Leptin is the principal regulator of body weight and is produced by adipose tissue when body energy needs are met. It acts on specific hypothalamic nuclei, inducing secretion of an anorexigenic neuropeptide, pro-opiomelanocortin, and of the suppressing orexigenic neuropeptide Y (62, 63). This makes it the most important regulator of body weight, producing satiety and inducing energy expenditure, its levels correlating with body adipose mass and BMI (64). Encoded by the LEP gene (homologous to the ob gene in mice) leptin is mainly produced by adipocytes and secreted into circulation. Its activity is exerted through a receptor, a member of the Class I cytokine family (LEPR or Ob-R) (65). Six isoforms del receptor LEPR with different physiological roles have been identified (66). After binding to its receptor, leptin activates a tyrosine kinase, Janus kinase 2 (JAK2), subsequently leading to phosphorylation of an extracellular signal-regulated kinase (ERK), and the signal transducer and activator of transcription (STAT)3 and STAT5. Later activation of cytokine signaling-3 (SOCS3) results in negative feedback which inhibits the LEPR (67). Notably, LEPR is expressed in CD4⁺ T cells, CD8⁺ T cells, Treg cells and NK cells (68–70). Aside from playing an important role in energy homeostasis, leptin is a potent modulator of the immune response, playing a significant role in regulating leukocyte extravasation in the CNS. Leptin receptor blockade prevents migration of immune cells into the CNS, attenuating EAE progression (71). Moreover, starvation produces a significant decrease in leptin levels, associated in turn with infection, as a result of a dysfunctional immune response, and reversed by exogenous administration of leptin (72). Leptin acts directly on macrophages, enhancing their proliferation and phagocytic activity, as well as the secretion of pro-inflammatory cytokines such as TNF-α, IL-1 and IL-6 (73, 74). Likewise, leptin stimulates proliferation of Th1 cells as well as production of IFN-γ and IL-2, suppressing production of IL-4 and IL-10. Additionally, in rat microglia, leptin induces the expression of IL1-β and TNF-α and enhances LPS effects (75). Leptin also inhibits proliferation of Treg cells and induces hypo-responsiveness (76), effects dependent on increased activity of the mammalian target of rapamycin (mTOR) signaling pathway. Transient inhibition of mTOR by rapamycin prior to TCR stimulation, makes Treg cells proliferate in the absence of IL-2, and recover their inhibitory capacity (69, 77).

Overall, these findings suggest leptin, under certain underlying metabolic conditions coinciding with the loss of immune self-tolerance, can modulate the immune response towards a more pro-inflammatory profile. Leptin-deficient (ob/ob) mice show increased susceptibility to infections, and resistance to induction of both active and passive EAE, associated in turn with progressive decline in autoreactive CD4⁺ T cell survival and reduced IFN-γ and IL-17 production (68). These effects were combined with the down-regulation of protein Bcl-2 survival and cell cycle arrest, determined by reduced degradation of p27 kip1 and impaired signaling of energy-sensing AKT-mTOR pathways (78). Administration of leptin to EAE-susceptible mice (no leptin deficient) worsened the clinical course, while anti-leptin-receptor antibodies ameliorated it (79). Also, prior to the onset of neurological symptoms following EAE induction, animals typically lost body weight, and show a marked increase in serum leptin levels. Immunohistochemical analysis has revealed parallel, in situ production of leptin in inflammatory infiltrates and in neurons, occurring only during acute/active phases of EAE (80). On the other hand, starvation delayed disease onset and attenuated clinical symptoms (80). In line with these results, studies in RRMS patients demonstrated increased levels of leptin in serum and cerebrospinal fluid (CSF), and transcriptional analysis of MS brain lesions showed increased leptin expression at sites of inflammation (81). These findings coincided with increased CSF IFN-γ concentrations, and reduced number of circulating Treg cells (82). Serum leptin levels and EDSS results have shown positive correlation in SPMS and PPMS patients (83, 84). Interestingly, in young boys with MS presenting increased leptin levels, longer periods between relapses were observed. Whereas in girls, a positive link between higher leptin levels and increased disability scores was recorded, suggesting varying, gender-specific leptin effects (61). In patients experiencing RRMS exacerbations, increased leptin receptor expression has been reported on CD8+ T cells and monocytes, compared to those in patients in remission or healthy controls (85). In summary, although several studies have found increased leptin levels in MS patients, signaling the hormone could somehow influence burden of disease, it is also important to note they are not strictly comparable, as many did not include important variables affecting the outcome (BMI, gender, or use of DMTs).

