Impact of obesity on autoimmune arthritis and its cardiovascular complications

Abstract

Obesity can instigate and sustain a systemic low-grade inflammatory environment that can amplify autoimmune disorders and their associated comorbidities. Metabolic changes and inflammatory factors produced by the adipose tissue have been reported to aggravate autoimmunity and predispose the patient to cardiovascular disease (CVD) and metabolic comorbidities. Rheumatoid arthritis (RA) and psoriatic arthritis (PsA) are autoimmune arthritic diseases, often linked with altered body mass index (BMI). Severe joint inflammation and bone destruction have a debilitating impact on the patient’s life; there is also a staggering risk of cardiovascular morbidity and mortality. Furthermore, these patients are at risk of developing metabolic symptoms, including insulin resistance resulting in type 2 diabetes mellitus (T2DM). In addition, arthritis severity, progression and response to therapy can be markedly affected by the patient’s BMI. Hence, a complex integrative pathogenesis interconnects autoimmunity with metabolic and cardiovascular disorders. This review aims to shed light on the network that connects obesity with RA, PsA, systemic lupus erythematosus and Sjögren’s syndrome. We have focused on clarifying the mechanism by which obesity affects different cell types, inflammatory factors and traditional therapies in these autoimmune disorders. We conclude that to further optimize arthritis therapy and to prevent CVD, it is imperative to uncover the intricate relation between obesity and arthritis pathology.

1. INTRODUCTION

Arthritis spans a range of pathogeneses and clinical manifestations, most commonly referring to osteoarthritis (OA), but also including other pathologies including rheumatoid arthritis (RA) or psoriatic arthritis (PsA). In the US alone, 54 million adults are affected by this progressive and debilitating disease [1]. It was found that the excessive joint inflammation in arthritis patients is often caused by underlying autoimmunity, as is believed to be the case in RA and PsA. Key players in these arthritic disorders are T helper 1 (Th1) and Th17 cells as well as the classical M1 macrophages and their associated cytokines such as tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6) and IL-23 [2–4]. In contrast, OA is predominantly considered a symptom of wear and tear, causing severe bone destruction combined with low-grade inflammation. OA results in progressive degeneration of articular joints, characterized by breakdown of joint homeostasis favoring the catabolic processes [5]. In comparison to RA and PsA, OA displays differences in risk factors, patient demographics, disease presentation and cytokine profile, yet, there are also common denominators.

Although mechanical factors potentiate OA progression, the primary risk factors for OA onset are believed to be body mass index (BMI) and fat tissue mass [6, 7]. Others postulate that OA can be triggered by acute injury, excessive mechanical overloading, obesity and type 2 diabetes mellitus (T2DM). A recent study demonstrated that in OA, patients with elevated BMI (≥25) have a 3-fold greater chance of cartilage damage compared to those with normal BMI [8]. Furthermore, dysfunction of the adipose tissue leads to potentiated levels of IL-1β, IL-6 and TNF-α [9, 10], cytokines which promote systemic inflammation. The adipose tissue also produces a unique set of so-called adipokines, including leptin. Leptin is an inflammatory adipokine that is believed to play an essential role in OA joint destruction. Consistently, plasma levels of leptin strongly correlate with BMI of OA patient, while the synovial levels are associated with OA severity [11, 12]. In OA, patients diagnosed with metabolic syndrome have higher levels of leptin in plasma and synovial fluid than those without metabolic syndrome [13]. In contrast, the protective anti-inflammatory adipokine; adiponectin, was reduced in OA patients with metabolic syndrome. Consistent with these findings, others have shown that weight loss decreases the risk of OA [7, 14]. Interestingly, there is also evidence that mechanical stress can lead to secretion of inflammatory cytokines within the joints [15]. Intra-articular biomechanical changes can activate inflammatory signals in the joint tissue and cartilage [16].

In addition to OA, obesity plays a critical role in autoimmune arthritic pathologies [17]. Incidence of obesity was reported to be 18% in the general population compared to 27%, 29% and 37% in RA, psoriasis and PsA, respectively [18]. Following a similar trend, average BMI is also highest in PsA at 29.6 kg/m² followed by 27.9 and 27.3 kg/m² in psoriasis and RA respectively compared to only 26.1 kg/m² in a healthy population [18]. In this review we summarize how RA and PsA pathogenesis are manipulated by obesity. We also recognize that additional studies are required to unravel the complex relationship between obesity and autoimmune inflammatory responses.

2. RA and OBESITY

2.1. Cardio-metabolic comorbidities associated with obesity in RA

2.1.1. Genetic and environmental risk factors for cardiovascular disease in RA

RA is a chronic autoimmune disease that encompasses joint inflammation and bone destruction leading to severe disability if untreated. In the US, approximately 1% of the population or 2.5 million people have RA and their life span is reduced by 5 to 15 years compared to healthy individuals [19–21]. Cardiovascular disease (CVD) is reported to be the primary cause of death in RA, which is 3-fold more prevalent in RA than in the general population [22, 23]. Elevated risk of CVD in RA is due to both genetic and environmental risk factors (Table 1). Susceptibility genes associated with developing RA, are among others, HLADR4, DRβ1, protein tyrosine phosphatase 22 (PTPN22) and peptidyl arginine deiminase (PAD)4 [24]. Additionally, environmental risk factors include infections, smoking, periodontitis, hormonal and dietary factors, physical inactivity, T2DM and obesity [25].

Table 1:

Risk factors, pathogenic markers and involved cell types in rheumatoid arthritis

Susceptibility genes	Environmental risk factors	Cells involved in RA pathogenesis	Other common pathogenic factors
HLA DR4 MHC IRF5/8 CSF2 CTLA4 CD28 PAD14 CD40 PTPN22 TYK2 CCL21 STAT4 TNFAIP3 TRAF1 IL-6R IRAK1 IL-2/IL-21 IL-2RB C5	Sex Age Smoking Hypertension Hyperlipidemia Metabolic Syndrome Family history Periodontitis Gut microflora Infection Stress Air Pollution	Neutrophils Monocytes, macrophages and foam cells Adipocytes Th1/Th17 cells B/plasma cells Endothelial cells Smooth muscle cells Platelets	Acute phase protein (CRP and fibrogen) Autoantibodies (α-CCP, α-RF, α-Ox-LDL, α-HSP) Inflammatory cytokines (TNF-α, IFNγ, IL-1) Matrix metalloproteinases Endothelial cell adhesion Defective apoptosis

Metabolic syndrome is defined as the coincidence of three or more of the following five characteristics: abdominal obesity, elevated serum triglycerides or fasting glucose levels, reduced HDL cholesterol and hypertension. Pathological BMI and adipose tissue mass is associated with macrophage infiltration into the adipose tissue and thus inflammatory response. Elevated BMI is linked to RA onset and obesity closely correlates with both disease activity and inferior response to treatment [26, 27]. Secondarily, RA patients with severe and chronic joint inflammation more frequently develop CVD than those who exhibit modest and shorter disease. This has been attributed to potentiated thrombogenic and inflammatory factors, oxidative stress markers, blood pressure and vascular dysfunction [28]. Patients with RA also have a greater incidence of metabolic syndrome, which supports an inflammatory and pro-thrombotic environment that triggers CVD and aggravates RA mortality [26, 27]. Markers and mechanisms characteristic of cardio-metabolic comorbidities in RA will be discussed in detail in this review.

