Introduction
In chronic inflammation, lipid abnormalities are important components of atherogenesis that determine the progression and clinical manifestations of atherosclerosis.1,2
During the acute phase of inflammation associated with trauma or infection, there are marked changes in the composition and concentration of plasma lipids and lipoproteins.3 Early studies have shown that experimental modeling of systemic inflammatory processes leads to changes in lipid metabolism, characterized by hypertriglyceridemia, increased levels of free fatty acids (FA), and decreased levels of low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C).4 Hypertriglyceridemia associated with acute-phase inflammatory proteins develops due to increased formation and decreased clearance of very-low-density lipoproteins.5 Reduction in lipoprotein lipase (LPL) and hepatic lipase activity also contributes to hypertriglyceridemia and suppresses the formation of a specific HDL subclass associated with cholesterol efflux capacity (CEC).6 Furthermore, significant changes occur in the protein and lipid composition of lipoproteins, which may not only redefine their function but also increase their atherogenic and inflammatory properties.7
Dyslipidemia often occurs in autoimmune diseases such as rheumatoid arthritis (RA), type 1 diabetes, psoriasis, inflammatory bowel disease, and other diseases. An imbalance in lipid metabolism accelerates inflammatory reactions and contributes to the development of atherosclerosis. While there have been numerous studies on the relationship between abnormal lipid metabolism and RA, and evidence suggests that lipid abnormalities occur in the preclinical stages of RA, it remains unclear whether dyslipidemia plays a unique role in the onset and progression of RA.
The purpose of this review is to present the latest data on the effect of inflammation on proatherogenic disorders of lipid and lipoprotein metabolism, with an emphasis on proinflammatory cytokines. It aimed to analyze lipid and lipidomic blood profiles based on disease phase, inflammation activity, and research methods. The review identifies the necessity to study high-density lipoprotein subfractions and their role in reverse cholesterol transport in RA patients as a promising direction for clarifying cardiovascular risk.
Effects of proinflammatory cytokines on lipid and lipoprotein metabolism
Numerous studies have shown that inflammation and inflammatory cytokines are closely related to lipid metabolism. Pro-inflammatory cytokines are involved in the regulation of adipocyte proliferation and apoptosis by stimulating lipolysis, inhibiting lipid synthesis, and reducing blood lipid concentration. Under the influence of pro-inflammatory cytokines during the inflammatory process, structural modifications and a decrease in HDL-C concentration occur, which are associated with the reduction of CEC proteins.8,9
Key cytokines affecting lipid metabolism include tumor necrosis factor alpha (TNF-α), interleukin 6 (IL-6), and IL-1β. The most important source of pro-inflammatory cytokines is adipose tissue, consisting mainly of adipocytes, macrophages, and endothelial cells.10 Increased TNF-α levels in adipose tissue are associated with increased lipolysis, disruption of insulin action on glucose transport, decreased LPL expression, inhibition of fat cell recruitment, and induction of apoptosis in fat cells.4
Several studies have demonstrated a negative effect of TNF-α on lipid metabolism.11,12 TNF-α affects LPL reduction and triglyceride (TG) hydrolysis and enhances lipid mobilization from fat depots, which probably causes proatherogenic changes in the blood lipid profile.13 In parallel, TNF-α stimulates the production of LDL in the liver, leading to the development of hypertriglyceridemia, which modulates the movement of cholesterol esters from HDL to LDL in exchange for TG mediated by cholesteryl ester transfer protein (CETP). Researchers found that lipolysis is mediated by TNF-α through the regulation of lipid droplet-associated perilipin protein expression, hormone-sensitive adipose tissue lipase, and triglyceride lipase.14,15 It has been suggested that the main physiological goal of increased TNF-α expression may be to limit adipose tissue overgrowth and subsequently induce insulin resistance and further disorders of glucose and lipid metabolism.16
The central inflammatory mediator IL-6 has a similar effect. IL-6 is known to be produced by visceral adipose tissue cells, including adipocytes and resident immune cells, primarily macrophages.17,18 In obesity, the number of macrophages significantly increases, and adipocyte hypertrophy is accompanied by increased IL-6 production.19
Administration of recombinant human IL-6 to rhesus macaques resulted in changes in lipid metabolism, including a reduction in apolipoprotein (apo) A1, apoA2, and apoB levels, a decrease in HDL-C and LDL-C concentrations, and an increase in TG.20 Under the influence of IL-6, the ratio of proatherogenic and antiatherogenic lipids, lipoproteins, and their protein components (apoB/apoA1) is disturbed, which obviously increases cardiovascular risk.21
It is important to note that overexpression of IL-6 and its receptors is found in areas of the vascular bed that are more exposed to atherosclerotic lesions (coronary, brachiocephalic, and peripheral arteries).22,23 Cytokine involvement and the protective function of IL-6 receptor inhibitors have been demonstrated in ischemia-reperfusion injury of human cardiac myocytes.24
There was a marked increase in plasma concentrations of IL-1, IL-6, IL-8, IL-10, and TNF-α receptor antagonists in men after marathon running.25 The increase in plasma concentrations of free FA, IL-6, and TNF-α during marathon running may be related to the release of angiopoietin-related protein 4. This mechanism is considered compensatory against lipotoxicity and oxidative stress.26 IL-6 and TNF-α can stimulate adipose tissue lipolysis during prolonged exercise independently of catecholamines and other factors.
