Adipose tissue & Immunity: The basics 2

Lymphoid structures in adipose tissue

Leukocytes present in the omentum are organized in clusters called "milky spots" (MS), which composition varies between species and is determined by antigenic exposure. The omentum has immunological and wound-healing properties, probably due to its angiogenic potential, large surface area and immunological activity (1). MS are composed primarily of macrophages and B1 cells. B1 cells are a subset of B cells that are different from conventional B cells (B2) in that they express different markers and antigen receptors that can bind common bacterial epitopes. These cells are also able to produce natural antibodies that provide a first protection to bacterial infections (2). The importance of MS as secondary lymphoid organs was highlighted by the work of Rangel-Moreno, et al. (3). They showed that SLP mice (spleen lymph node and Peyer's patch deficient) are able to generate B-cell responses to peritoneal antigens and present MS, which develop in the absence of LTi (lymphoid tissue inducer) cells. The local role for MS seems to be presenting  antigens derived from the peritoneal cavity to recirculating T and B lymphocytes, which enter MS from the blood. MS could also support somatic hypermutation and affinity maturation of B cells, and proliferation of T cells. Lymphocytes activated elsewhere were also found in MS, supporting the view of MS as an important part of the immune system. 

In the human mesentery, there is the presence of lymphoid clusters similar to MS called FALCs (fat-associated lymphoid clusters), in that both contain lymphocytes framed by adipose tissue in the peritoneal cavity. However, FALCs contain a great number (20-40% of total lymphocytes) of Lin- c-Kit+ Sca-1+ cells (4)*, which also express ILR7a. This suggests a progenitor potential. Moreover, these cells are capable of producing large amounts of IL2, IL4, IL5, IL6, GM-CSF and a moderate amount of IFN-gamma. Because of its properties (Th2-type lymphocytes with innate like characteristics), these FALC cells were termed as "natural helper cells" by the authors.

B cells

B1 cells, as has been mentioned, provides a first-line defense against bacterial and some viral infections. They have a less-diverse antibody repertoire, but respond to pathogen-associated molecules more rapidly than B2 cells.  B1 cells have been related to some metabolic diseases, due to the fact that they are responsive to LPS and cytokines (5). Importantly, these cells also produce IL-10 (6), which can regulate the activity of regulatory B cells (Bregs) (7). The implications of obesity in the function of B1 cells has yet to be discovered, but given that endotoxemia (8) and increased pro-inflammatory cytokine secretion are associated with obesity, overactivation of B1 cells could lead to the production of autoantibodies, promoting autoimmunity (9). Autoantibody production by B1 cells can also be promoted by endocrine disruptors (10). Finally, in contrast to their interaction with Bregs, B1 cells are unable to promote the conversion of naive Foxp3-CD4+ T cells into Foxp3+Treg cells (T-regulatory cells) compared to B2 cells, and promote Th1 and Th17 cell differentiation (see below) (11). 

The role of B2 cells in the pathogenesis of obesity has not been studied much, although there is an increase in total B cell count in adipose tissue from mice fed a high fat diet (HFD) (12). B cells might have a role mediating glucose intolerance in the obese, probably through IgG (12). Of note, some authors have suggested that B2 cells are atherogenic, while B1 cells are atheroprotective because BAFF-R deficient ApoE -/- mice (which are B2 depleted) show reduced inflammation and atherosclerosis (13, 14). 

T-cells

Infiltration of T lymphocytes to adipose tissue occurs rapidly in mice fed a HFD and correlates with the impairment in insulin sensitivity and glucose metabolism (15). Inducing obesity in mice produces an increase in CD8+ T cell levels in adipose tissue (epididymal fat), concomitantly with a reduction in CD4+ and Treg levels; accumulation which precedes macrophage infiltration (16). Furthermore, in vitro experiments have demonstrated that obese epididymal fat induces T cell proliferation and co-culturing obese adipose tissue with CD8+ T cells and monocytes induces the differentiation of macrophages to an inflammatory phenotype (16). Wu, et al. (17) found that obesity increases the level of regulated on activation, normal T cell expressed and secreted (RANTES) and CCR5 (one of RANTES receptors) in adipose tissue. This increases the migration of T-cells, as RANTES is chemotactic for these cells. The number of adipose tissue lymphocytes, as well as the expression of CCL20, increases with the degree of adiposity (18). Demonstrating a tight relationship between inflammation in adipose tissue and metabolic dysregulation, conditioned media from human adipose tissue lymphocytes is able to inhibit insulin-mediated FAS and LPL upregulation in mature adipocytes and to downregulate PIK3R1, effects probably mediated by IFN-gamma (18). In an elegant study, Winer, et al. (19) found that inducing obesity in mice produced the following changes in T-cell subsets in adipose tissue:

