Keywords

1 Introduction

The chronic low-grade activation of the innate immune system is evident even in childhood obesity . Thus, inflammatory markers are elevated in obese children as young as 3 years old. Furthermore, individuals are at risk for lifelong meta-inflammation (Skinner et al. 2010). Adipose tissue macrophage accumulation is directly proportional to adiposity in humans. Furthermore, adipocyte size is a strong predictor of the percentage of CD68-expressing macrophages in human subcutaneous adipose tissue . (Weisberg et al. 2003). The ratio of the macrophages is 5% in lean adipose tissue, whereas, during obesity this ratio rises up to 50% (Cinti et al. 2005). Although macrophages comprise 10–15% of stromal vascular cells (SVCs) in visceral adipose tissue of lean subjects, their numbers are increased to 40–50% of the SVCs of visceral adipose tissue in obese humans (Weisberg et al. 2003). During obesity immune cell population differs, not only in number, but also in inflammatory phenotypes . In this context, macrophage-elicited metabolic inflammation and adipocyte-macrophage interaction has the key importance in obesity (McNelis and Olefsky 2014).

2 Macrophage Infiltration into Adipose Tissue

Actually the initiation of macrophage infiltration into adipose tissue occurs with four different mechanisms (Sun et al. 2011). First of all, necrosis of adipocytes driven by hypertrophy is a prominent phagocytic stimulus that regulates adipose tissue macrophage infiltration which is gradually increased by obesity (Cinti et al. 2005). In the late stages of adipose tissue obesity, macrophages form crown-like structures surrounding necrotic adipocytes (Nishimura et al. 2007; Strissel et al. 2007). In this way, white adipose tissue macrophages are selectively localized to dead adipocytes. Clearance of free lipid appears to be an important function of galectin-3 (MAC-2)-expressing macrophages in white adipose tissue (Cinti et al. 2005). Galectin-3 is one of the major factors involved in the influx of macrophages to inflammatory sites (Sano et al. 2000). At first, galectin-3 expression at sites of adipocyte necrosis may be functionally linked to macrophage aggregation and crown-like structures formation (Cinti et al. 2005). Furthermore, galectin-3 possesses anti-apoptotic activity intracellularly and that is implicated in macrophage survival at sites of inflammation (Hsu and Liu 2004). Secondly, hypertrophic adipocyte-derived chemotactic monocyte chemoattractant protein-1 (MCP-1) /CC chemokine receptor 2 (CCR2) pathway also promotes more macrophage accumulation into the obese adipose tissue. Secretion of MCP-1 from adipocytes directly triggers the recruitment of macrophages to adipose tissue. The infiltrated macrophages may in turn secrete a variety of chemokines and other cytokines that further promote a local inflammatory response and affect gene expression in adipocytes, resulting in systemic insulin resistance (Kanda et al. 2006). Adiponectin expression may occur indirectly through interactions of adipocytes with CCR2-expressing adipose tissue macrophages and the alterations of obesity-induced inflammatory gene expression (Weisberg et al. 2006) (Fig. 14.1).

Fig. 14.1
figure 1

Macrophage-elicited metabolic inflammation and adipocyte-macrophage interaction has a primary importance in obesity . Large amounts of macrophages are accumulated by different mechanisms in obese adipose tissue (TLR4 Toll-like receptor 4, Akt Protein kinase-B, FOXO1 Forkhead box protein O1, NF-kappaB Nuclear factor-kappa B, JNK c-Jun N-terminal kinase, IKK Inhibitor kappa B kinase, PPAR-ɣ Peroxisome proliferator-activated receptor-gamma, IRS Insulin receptor substrate, MCP-1 Monocyte chemoattractant protein-1, MCSF Macrophage colony stimulating factor, CCL2: CCR2 Chemokine (C-C motif) receptor ligand 2, IRF5 Interferon regulatory factor-5, TNF-α Tumor necrosis factor-alpha, IL-1β Interleukin-1 beta, IL-6Interleukin-6, NLRP3, NLRP NOD-like receptor (NLR) family protein; PI3K Phosphatidylinositol 3-kinase, mtROS Mitochondrial reactive oxygen species, iNOS Inducible nitric oxide synthase, Mincle mRNA Macrophage-inducible C-type lectin mRNA, HIF-1 hypoxia-inducible factor-1, RAGE Receptor for advanced glycation end product, AGE Advanced glycation end product, TNFR Tumor necrosis factor receptor)

