Overweight and obesity are the consequence of a chronic imbalance between energy intake and energy expenditure, culminating in the excess of fat accumulation in AT. Since 1980, the global incidence of overweight and obesity has risen to the extent that almost one-third of the world population is now considered being overweight or obese (Chooi et al., 2019).

Obesity is a heterogeneous condition deriving from genetic and lifestyle interactions (Albuquerque et al., 2015; Hopkins and Blundell, 2016; MacLean et al., 2017; Schwartz et al., 2017) and is correlated with several pathological dysfunctions with important repercussions for individual and community health (Kyrou et al., 2018). Lifestyle and behavioral interventions (e.g., increased physical activity and decreased caloric intake) are fundamental parts for weight control (Butryn et al., 2011; Greenway, 2015). A gradual weight loss up to 16% of the original body weight is sufficient to improve β-cell function and insulin sensitivity in AT, liver, and skeletal muscle (Magkos et al., 2016). The improved glycemic control after weight loss is due, in part, to the dysregulated expression of genes involved in cholesterol flux, lipid synthesis, ECM remodeling, and oxidative stress (Magkos et al., 2016). In the light of the foregoing, obesity is the most frequent metabolic disorder in the world and the primary risk factor for IR and diabetes mellitus (Boles et al., 2017).

Diabetes mellitus refers to a group of conditions in which the body cannot use and store glucose correctly (American Diabetes Association, 2018). The proportion of individuals affected by diabetes mellitus has risen dramatically over the previous three decades, making it one of the major causes of death in the world. More than 300 million people are expected to develop T2D as a complication of obesity by 2025 (Ncd Risk Factor Collaboration [Ncd-RisC], 2016).

T2D is the most prevalent form of diabetes mellitus (American Diabetes Association, 2018), a chronic disease characterized by increased plasma glucose levels due to insulin secretion deficiencies (i.e., β-cell dysfunction) and IR (i.e., decreased target tissue capacity to react regularly to insulin) (American Diabetes Association, 2018).

Obesity is a risk factor for IR and a complete understanding of the mechanisms linking obesity to IR will enhance our knowledge of T2D pathogenesis and the capacity to manage obesity-related disorders (Choi and Cohen, 2017; Maksymets et al., 2018). For this purpose, several studies have been conducted on human and transgenic animal models demonstrating a correlative and causative association between dietary excess and activation of the innate and adaptive immune system in organs that control systemic energy homeostasis (Lumeng et al., 2007b; Lumeng and Saltiel, 2011; Lackey and Olefsky, 2016).

The initial mechanistic evidence supporting the inflammatory origin of obesity and diabetes comes from human and animal studies carried out in the early 1990s. In these studies, AT from obese rodents and humans show inflammatory modifications and enhanced secretion of pro-inflammatory cytokine TNF-α able to induce IR by inactivating the IRS-1 (Hotamisligil et al., 1993, 1995; Uysal et al., 1998). The pivotal role of TNF-α is significantly supported by evidence establishing that TNF-α neutralization in obese mice improves insulin sensitivity and glucose metabolism (Hotamisligil et al., 1993).

Low-grade chronic AT inflammation (also noted as meta-inflammation) is strongly and consistently associated with excess body fat mass and is characterized by infiltration and activation of pro-inflammatory macrophages and other immune cells that produce and secrete pro-inflammatory cytokines and chemokines (Chawla et al., 2011; Burhans et al., 2018).

Macrophages change not only their number during obesity (i.e., up to 40% of all AT cells in this context) but also their location and inflammatory phenotype (Weisberg et al., 2003). While in normal weight subjects the macrophages show anti-inflammatory properties, the polarization of AT macrophages (ATMs) in obese AT shifts to a pro-inflammatory phenotype (Lumeng et al., 2007a; Castoldi et al., 2016; Boulenouar et al., 2017). In obesity, macrophages surround dead adipocytes (i.e., forming crown-like structures) and secrete an array of pro-inflammatory cytokines that lead to local and systemic inflammation and IR (Lumeng et al., 2007a; Haase et al., 2014).