• IFN-gamma secreting Th1 lymphocytes were higher in visceral adipose tissue (VAT) compared to subcutaneous adipose tissue (SAT).
• The proportion of CD4+Foxp3+ T cells (Tregs) was 70% lower in VAT from obese mice, which produced an increase in the Th1:Treg ratio from 1.5:1 in lean mice to 6.5:1. 
• In absolute numbers, almost three times more Th1 cells accumulated per gram of fat in obese mice.
• In VAT from obese humans, the ratio of Th1:Treg was 12:1, compared to 6:1 in lean subjects. 
• T-cell expansion in VAT seems to be antigen-driven, given that OVA-specific T cells in OT2 TCR transgenic mice (20) showed secondary TCR-alpha rearrangements (thus, suggesting a positive selection towards an antigen present in VAT). These T-cells undergoing secondary TCR-alpha rearrangements in VAT showed a very narrow TCR-alpha diversity, implying a strong positive selection. Moreover, a negative pressure against selection of most of systemic TCR V-alpha (variable region in the alpha chain) was seen, given to the loss of some of these receptors**. 
• To characterize the role of T-cells in obesity, the authors used Rag1-null mice, which lack lymphocytes. These mice, when fed a HFD, gain more weight and visceral fat than HFD fed wild type mice. These mice also developed glucose intolerance, showing hyperglycemia, hyperinsulinemia and low insulin sensitivity. These mice also displayed a significant elevation of blood glucose when fed a normal diet, suggesting a physiological role of lymphocytes on metabolism.
• Reconstitution of these mice with CD4+ (but not CD8+) improved the metabolic abnormalities. It also induced Treg repopulation in VAT, albeit slowly. The effect of CD4+ cells was not mediated through differentiation into Tregs, nor by increasing IL-10 levels but by differentiation into Th2 cells. 
• Antibody treatment with anti-CD3 or F(ab')2 increased Treg levels in VAT and improved metabolic abnormalities. Additionally, F(ab')2 treatment increased the numbers of macrophage mannose receptor (MMR) positive cells (alternatively activated macrophages), increasing to almost 300% the production of IL-10.

The role of other inflammatory type of T-cell, Th17 cells, has been associated with obesity and metabolic complications. Diet-induced obese mice show an increased proliferation and differentiation towards Th17 cells, via an IL-6 dependent pathway (21). This increase in Th17 potentiated the severity of EAE and experimental colitis, reflecting the relationship between obesity and autoimmune risk. 

T-regulatory cells (Treg)

Treg cells have gotten much attention because they can regulate almost all immune responses. Further from being equal, different Treg populations display gene-expression differences (22), and subphenotype differences in Treg cells from different organs suggest that these tissue-specific Tregs modulate the immune response in a given organ. Of relevance to the topic, adipose-tissue-resident Tregs have been characterized (23). These fat Tregs had a different pattern of gene expression than their counterparts from other lymphoid organs: they over-represented genes coding for molecules involved in leukocyte migration and extravasation (CCR1, CCR2, CCR9, CXCL10, etc) and under-represented CCL5 and CXCR3. Fat Treg cells also seem to produce (and respond to, judging by expression of genes downstream of the IL-10 receptor) large amounts of IL-10, 136-fold higher than lymph node Tregs. The TCR repertoire of fat Treg cells was markedly different than that of adipose tissue T-cells and from other organs, suggesting that it is unlikely that these cells were the result of conversion of local T-cells. Finally, inducing T-cell conversion into Treg cells seem to ameliorate some metabolic abnormalities in rodent models (23). 

Adipose tissue effects on adaptive immune cells. Increases in adipose tissue mass  promotes changes in adaptive immune cell subtypes, such as T and B cells. The production of pro-inflammatory and chemotactic molecules (RANTES, IL-6, TNF-alpha, IFN-gamma, CCR1, CCR2, etc.) induces Th1 and Th17 cell polarization, as well as secretion of IgG by B-cells. 