Full size image

The inhibitor of differentiation or inhibitor of DNA binding (Id) proteins are negative regulators of cell differentiation and play key roles in the regulation of cell fate decisions, and in the timing of differentiation (Norton 2000). There are several type of Id proteins. In particular, Id3 attenuates visceral fat expansion by inhibiting high fat diet -induced visceral fat vascular endothelial growth factor (VEGF), and regulated upon activation normal T-cell expressed and secreted (RANTES) expression. Eventually, increase in capillary density is restricted (Cutchins et al. 2012). The helix-loop-helix transcription regulator Id3 promotes high fat diet -induced adipocyte progenitor cell (AdPC) accumulation in visceral adipose tissue . Id3-dependent AdPC expansion is responsible for high fat diet-induced MCP-1 production, M1 macrophage accumulation, and metabolic dysfunction during obesity . Adipose depot-specific differences in inflammation may be due to depot-specific differences in AdPCs (Kaplan et al. 2015). Hence MCP-1 expression is up-regulated in white adipose tissue of high fat diet-induced obesity. The 7.2-fold increase in MCP-1 as compared with normal subjects may alter adipocyte function by decreasing insulin-stimulated glucose uptake and the expression of lipoprotein lipase , adipsin, glucose transporter type 4 (GLUT-4), fatty-acid-binding protein (aP2), beta3-adrenergic receptor, and peroxisome proliferator-activated receptor-gamma (PPAR-gamma) (Sartipy and Loskutoff 2003; Takahashi et al. 2003). Furthermore, adipocyte hypertrophy creates poorly oxygenated areas in the human obese adipose tissue. As the third mechanism, hypoxic cells secrete chemokines, which attract macrophages, presumably to clear out necrotic cells (Murdoch et al. 2004). Eventually, decreased adipose tissue partial oxygen concentration is paralleled by an increase in the expression and secretion of the chemokine and markers of macrophage infiltration (Pasarica et al. 2009). Hypoxia -induced fibrosis in adipose tissue is a key factor that ultimately stimulates the local inflammatory responses (Halberg et al. 2009). Moreover, increased local extracellular lipid concentrations, which drive adipose tissue macrophage accumulation occurs as a final mechanism for macrophage infiltration. Thus the majority of free fatty acids released from necrotic adipocytes are transported across the plasma membrane of adjacent adipocytes and locally re-esterified (Thompson et al. 2010). In this context, free fatty acids are esterified with glycerol 3-phosphate. Stimulation of glyceroneogenesis and re-esterification of free fatty acids by increasing in phosphoenolpyruvate carboxykinase (PEPCK) activity leads to ultimate rise in adipocyte size and body weight (Franckhauser et al. 2002). The close relationship between adipocyte size and the abundance of macrophages in adipose tissue indicates that the relationship of adipocyte size with adipocyte function may be regulated through a paracrine pathway involving adipose tissue macrophages (Weisberg et al. 2003).

A paracrine loop involving free fatty acids and tumor necrosis factor-alpha (TNF-alpha) between adipocytes and macrophages establishes a vicious cycle that aggravates inflammatory changes in the adipose tissue. Initially TNF-alpha increases the release of free fatty acids from adipocytes. By contrast, high amount of saturated fatty acids which are released from adipocytes can induce inflammatory changes via increasing TNF-alpha production (Suganami et al. 2005). TNF-alpha-induced activation of caspase-1 is accompanied by induction of NOD-like receptor (NLR) family protein-3 (NLRP3) in adipocytes. But caspase-1 activation does not depend on the NLRP3 inflammasome (Furuoka et al. 2016). Caspase-1 activation in adipose tissue of obese animals is partly independent of macrophage infiltration. However, caspase-1 induces the release of IL-1beta, and finally leads to the development of insulin resistance . These results indicate a crosstalk between adipocyte-specific caspase-1 and IL-1beta produced by macrophages in adipose tissue (Stienstra et al. 2010). Additionally, saturated fatty acid also serves as a ligand for toll like receptor-4 (TLR4) . Thereby the inflammatory changes in both adipocytes and macrophages are induced by TLR4 through nuclear factor-kappaB (NF-kappaB) activation (Suganami et al. 2007). Free fatty acid-induced and TLR-mediated activation of c-Jun N-terminal kinase (JNK) -related pro-inflammatory pathways in CD11c+ immune cells plays a central role in the development of adipose tissue inflammation and insulin resistance (Nguyen et al. 2007). Although the adipocyte is the key player controlling local changes in the microenvironment, macrophages have pivotal role in remodeling events. Nevertheless, resident macrophages display remarkable heterogeneity in their activities and functions (Gordon and Taylor 2005) (Fig. 14.1).