The effects of pro-inflammatory cytokines on lipid metabolism may be mediated through increased formation of acute-phase proteins.27 Inflammation primarily increases the production of C-reactive protein (CRP), synthesized under the influence of IL-6, IL-1β, and TNF-α in liver and adipose tissue cells.28
The native circulating form of CRP is the pentameric form, which is released into the bloodstream after exposure to inflammatory stimuli.29 Native CRP is able to bind specifically to LDL when the pentameric structure of CRP changes to a monomeric structure or when LDL is modified/oxidized.30,31 Several in vitro studies have convincingly demonstrated CRP-mediated opsonization of LDL with phagocytic cells through interaction with the immunoglobulin receptors, Fc gamma receptors I and II.32–34
In earlier studies, native CRP was shown to induce the release of proinflammatory cytokines from endothelial cells (vascular cellular adhesion molecule-1 (VCAM-1), intercellular adhesion molecule 1 (ICAM-1), and E-selectin) and monocytes (Monocyte chemoattractant protein-1), endothelial dysfunction, and monocyte adhesion to the endothelium.35 The negative effect of CRP on the vascular wall has been confirmed by studies of its interaction with angiotensin II and its effects on the integrity of the endothelial glycocalyx,36,37 as well as on increased expression of the lectin-like oxidized low-density lipoprotein receptor-1 involved in the damaging effect of oxidized low-density lipoproteins (ox-LDL).38 Moreover, LDL-associated CRP deposits have been detected in foci of atherosclerotic lesions and myocardial damage induced by ischemia/reperfusion in experimental animals and humans.39 These observations served as a basis to consider CRP as an active participant in atherosclerosis. Further study of the structural and functional relations of CRP in vitro and in vivo, and clearer diagnostics with separation of pentameric and monomeric forms of CRP, led to an understanding of the atheroprotective role of CRP.33,40
On the surface of activated platelets, apoptotic cells, and circulating cell-derived microparticles, upon ligand binding or under conditions of high oxidative potential of the extracellular environment, pentameric CRP irreversibly dissociates into insoluble monomers that have different pathophysiological functions from pentameric CRP.41,42
Monomeric CRP is thought to be involved in innate immunity processes through activation of the complement cascade,43 angiogenesis, and atherothrombosis.44 Pentameric CRP is not involved in thrombogenesis, whereas monomeric CRP induces platelet activation and thrombus growth. Additionally, pentameric and monomeric CRP have different pathways for stimulating endothelial cells and neutrophils, and different ways of binding to ligands, including LDL and C1q.33,40
The atheroprotective role of monomeric CRP is evidenced by its ability to retard foam cell formation by suppressing the aggressive response of macrophages to ox-LDL and to remove native LDL from the extracellular space, reducing the risk of modifications.45,46 Moreover, monomeric CRP reduces the proatherogenic effects of macrophages in binding to lysophosphatidylcholine in ox-LDL and inhibits the association of ox-LDL with macrophages. These effects may partially slow the progression of atherosclerosis.47
Thus, CRP has conflicting proangiogenic and antiangiogenic effects that establish tissue remodeling in an atherosclerotic plaque (ASP) and in infarct-affected tissues.33
Myeloperoxidase (MPO) is considered a key element in oxidative damage to lipoproteins. MPO is a lysosomal enzyme that catalyzes the oxidation of various substrates, with hydrogen peroxide as a co-substrate. The specific oxidation products produced by MPO are found in ASP in large quantities.48
The products of MPO-catalyzed reactions are strong oxidants that initiate lipid peroxidation and cause protein modification.49 The enzyme increases the expression of P-selectin on the platelet surface and significantly raises the formation of oxygen radicals by platelets.50
MPO is thought to be a potent producer of oxidized LDL in vivo. Oxidized LDL accumulates in macrophages via class A1 scavenger receptors (SR) and SR-B1.51 Another target of MPO is apoA1 HDL; by oxidizing this lipoprotein, MPO reduces its atheroprotective functions.52 Modified HDLs are less effective in stimulating cholesterol efflux and are rapidly degraded by macrophages. Macrophages continue to engulf modified lipoproteins, and large amounts of lipids accumulate intracellularly, resulting in foam cell formation. Foam cells, in turn, have a pro-inflammatory effect by producing cytokines, chemokines, and growth factors, and by stimulating the secretion of adhesive molecules.53
Thus, systemic inflammation plays an important role in the development of disorders related to the blood cholesterol transport system. Data from numerous studies indicate the influence of cytokines (TNF-α, IL-6) on the proatherogenic potential of blood lipids and lipoproteins, supporting the concept of the contribution of chronic inflammation to the development of atherosclerosis.