Metabolic and immune integration

The function, differentiation and proliferation of several immune cell types is influenced by molecules that also play a key role in regulating metabolic homeostasis. Undoubtedly, leptin has gotten the most interest in recent years. Leptin is a master modulator of T-cell functioning: it promotes proliferation of peripheral blood lymphocytes and CD4+ T cells and induces the secretion of Th1 cytokines, such as IFN-gamma, while supressing the production of Th2 cytokines (24, 25). Leptin also modulates apoptosis of Th1 cells, promoting their survival and it is necessary for a Th2 response in vivo (26, 27). How does leptin modulates polarization and proliferation of lymphocytes? Compared to wild type mice and reconstituted ob/ob mice, autoreactive CD4+ T-cells from leptin-deficient mice have lower levels of Bcl-2 and phospho-ERK1/2, correlated with a reduction in the degradation of p27Kip1, a cell cycle inhibitor (28). Treatment with rapamycin mimics the effects of leptin deficiency (28), suggesting that leptin acts through the mTOR pathway. The effects of leptin in cells from the innate and adaptive immune system are shown in the following table (adapted from Procaccini, et al., 2012):

Cell type	Effect
Monocytes/macrophages	↑Phagocytosis ↑TNF-α, IL-6, IL-12 ↑Activation markers ↑Chemotaxis
Dendritic cells (DC)	↑TGFβ ↑Th1 priming ↑Survival ↑Immature DC migration ↑Stimulation of allogenic T cells
Neutrophils	↑Chemotaxis ↑ROS
NK cells	↑Activation marker ↑Cytotoxic activity ↑Perforin production
B cells	↑Lymphopoiesis and maturation ↑IL-6, IL-10, TNF-α ↑IgG2a
T cells	↑Activation markers ↑Proliferation ↑Th1 response ↑Adhesion molecules
Treg	↓Proliferation ↓Activity

mTOR

The mTOR pathway plays a very important role regulating immune function. Because of its role as a cellular energy sensor, it integrates metabolism and immunity, and provides a link for regulation of both by different molecules, such as leptin. 

Besides commonly known upstream activators of mTOR (for example, growth factors like insulin), immune-associated molecules are also able to induce mTOR activity. CD28, a T-cell costimulatory receptor, is a potent activator of mTOR through the activation of PI3K (29). Cytokines, such as IL-2, IL-4, IL-12, leptin, IFN-gamma and IL-1 are also able to activate mTOR (30). Conversely, PD-1 ligand 1 inhibits mTOR activity (31). Proteins normally thought to function only as metabolic sensors can also regulate mTOR activity, providing the first mechanism by which metabolism and immunity are related. For example, AMPK phoshporylates TSC, promoting the inhibition of mTOR (32, 33). The importance of AMPK for regulating the immune response is highlighted by the findings that loss of AMPK aggraviates EAE in mice (34). Aminoacids are potent activators of mTOR (35) and Treg induction from naïve T-cells is dependent on amino acid availability (36).

mTOR controls T-cell activation and differentiation 

mTOR plays a very important role integrating environmental cues for T-cell differentiation and activation. For a T-cell to become activated, two signals must be present. Signal 1 involves the recognition of antigens by TCR. On the other hand, Signal 2 has been defined as an species-specific accesory signal delivered by the stimulatory cell to the T-cell (37).  This means that for full T-cell activation, the two signals must be present. Signal 2, far from implicating only one type of interaction, is the result of the integration of multiple signals, which seem to be under the control of mTOR. For example, in Th1 cells, activation of mTOR is necessary for full T-cell activation and anergy reversal (37). Supporting the model of mTOR as Signal 2, experiments done on CD4+ T cells show that depending on the microenvironment (levels of specific cytokines), mTOR is necessary to differentiate T-cells into different subsets. Delgoffe, et al. (38) deleted mTOR specifically in T-cells. These mutant cells proliferated normally (albeit slower than wild type), but werent able to differentiate into Th1, Th2 or Th17 after skewing conditions. This correlated with decreased STAT activation (STAT4 for Th1, STAT6 for Th2, STAT3 for Th17) and lack of upregulation of T-bet, GATA-3 and ROR-gammat expression (specific transcription factors for Th1, Th2 and Th17 cells, respectively). Surprisingly, TCR engagement in the absence of mTOR induced the differentiation of Foxp3+ cells. 

mTOR also influences the activation of CD8+ cells. These cells are maintained in a quiescent state by several transcription factors, such as ELF4 and KLF4. mTOR controls CD8+ cell activation by inhibiting the expression of these proteins, reversing the quiescent state (39, 40). CD8+ cells lacking mTORC1 fail to become effector cells, and transition to memory cells seems to be also under control of mTOR (30). Besides controlling activation, the PI3K-mTOR axis regulates T-lymphocyte trafficking (41). 