2.1.2. Biomarkers of cardiovascular disease in RA

A number of biomarkers have been identified to predict the cardiovascular risk in the general population. These biomarkers include markers for metabolic breakdown, inflammation, endothelial cell dysfunction and vascular growth factors. Some are found to be particularly relevant to the pathogenesis of RA and obesity-induced complications.

Osteoprotegrin (OPG) is a decoy receptor, which prevents receptor activator of nuclear factor kappa-Β ligand (RANKL) binding to its receptor RANK, and thereby suppresses the differentiation of osteoclast precursor cells to mature osteoclasts. OPG is involved in the pathogenesis of CVD by potentiating traditional risk factors such as hyperlipidemia, endothelial activation, T2DM and hypertension. OPG levels also positively correlated with arterial calcification and cardiovascular mortality [29, 30].

Abnormal angiogenesis contributes to CVD pathology, hence, the angiogenic marker angiopoietin 2 (Ang-2) is associated with increased CVD risk in patients with hypertension [31]. Ang-2 levels were markedly higher in RA patients compared to normal individuals [32]. RA patients with left ventricular diastolic dysfunction and patients with increased carotid intima-media thickness (cIMT) in particular displayed higher Ang-2 expression [32].

Asymmetric dimethylarginine (ADMA) is a newly discovered marker of CVD in diseases associated with endothelial cell dysfunction, such as RA [33]. ADMA is a metabolic by-product generated from protein modification processes in the cytoplasm of human cells. ADMA inhibits L-arginine-induced nitric oxide (NO) production, which is essential for normal endothelial cell function and consequently for cardiovascular health [33]. RA disease activity was positively associated with ADMA levels [34]. It has been postulated that inflammation mediated by ADMA impacts NO metabolism and endothelial homeostasis in RA.

Serum uric acid and microalbuminuria are associated with hypertension, renal dysfunction and CVD. Serum uric acid is markedly elevated in RA compared to normal controls [35, 36] and is associated with cIMT [37]. Urinary albumin excretion is potentiated in RA compared to the controls and correlates with increased arterial stiffness, elevated vascular cell adhesion molecule-1 (VCAM-1) and reduced IL-10 levels [38].

2.1.3. Pathogenic changes in RA lipid profile

Several studies have shown that dyslipidemia can be detectable up to 10 years before clinical RA manifestation, though the observed changes are inconsistent among studies [39, 40]. A recent report reveals that levels of total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) are decreased to a greater extent in RA compared to non-RA individuals, 5 years prior to clinical detection [40]. In contrast, there were no significant differences in the concentration of high-density lipoprotein cholesterol (HDL-C) and triglyceride during the 5 years preceding RA diagnosis compared to the general population [40]. Intriguingly, it is shown that upregulation of inflammatory molecules and markers of infection is frequently associated with a concurrent decline in the TC and LDL-C levels [41].

The mechanism by which inflammatory disease modifies the lipid profile is not well understood. It has however been postulated that cytokines and Toll-like receptor (TLR) agonists can directly impact the lipid metabolism [42]. Blockade of TNF-α and IL-6 function in RA patients reverses the lipid profile as detailed in the therapeutic section, suggesting that TNF-α and IL-6 contribute to the reduced levels of cholesterol and LDL-C [25, 42]. TNF-α and IL-6 can also remodel macrophages into foam cells by engulfing modified LDLs, such as oxidized or acetylated LDL [43]. Seemingly contradictory though, high risk for CVD is linked to augmented levels of TC and LDL-C versus suppressed concentration of HDL-C, resulting in an increased TC/HDL-C ratio or atherogenic index as used clinically. The paradoxical reduction of LDL-C levels in RA would insinuate a lower risk of CVD in these patients [44, 45]. However, in addition to changing the absolute levels, inflammation also affects composition and activity of lipoproteins. In RA, small dense LDL particles which have a 3-fold greater ability to cause oxidative damage and atherogenicity than LDL-C, have been detected by NMR spectroscopy [46]. Similarly, in severe RA, HDL-C composition is altered and the modified lipoprotein is incapable of impairing oxidative damage and cholesterol transport [47].

The concentration of this pro-inflammatory HDL-C is associated with fibrinogen, serum amyloid A, haptoglobin and complement factor levels. These functional modifications clarify that, while absolute levels of LDL-C are reduced, its pro-atherogenic function is in fact amplified. Conversely, upregulated levels of HDL-C lose activity. As a result, RA patients are at a greater risk for CVD compared to the general population, despite their reduced atherogenic index. There are also other less established risk factors, such as lipoprotein a (Lpa) that may predict the risk of CVD in RA [44, 45, 48].

2.2. Obesity-associated adipokine production in RA and CVD

The adipose tissue and adipocytes in particular are responsible for producing a specific set of cytokines, collectively referred to as adipokines. These adipokines contribute to our physiological homeostasis. They regulate insulin sensitivity, vascular function and can also modulate our immune system. However, obesity affects adipose protein production. As the balance shifts, the adipose tissue plays an active pathological role that favors the expression of inflammatory adipokines over the protective ones, as has been observed in RA [49]. In this review we will discuss the most commonly studied adipokines which include adiponectin, leptin, resistin, visfatin and chemerin [26, 27, 49].

2.2.1. Adiponectin

Adiponectin has been described to have anti-inflammatory effects in individuals with metabolic syndrome, yet it can be pro-inflammatory in the context of RA [49]. This apparent discrepancy may be cell-related, depending on the overall inflammatory environment, or it could in part be regulated by posttranslational modification of the adipokine [50]. Adiponectin can suppress TNF-α-induced expression of adhesion molecules on endothelial cells [49]. Additionally, adiponectin can impair NF-κB signaling in endothelial cells as well as myeloid cell phagocytic activity and TNF-α production in response to LPS [51]. In LPS-stimulated human macrophages, adiponectin can activate IL-10 and IL-1RA and suppress interferon γ (IFNγ) secretion [52].

Adiponectin can also regulate insulin resistance and protect ApoE−/− mice from atherosclerosis. Consistently, adiponectin has shown to blunt macrophage activation both in adipose tissue and liver [53]. In contrast, adiponectin promotes production of inflammatory mediators (IL-6, IL-8), proangiogenic factors (vascular endothelial growth factor; VEGF) and matrix metalloproteinases (MMP-1/13) from RA synovial tissue fibroblasts [54]. Plasma levels of adiponectin in RA correlated with C reactive protein (CRP), an indicator of RA severity and CVD risk, and were associated with radiographic damage, further substantiating the impact of adiponectin on RA inflammation and bone destruction [55].

2.2.2. Leptin

Leptin is mainly produced by adipocytes and has structural similarities to IL-6, IL-12 and granulocyte macrophage colony-stimulating factor (GM-CSF) [56]. Leptin controls the appetite and its serum levels correlate with adipose mass in mice and humans [56]. Leptin levels in RA sera as well as the synovial fluid/sera leptin ratio are significantly upregulated in RA patients with more erosive disease [57]. Concentration of leptin is closely associated with RA disease duration and 28-joint disease activity score (DAS28). Leptin impacts neutrophil, myeloid and T cell function. Neutrophil chemotaxis and release of reactive oxidative species (ROS) from these cells is promoted by leptin [58].