Immunomodulatory effects of HDL
In recent years, there has been increasing evidence for the immunomodulatory properties of HDL. The determination of pro- or anti-inflammatory properties of HDL depends on the lipoprotein composition, the cholesterol content in macrophages, and the leading signaling pathways.54,55
HDLs undergo continuous remodeling processes in circulation, characterized by a wide variety of subtypes differing in size, density, shape, charge, and composition. This allows researchers to distinguish dysfunctional or even “proatherogenic” particles from “antiatherogenic” HDL.56 The inflammatory process helps to reduce HDL levels and eliminate cholesterol from cells, leading to the production of proinflammatory HDL.57,58 Proteins and enzymes responsible for the proinflammatory activity of HDL have been identified. The content of acute-phase proteins (such as serum amyloid A, fibrinogen, haptoglobin, apoJ, and complement factors B, C3, and C9) is elevated in proinflammatory HDL complexes compared to normal HDL.59 In chronic inflammation, elevated concentrations of amyloid A, apoJ, and pancreatic phospholipase A2 are found in serum and can displace the usual HDL components (apoA1, HDL, etc.). Moreover, changes in the acetylhydrolase activity of HDL enzymes were observed: a decrease in paraoxonase 1 (PON1) and an increase in platelet-activating factor acetylhydrolase.56 Together, these changes contribute to the proatherogenic potential of HDL.
On the other hand, HDLs inhibit the production of pro-inflammatory cytokines and chemokines, which may ultimately prevent immune cell activation and reduce inflammation.60 The ability of HDL to inhibit the expression of ICAM-1, VCAM-1, and E-selectin has been demonstrated in endothelial cells.61,62
HDL and apoA1 can suppress inflammation by binding directly to lipopolysaccharide or lipoteichoic acid, thereby neutralizing them.63,64 Several studies have clarified the immunomodulatory activity of HDL occurring without neutralization of lipopolysaccharide. HDL can block the ability of serum amyloid A to induce the production of reactive oxygen species and activate the NOD-like receptor protein 3 inflammasome involved in the formation of the active forms of IL-1β and IL-18.65 Pretreatment of human endothelial cells with HDL reduced TNF-α-induced expression of ICAM-1, VCAM-1, and E-selectin.66,67 Taborda et al.67 studied the ability of HDL to mediate immune-inhibitory effects on innate immunostimulatory compounds via Toll-like receptors and inflammasomes without affecting the expression of pattern recognition receptors. These results suggest that HDL may modulate the immune response induced by various stimuli, thereby influencing the inflammatory response.68
The comparative characterization of normal and proinflammatory HDL is presented in Table 1.
Table 1Comparative characteristics of normal and pro-inflammatory HDL
| Effects | Anti-inflammatory (normal) HDL | Pro-inflammatory HDL |
|---|
| Inflammatory | Apo AI reflects the antiatherogenic activity of lipoproteins and suppresses the atherogenic activity of ox-LDL; TNF-α regulation in macrophages; Prevention of EC IL-8, MCP-1 expression | Promote the oxidation of LDL; The synthesis of enzymes decreases under the influence of inflammation; Serum amyloid A, ceruloplasmin replace the usual components of HDL |
| Anti-oxidant properties | PON1 prevents the formation of oxidation products; Endothelial protection; Upregulation of EC and NO production; EC express receptors for apoA-I and HDL | The connection between PON1 and altered apoA1 is disrupted; PAPC interacts with the oxidation products of arachidonic acid to form biologically active oxidation products |
| Reverse cholesterol transport | ApoA1 and apoJ in HDL transport cholesterol from arterial vascular cells and macrophages to other lipids and liver cells for processing and disposal | The functions of apoA1 and apoJ are impaired by oxidative stress; Lipoprotein synthesis is reduced under the influence of inflammation |
| Platelet responses | Binding of mildly and native ox-LDL to apoER2′ and SR-BI on the platelet membrane induces signaling pathways that suppress platelet activation | Platelet activation through binding via CD36 to heavily oxidized HDL |
In recent years, understanding of the mechanism of cholesterol transfer from biomembranes to HDL has greatly advanced through studies of plasma factors and cellular proteins in mouse models.69 In 2022, researchers identified previously unknown membrane proteins associated with the regulation of lipoprotein metabolism and the cell’s ability to release membrane cholesterol, thereby clarifying mechanisms of reverse cholesterol transport.68,70 HDL directly transports cholesterol to SR-B1 receptors expressed in the liver. Alternatively, cholesterol from HDL transfers to LDL and very-low-density lipoproteins via CETP, then returns to the liver through LDL receptors, where it is metabolized into bile acids and excreted into the intestinal tract. Cholesterol exported from cells is transferred to large spherical HDL by the ATP-binding cholesterol transporter (ABC) G1. A related transporter, ABCA1, delivers cellular cholesterol to apoA1 and small dense HDL.68