The inflammatory triggers are still almost unknown; however, obesity-induced AT remodeling provides a plethora of intrinsic signals (e.g., adipocyte death, hypoxia, and mechanical stress) capable of initiating an inflammatory response (Reilly and Saltiel, 2017). The role of inflammation in T2D pathogenesis and associated metabolic complications has led to a growing interest in targeting inflammatory mediators or pathways to prevent and treat T2D (Shoelson et al., 2006; McLaughlin et al., 2016).

In this review, we address the primary role played by the loss of immune regulation in the AT inflammation and the development of obesity-associated disorders, providing details on molecular aspects. We highlight the cellular and molecular triggers for obesity-induced inflammation and finally give some insights into the new anti-inflammatory therapeutic strategies.

Insulin is an anabolic hormone secreted by β-cells that plays a crucial role not only in carbohydrate metabolism but also in protein and lipid anabolic regulation, cell growth, and proliferation (Fu et al., 2013). Blood glucose concentrations stimulate insulin synthesis and release; its effects on whole-body metabolism result from its binding to the cell membrane receptor, which is activated by autophosphorylation of specific tyrosine residues. The activated insulin receptor phosphorylates and recruits intracellular proteins, also known as IRSs. Downstream of IRS proteins, PI3K mediates insulin functions mainly by activating PKB and protein kinase C cascades (i.e., stimulating glucose uptake, glycogen synthesis and inhibiting hepatic gluconeogenesis). Insulin signaling also exerts mitogenic effects, most of which are mediated by PKB cascade and Ras/mitogen activated protein kinase (MAPK) pathway activation (Ramalingam et al., 2013).

Obesity association with T2D has long been recognized, and the primary reason is the ability of obesity to promote IR, the main pathophysiological aspect of T2D (Kahn and Flier, 2000).

IR is a metabolic complication in which the three major insulin-sensitive tissues (skeletal muscle, liver, and AT) become less responsive to insulin action. IR is characterized by serious failures in glucose uptake, glycogen synthesis, and, to a lesser extent, glucose oxidation (Ormazabal et al., 2018). In this scenario, the β-cells compensate for IR by increasing insulin secretion and restoring blood glucose concentration within the normal range. A further decline in insulin sensitivity makes the β-cells exhausted, and this results in persistent hyperglycemia and T2D (Shulman, 2000).

A number of studies have been performed to identify the causal factors responsible for obesity-induced IR. One of the most accepted theories considers chronic systemic inflammation induced by obesity as a preponderant mechanism (Weisberg et al., 2003; Xu et al., 2003; Luft et al., 2013). This theory is strongly supported by many findings and clinical evidence; for instance, inflammatory markers such as CRP, TNF-α, and interleukin 6 (IL-6) are elevated in obese and insulin-resistant subjects (Dandona et al., 1998; Kern et al., 2001; Vozarova et al., 2001; Phosat et al., 2017; Uemura et al., 2017). The first evidence of an association between IR and inflammation has been hypothesized when, following the administration of anti-inflammatory agents, an improvement in glucose homeostasis has been observed in T2D patients (Williamson, 1901; Reid et al., 1957; Yuan et al., 2001; Hundal et al., 2002). Further studies in the mid-1990s have shown that the white AT (WAT) of obese rodents and humans exhibited changes in the levels of pro-inflammatory molecules (e.g., TNF-α) (Hotamisligil et al., 1993, 1995; Uysal et al., 1998). Such inflammatory mediators modulate IR either directly by affecting insulin signaling or indirectly by stimulating inflammatory pathways (Tilg and Moschen, 2008). Other studies have shown that hypoxia, which occurs in AT during obesity, is directly responsible for IR induction in both human and murine models (Regazzetti et al., 2009; Yin et al., 2009).

Animal and human studies have identified WAT as the primary site where obesity-related chronic inflammation is initiated and exacerbated (Weisberg et al., 2003; Xu et al., 2003). AT remodeling during obesity provides a plethora of intrinsic and extrinsic signals capable of triggering an inflammatory response (Chawla et al., 2011; Huh et al., 2014; Reilly and Saltiel, 2017). These triggers, discussed later in the review, converge on the activation of the JNK and NF-κB signaling pathways (Nakatani et al., 2004; Shoelson et al., 2006; Bluher et al., 2009; Lee and Lee, 2014). The activation of these signaling pathways increases the production of pro-inflammatory cytokines, endothelial adhesion molecules, and chemotactic mediators that promote the infiltration of monocytes in AT and the differentiation into pro-inflammatory M1 macrophages (Shoelson et al., 2006). Infiltrating macrophages produce and secrete many inflammatory mediators that promote local and systemic pro-inflammatory status and impair insulin signaling (Haase et al., 2014).