Leptin also potentiates secretion of TNF-α, IL-6 and IL-12 from monocytes and macrophages [59], proliferation of monocytes and expression of myeloid IL-2Rα and transferrin receptor [60]. Interestingly, leptin fosters Th1 cell differentiation and inhibits Th2 cell polarization by activating IL-2 and IFNγ while suppressing IL-4 production [61]. Surprisingly, unlike adiponectin, leptin is incapable of triggering production of pro-inflammatory or erosive factors from RA synovial tissue fibroblasts [54]. Nevertheless, leptin has been reported to play a significant role in the immunopathogenesis of RA.

2.2.3. Resistin

Synovial fluid and serum levels of resistin are elevated in RA compared to OA or normal individuals [62, 63]. In RA synovial tissue, resistin co-localizes with macrophages, B and plasma cells, but not with T cells [63]. Others confirmed that resistin is secreted from human macrophages activated by IL-1, IL-6, TNF-α or LPS [64]. IFNγ or leptin do not stimulate resistin production. In turn, resistin is also capable of amplifying expression of several pro-inflammatory cytokines, chemokines and MMPs in human articular chondrocytes and macrophages in a NF-κB-dependent manner [65, 66]. Consistently, resistin promotes inflammation and cartilage degradation, contributing to the development and progression of RA. Additionally, resistin is involved in obesity-induced insulin resistance resulting in T2DM [67]. Notably, resistin could alter endothelial cell function by increasing the expression of endothelial adhesion molecules such as intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1), in contrast adiponectin dampens the resistin-induced effect [68].

2.2.4. Visfatin

Serum and synovial fluid levels of visfatin are associated with RA disease activity [69]. Interestingly, RA synovial lining and pannus express elevated levels of visfatin, which correlate with rheumatoid factor and radiographic damage [70]. Expression of visfatin by RA synovial tissue fibroblasts is regulated by TLR3 and TLR4 agonists. On the other hand, fibroblasts and macrophages are thought to be the effector cells that respond to visfatin and facilitate its pathological effect in RA. IL-6, IL-8, MMP-1 and MMP-3 are produced by visfatin-activated RA synovial tissue fibroblasts, while TNF-α and IL-6 are released by monocytes stimulated with visfatin [71]. Accordingly, knockdown of visfatin in RA synovial fibroblasts markedly reduced LPS- and poly IC-induced secretion of IL-6, IL-8, MMP-1 and MMP-3 [71].

2.2.5. Chemerin

Chemerin is a chemoattractant adipokine that is expressed in adipose tissue, RA synovial fluid, RA synovial tissue endothelial cells, synovial lining and sublining cells [72]. Chem R23 is a G protein-coupled receptor present on RA synovial tissue fibroblasts and macrophages that binds to chemerin. The expression levels of chemerin and Chem R23 are elevated in RA compared to OA synovial tissue [72]. Stimulation with TNF-α and IFNγ potentiates production of chemerin from RA synovial tissue fibroblasts [72]. Moreover, chemerin-activated RA synovial tissue fibroblasts secrete higher levels of IL-6, CCL2 and MMP-3 and RA fibroblast motility is also impacted by chemerin [72]. In a population of RA patients without clinical presentation of metabolic syndrome, chemerin levels had a stronger positive correlation with future metabolic risk than insulin or leptin concentrations [73]. More recently, levels of chemerin are shown to be a marker for endothelial activation, elevated angiogenesis and progression of atherosclerosis in RA [74].

2.3. Immune cells and their pathogenic function in obesity and RA

2.3.1. Neutrophils in obesity and RA

Unlike other immune cells, neutrophils are immunologically active the moment they egress from the bone marrow. Neutrophils are the first immune cells to detect pathogens through their TLRs (TLR2/TLR4) that recognize pathogen-associated molecular patterns (PAMPs) and activate NF-κB signaling [75–77]. They also produce monocyte chemoattractants that can exacerbate the local inflammation by recruiting the circulating monocytes and remodeling them into different macrophage subtypes [78, 79]. Subsequently, neutrophils and differentiated macrophages are capable of attracting T cells that polarize into different subtypes of T lymphocytes depending on the profile of the cytokines produced at the site of inflammation [80]. Thus, neutrophils typically initiate local inflammation, both in physiological or pathogenic conditions.

Notably, we demonstrate that collagen-induced arthritis (CIA) mice fed on 60% high fat diet (HFD) had an earlier disease onset compared to those on regular diet (RD) [81]. We documented that the early signs of inflammation in these obese CIA mice were due to an elevation of macrophage inflammatory protein 2 (MIP2), the murine homologue of IL-8, which potentiated neutrophil infiltration. Extending these results, levels of IL-8/MIP2 exceeded other neutrophil chemoattractants including CXCL1 and CXCL5 protein levels quantified in adipose condition media obtained from RA synovial tissue and obese mouse abdominal adipose tissue [81]. Consistent with the findings in CIA joints, immunoneutralization of IL-8 or MIP2, but not CXCL1 or CXCL5, in RA or mouse adipose tissue conditioned media impaired experimental neutrophil chemotaxis [81]. These results indicate that an earlier disease onset diagnosed in obese RA patients may be due to enhanced neutrophil infiltration instigated by higher levels of joint IL-8 compared to normal weight patients [81].

2.3.2. Macrophages in obesity and RA

Earlier studies have shown that obesity contributes to monocyte tissue infiltration and ensuing macrophage differentiation and inflammatory response. Accordingly, the fraction of macrophages/total immune cell infiltrate is markedly elevated in the adipose tissue of obese (45–60%) compared to lean mice (10–15%) [82]. Circulating monocytes can differentiate into either macrophages (mouse CD11b+Ly6C^hi CCR2⁺, human CD14^hi CD16⁻) or dendritic cells (mouse CD11b+Ly6C⁻ CX3CR1⁺, human CD14^hi CD16⁺), displaying distinct surface markers and chemokine receptors [83–86]. Inhibition of CCL2, but not CX3CR1, significantly suppresses recruitment of CD11b⁺Ly6C^hi CCR2⁺ monocytes into the adipose tissue, suggesting that CCL2 contributes to obesity-induced macrophage infiltration. Others have reported that myeloid cell trafficking to adipose tissue can also be due to CXCL5/CXCR2, osteopontin or BLT1, which is a high-affinity receptor specific for leukotriene B [82, 87–91]. In contrast to previous observations, while other monocyte-attractant chemokines such as CCL20 and CCL5 were elevated by high fat diet (HFD) in a TLR4-induced arthritis model, joint CCL2 levels were unaffected by the HFD, contesting the importance of the CCL2/CCR2 pathway for obesity-induced monocyte infiltration or M1 development [81].

Migration of macrophages into adipose tissue coincides with plasma insulin increase suggesting that macrophage-mediated inflammation contributes to insulin resistance and T2DM [92]. It was found that adipocytes treated with TNF-α have reduced glucose transporter 4 protein (GLUT4), a compensatory mechanism by which elevated monocyte trafficking normalizes the plasma insulin levels [93]. Notably, the potentiated extravasation of monocytes into the adipose tissue is linked to lipolysis and release of free fatty acids, rather than just exacerbating inflammatory response [93].