On the other hand, adipocyte-derived microparticles (MPs) are critical "find-me" signals for recruitment of monocytes and macrophages. Adipocytes exposed to saturated fatty acids show marked release of MPs, which are enriched in perilipin A. The release of MPs is highly dependent on caspase 3 and Rho-associated kinase. Thus hypertrophied and stressed adipocytes generate chemotactic signals that induce macrophage migration in a caspase 3 dependent manner (Eguchi et al. 2015). White adipose tissue is characterized by a continuous turnover of the adipocytes with approximately 10% of annual renewal (Spalding et al. 2008). In this context, old adipocytes undergoing programmed cell death are removed by macrophages (Duvall et al. 1985; Keuper et al. 2011). Removal of apoptotic adipocytes by macrophages is a process in which one cell takes bites out of another (trogocytosis). Therefore, adipocyte-derived DNA is not detected in the phagocytes (Sárvári et al. 2015). Apoptotic cells send an “eat me” signal to macrophages, triggering their own engulfment. Among the various molecules proposed to be involved in this process, in particular phosphatidylserine (PtdSer) is a strong candidate for the “eat me” signal (Krahling et al. 1999). PtdSer is transferred caspase-dependently from the inner leaflet to the outer leaflet of the plasma membrane (Martin et al. 1996). The specific signals of apoptotic cells are well recognized by the phagocytes. Thus milk fat globule-EGF-factor 8 (MFG-E8)-secreted active macrophages recognize aminophospholipids and specifically bind to apoptotic cells for engulfment (Hanayama et al. 2002). Actually the glycerophospholipid, PtdSer is perhaps the best-characterized “eat me” signal. In healthy cells, PtdSer is abundant in the inner leaflet of the plasmalemma, but during apoptosis it is quickly redistributed and exposed on the exofacial leaflet. It is recognized by specific phagocytic receptors such as T-cell immunoglobulin mucin receptor 4 (TIM4), which enhances engulfment (Miyanishi et al. 2007). Actually, MFG-E8 binds to the integrin alphavbeta3 complex in macrophages via the Rho and Rab family guaosine triphosphatases (GTPases) in MFG-E8’s EGF domain, hence bridging apoptotic cells and macrophages. TIM4 binds to PtdSer via the immunoglobulin-like domain in its extracellular region. The engulfment of apoptotic cells proceeds in two steps: TIM4 tethers apoptotic cells, and the integrin alphavbeta3 complex mediates engulfment in coordination with MFG-E8 (Toda et al. 2012). Thus, the expression of MFGE-8 and the alphavbeta5 integrin subunits are increased in adipose tissue of obese humans (Henegar et al. 2008).

3 Macrophage Phenotype in Adipose Tissue

When the visceral adipose tissue is excess, compared to subcutaneous adipose tissue , the absolute number of CD14+ adipose tissue macrophage is increased in visceral adipose tissue relative to subcutaneous adipose tissue. The absolute number of CD3+ T cells is increased in visceral adipose tissue relative to subcutaneous adipose tissue. An increase in adipose tissue macrophage predisposes to an M1 phenotype in visceral adipose tissue relative to subcutaneous adipose tissue in human obesity . Furthermore, the CD14-enriched cell population has a macrophage phenotype and is a dominant source of inflammatory cytokines in stromo-vascular cells (O’Rourke et al. 2009).

CD11c+ adipose tissue macrophages significantly increase in obesity . A novel F4/80+ CD11c+ population of macrophages in adipose tissue of obese mice is not found in lean mice (Lumeng et al. 2007a). CD11c is considered to be a classical cell-surface marker of dendritic cells (DCs) and F4/80 as substitute for function, have led to confusion on the exact contribution of DCs versus macrophages to tissue immunity in obesity. Indeed, the CD11c integrin can be expressed on macrophages. Therefore, the role of CD11c+ cells does not always reflect the role of DCs. Similarly, F4/80 is also expressed by eosinophils and neutrophils and is not a specific marker of macrophages (Hashimoto et al. 2011). However, in obese adipose tissue, the CD11c+ adipose tissue macrophages also express CD11b and F4/80 and are very different from the adipose tissue DCs, which express CD11c, but not CD11b. The obesity-induced changes in adipose tissue macrophage numbers are mainly due to increases in the triple-positive CD11b+ F4/80+ CD11c+ adipose tissue macrophage subpopulation (Lee and Lee 2014).