mTOR effects on T-cell subsets. mTORC1 is activated by growth factors (insulin, IGF-1, EGF, etc.), aminoacids (BCAAs), co-stimulatory molecules (CD-28) and several cytokines. Upon activation, mTORC1 promotes Th1 differentiation of CD4+ T-cells by activating STAT4, which in turns promotes transcription of T-bet; and Th17 by activating STAT3, which increases the expression of ROR-gammat. This effect is mediated by inhibition of SOCS3. Additionally, mTORC1 leads to activation of CD8+ T-cells by inhibiting the expression of ELF4 and KLF4, which maintains CD8+T-cells in a quiescent state. Upstream activation of mTORC2 is poorly understood, but growth factors and, recently, ribosomes have shown to activate this complex. mTORC2 promotes CD4+T-cell differentiation into Th2 cells, by inhibiting SOCS5, enhancing STAT6 phosphorylation. This leads to increased expression of GATA-3 transcription factor. Activation of mTORC2 also inhibits FOXO proteins. Both complexes inhibit Foxp3, thereby reducing Treg differentiation.

mTOR promotes survival and maturation of B-cells

The direct effect of mTOR signaling in B-cell function has not been addressed. However, conditional deletion of TSC1 (mTORC1 inhibitor) in murine B-cells caused an impairment on B-cell maturation and response to T-cell dependent and independent antigens (42). Moreover, survival and proliferation of activated B-cells requires PI3K and mTOR activation (43).

mTOR controls dendritic cell (DC) activation/maturation

Treatment of DC with rapamycin (mTOR inhibitor) inhibits maturation by IL-4, GM-CSF and IL-1b (37). These DC are poor stimulators of T-cells but promote the induction of T-cell tolerance by Treg induction (44, 45). Rapamycin also produces a decrease in the expression of receptors involved in antigen presentation and uptake (46) and synthesis of some cytokines, like IL-18 (47). In LPS-stimulated DC, rapamycin enhances IL-12 production  and supresses IL-10 expression (48). This suggests that mTOR TLR-induced activation promotes the secretion of IL-10 while inhibiting the production of IL-12. Finally, mTOR is critical for monocyte-derived DC survival and immunostimulatory potential (49).

mTOR effects on immune cells is related to their metabolism

Every cell needs ATP for performing their specific functions. Unlike most differentiated cells (which utilize the citric acid cycle and mitochondrial respiration in the presence of oxygen), lymphocytes show a similar metabolic phenotype as cancer cells: they utilize oxidative glycolysis (Warburg effect) (30). In the resting state, lymphocytes are in a catabolic state, and molecules required for protein synthesis and energy are provided by autophagy. Upon activation, when energy requirements increase, there is an upregulation in protein, nucleotide and lipid biosynthesis. This anabolic switch is mediated by immunostimulatory molecules. For instance, CD28 activates PI3K, which activates Akt, which in turn promotes membrane translocation of glucose transporters (principally GLUT1 in lymphocytes). This results in increased glucose uptake and glycolysis, in excess of what is required for sustaining an adequate cellular level of ATP or macromolecular synthesis (50, 51). CTLA-4, an inhibitory receptor, inhibits CD28-induced upregulation of glucose metabolism (50). Activation of T-cells also leads to an increase in other glucose utilization pathways, such as the pentose phosphate pathway, which provides substrates for nucleic acid synthesis and NADPH (52). 

The importance of these metabolic changes on T-cell functioning is shown by experiments in which blocking the PI3K-Akt-mTOR pathway inhibit T-cell activation. AICAR, which activates AMPK (and thereby inhibiting mTORC1) inhibits IL-2 synthesis and promotes T-cell anergy (53). NALA, a leucine antagonist, and 2-deoxyglucose inhibit IL-2 and IFN-gamma synthesis, as well as proliferation of Th1 cells (54). This effect was observed even in the presence of co-stimulation. 