The effects of these cytokines are mediated by stimulation of IκB kinase β (IKKβ) and JNK1, expressed in myeloid and insulin-targeted cells (McLaughlin et al., 2016).

JNK is one of the most investigated signal transducers in obesity models of IR. It is activated after exposure to many inflammatory stimuli including cytokines, free fatty acids, and activation of cellular pathways, such as UPR (Aguirre et al., 2000; Ozcan et al., 2004). Once activated, JNK starts a pro-inflammatory gene transcription and inhibits insulin signaling pathway through inhibitory serine–threonine phosphorylation of IRS-1, thereby decreasing PI3K/PKB signaling (Tanti et al., 1994; Gual et al., 2005). In obese mice (ob/ob and diet-induced obesity), JNK activity is increased in AT compared to control mice. The role of Jnk1 in adipocytes has been investigated using tissue-specific Jnk1-deficient mice. These mice are protected against the development of IR when fed a HFD. This effect is tissue specific because Jnk1 deficiency in adipocytes does not affect muscle insulin sensitivity (Hirosumi et al., 2002; Sabio et al., 2008).

Obesity is also associated with the activation of NF-κB inflammatory pathway. In physiological conditions, NF-κB proteins are retained in the cytoplasm of myeloid and insulin-targeted cells by a family of inhibitors called inhibitors of κB (IκBs) (McLaughlin et al., 2017). Activation of IKK kinase complex (that contains IKKα and IKKβ subunits) induces proteasomal degradation of IκBα, leading to NF-κB nuclear translocation. This culminates in the increased expression of several NF-κB target genes [e.g., IL-6, TNFα, interferon-γ (IFN-γ), transforming growth factor-β (TGF-β), monocyte chemotactic protein-1 (MCP-1), and interleukin-1β (IL-1β)], which further exacerbate IR progression (Shoelson et al., 2006; Panahi et al., 2018). IKKβ deficiency in adipocytes totally prevents the expression of IL-6 and TNF-α induced by free fatty acid, while its activation inhibits the expression of anti-inflammatory cytokines such as adiponectin and leptin (Jiao et al., 2011). Therapeutic approaches capable of targeting these pathways and improving insulin sensitivity in obese subjects will be further discussed below in this review.

Macrophages represent another important cell type in mediating the obesity-induced inflammation in the AT. During obesity, macrophages infiltrate the AT and secrete many pro-inflammatory cytokines (Weisberg et al., 2003). These mediators have local effects on adipocytes and resident immune cells (e.g., neutrophils, B cells, and T cells) and circulate in the periphery, where they affect the liver and skeletal muscle insulin sensitivity (Weisberg et al., 2003; Xu et al., 2003; Haase et al., 2014; McLaughlin et al., 2017).

Myeloid cells activate another molecular pathway, called inflammasome, in obesity (Lee and Lee, 2014). Macrophages and other innate immune cells may trigger inflammatory responses by detecting pathogen- or danger-associated molecular patterns (PAMPs or DAMPs) using a broad variety of pattern-recognition receptors such as TLRs and NLRs (Pedra et al., 2009; Vandanmagsar et al., 2011). Compelling evidence shows that NLRP3 (the most studied member of the NLR family) activation by DAMPs (generated by nutrient excess in obesity) plays a key role behind the chronic inflammation characteristic of obesity and IR (Stienstra et al., 2010, 2011, 2012; Zhou et al., 2010; Koenen et al., 2011; Vandanmagsar et al., 2011; Lee et al., 2013).