As mentioned above, two different populations of F4/80+CD11b+ macrophages are defined in adipose tissue; one of which also express CD11c. Obesity largely increases CD11c+ macrophage number in adipose tissue. F4/80+CD11b+CD11c+ cells are a specific population of adipose tissue macrophages recruited to adipose tissue upon high fat diet exposure (Patsouris et al. 2008). Moreover, F4/80 and CD11c are primarily regarded as cell surface markers of macrophages and dendritic cells, which are immune-phenotypically defined as F4/80+CD11c− and F4/80−CD11c+, respectively. In obese conditions, all these cells have not simultaneously completed their process of transitioning from a F4/80+CD11c− to a F4/80−CD11c+ state or cells that have transitioned from a CD11c− to CD11c+ phenotype. Some of them may show their original phenotypical characteristics and may have gradually acquired a CD11c+ phenotype (Nguyen et al. 2007). Adipose tissue macrophages from lean mice express many genes, which display characteristic of M2 or “alternatively activated” macrophages, including Ym1, arginase 1, and interleukin-10 (IL-10). Diet-induced obesity decreases expression of these genes in adipose tissue macrophages while increasing expression of genes such as those encoding TNF-alpha and inducible nitric oxide synthase (iNOS) that are characteristic of M1 or “classically activated” macrophages (Lumeng et al. 2007a). Indeed, TNF-alpha is a major macrophage-derived mediator of inflammation in adipocytes, whereas free fatty acids is an important adipocyte-derived mediators of inflammation in macrophages. TNF-alpha acts on TNF-alpha receptor of hypertrophied adipocytes. Briefly, a paracrine loop involving free fatty acids and TNF-alpha between adipocytes and macrophages establishes a vicious cycle that aggravates inflammatory changes in the adipose tissue (Suganami et al. 2005). Macrophages express some adipocyte-specific gene products such as ap2, meanwhile adipocytes secrete macrophage-specific gene products such as IL-6 or TNF-alpha. Lipid accumulation by macrophages in atherosclerotic lesions or by phagocytic capacities exhibited by adipocytes reveal an apparent coordinated activity between these two cell-types during the course of an innate immune response (Wolowczuk et al. 2008). An additional similarity between adipocytes and macrophages is further observed in the expression of TLR4 (Lin et al. 2000). The presence of functional TLR2 and TLR4 is demonstrated on human adipocytes isolated from subcutaneous fat tissue (Bès-Houtmann et al. 2007). “Adipocyte-macrophage-TLR4” pathway might be involved in the inflammatory process occurring in obesity (Wolowczuk et al. 2008). As mentioned above, diet-induced obesity is associated with the loss of tissue homeostasis and development of type 1 inflammatory responses in visceral adipose tissue . A key event is a shift of adipose tissue macrophages toward an M1 phenotype. In fact, obesity-induced adipocyte hypertrophy results in upregulated surface expression of stress markers. Adipose stress is detected by the phenotype of natural killer (NK) cells and CD8+ T cells, which produce interferon-gamma (IFN-gamma) , driving M1 macrophage polarization (Wensveen et al. 2015). The nuclear hormone receptor, PPAR-gamma is the first adipocyte-specific transcription factor, which is expressed in a highest level in adipose tissue (Chawla et al. 1994; Tontonoz et al. 1994). Further, PPAR-gamma is a critical signaling molecule in determining macrophage phenotype in adipose tissue (Charo 2007). The deficiency of PPAR gamma in immune cells favors expression of M1 and impairs M2 macrophage marker expression in adipose tissue (Bassaganya-Riera et al. 2009). Contrarily, PPAR-gamma activation induces monocyte differentiation into M2 macrophages (Bouhlel et al. 2007). The decreased PPAR-gamma protein level is associated with increased macrophage infiltration in visceral adipose tissue of obese subjects. In this case, pro-inflammatory macrophages suppress PPAR-gamma activity in adipocytes via S-nitrosylation (Yin et al. 2015). Actually, adipocytes are the source of Th2 cytokines. In response to adipocyte-derived Th2 cytokines , the expression of macrophage PPAR-beta/delta is up-regulated. Adipose tissue macrophage polarization ultimately could be modulated by adipocyte derived Th2 cytokines in a paracrine manner. PPAR-delta is induced by Th2 cytokines to control the transcriptional program of alternative activation in the macrophages. PPAR-delta may act as a factor for Th2 cytokine-induced M2 gene expression (Kang et al. 2008). In fact, obesity is accompanied by a transformation in the polarized states of macrophages from an anti-inflammatory “alternatively activated” M2 form (Kosteli et al. 2010), to a more pro-inflammatory “classically activated” M1 form (Lumeng et al. 2007a). The primary trigger for the recruitment of M1 macrophages is thought to be the secretion of TNF-alpha from hypertrophied adipocytes (Wellen and Hotamisligil 2003). The obesity-driven phenotypic adipose tissue macrophages can be characterized in these two broad classes based on the expression of particular antigens. Thus, diet-induced obesity leads to a shift in the activation state of adipose tissue macrophages from an M2-polarized state that may protect adipocytes from inflammation to an M1 pro-inflammatory state (Lumeng et al. 2007a).