The importance of the reverse cholesterol transport system has become evident from large cohort studies in humans, which consistently show an inverse association between the ability of apoB-containing lipoprotein-depleted plasma to uptake macrophage-derived cholesterol and cardiovascular disease (CVD) risk.71,72 In healthy subjects with varying HDL-C and apoA1 levels, the ability of HDL to CEC from macrophages in vitro correlates inversely with measures of carotid intima-media thickness, independent of HDL-C and apoA1 concentrations.73
However, even with efficient acceptors, there may be factors limiting cholesterol efflux from membranes, potentially contributing to various proatherogenic abnormalities.74–76
The ability of serum to induce ABCA1-mediated CEC was reduced in patients with type 2 diabetes mellitus with microalbuminemia and proteinuria. The authors attributed this impaired CEC to low levels of preβ-1 HDL.77 Additionally, depletion of preβ-1 HDL induced by chymase (a mast cell-secreted proteinase) therapy impaired ABCA1-mediated but not SR-B1-mediated, cholesterol efflux from J774 macrophages.78 Specific genomic variants of lipoprotein receptors, apoB-100, Proprotein Convertase Subtilisin/Kexin type 9, and other loci may contribute to lipid metabolism abnormalities complicating vascular disorders in type 2 diabetes mellitus.79,80
Disorders of blood lipid and lipoprotein metabolism in patients with rheumatoid arthritis
RA is an autoimmune disease of unknown etiology characterized by chronic erosive arthritis with a progressive course, leading to serious complications and severe comorbidity. RA is associated with early disability and reduced life expectancy among patients.81 CVDs play a significant role in the development of adverse outcomes in RA.82 Numerous studies have shown that the relative risk of cardiovascular events in RA patients ranges from 1.4 to 3.7.83 In a recent large cohort study, mortality rates remained high among RA patients (n = 16,047) observed from 2005 to 2018: the relative risk for death from all causes was 1.22 (95% confidence interval (CI) 1.15–1.30), from CVD - 1.63 (95% CI 1.51–1.75), from myocardial infarction - 2.00 (95% CI 1.78–2.26), and from stroke - 1.39 (95% CI 1.22–1.58). The risk of CVD in RA patients increases from the early years after diagnosis and does not decrease over time.84
The main mechanism leading to the development of CVD in RA is atherosclerotic vascular damage.85 The multifactorial pathogenesis of atherosclerosis involves vascular, metabolic, and inflammatory components.86 Disorders of blood lipid and lipoprotein metabolism are implicated in all three components and play a direct role in the formation of atherosclerotic vascular lesions in RA.87,88 According to meta-analyses, patients with RA and hypercholesterolemia have a 73% higher incidence of CVD compared to patients without it.87
Blood lipid profile of RA patients
Inflammation in RA occurs with increased catabolism of lipoproteins, often leading to decreased serum levels of HDL and LDL. Despite this so-called “lipid paradox,” patients remain at high risk of CVD.89,90 Many researchers have observed lower concentrations of total cholesterol, LDL-C, and HDL-C in patients with active, untreated RA compared to healthy controls and patients with osteoarthritis.91–96 In a prospective study, VanEvery et al.97 have found an association between low LDL-C concentrations and RA risk. Other authors suggest that early-stage RA is associated with normal or reduced serum TG and total cholesterol levels, which, combined with a more pronounced reduction in HDL-C, may increase atherogenic potential.98 Additionally, smaller, more atherogenic lipoprotein phenotypes have been detected in early RA patients compared to controls.94 The formation of small dense LDL particles has been attributed to the acute phase reaction.99
In recent years, the atherogenic index of plasma (AIP), the logarithm of the ratio of TG/HDL-C, has been well-established as a predictor of coronary heart disease in the general population and in type 2 diabetes mellitus.100–103 Hammam et al.104 found associations between AIP and 10-year CVD risk as determined by the Framingham scale in patients with RA and systemic lupus erythematosus (SLE). The potential role of AIP in predicting and managing CVD risk in autoimmune rheumatic diseases has been investigated in several other studies.105–108
Similar data were obtained when analyzing the blood lipid spectrum in patients with untreated early RA included in the REMARCA (Russian invEstigation of MethotrexAte and biologicals in eaRly aCtive inflammatory Arthritis) trial. Decreased total cholesterol and HDL-C levels, along with increased AIP, were found in RA patients compared to healthy controls (Table 2).