A new concept has emerged more recently, postulating that free fatty acids released from dead adipocytes can further increase macrophage infiltration by activating the TLR4 pathway [94]. It has also been shown that hyperplasia of the adipose tissue activates stress signaling in adipocytes which promotes secretion of free fatty acids and ROS. In obese mice, crown-like structures appear, formed by clusters of macrophages surrounding the dead adipocytes. The macrophages that make up these crown-like structures are classically activated M1 macrophages, that express TNF-α and IL-6 and are inducible nitric oxide synthase-positive (iNOS+) [95].

Extending these results, adipose conditioned media from RA patients and obese mice transform naïve myeloid cells into M1 macrophages. Blockade or deficiency in the TLR4 pathway impairs fat-induced M1 macrophage differentiation induced by mouse conditioned media [81]. Our findings support the notion that free fatty acids or endogenous TLR4 ligands participate in obesity-mediated M1 differentiation.

Intriguingly, in CIA mice, as disease progresses beyond day 30, differences in joint swelling between obese and lean mice are lost. Corroborating these results, the frequency of CIA leukocytes (F4/80, CD3 and B220) and Th cells (Th1 and Th17), as well as joint levels of monocyte and neutrophil chemoattractants were unaffected in either diet group in the long term [81]. Similarly, in CIA joints, monocyte migration, M1 and M2 macrophage differentiation and levels of TNF-α, IL-6, IFNγ, IL-17 and MIP2 were ultimately comparable in mice on HFD and those fed a RD. These findings indicate that, although the early CIA disease activity was exacerbated by HFD, the effector phase of CIA was not impacted [81]. In contrast, mice on HFD did show a delay in TLR4-induced arthritis compared to the mice on RD. The obese mice displayed a prolonged disease manifestation due to conversion of joint myeloid cells to iNOS+ M1 macrophages, while the lean mice predominately expressed arginase-1+ M2 macrophages [81].

Peroxisome proliferator-activated receptors (PPARs) also play a pivotal role in controlling macrophage polarization and insulin sensitivity. Three subtypes, namely PPARα, PPARβ/δ and PPARγ, form a family of nuclear hormone receptors. These ligand-dependent transcription factors connect metabolism with the immune system. Macrophage-specific deletion of PPARγ or PPARδ shifts macrophage polarization to the classical M1 phenotype [96, 97]. In contrast, macrophages evenly spread throughout the lean mouse adipose tissue are of the less inflammatory non-classical M2 subtype. M2 macrophages are associated with metabolic homeostasis, express arginase-1, CD206 and CD301 and secrete anti-inflammatory cytokines, including IL-10, IL-4, IL-13 [95].

2.3.3. T cells in obesity and RA

The number of T cells is significantly enhanced in the visceral adipose tissue of obese compared to lean individuals [88, 98]. Glucose, lipids, fatty acids and leptin are metabolic factors that modulate T cell proliferation [99]. HFD is shown to suppress expression of GATA3 (transcription factor of Th2 cells) and FOXP3 (master regulator of Tregs), while the number of the Th1 cells producing IFNγ is increased in obese mice. When lean mice are put on HFD, the resident Th2 cells and Tregs in the visceral adipose tissue transform into Th1 cells which can subsequently shift the M2 balance to a predominantly M1 macrophage phenotype [100].

RA is largely a Th17-mediated disease. In obesity, changes in the inflammatory milieu can alter the differentiation and pathogenicity of Th17 cells [100]. Th17 cells polarized in presence of IL-6 and transforming growth factor β are non-pathogenic and produce IL-17 plus IL-10 [101]. In contrast, IL-6, IL-1 and IL-23 induce pathogenic Th17 differentiation, characterized by expression of CXCL3, CCL4, CCL5, IL-3, IL-22, colony stimulating factor 2 and granzyme B [101]. More recently, the fatty acid synthetic enzyme acetyl-CoA carboxylase 1 (ACC1) was shown to facilitate obesity-induced Th17 cell differentiation [102]. Inhibition or genetic depletion of ACC1 in mice on HFD impaired this function. Moreover, obesity exacerbates early progression of preclinical colitis and multiple sclerosis by promoting IL-6-induced accumulation of Th17 cells in the spleen [103]. In contrast, HFD did not impact differentiation of Th17 cells in the joints or spleen of CIA mice [81]. Consistently, while IL-17 was detected in conditioned media extracted from mouse visceral fat, it was absent in the conditioned media prepared from adipose tissue obtained from RA synovium.

2.4. Role of microbiota in RA

Our body is populated by an abundant, genetically varied microbial ecosystem, the microbiota. Microbiota is considered a major physiological player and its influence is progressively accepted not to be confined to the gastrointestinal system. Disruption of the microbiota, affects local as well as systemic immunity and it has been associated with various immune disorders including RA, PsA, systemic lupus erythematosus and gout [104, 105]. Obesity and diet can influence the composition of the gastrointestinal microbiota which in turn has been reported to affect RA development. A so-called obesogenic diet consisting of high fat foods and low fiber content negatively affects the commensal flora [106, 107].

Despite the unclear connection between microbiota and RA, studies have shown that mice under germ free condition display diminished arthritic activity [108, 109]. It has also been reported that the normal balanced microbiota versus intestinal dysbiosis can differentially impact T cell differentiation. Aberrant intestinal microbiota in a preclinical IL-1Ra-deficient arthritis model can promote disease onset in an IL-17- and TLR4-dependent manner [110].

Furthermore, the periodontal pathogens such as Porphyromonas gingivalis and P. nigrescans distinctly exacerbate CIA disease severity [111]. Studies have primarily focused on P. gingivalis as it expresses PAD, an enzyme responsible for citrullination of endogenous peptides, human α-enolase peptides and fibrinogen [105, 112, 113]. RA is uniquely characterized by autoantibodies against citrulline [114]. Furthermore, citrullination can modify inflammatory activity of several proteins, cytokines and chemokines associated with RA pathogenesis [115]. In mice, prior colonization of oral P. gingivalis aggravates CIA by promoting Th17 cell responses [111]. Others have found equal abundance of subgingival microbiota in patients with early and chronic RA compared to healthy subjects [116]. Although the role of P. gingivalis in RA is not fully established, there are studies that demonstrated clinical efficacy when treating periodontitis as an adjuvant for RA drug therapy, further alleviating joint disease activity [117, 118].

2.5. Therapeutic implications for RA and its cardio-metabolic complications

An earlier study reported that over 5 years prior to their diagnosis, RA patients exhibit a more pronounced mean reduction in TC and LDL-C than control subjects, with no change in HDL-C, hence reducing the ratio of TC/HDL-C or atherogenic index [25, 28, 42, 119, 120]. Conversely, while disease-modifying anti-rheumatic drugs (DMARDs) and biological therapy suppress disease activity and risk for CVD in RA responders, these treatments increase TC levels. Reports suggest that in RA there is an inverse association between disease activity and lipid levels, with an emphasis on the importance of LDL-C [25, 28, 42]. It was postulated that in RA, a therapeutically altered lipid profile characterized by elevation of total lipid levels, especially cholesterol and triglycerides, does not reflect an increased risk for CVD. The changes in RA lipid profile observed subsequent to successful therapy may be in part secondary to resolution of inflammation (summarized in Fig. 1) [25, 42].

