibits CD40-TRAF6 interactions while leaving CD40-TRAF2/3/5 interactions intact, and were able to show that this compound could reduce AT inflammation and IR in a model of DIO (46). This example illustrates that targeting parts of the signal transduction cascade of costimulatory molecules has great potential for the treatment of chronic inflammatory diseases.

Over the years, the function of the individual immune cells within the course of the metabolic syndrome has become clearer. However, the role of individual immune cells in the course of obesity is still not representative for the immunometabolism of an organism. The main challenge of the next years is to unravel how the immune cells interact and mediate immune reactions that take place within the AT. First steps into the right direction have been accomplished. T-cell-APC interactions have been shown to play a major role in AT inflammation and IR, and the important APCs within the AT have been identified. One of the major challenges is the identification of the antigens involved. The other accomplishment is the discovery of an important but complex role of the costimulatory molecules, major regulators of immune cell interactions, in the metabolic syndrome.

Several studies have demonstrated an important function of costimulatory dyads CD80/86, CD40/CD40L, and CD137 in the development of obesity. The metabolic effects of costimulatory interactions as well as the underlying molecular mechanisms still warrant further investigation. The possible involvement of other costimulatory dyads, such as ICOS-ICOSL, PD-1-PD-L1/2, OX40-OX40L, CD27/CD70, and glucocorticoid-induced TNFR-related protein (GITR)-GITRL in the development of obesity and its complications has not yet been investigated and requires further attention.

Most studies described in this review used genetic-modified mice or antibody-mediated blockage to investigate the role of a single costimulatory dyad. Although these approaches are useful in experimental settings, reality is far more complex because the multiple costimulatory and coinhibitory dyads are not only dynamically expressed in time, but also regulate the expression of other costimulatory proteins. Elucidation of the complex interplay and downstream signaling cascades of the various costimulatory dyads may result in the development of more effective therapeutic strategies. Moreover, most of these studies looked at complete blockage of costimulatory dyads, whereas these are often expressed on a plethora of cells. Therefore, tissue-specific knockouts are required in future studies to evaluate the effects of costimulatory molecules on various cells.

Antibody-mediated inhibition of costimulatory interactions is clinically applied for severe autoimmune and inflammatory diseases. However, long-term antibody-mediated inhibition of costimulatory dyads may not be feasible for the treatment of obesity because it may result in severe side effects caused by suppression of the immune system. Therefore, future studies should focus on the cell type–specific effects and downstream signaling cascades that mediate the proinflammatory effects of costimulatory molecules because this may result in the development of therapeutic strategies that inhibit DIO-associated inflammation but preserve immunity.

See accompanying article, Diabetes 2014;63:2751–2760.

Funding. This work was supported by the Netherlands CardioVascular Research Initiative—the Dutch Heart Foundation, Dutch Federation of University Medical Centres, The Netherlands Organisation for Health Research and Development, and the Royal Netherlands Academy of Sciences—for the GENIUS (Generating the Best Evidence-Based Pharmaceutical Targets for Atherosclerosis) project (CVON2011-19); the Dutch Heart Foundation (Dr. E. Dekker MD grant to T.S. and Dr. E. Dekker Established Investigator grant to E.L.); the Deutsche Forschungsgemeinschaft (CH279/5-1 to T.C.); the European Research Council (to T.C.); and The Netherlands Organisation for Scientific Research (Vici grant to E.L.).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. T.S. and P.K. contributed to writing the manuscript. A.C. and T.C. contributed to reviewing and editing the manuscript. E.L. contributed to writing, reviewing, and editing the manuscript. E.L. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

1.
Osborn
O
,
Olefsky
JM
.
The cellular and signaling networks linking the immune system and metabolism in disease
.
Nat Med
2012
;
18
:
363
374
[PubMed]
2.
Ebstein W. Zur therapie des diabetes mellitus, insbesondere über die anwendung des salicylsauren natron bei demsellben. Berliner Klinische Wochenschrift 1876;13:337–340 [in German]
3.