NLRP3 is present in several tissues and cell types (Vandanmagsar et al., 2011). It is unclear which cell compartments in AT express the inflammasome components; however, immunostaining of AT sections of obese mice confirmed a strong co-localization of NLRP3 with macrophage marker F4/80 in crown-like structures (Vandanmagsar et al., 2011). Once activated, NLRP3 interacts with procaspase-1 through an adaptive protein forming the NLRP3 inflammasome (Schroder et al., 2010; Davis et al., 2011; Shao et al., 2015). This results in the processing and activation of caspase-1, which mediates the maturation and secretion of IL-1β and IL-18 by macrophages (Shoelson et al., 2003; Schroder et al., 2010; Davis et al., 2011; Shao et al., 2015). The primary role played by NLRP3 inflammasome is also supported by evidence that genetic ablation of NLRP3^–/– prevents the obesity-induced inflammasome activation in AT and protects against HFD-induced IR (Vandanmagsar et al., 2011). Caloric and exercise-mediated weight loss in obese people with T2D reduces NLPR3 and IL-1β gene expression in abdominal subcutaneous AT and improves systemic insulin sensitivity (Vandanmagsar et al., 2011).

Inflammasome-activated IL-1β is a major cytokine produced by macrophages (Sims and Smith, 2010). Its enhanced production in pancreatic islets and insulin-sensitive tissues is associated to T2D (Hotamisligil et al., 1993; Donath and Shoelson, 2011). In obesity, chronic rise in circulating nutrients such as glucose and free fatty acids (FFAs) resulted in over-expression of IL-1β in pancreatic β-cells (Maedler et al., 2002; Böni-Schnetzler et al., 2008; Fei et al., 2008; Böni-Schnetzler et al., 2009). It is now clear that IL-1β is a key cytokine in the etiology of T2D since it has been implicated in IR, β-cell dysfunction, and death (Eizirik and Mandrup-Poulsen, 2001; Donath et al., 2008, 2019). IL-1β alters the insulin sensitivity of AT by suppressing insulin signaling; exposure to IL-1β of murine and human adipocytes decreases insulin-stimulated glucose uptake and lipogenesis (Lagathu et al., 2006; Jager et al., 2007; Fève and Bastard, 2009), reduces glucose transporter type 4 (GLUT4) expression, and inhibits GLUT4 translocation to the plasma membrane (Jager et al., 2007; Ballak et al., 2015).

The pro-apoptotic effects of IL-1β on β-cells derive from a complex network of signaling events triggered by IL-1β binding to its cognate receptor, whose expression is higher in β-cells than in other tissues (Böni-Schnetzler et al., 2009). Once cytokine binds its receptor, the co-receptor is recruited, and this results in the formation of the heterodimer receptor transmembrane complex. Both receptors and co-receptors share a cytoplasmic motif, the Toll/IL-1 receptor (IL-1R) domain, which is required to initiate intracellular signaling by recruitment of different adaptor proteins and kinases, including the myeloid primary response differentiation-88 protein and the interleukin-1-associated kinase receptor. This leads to activation of MAPK and NF-κB signaling pathways. The activation of these two signaling pathways causes variations in gene expression, therefore triggering the apoptotic cell death program in β-cells (Donath et al., 2008, 2019). The pro-apoptotic effects mediated by NF-kB depend on the cell type, nature, and duration of the stimulus. Indeed, NF-kB activation in β-cells is more marked, rapid, and sustained than in other cell types (Ortis et al., 2006). MAPKs also take part in β-cell apoptosis through transcription-independent mechanisms, such as regulating B-cell lymphoma 2 protein activity (Donath et al., 2008, 2019). The combined use of IFN-γ and IL-1β induces the activation of an additional mechanism, the so-called non-canonical NF-kB pathway, also implicated in the pro-apoptotic effects of IL-1β on β-cells (Meyerovich et al., 2016).

IL-1β is also implicated in cardiovascular and microvascular long-term complications (nephropathy, retinopathy, and polyneuropathy) of diabetes (Herder et al., 2013, 2015; Agrawal and Kant, 2014; Stahel et al., 2016; Donath et al., 2019). Endothelial cell damage is a crucial and an early manifestation of diabetic-associated vascular complications (van den Oever et al., 2010; Gilbert, 2013; Liu et al., 2014). Among the multiple and potential mechanisms that contribute to this phenomenon, a crucial role is played by chronic low-grade inflammation. IL-1β has been reported to cause endothelial cell damage in isolated mesenteric rat micro-vessels (Vila and Salaices, 2005; Shashkin et al., 2006). Vallejo et al. (2014) have shown that the deleterious effects of IL-1β on endothelial cells are due to the IL-1R-mediated activation of NADPH oxidase, which stimulates the production of superoxide anion (Vallejo et al., 2014). Over-activation of NADPH oxidase has also been associated with excess ROS production and the development of atherosclerosis in diabetic vasculopathy (Olukman et al., 2010; Gray et al., 2013).