Figure 1:

Impact of anti-IL-6 and anti-TNF-α therapy on RA pathogenesis, associated dyslipidemia and cardiovascular comorbidity

Sulphasalazine monotherapy or combination therapy with methotrexate (MTX) and prednisolone in RA patients increased the serum HDL-C and TC, however the TC/HDL-C ratio was reduced [121]. In short-term therapy (less than 6 months) with anti-TNF-α, the TC/HDL-C ratio remained constant with a concomitant proportional increase in both TC and HDL-C levels [122–125]. Over time, further increases in TC and HDL-C levels shifted the balance, resulting in an increased atherogenic index.

It has been noted that there may be differences among TNF-α inhibitors, as it was indicated that etanercept may have a greater anti-atherogenic capacity compared to infliximab. This is in part because etanercept inhibits both TNF-α and lymphotoxin α, while infliximab can only impair TNF-α function [123, 126–128]. Additionally, inhibition of IL-6R by tocilizumab also improved clinical symptoms in RA responders. However, the atherogenic index remained unchanged, while increases in TC, triglycerides and LDL-C were observed [129–131].

3. PSORIATIC ARTHRITIS and OBESITY

Psoriasis is primarily known as an inflammatory skin disease, characterized by epidermal hyperplasia, leukocyte recruitment and in particular Th1, Th17 and Th22 cell activation. Yet, chronic autoimmune psoriatic disease goes further than just skin-deep and is increasingly recognized as a systemic disease [132]. Psoriatic arthritis is one of the serious comorbidities that affect psoriasis patients. Psoriatic arthritis is categorized as a seronegative spondyloarthritis, characterized by bone and cartilage destruction, associated with pathologic new bone formation and an extensively vascularized synovium [133].

Multiple studies have identified BMI as a risk factor for PsA [134–136]. Although the first study that reported an increased prevalence of obesity, claimed obesity to be a consequence rather than cause of psoriasis [137], subsequent research has contested this claim. Even BMI at the age of 18 was revealed to predict later incidence of joint involvement in psoriasis patients [134]. PsA is characterized by a significantly higher BMI, not only compared to a healthy population but also relative to RA or psoriasis patients [18]. Furthermore, a larger proportion of patients classifies as obese. Among postmenopausal women, PsA patients display a higher body fat percentage than psoriatic or healthy controls [138].

In obese patients, the adipose tissue can establish a low-grade inflammatory milieu that favors the progression of inflammatory disorders. While many have reported a strong association between psoriatic disease and obesity, only few have examined the underlying mechanism that connects these two components (Fig. 2). In this review, we will elaborate on the observational as well as the mechanistic studies that have been published to date.

Figure 2:

Impact of obesity and metabolism on PsA pathogenesis

3.1. Comorbidities associated with obesity in PsA

3.1.1. Obesity and metabolic complications in PsA

Complications of PsA often include obesity-related pathologies such as CVD, T2DM and metabolic syndrome [139, 140]. Patients with PsA have an increased risk of developing metabolic syndrome with an incidence as high as 38%, compared to 20% in RA or 11% in ankylosing spondylitis [141]. The prevalence of coronary artery disease, hypertension, hyperlipidemia and metabolic syndrome in PsA significantly exceeds the risk observed in non-psoriatic spondyloarthritis [142]. Diabetes mellitus and hypertriglyceridemia also develop more frequently in PsA compared to RA [143]. While several inflammatory mediators involved in PsA are known to modulate insulin sensitivity (detailed in section 3.2), the increased diabetes incidence rate in PsA may largely be attributable to obesity and increased BMI [144].

Psoriasis at a younger age could be driven primarily by genetic factors, while obesity is an acquired risk factor that promotes the pathogenesis of psoriasis at a later age [145]. Accordingly, elevated BMI at the age of 18 was demonstrated to predispose patients to develop PsA independently of BMI at disease onset [134]. Later PsA onset is associated with a reduced familial history of psoriasis, whereas there is a positive correlation between increasing age at onset and the prevalence of hypertension, diabetes and dyslipidemia [146, 147]. Nevertheless, over a third of children diagnosed with juvenile psoriatic arthritis were also overweight, still exceeding the general frequency of overweight in children or among juvenile idiopathic arthritis patients [146].

3.1.2. Obesity and cardiovascular complications in PsA

Psoriasis patients present an increased risk of cardiovascular complications. Furthermore, cardiovascular morbidity and mortality are prominent features of inflammatory rheumatic diseases, which can be impacted by patient’s body weight and BMI. Consistently, the risk of cardiovascular morbidity, including ischemic heart disease, congestive heart failure, cerebrovascular and peripheral vascular disease is increased in PsA patients [148, 149]. Indeed, PsA patients display classic CVD risk factors like obesity, T2DM and dyslipidemia, that are accompanied by activation of inflammatory pathways [148, 150]. Even when adjusted for typical CVD risk factors like age and smoking, PsA patients display a markedly elevated prevalence of subclinical artherosclerosis compared to age- and sex-matched controls [151, 152]. However, this risk is often underestimated, since the predictive systemic coronary risk evaluation (SCORE) algorithm does not account for variables like BMI, triglycerides and CRP that are typically increased in PsA patients [153].

Unfortunately, only few studies have compared the risk of CVD complications in psoriasis versus PsA. A comparative literature review concluded that PsA patients are at a higher risk of developing a myocardial infarction [154]. Evaluation of a large psoriasis patient population confirmed that multiple major cardiovascular risk factors were more prevalent in patients with psoriatic arthritis than in the general psoriasis population [155]. In a Canadian cohort study, obesity, hyperlipidemia and T2DM were all shown to be more frequent among PsA relative to psoriasis patients [156]. Moreover, these investigators confirmed that CVD and hypertension are more commonly displayed in psoriasis with joint involvement [156].

It has been postulated that adipokines contribute to the development of cardiovascular symptoms. High levels of the pro-angiogenic leptin in PsA patients may promote atherosclerosis. Conversely, adiponectin levels were decreased in PsA patients, thus interfering with its protective function in atherosclerosis and other forms of endothelial dysfunction [157, 158]. PsA patients also present a shifted lipid profile with decreased HDL-C and increased LDL-C levels [159, 160]. This distorted balance is considered to be atherogenic. Coronary plaque formation is accelerated in PsA, however, the plaque burden does not correlate with metabolic syndrome [161]. Taken together, the prevalence of obesity-linked cardiovascular risk factors supports a role for obesity-driven CVD in PsA.

3.2. Cytokines and inflammatory pathways in PsA

3.2.1. Adipokines

Both pro- and anti-inflammatory adipokines produced by adipocytes and immune cells residing within the adipose tissue have distinct functions in PsA; regulating metabolism, insulin sensitivity and inflammation. The dominance of the inflammatory adipokines such as resistin, leptin and visfatin over the protective adiponectin contributes to low-grade inflammation and vascular changes [162]. The adipose tissue can thus actively contribute to obesity-related inflammatory disorders such as PsA and associated metabolic and cardiovascular complications.

IL-6 and TNF-α suppress adiponectin production by adipocytes [163, 164]. Adiponectin concentrations correlate inversely with psoriasis severity [165–167]. This is in accordance with the negative correlation between adiponectin and BMI [168]. Despite some conflicting reports [169], a decrease in adiponectin levels was observed in PsA patients, correlating inversely with osteoclastogenesis [170].