Resistin is produced during adipocyte differentiation and was initially described as an adipokine, linking obesity to insulin resistance (86). Resistin increases the expression of cytokines and adhesion molecules in murine vascular endothelial cells, and has been associated with atherogenesis (87). Notably, resistin is mainly produced by adipocytes in rodents, whereas in humans it is produced for the most part by macrophages, and peripheral blood monuclear cells (PBMCs; 91). Human and rodent resistin share only 59% amino-acid identity (88), but their function is similar, despite differences in production source. Resistin increases production of TNF-α, IL-1β, and IL-6, in PBMCs, and macrophages. Conversely, exposure of PBMCs to TNF-α, but not IL-6 or IL-1β, induces expression of resistin (89, 90), further enhancing its own activity through positive feed-back. It also promotes expression of cell adhesion molecules, including VCAM-1, ICAM-1, and MCP-1, and the chemokine (C-C motif) ligand 2 (CCL2), contributing to chemotaxis and recruitment of leukocytes to inflammation sites (87, 91–93). As mentioned above, resistin induces its own production in PBMCs through positive feedback, which is why different studies have highlighted its pro-inflammatory role, by binding to TLR4 and activating NF-κB in human macrophages. In mice hypothalamic cells, its action is exerted via c-Jun N-terminal kinase (JNK) p38 and mitogen-activated protein kinase (MAPK) pathways (94). Given its role in insulin resistance and inflammation, resistin is likely to explain, at least in part, the relationship between the inflammatory processes observed in obesity and metabolic diseases.

Increased serum levels of resistin, leptin, and visfatin, as well as decreased expression of forkhead box P3 (FoxP3) mRNA in T cells,have been observed in MS patients, correlating with circulating levels of TNF-α, and IL-1β (95), as well as with disability progression (96), indicating resistin could be an important driver of chronic inflammation. These results support the view that resistin plays a role in the pathogenesis of MS, although additional studies are required to confirm the findings.

Visfatin is highly expressed by both leukocytes and adipocytes. It is the rate-limiting step in a salvage pathway of nicotinamide adenine dinucleotide (NAD; 101), and in this manner regulates intracellular metabolism. Visfatin binds to the insulin receptor, triggering insulin-mimetic activities. However, it also binds to TLR4, inducing a pro-inflammatory response (97). Visfatin synthesis is promoted by IL-6 and TNF-α (98, 99), while the PPAR-γ agonist rosiglitazone inhibits synthesis by adipocytes (98).

Visfatin exerts pro-inflammatory effects through different mechanisms. In both humans and mice, it induces synthesis of TNF-α, IL1β and IL-6 (100) and acts as a chemotactic factor for monocytes and lymphocytes, by upregulating the chemokines CCL2, CXCL2 and CXCL8 expression in endothelial cells, as well as expression of adhesion molecules like ICAM-1, and VCAM-1 (101). Visfatin also mediates LPS-induced synthesis of IL-6, IL-1β, inducible nitric oxide synthase (iNOS), nitric oxide and reactive oxygen species (ROS) in microglial cells (102). Moreover, it promotes activation of T cells by inducing the expression of co-stimulatory molecules CD54, CD40 and CD80 on monocytes (100). Finally, in mouse endothelial cells, visfatin downregulates expression of tight junction-associated proteins such as zonula occludens 1 and 2, cadherin, and occludin, and therefore increases BBB permeability (103).

In EAE, the inhibition of visfatin after symptom onset significantly reduces disability and demyelination of the spinal cord. Furthermore, in vitro studies have demonstrated suppression of T cell proliferation and depletion of T cells, probably through induced inhibition of NAD+ and subsequent ATP depletion (104). In MS patients, levels of visfatin are increased, particularly in RRMS cases, correlating positively with TNF-α and IL-1β levels and negatively with mRNA FoxP3 expression in T cells. Circulating levels of visfatin in RRMS patients are significantly higher compared to those observed in SPMS and PPMS patients (95).