IL-18 is another pro-inflammatory mediator activated by inflammasome and produced and released by human AT (Wood et al., 2005). IL-18 plasma levels are increased in obese people and in individuals with T2D (Moriwaki et al., 2003; Evans et al., 2007) while being restored in subjects who have lost weight following bariatric surgery (Schernthaner et al., 2006). It is a powerful pro-inflammatory cytokine that increases the maturation of T and NKs, as well as the production of other pro-inflammatory cytokines, exacerbating the obesity-induced systemic inflammation (Weisberg et al., 2003).

Likewise, IL-6 has been suggested to be involved in the development of obesity-related and T2D-related IR (Fève and Bastard, 2009). IL-6 leads to impaired insulin signaling, and this occurs primarily by inhibition of insulin-stimulated tyrosine phosphorylation of IRSs both in the liver and in AT (Senn et al., 2002; Klover et al., 2003; Lagathu et al., 2003; Rotter-Sopasakis et al., 2004; Fève and Bastard, 2009). Nonetheless, conflicting results have been reported for IL-6 action on skeletal muscle (Fève and Bastard, 2009). Carey et al. (2006) have shown that IL-6 increases GLUT4 translocation on plasma membrane and promotes insulin-stimulated glucose uptake in myotubes (Carey et al., 2006). Nevertheless, it has also been shown that in murine skeletal muscle cells, IL-6 is capable of reducing insulin-stimulated glucose uptake through JNK activation (Nieto-Vazquez et al., 2008). In pancreas, IL-6 impairs insulin secretion and has pro-apoptotic effects on β-cells (Ellingsgaard et al., 2008). An opposite effect is carried out on α-cells; IL-6 prevents α-cells apoptosis and induces the secretion of glucagon-like peptide-1. This could be considered an adaptive mechanism to compensate for β-cell failure (Ellingsgaard et al., 2008, 2011; Akbari and Hassan-Zadeh, 2018). Such findings support the tissue-specific effect of IL-6 on glucose homeostasis, which depends on several factors, such as concentrations, targets, and signaling pathways activated.

The IL-6 signaling cascade involves activation of the Janus kinase (JAK)-signal transducer and activator of transcription (STAT) pathway (Fève and Bastard, 2009; Dodington et al., 2018). It serves as a crucial downstream mediator for a variety of hormones, cytokines (Gadina et al., 2018), and growth factors, including growth hormone, leptin, IL-6, and IFN-γ (Dodington et al., 2018). There are four identified members in the JAK kinase family (JAKs 1-3 and Tyk2), which associate with cytokine and growth factor receptors. JAK-mediated signaling leads to the activation of seven STAT family members (STATs 1-4, 5A, 5B, and 6). STAT proteins have cell- and tissue-specific distribution that influences their specificity and function (Schindler and Darnell, 1995; Richard and Stephens, 2011, 2014). The regulation of tissue-specific genes and the ability to have cell-specific tasks appear to be important physiological roles of the JAK/STAT pathway (Richard and Stephens, 2011). JAK/STAT signaling in the peripheral metabolic organs modulates a multitude of metabolic processes, including adiposity, energy expenditure, glucose tolerance, and insulin sensitivity (Dodington et al., 2018). This signaling pathway mediates the action of several hormones that have profound effects on adipocyte development and function. Adipocytes also produce hormones that utilize this pathway (Richard and Stephens, 2011). The expression of several STATs is modulated during adipogenesis (Richard and Stephens, 2011). Additional functions of JAK/STAT signaling in adipocytes include the transcriptional regulation of genes involved in insulin action and lipid and glucose metabolism (Richard and Stephens, 2011). JAK2, STAT3, and STAT5 are essential for signaling through both the growth hormone and leptin receptors and have been characterized in WAT (Dodington et al., 2018). As the major upstream kinases required for STAT activity, it is not surprising that JAK proteins also play important roles in the control of AT function (Gurzov et al., 2016). Adipose-specific Jak2 KO mice have demonstrated defective lipolysis, increased body weight and adiposity compared to controls, leading to IR (Nordstrom et al., 2013; Shi et al., 2014; Corbit et al., 2017). Similarly, loss of either Stat3 or Stat5 in AT contributes to increased weight gain, adiposity, and impaired lipolysis (Dodington et al., 2018). There is a controversy over the effects of adipocyte JAK2/STAT5 on insulin sensitivity. Some studies have shown IR (Shi et al., 2014) while others have demonstrated enhanced whole-body insulin sensitivity in the absence of JAK2 or STAT5 (Nordstrom et al., 2013; Corbit et al., 2017). This inconsistency might be due to a variety of factors including tissue specificity and cell stage-dependent expression of the cre transgene, mouse genetic background, physiologic status, and other environmental factors in which the experiments were performed (Dodington et al., 2018). Although the direct role of STAT1 in the anti-adipogenic action of IFN-γ was not investigated, experiments using pharmacological inhibitors show that the JAK-STAT1 pathway plays a key role in the ability of IFN-γ to induce IR, decline triglyceride stores, and down-regulate expression of lipogenic genes in mature human adipocytes (Richard and Stephens, 2014). The increased IFN-γ levels and JAK-STAT1 signaling in obesity contribute to AT dysfunction and IR (Gurzov et al., 2016).