Hotamisligil
GS
,
Shargill
NS
,
Spiegelman
BM
.
Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance
.
Science
1993
;
259
:
87
91
[PubMed]
4.
Gregor
MF
,
Hotamisligil
GS
.
Inflammatory mechanisms in obesity
.
Annu Rev Immunol
2011
;
29
:
415
445
[PubMed]
5.
Mathis
D
.
Immunological goings-on in visceral adipose tissue
.
Cell Metab
2013
;
17
:
851
859
[PubMed]
6.
Ouchi
N
,
Parker
JL
,
Lugus
JJ
,
Walsh
K
.
Adipokines in inflammation and metabolic disease
.
Nat Rev Immunol
2011
;
11
:
85
97
[PubMed]
7.
Weisberg
SP
,
McCann
D
,
Desai
M
,
Rosenbaum
M
,
Leibel
RL
,
Ferrante
AW
 Jr
.
Obesity is associated with macrophage accumulation in adipose tissue
.
J Clin Invest
2003
;
112
:
1796
1808
[PubMed]
8.
Lumeng
CN
,
Bodzin
JL
,
Saltiel
AR
.
Obesity induces a phenotypic switch in adipose tissue macrophage polarization
.
J Clin Invest
2007
;
117
:
175
184
[PubMed]
9.
Patsouris
D
,
Li
P-P
,
Thapar
D
,
Chapman
J
,
Olefsky
JM
,
Neels
JG
.
Ablation of CD11c-positive cells normalizes insulin sensitivity in obese insulin resistant animals
.
Cell Metab
2008
;
8
:
301
309
[PubMed]
10.
Phieler
J
,
Chung
K-J
,
Chatzigeorgiou
A
, et al
.
The complement anaphylatoxin C5a receptor contributes to obese adipose tissue inflammation and insulin resistance
.
J Immunol
2013
;
191
:
4367
4374
[PubMed]
11.
Talukdar
S
,
Oh
Y
,
Bandyopadhyay
G
, et al
.
Neutrophils mediate insulin resistance in mice fed a high-fat diet through secreted elastase
.
Nat Med
2012
;
18
:
1407
1412
[PubMed]
12.
Wu
D
,
Molofsky
AB
,
Liang
H-E
, et al
.
Eosinophils sustain adipose alternatively activated macrophages associated with glucose homeostasis
.
Science
2011
;
332
:
243
247
[PubMed]
13.
Liu
J
,
Divoux
A
,
Sun
J
, et al
.
Genetic deficiency and pharmacological stabilization of mast cells reduce diet-induced obesity and diabetes in mice
.
Nat Med
2009
;
15
:
940
945
[PubMed]
14.
Winer
S
,
Chan
Y
,
Paltser
G
, et al
.
Normalization of obesity-associated insulin resistance through immunotherapy
.
Nat Med
2009
;
15
:
921
929
[PubMed]
15.
Feuerer
M
,
Herrero
L
,
Cipolletta
D
, et al
.
Lean, but not obese, fat is enriched for a unique population of regulatory T cells that affect metabolic parameters
.
Nat Med
2009
;
15
:
930
939
[PubMed]
16.
Chatzigeorgiou
A
,
Karalis
KP
,
Bornstein
SR
,
Chavakis
T
.
Lymphocytes in obesity-related adipose tissue inflammation
.
Diabetologia
2012
;
55
:
2583
2592
[PubMed]
17.
Priceman
SJ
,
Kujawski
M
,
Shen
S
, et al
.
Regulation of adipose tissue T cell subsets by Stat3 is crucial for diet-induced obesity and insulin resistance
.
Proc Natl Acad Sci U S A
2013
;
110
:
13079
13084
[PubMed]
18.
Nishimura
S
,
Manabe
I
,
Nagasaki
M
, et al
.
CD8+ effector T cells contribute to macrophage recruitment and adipose tissue inflammation in obesity
.
Nat Med
2009
;
15
:
914
920
[PubMed]
19.
Cipolletta
D
,
Feuerer
M
,
Li
A
, et al
.