Treg cells display a different metabolic phenotype than effector CD4+ T cells. Whereas the latter exhibit a glycolytic dependent metabolism and high surface expression of GLUT1, Treg cells are dependent on lipid metabolism (55). Michalek et al. found that glucose deficiency inhibits Th17 polarization, without affecting Treg differentiation. Moreover, blocking CPT-I supressed Treg generation and the addition of exogenous fatty acids (oleate/palmitate mixture) inhibited the production of Th1, Th2 and Th17 cytokines (specifically, Th1 differentiation was reduced) but enhanced Treg expression of Foxp3. Finally, rapamycin and metformin increased lipid oxidation in CD4+ cells and Treg generation, while decreasing GLUT1 expression in CD4+ T-cells in a murine model of allergic asthma. 

Summary

• Obesity alterates the immune balance in all immune cell types, by increasing the secretion of several cytokines and chemotactic molecules. Specifically, obesity is associated with Th1 and Th17 skewing, CD8+ T-cell activation and proliferation and reduction in Treg cell numbers. It also promotes IgG production by B-cells, which seems to be important for the metabolic alterations observed.  
• Leptin, the most famous adipocytokine, is central to the relationship between immunity and metabolism. 
• Leptin effects are mediated by mTOR, which in turn, is the link between metabolism and immune function. mTOR is now being proposed as Signal 2, which is required for full T-cell activation.
• mTOR controls T-cell activation, proliferation and differentiation. Activation of mTOR promotes Th1, Th17 and Th2 differentiation of CD4+ T-cells and inhibits Treg generation. It also controls survival, activation and maturation of B-cells and dendritic cells. 
• Lymphocytes display a glycolytic metabolism, oxidizing glucose preferentially via aerobic glycolysis. Upon activation, mTOR and Akt increase the surface expression of GLUT1 and several enzymes involved in glucose uptake and oxidation. Conversely, Treg cells display a lipolytic metabolism, which is essential for their proliferation, differentiation and activity. 
• The function, differentiation and activity of the different subsets of lymphocytes is dependent on nutrient and energy availability. 

The results discussed above provide the basis for the potential of nutrition as a therapeutic tool in inflammatory and autoimmune conditions. In the next posts, I will expose my nutritional recommendations for treating these disorders, offering an up-to-date and detailed science-based nutritional immunotherapy.


* For flow cytometric analysis, specific cell surface markers are used for identifying distinct cell populations. Usually, a cell positive for a given marker is denoted by the marker and a plus sign (+), and a negative cell is written as the marker and a negative sign (-). So, for example, these cells present in FALCs express c-Kit and Sca-1, but not Lin markers.  
** For understanding the importance of these results, I strongly suggest some basic reading on TCR and V(D)J recombination. 

Addendum

In the first part of these series, I ommitted the evidence linking mast cells in the development of obesity and diabetes. I find utterly surprising that common mast cell stabilizers are able to improve these conditions. 

Powell JD, Pollizzi KN, Heikamp EB, & Horton MR (2011). Regulation of Immune Responses by mTOR. Annual review of immunology PMID: 22136167

Winer S, Chan Y, Paltser G, Truong D, Tsui H, Bahrami J, Dorfman R, Wang Y, Zielenski J, Mastronardi F, Maezawa Y, Drucker DJ, Engleman E, Winer D, & Dosch HM (2009).Normalization of obesity-associated insulin resistance through immunotherapy. Nature medicine, 15 (8), 921-9 PMID: 19633657

Procaccini C, Jirillo E, & Matarese G (2012). Leptin as an immunomodulator. Molecular aspects of medicine, 33 (1), 35-45 PMID: 22040697

Ancestral Health Symposium 2012

I have been selected for giving a talk this August, in the Ancestral Health Symposium 2012, held in Harvard Law School. My 20-minute talk will be about Immunometabolism, where I will cover some basics of the immune system, the relationship between metabolism and immunity at a molecular level and how a paleo-type diet, coupled with other approaches, can be a powerful tool for controlling and preventing inflammatory and autoimmune diseases. I will discuss the effects of specific nutrients and food components on immune cell functioning, as well as some dietary suggestions based on individual symptoms. 


Given that Im a post-graduate student travelling from Peru and Im working full time on my thesis, I am looking for some financial help for travel expenses. Additionally, Im looking for some help finding places to stay for a reasonable price. 


If you are interested and find my work helpful, please send me an email to lucas@lucastafur.com for further information. Any help is greatly appreciated. 


Alternatively, you can send any donation through PayPal, to lucas@lucastafur.com. I have eliminated the "Donate" button because PayPal charges a commission that is not charged if you send the money directly to my email from the PayPal website.