TLR-dependent polarization mediators of M1 macrophages include different transcription factors such as NF-kappaB, activator protein-1 (AP-1), transcription factor PU.1 (PU.1), CCAAT/enhancer-binding protein alpha (C/EBP-alpha), signal transducer and activator of transcription 1 (STAT1) as well as interferon regulatory factor-5 (IRF5). AP-1 together with NF-kappaB in signal-dependent gene expression is necessary for innate immunity (Juhas et al. 2015). Among these transcription factors, IRF5 expression in macrophages is reversibly induced by inflammatory stimuli and contributes to the macrophage polarization . In this context, high expression of IRF5 is characteristic of M1 macrophages (Krausgruber et al. 2011). Thereby, IRF5 has been implicated in polarizing macrophages towards an inflammatory phenotype. In obese individuals, IRF5 expression is negatively associated with insulin sensitivity and collagen deposition in visceral adipose tissue (Dalmas et al. 2015), whereas another interferon regulatory factor , IRF4 is a key transcription factor that controls M2 macrophage polarization (Satoh et al. 2010). Consequently, differentiation of M1 macrophages is connected with the upregulation of IRF5 levels. The overexpression of IRF5 in M2 polarized macrophages leads to their phenotypic switch. It makes the M2 macrophages functionally similar to M1 (Juhas et al. 2015). However, Prieur et al. showed that early stages of adipose tissue expansion are characterized by M2-polarized adipose tissue macrophages and progressive lipid accumulation within adipose tissue macrophages (Prieur et al. 2011). The microenvironment in a lean adipose tissue is composed of a 4:1; M2:M1 ratio (Lumeng et al. 2008). Actually, a delicate balance of polarized populations of macrophages is necessary to maintain adequate adipocyte function. Inflammatory signals in adipose tissue of obese individuals is eliminated by retaining M2 polarization of adipose tissue macrophages or by triggering the phenotypic switch from M1 to M2 (Sun et al. 2011). Diet-induced obesity increases the number of M1 macrophages by 65-fold, whereas the number of M2 macrophages per weight basis is also increased by six-fold. Eventually, the ratio of M1-to-M2 macrophages is increased in obesity (Fujisaka et al. 2009). However, Fjeldborg et al. found a relatively higher expression of the anti-inflammatory markers, CD163 and IL-10, and a relative reduction of the pro-inflammatory markers, TNF-alpha and IL-6 , in adipose tissue from obese subjects compared to lean individuals. Thus, human adipose tissue macrophages change polarization to a more anti-inflammatory profile in obesity than towards a pro-inflammatory profile (Fjeldborg et al. 2014). There is a shift towards a M2 phenotype in non-crown like structure macrophages in adipose tissue from obese subjects compared with lean ones. Macrophages in crown like structure are predominantly M1, but most other macrophages, particularly those in fibrotic areas, are M2 (Spencer et al. 2010). At initial stage the expression of IL-10 by M2 macrophages is approximately 100-fold higher than that in adipocytes. Hence M2 macrophage-derived IL-10 constitutes the major portion of the IL-10 in adipose tissue. In this context, the phenotypic shift of adipose tissue macrophages toward an M1-dominant state cannot be explained simply by the conversion of M2 macrophages to M1. De novo accumulation of circulating monocytes to adipose tissue followed by their differentiation into M1 and M2 macrophages (Fujisaka et al. 2009). This situation can be explained as follows; first of all, transient accumulation of M2 macrophages could be essential for the control of tissue fatty acid levels during activation of lipolysis . M2 macrophage-borne signaling molecules could inhibit lipolysis and re-esterification of lipolyzed fatty acids form triacylglycerols (triacylglycerols/fatty acid cycle). Thus, M2 macrophages initially support the metabolically flexible adipocytes, which have a high capacity of both triacylglycerols/fatty acid cycling and oxidative phosphorylation in obese white adipose tissue (Masoodi et al. 2015). The M1 adipose tissue macrophage polarization in obesity is regarded as the optimization of the fat deposition and repartitioning toward adipocytes. It prevents adipose tissue macrophage lipotoxicity (Prieur et al. 2011). However, the expression of carnitine palmitoyltransferase 1A (CPT1A), the rate-limiting enzyme in mitochondrial fatty acid oxidation, is higher in human adipose tissue macrophages than in adipocytes. It is differentially expressed in visceral versus subcutaneous adipose tissue in obese individuals. Enhanced fatty acid oxidation in saturated fatty acid -incubated adipocytes and macrophages reduces triglyceride content and inflammation . While insulin sensitivity increases in adipocytes, endoplasmic reticulum (ER) stress and reactive oxygen species (ROS) damage decreases in macrophages (Malandrino et al. 2015). M2 macrophages sustain insulin sensitivity by secreting IL-4 and IL-10, while M1 macrophages induce insulin resistance through the secretion of pro-inflammatory cytokines, such as TNF-alpha (Tateya et al. 2013). M2 macrophages secrete anti-inflammatory cytokines and utilize oxidative metabolism to maintain adipose tissue homeostasis (Castoldi et al. 2015). M1 macrophages provide their energy from aerobic glycolysis. In this condition, glucose uptake as well as the conversion of pyruvate to lactate is increased. Meanwhile, the activities of the respiratory chain are attenuated, allowing for ROS production. However, M2 macrophages obtain much of their energy from fatty acid oxidation and oxidative metabolism (Galván-Peña and O’Neill 2014). In response to IL-4, STAT6 and PPARgamma-coactivator-1beta (PGC-1beta) induce the fatty acid oxidation and mitochondrial biogenesis in macrophages (Vats et al. 2006). IL-4 activates the transcriptional responses of STAT6. Active STAT6 can induce the coactivator PGC-1beta. PGC-1beta can induce mitochondrial respiration as well as mitochondrial biogenesis (Galván-Peña and O’Neill 2014; St-Pierre et al. 2003). In fact, PPAR-gamma is required for IL-4-induced increase in beta-oxidation of fatty acids. Despite IL-4 stimulation, the rate of fatty acid oxidation is reduced by approximately 70% in PPAR-gamma null macrophages. By contrast, increase in PPAR-gamma expression doubles the rate of beta-oxidation (Odegaard et al. 2007).