Chemerin, an adipokine synthesized as a precursor in the liver and in adipocytes, regulates adipocyte differentiation, and is associated with obesity and metabolic syndrome (105). Different chemerin isoforms exert contrasting effects on immune cells, making it difficult to establish their exact role in autoimmune responses. It has been identified as a chemoattractant for plasmacytoid dendritic cells and macrophages, activated via TLR9 and alarmin high-mobility group box 1 (HMGB1) factor, to produce type I interferons, which in turn induce pro-inflammatory polarization of adipose tissue (106). In addition, chemerin can also bind to the chemokine receptor CCRL2, which when deficient, promotes macrophage infiltration of adipose tissue and accelerates insulin resistance (107). In the CNS, it is expressed on endothelial cells in the meninges as well as in white matter MS lesions, whereas its receptor, the chemerin chemokine-like receptor 1 (CMKLR1), a G protein-coupled receptor, is expressed mainly on infiltrating lymphocytes, dendritic cells and macrophages. These observations suggest chemerin is involved in processes leading to migration of peripheral cells into the CNS, contributing to the inflammatory process. Indeed, CMKLR1+ leukocytes, and dendritic cells can be found in the leptomeninges and perivascular cuffs of active and chronic MS lesions (108). Chemerin is up-regulated in the spinal cord of EAE animals, and CMKLR1 KO mice develop less severe EAE, failing to induce EAE adoptive transfer (109, 110). Similarly, 2-(alpha-Naphthoyl) ethyltrimethylammonium iodide (α-NETA), a CMKLR1 antagonist, inhibits infiltration of the CNS by inflammatory cells in EAE animals, reducing myelin damage, and delaying onset of disease. It also alters leukocyte distribution in peripheral lymphoid organs (111). Collectively, these findings suggest chemerin is a pro-inflammatory adipokine, possibly contributing to inflammation in obese MS subjects. Increased levels of chemerin have been associated with obesity and excess weight in patients with MS compared to non-obese MS patients and healthy controls, particularly females (112, 113). Conversely, more recent studies have shown that although obese MS patients also present insulin resistance, chemerin levels are not influenced by BMI, nor are chemerin levels related to disease progression, or cognitive dysfunction (114). It has been postulated that chemerin competes with resolvin (an anti-inflammatory molecule synthesized from ω-3 polyunsaturated fatty acids) for binding to the CMKLR1 receptor, a factor promoting resolution of the anti-inflammatory process (115). In the initial stages of inflammation, chemerin stimulates different immune cells on site. During the resolution phase, resolvin activates the CMKLR1 receptor, and macrophages increase production of IL-10, generating an anti-inflammatory profile (116). This mechanism may represent a chemerin/CMKLR1/resolvin control loop, through which chemerin regulates its pro- and anti-inflammatory effects.

Adipocyte-fatty acid binding protein 4 (A-FABP; also known as FABP-4 and adipocyte protein 2, aP2) is produced by adipocytes, monocytes and macrophages, and its expression is enhanced by TLR2 stimulation (117). FBPA4 regulates lipolysis and FBPA4 deficiency diminishes pro-inflammatory cytokines, via attenuation of IKK-β/NF-κB pathway (118). Conversely, recombinant FBPA4 promotes secretion of pro-inflammatory cytokines in adipocytes via the p38/NF-κB pathway (119), supporting its pro-inflammatory role in obesity. Given that FBPA4 is strongly regulated by lipolysis, it could function as a specific lipid sensor in adipocytes, transporting certain plasma lipids to specific organs, including the CNS.

FBPA4 knockout mice exhibit reduced clinical symptoms of EAE and impaired pro-inflammatory cytokine production by dendritic cells (120, 121). Likewise, in MS, FBPA4 has been associated with increased disability (122). Indeed, high plasma levels have been found in SPMS, suggesting a possible pathogenic role for this protein (119). Increased levels of FBPA4 have also been reported in pediatric-onset MS (61, 119). Interestingly, a positive correlation between FBPA4 and leptin in pediatric RRMS patients was reported, suggesting that in the initial stages of the disease, both adipokines play a role in early inflammation as well as in later progression (119). Higher FBPA4 levels have been observed in females, as in the case of leptin (61), and differential expression of microRNAs of FABP4 has been considered a prognostic biomarker of RRMS (123).