Emerging evidence demonstrates that the highly conserved and potent JAK/STAT signaling pathway is dysregulated in metabolic diseases, including obesity and T2D (Gurzov et al., 2016; Dodington et al., 2018). Studies show that many STAT activators play an important role in the regulation of adipocyte gene expression and exhibit differential expression in the condition of obesity and/or IR (Richard and Stephens, 2014). Obesity increases levels of IL-6 in WAT that, in turn, chronically activate intracellular JAK-STAT3 signaling. Chronic JAK-STAT3 signaling induced by IL-6 leads to the increased expression of suppressor of cytokine signaling-3 that not only negatively regulates IL-6 signaling but also hinders insulin action, eventually resulting in obesity and IR (Wunderlich et al., 2013). JAK/STAT signaling can have both physiological and pathological roles depending on the context. It is difficult to speculate how JAK/STAT inhibition will affect individuals with obesity and diabetes (Dodington et al., 2018). This complexity highlights the need for validation of the relative contribution of STAT proteins in human samples. Further studies will also be required to reveal the complex roles of the JAK-STAT pathway in adipocytes, obesity, and IR. Manipulation of this pathway within AT is a novel therapeutic approach for the treatment of obesity and diabetes.

Systemic inflammation is characterized by high circulating levels of inflammatory mediators and immune cells that infiltrate insulin-dependent tissues (Weisberg et al., 2003). As has already been discussed in the review, WAT is the main site where low-grade systemic inflammation begins (Weisberg et al., 2003; Xu et al., 2003). Accumulation of lipids that occurs in AT during obesity triggers an inflammatory response that results in an increased secretion of several inflammatory cytokines (Haase et al., 2014; Raciti et al., 2017). Such molecules can also activate JNK and NF-κB signaling pathways in the liver and skeletal muscle, thus inhibiting systemic insulin signaling (Hotamisligil et al., 1993; Ciccarelli et al., 2016).

Obesity-induced inflammation initiates in WAT and then spreads to other tissues, resulting in low-grade systemic inflammation. In obesity, both liver and skeletal muscle exhibit signs of local inflammation (Figure 1).

Skeletal muscle is the principal organ for insulin-stimulated glucose uptake (i.e., capable for 80% of glucose disposal in human), and muscle IR plays a key part in T2D etiology (DeFronzo and Tripathy, 2009; Honka et al., 2018). Obesity contributes to the development of chronic muscle inflammation, characterized by increased pro-inflammatory M1 macrophage infiltration (Fink et al., 2013, 2014). These macrophages secrete many cytokines, which have been shown to trigger inflammatory pathways within myocyte, culminating in decreased insulin signaling (Varma et al., 2009; Pillon et al., 2012; Patsouris et al., 2014).

In the liver, obesity leads to increased infiltration and pro-inflammatory activation of two major macrophage groups: Kupffer cells (i.e., resident specialized hepatic macrophages) and monocyte-derived recruited hepatic macrophages (Tencerova et al., 2015). Further work indicates that, during obesity, the number of Kupffer cells remains unaffected, whereas the accumulation of monocyte-derived recruited macrophages increases several times (Morinaga et al., 2015). It has been demonstrated that ATM-released inflammatory mediators lead to hepatic IR by reducing insulin signaling (Morinaga et al., 2015). These inflammatory mediators also contribute to liver steatosis by promoting lipogenesis and toxic ceramide biosynthesis (Schubert et al., 2000; Obstfeld et al., 2010).