Adipose tissue & Immunity: The basics 1

Obesity has been, and its still seen, as primarily a metabolic disease. Obesity results from increased energy intake and decreased energy expenditure, that is, a positive energy balance for a prolonged time. By this logic, to cure obesity and associated diseases, one must restrict calories and/or do more exercise. While this approach works (calorie restriction being the key player), it does not solve the cause of obesity in the first place. This has been shown by numerous evidence that finds that even after weight loss, there is still some metabolic dysfunction in previously obese people. Some obese people can't fully recover. This underscores a common problem in modern health science: clinicians and health practitioners only focus on the proximate cause and not in the ultimate cause (for an interesting read on the subject, see this article). Moreover, obesity is just the tip of the iceberg. We have evolved mechanisms to prevent the development of a rather unadvantageous phenotype. Obesity occurs when these mechanisms start to fail, such as when pathological insulin resistance and leptin resistance develop. 

The obvious cause of obesity is the storage of excess energy as fat tissue. In this manner, excess energy causes an increased fat mass and problems start to arise due to the accumulation of excess body fat. While this statement is true, there is recent evidence that suggests that energy excess has also peripheral effects in cells that were previously unrelated to obesity; in particular, immune cells. 

Immunometabolism

Immunometabolism refers to "the interplay between immunological and metabolic processes" (1). Traditionally, the immune system and metabolic processes have been viewed as different, non-related systems. Now, research findings suggest that this is not the case, on the contrary, both are very related and understanding their interplay is essential for preventing and treating metabolic disorders.

The hypothesis that the immune system is involved in the pathogenesis of obesity started from the findings that targeting proteins which are part of inflammatory cellular pathways ameliorated or prevented the development of obesity and insulin resistance. For instance, Uysal, et al. (2) showed that a null mutation of the TNF-alpha and its two receptor genes improved insulin sensitivity in diet-induced obesity and in ob/ob mice. This confirmed previous in vitro evidence linking TNF-alpha with insulin resistance in adipocytes (3). Interest began to increase with the discovery and characterization of adipocytokines, as well as the finding that the adipose tissue secretes inflammatory cytokines. Adipocytokines are cytokines produced mainly (but not exclusively) in the adipose tissue, and include adiponectin, leptin, resistin and visfatin; being the first two the main adipocytokines produced. Other cytokines secreted by adipocytes are TNF-alpha, IL-6, IL-1 and CCL2; as well as other proteins, including PAI-1 and some complement factors. The table below shows some immune and metabolic effects of the main cytokines discovered, as well as their potential role on inflammation (modified from Tilg & Moschen, 2006). 

Adipocytokine	Effect on inflammation	Levels in obesity	Immunity	Metabolism
Adiponectin	Anti-inflammatory	Decreased	↓NFκB ↓TNF ↓Phagocytic activity (macrophages) ↑IL-10 ↑IL-1RA ↓IFN-γ	↓Hyperglycemia ↓FFA ↑Insulin sensitivity ↑β-oxidation ↓SREBP1c ↑AMPK
Leptin	Pro-inflammatory	Increased	↑TNF ↑IL-6 ↑IL-12 ↑CCL2 ↑Th1 (IL-2, IFN-γ) ↓Th2 (IL-4) ↑ROS ↑Chemotaxis ↑NK-cell function ↑Lymphopoiesis ↑Thymocyte survival ↑T-cell proliferation	↑Energy expenditure ↑Satiety ↑Insulin sensitivity
Resistin	Pro-inflammatory	Increased	↑NFκB ↑TNF ↑IL-6 ↑IL-1 ↑IL-12 ↑CCL2 ↑VCAM1 ↑ICAM1	↑Hepatic insulin resistance
Visfatin	Pro-inflammatory	Increased	↑IL-6 ↑IL-8 ↑IL-1β ↑TNF ↑ICAM-1 ↑IL-10, IL-1Ra (high concentrations)	↓Insulin resistance

The adipose tissue as an immune organ

Besides the role of pro-inflammatory cytokines produced by the adipose tissue, research has shown that obesity alters the function of several immune cells. In an elegant study, Caspar-Bauguil, et al. (4) found that immune cells are present in adipose tissue from mice, but their characteristics are different from other tissues, sharing some common ancestral features with hepatic immune cells. Specifically, the adipose tissue (levels of specific cell populations varying in different anatomical sites) shows both innate and adaptive features, such as the presence of Natural Killer cells (NK), NKT cells and delta-gamma T cells for the former and the presence of lymph nodes, B-cells and alpha-beta T-cells for the latter. This led the authors to propose that the adipose tissue (specially the epididymal adipose tissue) is an ancestral immune organ, due to the fact that delta-gamma T cells are thought to represent an evolutionary and functional bridge between the innate and adaptive immune systems. In addition, inguinal fat contained more adaptive immune cells. Not surprisingly, inducing obesity produced some changes in immune cells: NK cells in epididymal fat were decreased, whereas delta-gamma T cells were increased in inguinal fat and lymph nodes. 