Table 2Blood lipid spectrum in patients with early RA and in control (REMARCA trial), Me [25th; 75th percentile]
| Blood lipids | RA (n = 154) | Сontrol (n = 104) | p |
|---|
| Total cholesterol, mmol/L | 5.22 [4.63; 5.98] | 5.88 [5.45; 6.63] | 0.036 |
| LDL-C, mmol/L | 3.41 [2.84; 4.14] | 3.78 [3.53; 4.26] | 0.01 |
| HDL-C, mmol/L | 1.32 [1.03; 1.62] | 1.73 [1.43; 1.97] | 0.023 |
| TG, mmol/L | 1.13 [0.86; 1.63] | 1.03 [0.85; 1.39] | 0.12 |
| AIP (TG/HDL-C) | 1.1 (0.6; 1.9) | 0.6 (0.4; 1.5) | <0.001 |
In contrast, other studies have found higher levels of total cholesterol, LDL-C, TG, and, to a lesser extent, low HDL-C, resulting in significantly higher AIP in patients with early RA compared to healthy individuals.109,110 It is important to note that total cholesterol and TG levels increase even before the clinical manifestations of RA, and the subsequent development of chronic inflammation can explain the decrease in these indicators.98
Lipidomics has shown that serum lipid profiles in patients with active RA are similar to those of patients with preclinical RA, confirming alterations in lipid metabolism before disease onset.111,112 In a prospective cohort study of anti-citrullinated protein antibody (ACPA) positive individuals, higher levels of monohydroxyeicosatetraenoic acids (5-HETE, an important precursor to leukotrienes) appeared to be associated with the subsequent development of inflammatory arthritis.111 Other predictors of RA development in ACPA-positive individuals were decreased levels of apoA1 and ω-3 FA.113,114 A body of research suggests a role for polyunsaturated FA metabolites in preclinical RA. However, the mechanism of their action is still poorly understood.115,116
A recent study by Koh et al.117 demonstrated a marked shift in the lipidome profile in the synovial fluid of RA patients, correlating with disease activity and the degree of synovitis on ultrasound. In particular, lipid subclasses were decreased in lysophosphatidylcholine (LPC) and increased in phosphatidylcholine, its ester, triacylglycerol, and sphingomyelin. Similar changes, although less pronounced, were found in the serum lipidome profile. It is believed that the immunomodulatory effect of LPC depends on its biochemical structure: saturated and monounsaturated LPC species have a pro-inflammatory effect, whereas polyunsaturated species have an anti-inflammatory effect.118
Additionally, depletion of short-chain acylcarnitine was detected in the serum of women predisposed to RA.119 The blood lipid spectrum in patients with active RA and high laboratory activity (CRP and erythrocyte sedimentation rate (ESR) levels) was similar to the lipid profile in patients with preclinical RA and normal CRP/ESR levels. This confirms the occurrence of dyslipidemia at the preclinical stage of RA.119
The relationship of lipid abnormalities with RA activity and acute-phase blood indices is also indicated in several other studies. Boers et al.120 found the lowest levels of LDL-C and HDL-C in high RA activity. An association was found between levels of total cholesterol, CRP, and DAS28 activity index.121 Park et al.,122 in their study of lipid disorders in untreated RA patients, revealed a negative association of HDL-C and apoA1 concentrations with CRP. In the work of White et al.,123 they confirmed the relationship between low HDL-C and high CRP/ESR levels. According to other data, lower serum HDL-C and apoA1 levels were observed in RA patients not only with high CRP levels but also with RF positivity and hand synovitis.124,125
During inflammation, lipoprotein (a) (Lp(a)) synthesis and expression increase, similar to acute-phase proteins.126 Lp(a) is a cholesterol-rich plasma lipoprotein subclass consisting of LDL particles attached to apoA. In RA patients, blood Lp(a) concentration and apoB/apoA1 ratio were significantly higher than in the control group.122,124 Higher levels of Lp(a), proinflammatory, and proatherogenic modified LDL associated with RA are independent risk factors for CVD.127
The comparative characterization of serum blood lipids in RA patients and healthy controls is presented in Table 3.91-93,96,99,109,110,112,122-125,128-139