PPAR-γ is a major driver of the accumulation and phenotype of adipose tissue Treg cells
.
Nature
2012
;
486
:
549
553
[PubMed]
20.
Fabbrini E, Cella M, McCartney SA, et al. Association between specific adipose tissue CD4+ T-cell populations and insulin resistance in obese individuals. Gastroenterology 2013;145:366–374
21.
Kotas
ME
,
Lee
H-Y
,
Gillum
MP
, et al
.
Impact of CD1d deficiency on metabolism
.
PLoS One
2011
;
6
:
e25478
[PubMed]
22.
Ji
Y
,
Sun
S
,
Xu
A
, et al
.
Activation of natural killer T cells promotes M2 macrophage polarization in adipose tissue and improves systemic glucose tolerance via interleukin-4 (IL-4)/STAT6 protein signaling axis in obesity
.
J Biol Chem
2012
;
287
:
13561
13571
[PubMed]
23.
Schipper
HS
,
Rakhshandehroo
M
,
van de Graaf
SFJ
, et al
.
Natural killer T cells in adipose tissue prevent insulin resistance
.
J Clin Invest
2012
;
122
:
3343
3354
[PubMed]
24.
Strodthoff
D
,
Lundberg
AM
,
Agardh
HE
, et al
.
Lack of invariant natural killer T cells affects lipid metabolism in adipose tissue of diet-induced obese mice
.
Arterioscler Thromb Vasc Biol
2013
;
33
:
1189
1196
[PubMed]
25.
Martin-Murphy
BV
,
You
Q
,
Wang
H
, et al
.
Mice lacking natural killer T cells are more susceptible to metabolic alterations following high fat diet feeding
.
PLoS One
2014
;
9
:
e80949
[PubMed]
26.
Mantell
BS
,
Stefanovic-Racic
M
,
Yang
X
,
Dedousis
N
,
Sipula
IJ
,
O’Doherty
RM
.
Mice lacking NKT cells but with a complete complement of CD8+ T-cells are not protected against the metabolic abnormalities of diet-induced obesity
.
PLoS One
2011
;
6
:
e19831
[PubMed]
27.
Ohmura
K
,
Ishimori
N
,
Ohmura
Y
, et al
.
Natural killer T cells are involved in adipose tissues inflammation and glucose intolerance in diet-induced obese mice
.
Arterioscler Thromb Vasc Biol
2010
;
30
:
193
199
[PubMed]
28.
Satoh
M
,
Andoh
Y
,
Clingan
CS
, et al
.
Type II NKT cells stimulate diet-induced obesity by mediating adipose tissue inflammation, steatohepatitis and insulin resistance
.
PLoS One
2012
;
7
:
e30568
[PubMed]
29.
Wu
L
,
Parekh
VV
,
Gabriel
CL
, et al
.
Activation of invariant natural killer T cells by lipid excess promotes tissue inflammation, insulin resistance, and hepatic steatosis in obese mice
.
Proc Natl Acad Sci U S A
2012
;
109
:
E1143
E1152
[PubMed]
30.
Winer
DA
,
Winer
S
,
Shen
L
, et al
.
B cells promote insulin resistance through modulation of T cells and production of pathogenic IgG antibodies
.
Nat Med
2011
;
17
:
610
617
[PubMed]
31.
Kintscher
U
,
Hartge
M
,
Hess
K
, et al
.
T-lymphocyte infiltration in visceral adipose tissue: a primary event in adipose tissue inflammation and the development of obesity-mediated insulin resistance
.
Arterioscler Thromb Vasc Biol
2008
;
28
:
1304
1310
[PubMed]
32.
Chen
L
,
Flies
DB
.
Molecular mechanisms of T cell co-stimulation and co-inhibition
.
Nat Rev Immunol
2013
;
13
:
227
242
[PubMed]
33.
Bertola
A
,
Ciucci
T
,
Rousseau
D
, et al
.
Identification of adipose tissue dendritic cells correlated with obesity-associated insulin-resistance and inducing Th17 responses in mice and patients
.