Disease severity in psoriasis patients was shown to correlate with increased leptin and resistin levels [171–173]. Leptin serum concentrations were even higher in PsA compared to psoriasis patients [149]. Serum levels of leptin and omentin in PsA, similar to TNF-α and RANKL, positively correlate with the number of mature osteoclasts [170]. Interestingly, TNF-α could contribute both to the increased production of leptin and the decreased expression of adiponectin, explaining in part how TNF-α induces insulin resistance [164, 174].

One study describes a decrease in chemerin levels [170], while several others have reported increased chemerin production in psoriasis, PsA and other arthritic diseases [175–177]. In initial stages of psoriasis, chemerin plays a crucial role in promoting plasmacytoid dendritic cell recruitment [178]. Nevertheless, chemerin levels are further increased in PsA patients compared to psoriasis patients without joint involvement, which suggests chemerin is also pivotal in the progression of complicated psoriasis [175].

3.2.2. TNF-α

Psoriatic skin is characterized by high production of cytokines like IFNγ, IL-17, IL-22 and TNF-α [179]. Similar to RA, TNF-α levels are also significantly higher in PsA compared to OA synovial fluids [180]. TNF-α plays a central role in RA and PsA joint bone resorption, by inhibiting proteoglycan synthesis and activating MMPs and prostaglandin E2 production [181]. Emphasizing the critical role of TNF-α in skin and joint disease, TNF-α polymorphism (-308) was reported as a prognostic marker associated with age of psoriasis onset and progression of bone erosion in PsA [182].

Interestingly, TNF-α is not only responsible for promoting inflammation and bone destruction in PsA patients, but can also promote insulin resistance [183]. Insulin signaling is inhibited as TNF-α prevents insulin receptor autophosphorylation and following insulin receptor substrate-1 (IRS-1) phosphorylation [184]. TNF-α also reduces expression of IRS-1, PPARγ, GLUT4 and adiponectin [164, 185, 186]. TNF-α alters expression of other adipokines such as leptin, which can impact metabolism and insulin sensitivity [174]. Finally, increased free fatty acid levels attributed to TNF-α-mediated lipolysis promote insulin resistance [187, 188].

3.2.3. IL-12 and IL-23

IL-12 is a heterodimeric glycoprotein that is composed of a unique p35 subunit, as well as a p40 subunit that is shared with IL-23 [189]. IL-12 binds to IL-12R and stimulates differentiation of naïve CD4+ T cells to Th1 cells and production of IFNγ [189]. Levels of IL-12 are markedly elevated in PsA compared to psoriasis and normal counterparts [190]. In parallel, increased secretion of IL-12 was detected in obese individuals and patients with T2DM [191, 192]. IL-23 is a family member of IL-12 which is composed of a p40 and a p19 subunit. IL-12 and IL-23 are mainly produced by macrophages, dendritic cells and adipocytes in response to TNF-α and microbial or host immune stimuli [189].

Interestingly, mechanical stress triggers phosphorylation of kinases such as ERK and p38 MAPK, and can consequently promote IL-23, TNF-α and IL-17 expression [193, 194]. Cases have been reported in which injury was associated with PsA risk [195]. This pathway provides insight into the mechanistic link between joint trauma or, by extension, repeated strain associated with obesity and the pathogenesis of psoriatic arthritis [194].

IL-23, together with other pro-inflammatory cytokines such as IL-6, IL-1 and TNF-α play an integral role in the transformation of naïve CD4+ T cells to the Th17 phenotype [189]. Ustekinumab, a monoclonal antibody that recognizes the p40 subunit of IL-12 and IL-23, can impair Th1 and Th17 cell differentiation and is used in the clinical care of psoriasis and PsA patients [196]. Additionally, IL-23 is involved in PsA osteoclastogenesis both through RANK and TNF-α as well as through induction of IL-17 secretion [197]. Single nucleotide polymorphisms in the IL-12B and IL23R gene have been identified and confirmed to be associated with psoriasis and PsA susceptibility [198].

3.2.4. IL-17

In psoriasis and PsA, released IL-23 can trigger Th17 cells to produce high levels of IL-17, IL-17F, IL-26 and IL-29 [197, 199]. Th17 cells are elevated in PsA synovial fluid and synoviocytes compared to counterpart OA samples [200, 201]. Th17 cells and IL-17 form a crucial link between obesity and psoriasis [202]. Both in mice and humans, obesity is associated with elevated IL-17 concentrations [203, 204]. IL-17 alone or in synergy with other inflammatory factors, such as TNF-α, IL-6, IL-23 and IL-1β, can promote angiogenesis, bone erosion, production of neutrophil-mobilizing chemokines, granulocytopoiesis factors and MMPs [205]. In PsA, IL-17 fosters osteoclast formation via induction of TNF-α and RANKL [206]. These findings were supported by the positive impact of IL-17 and IL-17R antibody therapy on PsA joint inflammation and radiographic progression [207, 208].

In psoriasis, circulating IL-17 activates VEGF production from adipose tissue, which provokes inflammation, neovascularization and endothelial dysfunction [209, 210]. In preclinical models of psoriasis, sustained levels of IL-17, CCL2 and TNF-α in skin were capable of inducing aortic inflammation and thrombosis [211]. Consistently, in severe psoriasis, IL-17 potentiates hyperlipidaemia and atherosclerosis [212]. Although, therapies targeting IL-17 did not impact patient’s weight, they could moderate inflammation and associated CVD risk [213].

3.2.5. Serum complement C3

Insulin resistance in PsA patients, as estimated by homeostasis model assessment, was shown to correlate with serum complement C3 [214]. C3 has previously been shown to predict insulin resistance in other settings as well, such as diabetes, polycystic ovary syndrome and in elderly populations. C3 is crucial for complement system activation, but more interestingly, its component C3a also has a major metabolic function. Obesity is associated with increased levels of acylation-stimulating protein (ASP), a desarginated processed form of C3a, which enhances fat storage [215–217].

Accordingly, complement C3 has emerged as a novel risk factor for CVD. Production of C3 is promoted by pro-inflammatory cytokines, including TNF-α, and anti-TNF-α treatment appears to normalize serum complement C3 concentrations [218]. More recent studies confirmed that both the insulin sensitivity and predictive oral disposition index in non-diabetic PsA patients correlated inversely with complement C3 levels [219]. C3 is described as a novel predictor of insulin sensitivity in PsA. C3 concentrations >1.32 g/L identify the insulin-resistant patients with a 56% sensitivity and 96% specificity.

3.2.6. Peroxisome proliferator-activated receptors

PPARs are important regulators of both metabolism and inflammation. Endogenous ligands of the PPARs include unsaturated fatty acids, components of oxidized LDL and very low-density lipoproteins (VLDL), eicosanoids and linoleic acid derivatives [220]. PPARα promotes fatty acid oxidation and glucose accumulation. PPARβ also modulates fatty acid oxidation. PPARγ mainly promotes glucose and lipid uptake. On the other hand, PPAR ligand binding and activation can suppress expression of inflammatory factors like IL-1 and TNF-α [220].