Adiponectin is the most abundant circulating adipokine. It regulates glucose and lipid metabolism through fatty acid oxidation and inhibition of gluconeogenesis via activation of the AMP-activated kinase (AMPK) pathway (124). The latter in turn activates PPAR-γ and PPAR-α signaling. Adiponectin is produced as a monomer, circulating as three different oligomers of varying molecular weight (125). Adiponectin mediates its actions through 3 receptors: AdipoR1, which is located predominantly in skeletal muscle; AdipoR2, expressed mainly in the liver; and T-cadherin, which mediates its actions in the cardiovascular system (126). AdipoR1 and AdipoR2 are also found on monocytes, human B, and NK cells, but only on a small percentage of T cells (127). Adiponectin binding to AdipoR1 promotes activation of the AMPK pathway, while AdipoR2 activates the PPARα pathway (128). Contrary to leptin, adiponectin is decreased in obese individuals, and increases with weight loss (129). While leptin induces pro-inflammatory activity, adiponectin exhibits anti-inflammatory effects on the cells of the immune system. At the endothelial level, adiponectin significantly reduces expression of: TNF-α-induced ICAM-1, VCAM-1, and endothelial-leukocyte adhesion molecule (E-selectin), limiting leukocyte rolling and adhesion (130, 131). In contrast to leptin, adiponectin inhibits maturation, proliferation and phagocytic activity of macrophages, as well as synthesis of IFN-γ and TNF-α after stimulation with LPS (132–134). Its anti-inflammatory effect is mediated by increased production of IL-10 (particularly by Treg cells), as well as of the IL-1 receptor antagonist, and of TGF-β release by monocytes, dendritic cells and macrophages. It also increases Treg cell numbers (134, 135). Notably, overexpression of adiponectin is associated with induction of Th2 and Treg responses via PPRA−γ pathway activation, consequently amplifying IL-4 and IL-10 production (136). In addition, anti-inflammatory effects of adiponectin also include enhancement of apoptosis and inhibition of proliferation of antigen-specific T-cells, and inhibition of IL-2-induced NK cell cytotoxic activity (137, 138). However, adiponectin also appears to exert pro-inflammatory effects, activating dendritic cells, and leading to Th1 and Th17 polarization (120). This dual activity (anti- and pro-inflammatory) might depend on specific levels of different circulating isoforms.

In EAE, adiponectin-deficient mice develop more severe disease. T cells from these animals produce greater amounts of TNF-α, IFN-γ, IL-17, and IL-6, and fewer Treg cells, suggesting adiponectin exerts anti-inflammatory effects in vivo. Administration of exogenous adiponectin increases Treg cell numbers, ameliorating disease severity (139), whereas calorie-restricted diets decrease EAE severity and are associated with higher plasma adiponectin levels and lower concentration of leptin (140).

Some studies found lower serum levels of adiponectin in males with RRMS compared to healthy controls, with no difference found in females, suggesting adiponectin levels are gender-dependent (61, 101). Conversely, in other cohorts, a predominant decrease in adiponectin was found only in females (141). Some of these discrepancies can probably be explained by differences in age of the populations studied. In studies conducted in twins, higher levels of adiponectin were detected in the CSF of symptomatic MS siblings during remission, compared to the asymptomatic twin. CSF levels did not correlate with serum levels, suggesting either intrathecal synthesis or increased transport across the BBB following enhanced systemic production (142). Unfortunately, these results could not be reproduced by other authors. Larger, more extensive studies will be needed to clarify the real significance of adiponectin in MS.

Some authors observed the presence of high and middle molecular weight isoforms of adiponectin at the time of MS diagnosis, which were associated with a greater risk of disease progression and severity (143, 144) subsequent to IL-6 secretion by monocytes (145). Oligomerization and secretion of adiponectin during inflammation can be modulated by TNF-α (146). Therefore, depending on the isoform, adiponectin may exert pro- or anti-inflammatory effects. All these findings contribute to keeping the discussion open on the significance of adiponectin as a biomarker in MS.

Apelin plays an important role in macrophage function and development in different tissues. In rodents, it down-regulates IL-6, TNF-α and MCP-1 expression in macrophages, as well as macrophage inflammatory protein (MIP)1, ROS formation, and phagocytic activity (147–149). In N9 microglia cell lines, it decreases LPS-induced iNOS and IL-6 production, while up-regulating production of IL-10 and Arginase 1 (150). Overall, apelin is considered an anti-inflammatory adipokine. Even though apelin has not yet been studied in EAE, it has been reported to suppress neuroinflammation in experimental ischemic stroke, by suppressing microglia recruitment and reducing ROS production (151, 152). Neuroprotective effects of apelin on brain ischemia appear to be mediated by the nuclear erythroid 2-related factor (Nrf-2), which is also the target of dimethyl fumarate, a treatment shown to reduce MS relapse rates and disability progression (152). In MS patients, plasma levels of apelin were significantly lower compared to healthy controls, showing a statistically significant negative correlation with disability scores and the number of relapses (153). In addition, apelin and its APJ receptor are widely expressed in the central nervous system, especially in neurons and oligodendrocytes. Notably, in recent studies, apelin/APJ complex expression in oligodendrocytes correlated with age-associated changes in remyelination efficiency through the translocation of myelin regulatory factors (154). Apelin promotes the differentiation of neural stem cells (155) and can therefore represent not only an anti-inflammatory factor but also contribute to repair processes observed during the course of MS.