The adipose tissue has site-specific properties and adipocytes interact in a paracrine fashion with adjacent lymphoid cells. Adipocytes near a lymph node are called "perinodal", and show differences from adipocytes far from lymph nodes (5, 6):

• They are smaller and size increases in a gradient manner from the node.
• Lipids extracted from perinodal adipose tissue contain proportionately more PUFAs and less SFA than those further from nodes or nodeless depots, proportion which is not significantly affected by diet (perinodal adipose tissue still has more PUFAs than nodeless adipose tissue). 
• Perinodal adipocytes influence the lipid composition of dendritic cells (and other lymphoid cells) found in lymph nodes, which suggests that perinodal adipocytes provide energy to immune cells for their activity.
• Following chronic immune stimulation, ratios of omega-6/omega-3 PUFA converge in perinodal adipocytes, probably for providing more substrates for ecosanoid and docosanoid synthesis.  
• Perinodal adipocytes are very sensitive to cytokines and noradrenaline, compared to adipocytes from other sites. 

Diet can influence the activity of perinodal adipocytes and associated immune cells. For example, Mattacks, et al. (7) compared the effect of feeding beef suet (mostly saturated and monounsaturated fat), sunflower oil (mostly omega-6 PUFA) and fish oil (mostly omega-3 PUFA) on the response of mesenteric, omental, popliteal and perirenal adipocytes to experimentally-induced local inflammation in guinea pigs. They found that basal lipolysis from sunflower oil-fed pigs was higher and lipolysis from perinodal adipocytes after incubation with noradrenaline was increased, compared with the other groups. The same authors found that the addition of sunflower seed oil (20%) to chow increased the number of dendritic cells in all adipose tissue samples, after stimulation with LPS (8).

Infiltration of immune cells to adipose tissue is now an accepted phenomenon during obesity. It seems that CD4+ T lymphocytes are recruited to adipose tissue first, coinciding with the appearance of glucose intolerance and reduced insulin sensitivity, while macrophages accumulate at late stages of obesity-induced insulin resistance (9, 10). Infiltration of B-cells occur rapidly in mice, before any significant change in body fat mass (10).

Innate immunity and adipose tissue

As has been mentioned, some adipose tissue show the presence of innate immune cells. One striking fact is that adipocytes and macrophages show similar characteristics. Weisberg, et al. (11) found that the expression of 1,304 transcripts in perigonadal adipose tissue from different mice correlated significantly with body mass. Of the 100 most significantly correlated genes, 30% encoded macrophage specific proteins. In mice, the adipose tissue is a major source of IL-6 during systemic inflammation produced by LPS (12). The tight relationship between adipocytes and monocytes/macrophages is exemplified by C3a. After activation of the alternative complement pathway, C3a induces mast cell degranulation and an immune response. This protein is also produced by adipocytes and the N-terminal cleavage of its alpha chain through the interaction of complement factors B and adipsin, followed by C-terminal arginine cleavage by serum carboxipeptidase N produces acylation stimulating protein (ASP) or C3adesArg, which is an important regulator of triglyceride synthesis. Moreover, C3a (ASP precursor) can also have metabolic effects: its receptor, C3aR, is expressed on both monocytes-macrophages and adipocytes. C3aR null mice are transiently resistant to diet-induced obesity, and are protected from diet-induced insulin resistance and hepatic steatosis, showing improved insulin sensitivity compared to wild-type mice (13). This was accompanied by a decrease in macrophage infiltration to adipose tissue, plasma cytokine levels and a polarization of macrophages towards a M1 phenotype (see below).