In the psoriatic epidermis, the expression of PPARα and PPARγ is reduced, while that of PPARβ is increased. In the context of T2DM, a collection of synthetic exogenous PPARγ agonists, referred to as the thiazolidinediones, have exhibited therapeutic potential and can restore of insulin sensitivity [221]. Furthermore, PPARs are a novel therapeutic target in CVD [222]. Given the metabolic and cardiovascular complications associated with psoriasis and PsA, these patients may benefit from PPAR-targeted therapy. Treatment with PPARγ agonist pioglitazone has shown significant clinical improvement of the psoriasis area and severity index (PASI) and a reduction in the number of tender and swollen joints in PsA patients [223]. Yet, side effects such as weight gain and fluid retention may limit its therapeutic application [223].

3.2.7. Role of microbiota in PsA

The gastrointestinal microbiome has an important physiological role that regulates digestion and uptake of nutrients, supports a barrier function and prevents gut colonization by more pathogenic micro-organisms. It modulates local immunity, but also affects our systemic immune system. A link between microbiota and PsA has been established, uncovering that in psoriatic patients the microbial diversity is decreased [224, 225]. Both psoriasis and PsA are associated with reduced number of intestinal bacteria. However, similar to the pathogenic profile described for inflammatory bowel disease, patients with PsA lacked the beneficial taxa such as Akkermansia, Ruminococcus and Pseudobutyrivibrio [224]. The interaction between microbiota and inflammatory joint pathology has been discussed in the context of RA in section 2.4.

3.3. Therapeutic implications for PsA and its cardio-metabolic complications

TNF-α is a key player in various types of chronic arthritis and thereby a much preferred target for therapy. Besides promoting articular and skin inflammation, TNF-α is thought to regulate metabolic symptoms such as dyslipidemia and insulin resistance. Thus, TNF-α inhibitors can help relieve metabolic complications as well as the primary manifestation of RA and PsA. However, the effect of TNF-α blockade on body weight and BMI is still inconclusive. Some studies have suggested that use of these therapeutic antagonists in PsA and RA results in weight gain, reflecting an increase in both fat and lean mass [226, 227]. Others have evaluated the effects of infliximab and adalimumab and have not observed any change in BMI after 4–8 weeks of therapy [157]. In addition, several short-term case reports documented elevated levels of triglyceride in PsA patients undergoing infliximab [127, 228] or adalimumab therapy [229].

Prolonged retrospective studies conducted for 48 weeks on various TNF-α antagonists did not report any lipid changes though [230]. Others have reported that adalimumab or etanercept treatment in PsA patients increases serum HDL concentrations, while reducing the waist circumference, triglyceride and glucose levels and finally also risk of metabolic syndrome [231].

Resolution of PASI scores, through anti-TNF-α or narrowband ultraviolet B (UV-B) irradiation therapy, is associated with elevated serum adiponectin concentrations as demonstrated following 4–8 weeks of treatment [157]. In a different study, onercept-mediated TNF-α blockade in PsA patients did not impact adiponectin production measured 12 weeks after baseline [232].

A different case report on a PsA and diabetes mellitus patient treated with infliximab showed decreased serum glucose, while discontinued treatment resulted in a relapse of diabetes [233]. Similarly, others observed manifestation of T2DM in a PsA patient after discontinuing infliximab treatment upon clinical PsA remission [234]. Both reports support the hypothesis that infliximab can improve insulin sensitivity.

In RA, anti-TNF-α responders demonstrated a reduced risk of myocardial infarction [235]. This raises the question whether TNF-α blockade could also benefit psoriasis patients and reduce their risk of CVD complications. Perhaps, efficient treatment of psoriasis can also secondarily improve CVD and metabolic co-morbidities. Alternatively, therapies aimed at resolving CVD and metabolism could benefit psoriasis and PsA.

From a different perspective, obesity also influences response to therapy [236]. An observational study performed on PsA patients, revealed that adherence to anti-TNF-α therapy was considerably shorter in obese compared to normal weight patients [237]. Moreover, non-obese patients proved more likely to exhibit a moderate to good response to TNF-α blockade. Therapy withdrawal was more common in obese patients. Several studies substantiated that obesity lowers the chance of achieving and maintaining minimal disease activity upon anti-TNF-α therapy [238, 239].

Interestingly, one study found that patients with a fixed dose of TNF-α inhibitor were at a higher risk of interrupting therapy compared to cases in which the dose was based on the patient’s weight [238]. Body weight and fat mass have pharmacological consequences, altering drug distribution and efficacy. Notably, weight loss concurrent with anti-TNF-α therapy enhanced the chance of achieving minimal disease activity in patients with PsA [240]. In addition to traditional treatment, psoriasis patients should be encouraged to alter their lifestyle and consider the added risk of obesity in the context of their disease. In spite of this, the potential benefit of bariatric surgery on reducing incidence or severity of psoriasis and PsA is still controversial [241, 242].

An earlier study validated that treatment with synthetic DMARDs, similar to TNF-α inhibition, was more successful in decreasing disease activity in patients with a BMI below 25 kg/m² [243]. Conversely, obesity was described to increase the risk of renal and liver toxicity from MTX or cyclosporine treatment [244].

4. OTHER AUTOIMMUNE RHEUMATIC DISEASES

The role of obesity in RA and PsA has been thoroughly investigated, as discussed above. Other arthritis diseases have shown increased prevalence of obesity and metabolic syndrome as well, though the role of obesity in the pathogenesis of these disorders has not been elucidated as clearly. These include systemic lupus erythematosus (SLE) and Sjögren’s syndrome (SS), both distinctive autoimmune diseases associated with destructive joint inflammation.

3.4. Systemic lupus erythematosus

SLE is a chronic autoimmune disorder with a heterogeneous clinical presentation that frequently includes joint involvement. Incidence of obesity among SLE patients is high [245, 246]. In a cohort of female SLE patients, for example, 27.8% were obese at a BMI exceeding 30 [245]. Some studies have disputed the independent association of patient BMI with SLE manifestation, or were inconclusive [246–248]. In contrast, others have reported impaired functional capacity, increased fatigue and elevated levels of inflammatory markers in obese SLE patients [249]. A significant association was also described between BMI and CRP [249]. Furthermore, obesity predicted greater functional decline, evaluated after a 4-year follow-up [245].

Many studies focus on obesity as defined by patient BMI. Others have made the distinction between fat and fat-free or lean body mass and have observed a more nuanced change in body composition in SLE patients [250]. These patients may have a lower lean body mass, which correlates with reduced bone mineral density and in turn negatively correlates with disease severity [251]. SLE patients often present dyslipidemia [252]. One study found that within 3 years of diagnosis, over 60% of SLE patients have hypercholesterolemia [253]. Other typical changes in the lipoproteins include increased VLDL and triglyceride concentrations, as opposed to decreased levels of HDL-C [252]. These low levels of HDL-C, together with an increased incidence of hypertension, contribute to a higher prevalence of metabolic syndrome in SLE [254]. Additionally, metabolic syndrome contributes to endothelial injury that can result in coronary atherosclerosis in obese SLE patients [255].

In a SLE preclinical model, body weight and visceral adipose tissue weight are increased [256]. Simultaneously, plasma leptin levels and the infiltration of macrophages into renal and adipose tissue were elevated. Increased fasted insulin levels and impaired glucose tolerance in SLE mice support a proposed SLE-associated risk for developing insulin resistance [256]. Another strain of lupus-prone mice exhibited weight gain associated with an augmented production of anti-nuclear antibodies and neutrophil-derived ROS, which contribute to SLE onset, after being put on a high caloric diet containing high fat and sugar levels [257]. Interestingly, these mice also display lower triglyceride concentrations versus elevated glucose and TC levels.