Diabetes
2012
;
61
:
2238
2247
[PubMed]
34.
Morris
DL
,
Cho
KW
,
Delproposto
JL
, et al
.
Adipose tissue macrophages function as antigen-presenting cells and regulate adipose tissue CD4+ T cells in mice
.
Diabetes
2013
;
62
:
2762
2772
[PubMed]
35.
Deng
T
,
Lyon
CJ
,
Minze
LJ
, et al
.
Class II major histocompatibility complex plays an essential role in obesity-induced adipose inflammation
.
Cell Metab
2013
;
17
:
411
422
[PubMed]
36.
Winer
DA
,
Winer
S
,
Chng
MHY
,
Shen
L
,
Engleman
EG
.
B Lymphocytes in obesity-related adipose tissue inflammation and insulin resistance
.
Cell Mol Life Sci
2014
;
71
:
1033
1043
[PubMed]
37.
Gerdes
N
,
Zirlik
A
.
Co-stimulatory molecules in and beyond co-stimulation - tipping the balance in atherosclerosis
?
Thromb Haemost
2011
;
106
:
804
813
[PubMed]
38.
Seijkens
T
,
Kusters
P
,
Engel
D
,
Lutgens
E
.
CD40-CD40L: linking pancreatic, adipose tissue and vascular inflammation in type 2 diabetes and its complications
.
Diab Vasc Dis Res
2013
;
10
:
115
122
[PubMed]
39.
Engel
D
,
Seijkens
T
,
Poggi
M
, et al
.
The immunobiology of CD154-CD40-TRAF interactions in atherosclerosis
.
Semin Immunol
2009
;
21
:
308
312
[PubMed]
40.
Rudd
CE
,
Taylor
A
,
Schneider
H
.
CD28 and CTLA-4 coreceptor expression and signal transduction
.
Immunol Rev
2009
;
229
:
12
26
[PubMed]
41.
Dong
C
,
Juedes
AE
,
Temann
UA
, et al
.
ICOS co-stimulatory receptor is essential for T-cell activation and function
.
Nature
2001
;
409
:
97
101
[PubMed]
42.
Shilling
RA
,
Bandukwala
HS
,
Sperling
AI
.
Regulation of T:B cell interactions by the inducible costimulator molecule: does ICOS “induce” disease
?
Clin Immunol
2006
;
121
:
13
18
[PubMed]
43.
Okazaki
T
,
Wang
J
.
PD-1/PD-L pathway and autoimmunity
.
Autoimmunity
2005
;
38
:
353
357
[PubMed]
44.
Agata
Y
,
Kawasaki
A
,
Nishimura
H
, et al
.
Expression of the PD-1 antigen on the surface of stimulated mouse T and B lymphocytes
.
Int Immunol
1996
;
8
:
765
772
[PubMed]
45.
Poggi
M
,
Engel
D
,
Christ
A
, et al
.
CD40L deficiency ameliorates adipose tissue inflammation and metabolic manifestations of obesity in mice
.
Arterioscler Thromb Vasc Biol
2011
;
31
:
2251
2260
[PubMed]
46.
Chatzigeorgiou
A
,
Seijkens
T
,
Zarzycka
B
, et al
.
Blocking CD40-TRAF6 signaling is a therapeutic target in obesity-associated insulin resistance [published correction appears in Proc Natl Acad Sci U S A 2014;111:4644]
.
Proc Natl Acad Sci U S A
2014
;
111
:
2686
2691
[PubMed]
47.
Ishii
N
,
Takahashi
T
,
Soroosh
P
,
Sugamura
K
.
OX40-OX40 ligand interaction in T-cell-mediated immunity and immunopathology
.
Adv Immunol
2010
;
105
:
63
98
[PubMed]
48.
Tu TH, Kim CS, Goto T, Kawada T, Kim B-S, Yu R. 4-1BB/4-1BBL interaction promotes obesity-induced adipose inflammation by triggering bidirectional inflammatory signaling in adipocytes/macrophages. Mediators Inflamm 2012;2012:972629
49.