Table 3Lipid spectrum of the blood of rheumatoid arthritis (RA) patients, healthy controls (HC)
| Researchers | Year | Number of RA patients/HC | Results |
|---|
| Lazarevic et al.128 | 1992 | 69/65 | ↓ LDL-C, TC, HDL-C, more pronounced in patients with high RA activity |
| Park et al.122 | 1999 | 42/42 | ↓ HDL-C и apoА1, ↑ Lp(а) and AI; correlations between HDL-C and CRP (r = −0.35), apoA1 and CRP (r = −0.44) |
| Hurt-Camejo et al.99 | 2001 | 31/28 | TC, LDL-C, TG, HDL-C, apoB, apoA1 did not change. ↑ subfractions LDL1, ↓ HDL |
| Kim et al.91 | 2004 | 54/50 | ↓ TC, LDL-C |
| Choi et al.125 | 2005 | 104/NHANES (>60 лет) | ↓ HDL-C and apoA1; inverse correlations HDL-C levels with CRP, RF, and synovitis of the hands |
| White et al.123 | 2006 | 204/75 | ↑ LDL-C; ↓ HDL-C; ↑ AI; negative correlations HDL-C and CRP, HDL-C and ESR and positive – LDL-C and ESR |
| Georgiadis et al.109 | 2006 | 58/63 | ↑ TC, LDL-С, TG; ↓ HDL-С; ↑ AI |
| Popkova et al.129 | 2007 | 84/15 | ↑ TG, apoА1; ↓ LDL-С; TC and HDL-С did not differ |
| Gonzalez et al.130 | 2008 | 603/603 | TC, HDL-C, LDL-C, TG, apo did not change |
| Myasoedova et al.92 | 2010 | 577/540 | ↓ TC, LDL-С: 5 years before the development of RA |
| Boyer et al.131 | 2011 | 2,956/3,713 | ↓ HDL-С |
| Liao et al.93 | 2013 | 2,005/NHANES | ↓ TC and LDL-С |
| Pozzi et al.132 | 2015 | 30/30 | TC, LDL-C, TG, apoA1 did not differ, ↑ HDL-C, apoЕ, ↓ apoВ |
| Chavan et al.110 | 2015 | 50/50 | ↑ TC, LDL-C; ↓ HDL-C |
| Govindan et al.124 | 2015 | 30/30 | ↓ HDL-C and apoA1, ↑ AI |
| Kim et al.133 | 2016 | 33/13 | ↓ HDL-C, ↑TG, СЕТР, AGEs, apoА1 |
| Ormseth et al.134 | 2016 | 169/92 | ↓ LDL-C; CEC did not differ |
| Rodríguez-Carrio et al.135 | 2017 | 113/113 | ↑ HDL-C, TG, ↓PON1; TGhighHDLlow profile did not differ and was associated with increased TNF-α, MCP-1, leptin, and INF- γ-inducible protein-10 (p < 0.05 in all cases) |
| Tejera-Segura et al.96 | 2017 | 178/223 | ↓ HDL-C, LDL-C, apoА; ↑ Lp(a). In patients with RA activity, a decrease in CEC was detected, in contrast to patients with remission. |
| Burggraaf et al.136 | 2018 | 312/57 | No differences were detected in the lipid profile; ↑ аpoAI, аpoB48. ApoB48 correlated with TG (r = 0.645, p < 0.001) |
| Dessie et al.137 | 2020 | 73/40 | ↑ TC, TG; ↑TC/HDL and LDL/HDL ratios; ↓ HDL-C; LDL-C did not differ. HDL-C negative correlation with CRP, and TC/HDL-C and LDL/HDL-C had а direct association with CRP (p < 0.05). |
| Alisik et al.138 | 2021 | 130/67 | No differences were detected in the lipid profile; ↑ MPO/PON1 ratios in RA patients, especially high with CVD history. MPO/PON1 ratios were significant positive correlated with DAS28 scores (r = 0.357, p < 0.001) |
| Rodríguez-Carrio et al.112 | 2021 | 60/28 | In RA patients, 2 oxylipin clusters (cluster I and cluster II) were identified in contrast to the control group (p = 0.003). |
| Chang et al.139 | 2023 | 80/15 | ↑ particle number, cholesterol ester, free cholesterol, and phospholipids in large/very large-sized HDL; small-sized HDL negative correlation with CRP; ACPA-positivity was associated with HDL- metabolites. |
In a study by Memon et al.,140 it was observed that experimentally induced infection increases the level of oxidized lipids in serum, which may induce LDL oxidation in vivo and be a mechanism leading to increased CVD in patients with chronic infections and inflammatory diseases. The results of several subsequent studies confirmed higher blood levels of ox-LDL in RA patients compared to healthy controls.141,142 LPC is known to be a major component of оx-LDL.143 Recently published research in RA patients demonstrated a possible atherogenic effect of IgG anti-ox-LDL antibodies.144,145 Specifically, it was shown that anti-ox-LDL titers (p = 0.020) may predict the presence of coronary ASP (noncalcified, partially calcified, and high-risk plaque) in RA patients with lower LDL-C levels (<1.8 mmol/L).144
The influence of pro-inflammatory cytokines on lipid metabolism and the pro-atherogenic potential of immune cells in RA is presented in Figure 1.