Obesity favors excessive inflammation and oxidative stress in SLE patients. Neutrophil extracellular traps (NETs) produce a variety of enzymes that contribute to oxidative stress and formation of oxidized HDL-C [258]. Elevated NET numbers in SLE thus play a pro-atherogenic role, promoting oxidized HLD-C-mediated cardiovascular risk. On the other hand, high levels of LDL-C, have also been associated with atherosclerosis in SLE [259, 260].

Serum concentrations of pro-inflammatory factors, such as IL-6 and TNF-α, are increased in childhood-onset SLE [261]. Furthermore, TNF-α levels correlate with the patient’s body fat percentage. IL-23 is similarly highly produced in SLE patients and was also found to be associated with obesity [262]. Obesity-induced IL-23 could promote SLE pathogenesis and increase the risk for complications, including lupus nephritis or atherothrombosis [262].

SLE patients, like other patients suffering from autoimmune arthritis, tend to live a more sedentary life [263]. Introduction of regular physical activity into the SLE patient’s routine has been shown to significantly improve quality of life [264]. Both physical inactivity and obesity are independent risk factors for cognitive dysfunction in SLE [265]. The benefit of weight loss after bariatric surgery was illustrated by the following decreased need for SLE immunosuppressive medication [266]. However, one should carefully consider the increased risk of the surgery itself in SLE patients [266].

3.5. Sjögren’s syndrome

SS is a rheumatic autoimmune disease in which chronic inflammation leads to destruction of lacrimal and salivary glands through lymphocyte infiltration. The prevalence of SS is 0.01–0.6% and the characteristic manifestations include persistent dry eye and mouth [267]. SS patients have a higher incidence of metabolic syndrome; hence it was postulated that the salivary gland dysregulation may be linked to adipose tissue expansion [268, 269]. Consistently, in SS patients and a preclinical model, a progressive lipid disposition is found in the lacrimal and salivary glands [270]. It was also uncovered that lipid droplets could trigger migration of CD68+ cells into the lacrimal grands of a SS experimental model, contributing to elevated levels of TNF-α, IL-6, IL-1β, Cathepsin S and H [271]. Others have shown that anti-phospholipid auto-antibodies are closely correlated with thrombotic events observed in SS patients [272].

Furthermore, inflammation was exacerbated in lacrimal and salivary glands of SS mice that were fed with HFD compared to RD, due to elevated auto-antibody concentrations [273]. It was determined that saturated fatty acids such as palmitic acid promotes SS pathogenesis by activating IL-6 production from epithelial cells in the glands. Subsequently, the secreted IL-6 differentiates the effector T cells and activates the maturation of B cells to plasma cells [274]. Additionally, high plasma concentration of palmitic acid provokes apoptosis in SS salivary gland epithelial cells, resulting in release of auto-antigens such as α-fodrin [274]. Antigen-presenting cells engulf α-fodrin and production of IL-1β from myeloid cells can potentiate disease severity by amplifying infiltration of monocytes into the salivary gland [274]. Some studies have demonstrated that lipid-related molecules such as docosahexaenoic acid may have a beneficial impact on salivary glands by inhibiting the palmitic acid-induced pro-inflammatory cytokine production [275].

5. CONCLUSION AND FUTURE PERSPECTIVE

Obesity sustains a systemic low-grade inflammatory environment that can trigger or amplify other inflammatory disorders and comorbidities. Metabolic changes and inflammatory factors produced by the adipose tissue aggravate autoimmunity and predispose patients to CVD and secondary metabolic comorbidities. Severity and progression of the autoimmune arthritic diseases, RA and PsA, are impacted by the patient’s BMI. Moreover, recent studies reveal that obesity is an important risk factor that mainly dysregulates RA onset and remission, however has no impact on the effector phase of the disease.

Intriguingly, prior to RA onset, patients exhibit a reduction in the ratio of total cholesterol to HDL-c. Conversely, RA patients whose disease activity is under control, exhibit an increase in total cholesterol levels indicating a reverse correlation between lipid levels and disease activity. Although several studies have revealed a crucial association between obesity and arthritis, additional translational research is required to fully understand the mechanistic link between obesity, CVD as well as skin and joint diseases.

TAKE-HOME MESSAGE.

• Obesity is an environmental risk factor that contributes to arthritic disease onset and severity and increases the risk of CVD.

• There is a complex crosstalk between obesity, inflammatory arthritis and glucose and lipid metabolism.

• Obesity impacts various forms of arthritis; OA as well as autoimmune-driven RA and PsA. Although RA and PsA pathogenesis and clinical presentation display similarities, they are unique and distinct diseases.

• Obesity impacts RA onset by triggering neutrophil migration through IL-8.

• Resolution of RA disease activity is delayed by obesity through a mechanism that is dependent on TLR4-induced M1 macrophage differentiation.

• Disease progression in PsA is modulated by TNF-α, IL-12/IL-23 and IL-17 pathways and adipokine production, which can be potentiated by obesity.

ABBREVIATIONS

ACC1

Acetyl-CoA carboxylase 1

ADMA

Asymmetric dimethylarginine

Ang-2

Angiopoietin 2

ASP

Acylation-stimulating protein

BMI

Body mass index

CIA

Collagen-induced arthritis

cIMT

Carotid intima-media thickness

CRP

C reactive protein

CVD

Cardiovascular disease

DAS28

Disease activity as determined by 28 joint count

DMARD

Disease-modifying anti-rheumatic drug

GLUT4

Glucose transporter 4 protein

GM-CSF

Granulocyte macrophage colony-stimulating facto

HDL-C

High-density lipoprotein cholesterol

HFD

High fat diet

ICAM-1

Intercellular adhesion molecule-1

IFNγ

Interferon γ

IL

Interleukin

iNOS

Inducible nitric oxide synthase

IRS-1

Insulin receptor substrate-1

LDL-C

Low-density lipoprotein cholesterol

Lpa

Lipoprotein a

LPS

Lipopolysaccharide

MIP2

Macrophage inflammatory protein 2

MMP

Matrix metalloproteinase

MTX

Methotrexate

NET

Neutrophil extracellular traps

NO

Nitric oxide

OA

Osteoarthritis

OPG

Osteoprotegrin

PAD

Peptidyl arginine deiminase

PAMP

Pathogen-associated molecular pattern

PASI

Psoriasis area and severity index

PPAR

Peroxisome proliferator-activated receptor

PsA

Psoriatic arthritis

PTPN22

Protein tyrosine phosphatase 22

RA

Rheumatoid arthritis

RANK

Receptor activator of nuclear factor κB

RANKL

Receptor activator of nuclear factor κB ligand

RD

Regular diet

ROS

Reactive oxidative species

SLE

Systemic lupus erythematosus

SS

Sjögren’s syndrome

T2DM

Type 2 diabetes mellitus

TC

Total cholesterol

Th

T helper cell

TLR

Toll-like receptor

TNF-α

Tumor necrosis factor-α

Treg

Regulatory T cell

VEGF

Vascular endothelial growth factor

VCAM-1

Vascular cell adhesion molecule-1

VLDL

Very low-density lipoprotein