Kim
C-S
,
Kim
JG
,
Lee
B-J
, et al
.
Deficiency for costimulatory receptor 4-1BB protects against obesity-induced inflammation and metabolic disorders
.
Diabetes
2011
;
60
:
3159
3168
[PubMed]
50.
Denoeud
J
,
Moser
M
.
Role of CD27/CD70 pathway of activation in immunity and tolerance
.
J Leukoc Biol
2011
;
89
:
195
203
[PubMed]
51.
Kim
H-J
,
Kim
H-M
,
Kim
C-S
, et al
.
HVEM-deficient mice fed a high-fat diet are protected from adipose tissue inflammation and glucose intolerance
.
FEBS Lett
2011
;
585
:
2285
2290
[PubMed]
52.
Chatzigeorgiou
A
,
Chung
KJ
,
Garcia-Martin
R
, et al
.
Dual role of B7 costimulation in obesity-related non-alcoholic steatohepatitis and metabolic dysregulation
.
Hepatology
2014
;
60
:
1196
1210
[PubMed]
53.
Zhong
J
,
Rao
X
,
Braunstein
Z
, et al
.
T-cell costimulation protects obesity-induced adipose inflammation and insulin resistance
.
Diabetes
2014
;
63
:
1289
1302
[PubMed]
54.
Zeng
M
,
Guinet
E
,
Nouri-Shirazi
M
.
B7-1 and B7-2 differentially control peripheral homeostasis of CD4(+)CD25(+)Foxp3(+) regulatory T cells
.
Transpl Immunol
2009
;
20
:
171
179
[PubMed]
55.
Fujii
M
,
Inoguchi
T
,
Batchuluun
B
, et al
.
CTLA-4Ig immunotherapy of obesity-induced insulin resistance by manipulation of macrophage polarization in adipose tissues
.
Biochem Biophys Res Commun
2013
;
438
:
103
109
[PubMed]
56.
Smeets
E
,
Meiler
S
,
Lutgens
E
.
Lymphocytic tumor necrosis factor receptor superfamily co-stimulatory molecules in the pathogenesis of atherosclerosis
.
Curr Opin Lipidol
2013
;
24
:
518
524
[PubMed]
57.
Poggi
M
,
Jager
J
,
Paulmyer-Lacroix
O
, et al
.
The inflammatory receptor CD40 is expressed on human adipocytes: contribution to crosstalk between lymphocytes and adipocytes
.
Diabetologia
2009
;
52
:
1152
1163
[PubMed]
58.
Baena-Fustegueras
JA
,
Pardina
E
,
Balada
E
, et al
.
Soluble CD40 ligand in morbidly obese patients: effect of body mass index on recovery to normal levels after gastric bypass surgery
.
JAMA Surg
2013
;
148
:
151
156
[PubMed]
59.
Klein
D
,
Timoneri
F
,
Ichii
H
,
Ricordi
C
,
Pastori
RL
.
CD40 activation in human pancreatic islets and ductal cells
.
Diabetologia
2008
;
51
:
1853
1861
[PubMed]
60.
Missiou
A
,
Wolf
D
,
Platzer
I
, et al
.
CD40L induces inflammation and adipogenesis in adipose cells—a potential link between metabolic and cardiovascular disease
.
Thromb Haemost
2010
;
103
:
788
796
[PubMed]
61.
Wolf
D
,
Jehle
F
,
Ortiz Rodriguez
A
, et al
.
CD40L deficiency attenuates diet-induced adipose tissue inflammation by impairing immune cell accumulation and production of pathogenic IgG-antibodies
[published correction appears in PLoS One 2012;7: doi:10.1371/annotation/84102aae-770a-41ae-9c15-4736ed129c17].
PLoS One
2012
;
7
:
e33026
[PubMed]
62.
Wolf D, Jehle F, Michel NA, et al. Coinhibitory suppression of T cell activation by CD40 protects from obesity and adipose tissue inflammation in mice. Circulation 2014;129:2414–2425
63.
Guo
C-A
,
Kogan
S
,
Amano
SU
, et al
.