Recently, as specified above, particular attention has been paid to the role played by macrophages in AT inflammation. Nevertheless, many other immune cells (both innate and adaptive immune systems) are involved in the development of local and systemic inflammation and IR.

There is a limited understanding of how obesity-induced inflammation in AT is triggered. However, potential mechanisms identified include dysregulation of fatty acids homeostasis, increased adipose cell size and death, local hypoxia, mitochondrial dysfunction, ER stress, and mechanical stress (Figure 2) (Heilbronn and Liu, 2014; Reilly and Saltiel, 2017). These mechanisms are recognized as the link between chronic caloric excess and AT inflammation or as factors that may perpetuate chronic tissue inflammation (Burhans et al., 2018). The list of potential mechanical links mentioned here is not complete, and it is likely that the triggers leading to AT inflammation have not yet been identified.

The role of chronic inflammation, particularly in the AT, in the pathogenesis of T2D and associated complications, is now well established. The association between obesity, AT inflammation, and metabolic disease makes inflammatory pathways an appealing target to treat metabolic disorders. Inflammation is recognized as the pathologic mediator of these frequently common comorbidities. Several anti-inflammatory approaches have been tested in clinical studies of obese individuals with IR, but clinical trials to confirm the therapeutic potential are still ongoing (Goldfine and Shoelson, 2017). The number of available drugs that can target different components of the immune system and improve different metabolic aspects is increasing rapidly (Donath, 2014).

Based on the mechanism of action, therapeutic approaches to target inflammation in IR and T2D can be divided into (i) pharmacologic approaches that directly target inflammation and (ii) diabetes drugs with anti-inflammatory properties.

The global obesity epidemic results in a higher incidence of metabolic disorders. The mechanisms underlying the association between obesity and IR have not yet been fully explained. Therefore, further well-designed clinical and basic research studies are needed to establish this relationship. From our point of view, inflammation occurring in the AT during obesity is the primary mechanism for developing local and systemic IR. AT is the primary whole-body regulator of lipid and glucose homeostasis and is no longer considered merely a storage tissue.

Obesity leads to severe adipocyte disorders by altering the amount and activity of almost all resident immune cells. The imbalance of immunological phenotypes is correlated with the development of persistent local inflammation during which several biologically active molecules are released. These molecules affect distal tissues and organs, such as skeletal muscle and liver.

The inflammatory nature of obesity opens new prospects in the development of therapeutic strategies for the treatment of its related metabolic complications. However, there are still a lot of issues that need to be addressed.

Anti-inflammatory strategies have proven to be effective in improving obesity-induced IR in murine models. However, clinical studies are still ongoing to confirm the therapeutic potential in obese and insulin-resistant individuals. Another issue is the modest effects of anti-inflammatory therapies observed in these studies. Targeting only one inflammatory molecule may not be sufficient to have a beneficial effect; therefore, we could hypothesize the combined use of more anti-inflammatory therapies. In addition, a recent study showed that acute and transient inflammation is essential for healthy AT expansion and remodeling in obesity (Asterholm et al., 2014). This finding raises further questions on the effectiveness of anti-inflammatory therapies in the treatment of obesity-induced metabolic disorders. Inflammation is a finely regulated mechanism, and all defects in its balance can cause AT dysfunction.

In the era of personalized and precision medicine, increasing our knowledge of the obesity-induced inflammation mechanisms might enable us to overcome the limitations of the traditional anthropometric indices of obesity. These anthropometric indices are not correlated with obesity-induced metabolic complications and additional clinical parameters need to be identified for risk assessment (Longo et al., 2019). From our point of view, given the strong association between inflammation and obesity complications, circulating inflammatory biomarkers may be used for the risk assessment of these diseases in the future. The identification and evaluation of these biomarkers in obese patients will allow the prediction of those who will develop obesity-associated metabolic complications.

FB and CM conceived the idea and edited the manuscript. FZ, ML, JN, GR, and AD wrote the manuscript. FZ and ML prepared the figures. All authors reviewed the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.