The most negatively charged LDL subcomponent associated with atherosclerosis is L5.146 In RA, L5 may contribute to atherosclerosis by enhancing foam cell formation by macrophages and increasing the expression of M1 macrophage-associated markers (TNF-α, IL-6, and IL-8).147
Heterogeneity and functions of HDL
The first study on the determination of HDL subfractions in RA patients was carried out by Hurt-Camejo et al. in 2001.99 The authors found a decrease in HDL2 in RA patients compared to healthy individuals, while the levels of total cholesterol, LDL-C, HDL-C, TG, apoB, and apoA1 did not differ.99 In another study, lower levels of HDL2 and HDL3 were found in 45 patients with RA (especially women) compared to healthy controls.148 The study demonstrated a moderate effect of RA activity on HDL2 levels. HDL2 concentration was reduced by 0.06 mmol/L with increasing DAS28 adjusted for sex, age, RA duration, and glucocorticoid use.
The inflammatory process in RA may attenuate the potential anti-atherogenic effect of HDL, causing a decrease in HDL-C levels and a reduction in antioxidant capacity. This reduction in antioxidant capacity may reduce the ability to eliminate cholesterol from cells, potentially leading to proinflammatory properties of lipoproteins.57,58 Proinflammatory HDL has been detected in the blood of patients with CVD,149 RA, and SLE.150 Charles-Schoeman et al.151 showed that proinflammatory HDLs were more frequently found in RA patients than in healthy individuals (20% and 4%, respectively, p < 0.006). Age, disease activity, erosive process, smoking, and non-Caucasoid race were correlated with proinflammatory HDL.
Further examination of the composition of proinflammatory HDL in RA patients revealed elevated levels of acute-phase proteins and complement factors compared to normal HDL.59 Moreover, changes in the acetylhydrolase activity of HDL enzymes were observed: a decrease in PON1 and an increase in platelet-activating factor acetylhydrolase.
MPO may contribute to the atheroprotective effect of HDL in RA. MPO is an enzyme found in macrophages and neutrophils, acting as a mediator of inflammation and oxidative stress.152,153 Several studies have found increased MPO concentration and activity in RA patients compared to healthy controls and patients with other rheumatic diseases (osteoarthritis, ankylosing spondylitis, SLE, and Sjögren’s syndrome).154–157 The high inflammatory status of RA and markers of acute-phase response (CRP, ESR) were associated with increased MPO activity.154,156,158 MPO content in synovial fluid was higher in untreated RA patients compared to treated patients; moreover, in untreated patients, MPO concentration and activity correlated with IL-8 and IL-18 content.157
In a recent study by Alisik et al.,138 MPO levels were higher in RA patients with concomitant CVD compared to RA patients without CVD. According to receiver operating characteristic analysis, the ratio of MPO to PON1, which characterizes HDL dysfunction, appeared to be more associated with CVD in RA.
The best-studied effects of MPO on CVD include the formation of dysfunctional atherogenic lipoproteins, decreased nitric oxide availability, and endothelial dysfunction leading to vasoreactivity and ASP instability.159 These effects strongly suggest that MPO is directly involved in the pathophysiology of CVD in RA.160,161
Thus, inflammation and high RA activity may reduce antioxidant properties and contribute to the proatherogenic potential of HDL. However, there is no clear consensus on whether this mechanism affects CVD risk in RA.161
HDL becomes proatherogenic against the backdrop of inflammation due to decreased CEC from macrophages, which is an independent risk factor for CVD.96,162
The efficacy of CEC is known to correlate directly with the properties of plasma lipid transport proteins. Decreases in CETP mass and activity were found in RA patients compared to healthy subjects,96 with the lowest rates observed in RA patients on glucocorticoids.163
Studies have shown no difference or a decrease in CEC in RA patients compared with healthy donors.96,155,161,164,165 A meta-analysis confirmed no differences in HDL-C efflux capacity and concentration between patients with RA and controls.166
In a recent study including 195 SLE patients and 265 RA patients, more severe CEC impairment was noted in SLE patients compared to RA patients.167 Although both diseases share common features of inflammatory dyslipidemia, RA patients had lower HDL-C levels and higher apoB levels than SLE patients.
Importantly, patients with high RA activity (DAS28 > 5.1) had lower CEC than patients in remission.96,164 The association of high RA activity with suppression of CEC may be indicated by inverse correlations of cholesterol efflux with DAS28 (r = −0.39, р = 0.01) and ESR (r = −0.41, р < 0.001).164 Multivariate analysis revealed that smoking, diabetes, ESR, and glucocorticoid use were correlated with CEC. However, according to Ormseth et al.,161 cholesterol efflux was not associated with RA activity, the systemic inflammation index CRP, oxidative stress (urinary F2-isoprostanes), or insulin resistance measured by the HOMA index. The authors concluded that net cholesterol efflux capacity by HDL-enriched serum is not altered in patients with well-controlled RA.