CD40 deficiency in mice exacerbates obesity-induced adipose tissue inflammation, hepatic steatosis, and insulin resistance
.
Am J Physiol Endocrinol Metab
2013
;
304
:
E951
E963
[PubMed]
64.
Jabara
HH
,
Angelini
F
,
Brodeur
SR
,
Geha
RS
.
Ligation of CD46 to CD40 inhibits CD40 signaling in B cells
.
Int Immunol
2011
;
23
:
215
221
[PubMed]
65.
Vinay
DS
,
Kwon
BS
.
Role of 4-1BB in immune responses
.
Semin Immunol
1998
;
10
:
481
489
[PubMed]
66.
Tu TH, Kim CS, Kang JH, et al. Levels of 4-1BB transcripts and soluble 4-1BB protein are elevated in the adipose tissue of human obese subjects and are associated with inflammatory and metabolic parameters. Int J Obes (Lond) 2014;38:1075–1082
67.
Kim
C-S
,
Tu
TH
,
Kawada
T
,
Kim
B-S
,
Yu
R
.
The immune signaling molecule 4-1BB stimulation reduces adiposity, insulin resistance, and hepatosteatosis in obese mice
.
Endocrinology
2010
;
151
:
4725
4735
[PubMed]
68.
Bassols
J
,
Moreno-Navarrete
JM
,
Ortega
F
,
Ricart
W
,
Fernandez-Real
JM
.
LIGHT is associated with hypertriglyceridemia in obese subjects and increased cytokine secretion from cultured human adipocytes
.
Int J Obes (Lond)
2010
;
34
:
146
156
[PubMed]
69.
Kim
H-M
,
Jeong
C-S
,
Choi
H-S
,
Kawada
T
,
Yu
R
.
LIGHT/TNFSF14 enhances adipose tissue inflammatory responses through its interaction with HVEM
.
FEBS Lett
2011
;
585
:
579
584
[PubMed]
70.
Lo
JC
,
Wang
Y
,
Tumanov
AV
, et al
.
Lymphotoxin beta receptor-dependent control of lipid homeostasis
.
Science
2007
;
316
:
285
288
[PubMed]
71.
Seijkens
T
,
Engel
D
,
Tjwa
M
,
Lutgens
E
.
The role of CD154 in haematopoietic development
.
Thromb Haemost
2010
;
104
:
693
701
[PubMed]
72.
Kawai
T
,
Andrews
D
,
Colvin
R
,
Sachs
D
,
Cosimi
A
.
Thromboembolic complications after treatment with monoclonal antibody against CD40 ligand
.
Nat Med
2000
;
6
:
114
73.
Kasran
A
,
Boon
L
,
Wortel
CH
, et al
.
Safety and tolerability of antagonist anti-human CD40 Mab ch5D12 in patients with moderate to severe Crohn’s disease
.
Aliment Pharmacol Ther
2005
;
22
:
111
122
[PubMed]
74.
Hassan
SB
,
Sørensen
JF
,
Olsen
BN
,
Pedersen
AE
.
Anti-CD40-mediated cancer immunotherapy: an update of recent and ongoing clinical trials
.
Immunopharmacol Immunotoxicol
2014
;
36
:
96
104
[PubMed]
75.
Kremer
JM
,
Peterfy
C
,
Russell
AS
, et al
.
Longterm safety, efficacy, and inhibition of structural damage progression over 5 years of treatment with abatacept in patients with rheumatoid arthritis in the abatacept in inadequate responders to methotrexate trial
.
J Rheumatol
2014
;
41
:
1077
1087
[PubMed]
76.
Leibler
C
,
Matignon
M
,
Pilon
C
, et al
.
Kidney transplant recipients treated with belatacept exhibit increased naïve and transitional B cells
.
Am J Transplant
2014
;
14
:
1173
1182
[PubMed]
77.
Snanoudj
R
,
Zuber
J
,
Legendre
C
.
Co-stimulation blockade as a new strategy in kidney transplantation: benefits and limits
.
Drugs
2010
;
70
:
2121
2131
[PubMed]