Progression of obesity increasingly induces a phenotypic switch from the M2 macrophages, to the M1 macrophages . Hence pro-inflammatory CD11c+ M1 macrophages is markers of insulin resistance in human obesity (Wentworth et al. 2010). Thus the enhanced macrophage-adipocyte crosstalk in obesity disrupts insulin action in adipocytes. Macrophage-secreted factors block insulin action in adipocytes via downregulation of GLUT4 and insulin receptor substrate-1 (IRS -1) , leading to a decrease in Akt phosphorylation and impaired insulin-stimulated GLUT4 translocation to the plasma membrane (Lumeng et al. 2007c). The secretion of inflammatory cytokines and chemokines by adipose tissue macrophages includes TNF-alpha , IL-6 , IL-1beta, MCP-1 , CCL2, and macrophage inhibitory factor (MIF) (Olefsky and Glass 2010). In particular IL-1beta is produced largely by macrophages. It provokes the development of obesity-associated insulin resistance by inhibition of insulin signal transduction in adipocytes. Contrarily, blocking the activity of IL-1beta improves insulin signaling in human adipocytes. This is in parallel with a reduction in macrophage-stimulated pro-inflammatory profile and lipolysis (Bing 2015). Thus the reduction in protein expression of IRS -1 , phosphatidylinositol 3-kinase (PI3K) p85alpha, and GLUT4 by macrophage-conditioned-medium is reversed when macrophage-derived IL-1beta production is inhibited. In contrast, macrophage-stimulated release of IL-6, IL-8, and chemokine (C-C motif) ligand 5 (RANTES) by adipocytes is also significantly reduced by blocking IL-1beta production (Gao et al. 2014). PGC-1beta plays a key role in saturated fatty acids metabolism and in the regulation of inflammatory signaling. PGC-1beta expression is significantly decreased in response to saturated fatty acid in macrophages in a dose dependent manner. PGC-1beta inhibits saturated fatty acid-induced TNF-alpha , MCP-1 , and IL-1beta mRNA and protein expressions. Furthermore, PGC-1beta significantly antagonizes saturated fatty acid-induced macrophage NF-kappaB-p65 and JNK activation. PGC-1beta overexpression in saturated fatty acid treated macrophages improves adipocytes PI3K-Akt insulin signaling in a paracrine fashion (Chen et al. 2016). In response to high-fat diet the tumor necrosis factor receptor (TNFR) -associated factor-3 (TRAF-3) , binds to and activates TGF-beta-kinase 1 (TAK1). Thereby the activation of downstream inhibitor of nuclear factor kappa-B kinase subunit beta (IKKbeta)–NF-κB and mitogen-activated protein kinase kinase (MAPKK)–JNK–IRS1 signaling cascades are enhanced, while disrupting AKT–glycogen synthase kinase 3 beta (GSK3beta)/forkhead box protein O1 (FOXO1) phosphorylation cascade. Eventually insulin resistance and inflammatory response is facilitated (Wang et al. 2016) (Fig. 14.1).

The release of obesity -related danger signals such as ROS , lysosomes, and other obesity-induced danger signals result in the oligomerization of NLRP3 in adipose tissue (Gurung et al. 2015; Zhou et al. 2011). Activation of inflammasomes by damaged mitochondria results in caspase-1-dependent secretion of the inflammatory cytokines IL-1beta and IL-18, and leads to an inflammatory form of cell death. In addition, regulation of mitochondria-induced inflammasome activation centrally contributes to the inflammatory process that is responsible for obesity (Gurung et al. 2015). Actually NLRP3 with macrophage marker F4/80 co-localizes in crown-like structures. Hence, obesity is associated with significant increase in serum IL-18 concentrations which is blocked upon ablation of NLRP3. Indeed, NLRP3 inflammasome is activated in response to high fat diet and controls the production of IL-1beta in adipose tissue and IL-18 in obesity. Induction of high-fat diet induced obesity causes marked caspase-1 activation in adipose tissue. Ceramides are lipid metabolites that accumulates in tissues in response to obesity, and induce caspase-1 activation in an NLRP3-dependent mechanism. In contrast, elimination of NLRP3 inflammasome reduces M1-like macrophage gene expression and increases M2-like expressed cytokines (Youm et al. 2011). Partial depletion of macrophages from adipose tissue of obese subjects decreases the expression of the macrophage marker CD68, with no significant alteration in the expression of caspase-1. This indicates that the effects of the NLRP3-apoptosis-associated speck-like protein (ASC)-caspase-1 protein complex on adipose tissue are not exerted though infiltrating macrophages (Benetti et al. 2013; Stienstra et al. 2011). Actually, innate pattern recognition receptors inflammasomes (NLRs) are cytosolic sensors that detect endogenous metabolic stress and activates caspase enzymes. Activated caspase-1 processes the cytosolic precursors of the related cytokines IL-1beta and IL-18 (Li et al. 2014; Stienstra et al. 2011).