Toll-like receptors (TLR) are pattern recognition molecules with an essential function recognizing pathogens via pathogen-associated molecular patterns (PAMPs). Several types of TLRs are expressed by pre- and mature murine adipocytes, but mature adipocytes seem to be more responsive to a broader spectrum of TRL ligands (14). LPS triggers the secretion of IL-6 and different chemokines (CCL2, CCL5 and CCL11) and this inflammatory response appears to be based mainly on preadipocytes. The ultimate result of the activation of TLRs in adipocytes is the secretion of inflammatory cytokines via activation of NFkB signaling. Accordingly, LPS increases the expression of TLR2, TRAF-6 and NFkB in human adipose tissue, and increased levels of these markers, as well as LPS, is observed in type 2 diabetic patients (15). The finding that mice with defects on different TLRs are protected from obesity and insulin resistance supports the role of TRLs in the development of metabolic dysregulation (16, 17, 18, 19). Additionally, it suggests that different inflammatory stimuli act in the adipose tissue via different TLRs.

Macrophages infiltrating the adipose tissue can have two potential sources: those differentiated from bone-marrow-derived monocytes which reach the adipose tissue from the systemic circulation or by trans-differentiation from local adipose tissue preadipocytes and mesenchymal stem cells (14). Diapedesis of monocytes is stimulated by chemoattractants secreted by adipocytes (CCL2, CCL5, MIF and MIP1a) and locally produced macrophage colony stimulating factor (M-CSF) supports differentiation and maturation of monocytes into macrophages. On the other hand, both adipocyte and macrophage differentiation and function is controlled by PPAR-gamma. In addition, adipocytes also express macrophage-specific genes (20). This suggests that both cell types arise from a common precursor cell, and trans-differentiation of adipocytes into macrophages is supported by the findings of Charriere, et al. (21): preadipocytes are able to convert into macrophage-like cells, judging by specific antigens and the phagocytic index. Similar findings have been observed in other studies (22, 23), which report that preadipocytes can phagocyte and kill micro-organisms. 

Macrophages can show different activities depending on their phenotype. Classically activated macrophages (M1) respond to products derived from or associated with bacterial infections, like LPS and IFN-gamma. These macrophages are characterized by a highly inflammatory phenotype, displaying high phagocytic and bactericidal potential. Alternatively activated macrophages (M2) are induced in response to products from or associated with parasitic infections, such as Schistosoma egg antigen and Th2-type cytokines like IL-4 and IL-13 (24). M2 stimulate tissue repair and remodeling. Contrary to what was thought, adipose tissue from lean animals does have macrophages. However, obesity, besides promoting infiltration and migration of macrophages, induces a shift in macrophage balance towards M1 phenotype (25). In fact, obesity shifts the adipose M2:M1 ratio  from 4:1 in normal mice to 1.2:1 (26). Th2-type cytokines derived from the adipose tissue (IL-13 and IL-4) regulate macrophage polarization, favoring alternative activation (27). M2 development is also promoted by IL-10 (28). 

The role of other innate cells is not well characterized. It has been seen that peripheral natural killer (NK) cell levels in unhealthy obese patients is reduced compared with healthy obese, and NK cells from these patients show increased levels of inhibitory markers (CD158b and NKB1), but expression of CD69, a marker of NK activation (29). This suggests that although activated, NK cells from unhealthy obese patients cannot function properly. In visceral adipose tissue from obese subjects, there is an increase in NK cells compared to subcutaneous fat (30). Another type of innate cells, natural killer T (NKT) cells , which are T-like innate cells capable of producing both Th-1 and Th-2 type cytokines have been implicated in the inflammatory environment seen in obesity. Mice lacking NKT cells show reduced macrophage infiltration in response to a high-fat diet, and activation of NKT cells by alpha-galactosylceramide exacerbates glucose intolerance, macrophage infiltration and cytokine gene expression in diet-induced obese mice (31). However, the effects of NKT cells on obesity and insulin resistance seem to be dependent on CD8+ T cells (32). 

Summary

Interrelationship between the adipose tissue and the immune system. The adipose tissue (AT) secretes  adipocytokines (adiponectin, resistin, leptin, visfatin, among others) and conventional cytokines (TNF-a, IFN-gamma. IL-6, etc.), which influence metabolism and immunity. Cells from the innate and adaptive immune system are present in AT, and are fueled by perinodal adipocytes. Adipocytokines influence the function of immune cells, and cytokines secreted both by adipocytes and immune cells regulate the inflammatory milieu of the AT. Nutrition, by regulating fat mass and lipid composition, has direct effects on the function of immune cells.

Schäffler A, & Schölmerich J (2010). Innate immunity and adipose tissue biology. Trends in immunology, 31 (6), 228-35 PMID: 20434953 

Pond CM (2005). Adipose tissue and the immune system. Prostaglandins, leukotrienes, and essential fatty acids, 73 (1), 17-30 PMID: 15946832