Liao et al.165 studied the blood lipid spectrum and CEC in RA patients with CRP concentration ≥10 mg/L. After one year of therapy and a 23.5 mg/L decrease in CRP, patients showed a 7.2% increase in HDL-C (р = 0.02) and a 5.7% increase in CEC (р = 0.002). The decrease in CRP concentration was accompanied by an increase in apoA1 (r = 0.27, p = 0.01) and CEC (r = 0.24, р = 0.002).
When CEC mediated by SR-B1, ABCG1, and ABCA1 were studied using apoB-depleted serum, Ronda et al.168 found that only ABCG1-mediated CEC was lower in active RA patients compared with controls. The same authors later conducted a comparative study of CEC before and after 6 months of methotrexate as monotherapy and in combination with the TNF-α inhibitor adalimumab.169 Methotrexate monotherapy moderately increased SR-B1- and ABCG1-mediated CEC by 6% and 7%, respectively. Multidirectional associations of ABCG1-mediated CEC with serum HDL-C concentration (r = 0.18, p = 0.047) and DAS28 (r = −0.247, p = 0.018) were observed in RA patients after 6 months of combination therapy with methotrexate and a TNF-α inhibitor. The study found no correlation between CEC and inflammatory markers.
As noted by Ormseth & Stein in their review, there is no uniform data on the significance of impaired CEC in RA patients and the influence of disease activity or inflammation on it.161 A meta-analysis of HDL cholesterol efflux capacity and concentration in RA showed improvement in CEC with early administration of antirheumatic therapy and control of inflammation.166 However, the authors noted that the results should be interpreted with caution due to the heterogeneity of groups and methodologies in those studies.
CEC is considered a sensitive predictor of cardiovascular risk in the general population.170,171 Nevertheless, the association of CEC with CVD risk in RA appears to be ambiguous.
In a study by Vivekanandan-Giri et al.,155 no differences were found between RA patients with and without CVD. According to Ormseth et al.,134 an increase in net CEC by HDL-enriched serum was not significantly associated with a decrease in coronary calcium score (odds ratio (OR) = 0.78 [95% CI 0.51–1.19], p = 0.24), adjusting for age, sex, race, the Framingham cardiovascular risk scale, and the presence of diabetes. Another study demonstrated the association of high CEC with a low risk of carotid ASP development in RA patients (OR = 0.94 [95% CI 0.89–0.98], p = 0.015).96
Convincing data on the impact of rheumatoid inflammation on CEC and its contribution to the development of atherosclerosis and associated CVD were reported in 2023. Karpouzas et al.172 confirmed the inverse association of systemic inflammation measured by CRP and disease activity measured by DAS28-CRP with ABCA1-mediated CEC. ABCA1-mediated CEC was associated with less progression and fewer coronary ASPs in patients with low baseline and cumulative CRP levels, in patients not receiving glucocorticoids, and in patients on disease-modifying antirheumatic drug therapy. In contrast, ABCA1-mediated CEC was associated with accelerated development of coronary atherosclerosis in patients with baseline high CRP levels and in those receiving glucocorticoids but not methotrexate or biologic disease-modifying antirheumatic drugs.173
In another study by these authors,172 ABCG1-mediated CEC was inversely correlated with a large number (≥5) of carotid ASPs (adjusted OR 0.50 [95% CI 0.28–0.88]), numbers of partially calcified (rate ratio 0.71 [0.53–0.94]), and low-attenuation plaques (rate ratio 0.63 [0.43–0.91] per standard deviation increment). Low-attenuation plaques on coronary computed tomography angiography are known to be associated with ASP progression and instability.
The general form of atherothrombosis in RA is shown schematically in Figure 2.
Thus, in well-treated RA with controlled disease activity, ABCA1-CEC may exhibit atheroprotective actions. Conversely, in uncontrolled RA, inflammation promotes the development of proatherogenic properties of ABCA1-CEC.166,173
Conclusions
Interest in the study of lipids and lipoproteins in RA remains high. New data on the development of lipid abnormalities have been obtained in the preclinical phase of RA research. Studies have demonstrated differences in lipid and lipidome profiles depending on disease phase, inflammation activity, and study methods. Lipids may contribute to progression or exert a protective effect in RA depending on their type and metabolites. Inflammation or disease activity appears to have a deleterious effect on HDL-mediated reverse cholesterol transport activity. Effective treatment of inflammation by adequate control of RA activity can offset proatherogenic abnormalities in lipid and lipoprotein metabolism.
Declarations
Funding
This work was supported by the Russian Science Foundation (Grant # 22-15-00199).
Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Authors’ contributions
Study concept and design (EG, TP), acquisition of data (EG, MS), analysis and interpretation of data (EG, TP, DG), drafting of the manuscript (EG, MS, DG), critical revision of the manuscript for important intellectual content (TP, DG), administrative, technical, or material support (TP), and study supervision (TP). All authors have made significant contributions to this study and have approved the final manuscript.