One important molecule regulating glycolysis and macrophage activation is hypoxia -inducible factor 1-alpha (HIF-1alpha). However, hypoxia induces the inflammatory phenotypes of macrophages via HIF-1alpha-dependent and -independent mechanisms (Fujisaka et al. 2013). PI3K/Akt contributes hypoxic stress-induced TLR4 expression through the regulation of HIF-1 activation (Kim et al. 2012). Moreover, hypoxia increases the activation of JNK and p38 mitogen-activated protein kinase signaling in saturated fatty acid-treated macrophages. Inhibition of JNK blocks the hypoxic induction of pro-inflammatory cytokine expression. In any case, inhibition of hypoxia-induced transcription factors fails to reduce the expression of cytokines by saturated fatty acid -treated hypoxic macrophages. Enhanced pro-inflammatory cytokine production and JNK activity under hypoxia are prevented by inhibiting ROS generation. It is well known that hypoxia along with higher concentrations of free fatty acids exacerbates macrophage-mediated inflammation in obesity (Snodgrass et al. 2016). In white adipose tissue , increased oxidative stress due to the mRNA over-expression of NADPH oxidase subunits in ectopic fat leads to enhanced plasminogen activator inhibitor-1 (PAI-1) , IL-6 , and MCP-1 mRNA expression. Furthermore, the selective increase in ROS production in obesity eventually leads to elevation of systemic oxidative stress (Furukawa et al. 2004). In M2 macrophages, overexpression of HIF-2alpha decreases nitric oxide production and suppresses expression of pro-inflammatory cytokines through induction of arginase-1. HIF-2alpha-overexpressing macrophages alleviate pro-inflammatory responses and improve the insulin resistance in adipocytes (Choe et al. 2014). Additionally, recruited adipose tissue macrophages overexpress IL-6, iNOS , and CCR2 (Lumeng et al. 2007b). In particular, CCR2 is expressed by obese adipose tissue macrophages, which are derived from bone marrow cells (Ito et al. 2008). A high level of de novo IL-6 secretion by macrophages is a result of engulfment of the lipid content of adipocytes by macrophages. Thus, IL-6 secretion during interaction of adipocytes and macrophages might have an anti-inflammatory role in the inflamed adipose tissue by downregulating the induction and release of pro-inflammatory cytokines (Sárvári et al. 2015). In total, production of these inflammatory factors in obesity is under the transcriptional control of two key intracellular inflammatory pathways; JNK-activator protein 1 (AP1) and IKKbeta. Extracellular free fatty acids activate JNK and IKKbeta via TLR signaling pathway. Elevation of the pro-inflammatory cytokines in obesity activates JNK and IKKbeta by a specific signaling pathway (Solinas and Karin 2010). Endogenous fatty acids, which are released from adipocytes via the beta3-adrenergic stimulation, result in the activation of the TLR4 /NF-kappaB pathway. In this process, saturated fatty acids , which are released in large quantities from hypertrophied adipocytes via the macrophage-induced adipocyte lipolysis, serve as a natural ligand for TLR4. Thereby the inflammatory changes in both adipocytes and macrophages are induced through NF-kappaB activation (Suganami et al. 2007). Macrophage-inducible C-type lectin (Mincle ; also called Clec4e and Clecsf9), is a type II transmembrane C-type lectin. It is induced selectively in macrophages during the interaction between adipocytes and macrophages. Saturated fatty acid released from adipocytes induces Mincle mRNA expression in macrophages through the TLR4/NF-kappaB pathway. Macrophage-induced adipocyte lipolysis aggravates obesity-induced adipose tissue inflammation by this way (Ichioka et al. 2011). Engagement of either TLR4 or TLR2 with long-chain saturated fatty acids leads to recruitment of the myeloid differentiation primary response gene 88 (MyD88) adapter protein and formation of signaling complexes containing the IL-1 receptor associated serine/threonine kinase (IRAK1) and the TRAF family members, TRAF3 and TRAF6. The latter involves the activation of the MAP kinase kinase kinase (MAP3K) TAK1, which is required for JNK and IKKbeta activation (Kawai and Akira 2007). In addition to these pathways, IRAK-1 is also a tissue marker of meta-inflammation in obesity and a key component of the TLR2/IL-1receptor/MyD88 signaling pathway. Enhanced IRAK-1 gene expression correlates with adipose tissue infiltration by macrophages in obesity-associated chronic low-grade metabolic inflammation (Ahmad et al. 2015) (Fig. 14.1).

Inflammation leads to an increased number of microRNAs (miRNAs) detectable in both adipocytes and M1 macrophages . Indeed, under inflammatory conditions, adipocytes and M1 macrophages share the expression of 147 miRNAs, and 100 common miRNAs . Additionally, miR-221 by 2-fold, miR-222 by 2.5-fold, and miR-155 by 5-fold increased in inflamed adipocytes (Ortega et al. 2015). Although there is no significant change in the level of chemokine (C-C motif) ligand 2 (CCL2) either in adipose tissue or in circulation, the cellularity of adipose tissue macrophages (ATMs) is dramatically elevated at the early stage of obesity . Thus alternatively activated macrophages at the early stages of obesity are the major population of resident macrophages, with more specialized lipid catabolism than that of immigrant macrophages (Zheng et al. 2016). In subcutaneous white adipose tissue obtained from 56 subjects, 11 miRNAs are present in all subjects and downregulated in obesity. Of these, ten affect adipocyte CCL2 secretion for miR-126 and miR-193b. The levels of miR-193b in subcutaneous white adipose tissue are significantly associated with CCL2 secretion, whereas expression of integrin-alpha-X, an inflammatory macrophage marker, is associated with miR-193b and miR-126. In this context, miRNAs may be important regulators of adipose inflammation through their effects on CCL2 release from human adipocytes and macrophages (Arner et al. 2012). Overexpression of miR-126/-193b/-92a in different pairwise combinations reduce CCL2 secretion more efficiently than either miRNA alone. However, although effects on CCL2 secretion by co-overexpression of miR-92a/-193b and miR-92a/-126 are additive in adipocytes, the combination of miR-126/-193b is primarily additive in macrophages. Furthermore signals for miR-92a and -193b converge on the NFkappaB pathway (Kulyté et al. 2014).

The expression of insulin-like growth factor-1 (IGF1) by adipose tissue is derived from adipocytes and macrophages. In lean animals, adipocytes are the primary source of IGF1, but in obesity expression by adipocytes is reduced and by macrophages increased, as to maintain overall adipose tissue IGF1 expression (Chang et al. 2016). IGF1 reduces free fatty acid -induced JNK1 activation and TNF-alpha expression in human subcutaneous but not omental preadipocytes. Impaired anti-inflammatory action of IGF1 in omental preadipocytes is a result of the chronic inflammation in visceral adipose tissue (Neacsu et al. 2013). Actually, preadipocytes could function as macrophage-like cells and raise the possibility of a potential direct involvement of adipose tissue in inflammatory processes (Cousin et al. 1999). Moreover, preadipocyte and macrophage phenotypes are very similar and that preadipocytes have the potential to be very efficiently and rapidly converted into macrophages (Charrière et al. 2003). In this respect, both macrophages and adipocytes are the sites for active lipid metabolism and signaling. Macrophage/adipocyte aP2 expression is controlled with the same promoter or enhancer elements that confer expression in adipocytes. Hence, there is a striking overlap between the biology of adipocytes and macrophages (Makowski et al. 2001). The macrophage/adipocyte aP2 has also been expressed exclusively by differentiated adipocytes. aP2 plays a vital role in the local macrophage responses and binds a number of hydrophobic ligands which are known to influence macrophage function (Makowski et al. 2005). As lipid chaperones, fatty acid-binding proteins (FABPs) may actively facilitate the transport of lipids to specific compartments in the cell. Adipocyte/macrophage aP2 and FABP5, act at the interface of metabolic and inflammatory pathways. Although being the minor isoform in adipocytes, FABP5 protein levels are comparable to aP2 in normal macrophages (Makowski and Hotamisligil 2005). PPAR-gamma induces the expression of the adipocyte/macrophage aP2 and increases aP2 mRNA of primary human monocytes in time- and dose-dependent manner (Pelton et al. 1999). FABP -mediated lipid metabolism is closely linked to both metabolic and inflammatory processes through modulating critical lipid-sensitive pathways in adipocytes and macrophages (Furuhashi and Hotamisligil 2008). Deletion of FABPs in adipocytes results in reduced expression of inflammatory cytokines in macrophages, whereas the same deletion in macrophages led to enhanced insulin signaling and glucose uptake in adipocytes. The metabolic status of adipocytes is a major determinant of macrophage inflammatory output. Neither macrophages nor adipocytes individually could account for the total impact of FABPs on systemic metabolism (Furuhashi et al. 2008).

In adipose tissue, eosinophils are responsible for 90% of IL-4 expression and accelerate M2 macrophage polarization by secreting Th2 type cytokines such as IL-4 and IL-13. Hence, eosinophils might act as anti-inflammatory immune cells in obesity -induced adipose tissue inflammation (Huh et al. 2014). Th1 cells primarily secret IFN-gamma which stimulates monocyte differentiation into M1 type macrophages. IFN-gamma exacerbates adipose tissue inflammation in obesity (Rocha et al. 2008). IFN-gamma might also play an important role in growing adipose tissue, where adipocytes respond to inflammatory products derived from infiltrating macrophages (Wellen and Hotamisligil 2003).

Adipose tissue synthesizes complement proteins and is a target of complement activation. C3a-desArg/acylation-stimulating protein induces lipogenesis and affects lipid metabolism . The C3a receptor and C5a receptor are involved in the development of insulin resistance in adipocytes through macrophage infiltration and the activation of adipose tissue (Vlaicu et al. 2016). C5a receptor-like receptor 2 (C5L2) has been identified as a receptor for acylation-stimulating protein (ASP) and the inflammatory factor C5a. While adipocyte-conditioned medium increases C5L2-C5a receptor co-localization in macrophages, this is blocked by C5a. Induction of ASP increases Akt phosphorylation in both cell types. However, C5a induces Akt phosphorylation in adipocytes with less effect in macrophages (Poursharifi et al. 2013). ASP may induce specific inflammatory cytokines in adipocytes through PI3K- and NF-kappaB-dependent pathways, thus further promoting macrophage infiltration and local inflammation in obese adipose tissue (Tom et al. 2013).

4 Conclusion

Adipocyte death defines macrophage localization and function in adipose tissue of obese humans. However, mechanistic details of the macrophage accumulation surrounding the adipocytes and other parts of the adipose tissue is still vague. In addition, the source of TNF-alpha and the other cytokines requires a detailed analysis by taking the inflammatory alterations in obesity into account. The triggering mechanisms of lipid mediators in adipose tissue has been described by indirect evidences. Therefore, further investigations are necessary in order to clarify the molecular nature of the macrophage-adipocytes interactions in obesity-related diseases.