High-Density Lipoprotein in Lupus: Disease Biomarkers and Potential Therapeutic Strategy

1. Introduction

Systemic lupus erythematosus (SLE) is an autoimmune syndrome with pleiotropic clinical manifestations, characterized by the synthesis of autoantibodies, and the development of significant immune dysregulation, and organ damage. While life expectancy in SLE has improved due to the advancement of immunosuppressive therapies and improved treatments for infections and renal disease, mortality rates remain approximately 3-fold higher than in the general population.¹ Of the many causes of mortality in SLE, accelerated atherosclerosis and cardiovascular disease (CVD) is recognized as one of the most prevalent.² SLE patients showed a higher prevalence of CAD and atherosclerosis compared to controls, which could not be predicted by traditional risk factors alone.³ As CVD-inflicted mortality accounts for more than one-third of all deaths in SLE patients⁴, it is clear that CVD continues to be a significant threat to SLE patients.

High-density lipoprotein (HDL) is the smallest lipoprotein and is well-known to exhibit various atheroprotective effects independent of cholesterol mobilization, including its anti-oxidative, anti-inflammatory, anti-thrombotic, and anti-apoptotic abilities.⁵ Thus, the low levels of HDL had been associated with an increased risk of CVD. SLE patients with accelerated atherosclerosis exhibit decreased levels of HDL and the development of dysfunctional HDL.^6–8 Such evidence suggests that HDL is likely a novel target for minimizing the risk of CVD in SLE patients, and several studies have recently proposed and investigated HDL-targeted therapies as a potential therapeutic intervention SLE patients with CVD. In this review, we will discuss the quantitative and qualitative roles of HDL in both normal and SLE conditions, as well as the potential of HDL-targeted therapeutic interventions.

2. Structure and Composition of High-Density Lipoproteins

HDL is constantly remodeled in the bloodstream through interactions with other lipoproteins, enzymes, and contact with target cells, resulting in significant particle heterogeneity. HDL consists of a core of hydrophobic lipids, including cholesteryl esters and triglycerides, and a surface monolayer containing phospholipid, free cholesterol, and apolipoproteins. Apolipoprotein A-I (apoA-I) is the most abundant protein associated with HDL, comprising 70% of the total HDL protein content.⁹ ApoA-I is a 28.1 kDa, highly α-helical and amphipathic scaffold protein consisting of 243 amino acids that interact with lipids to ultimately define the size and shape of HDL species.⁹

In addition to proteins, lipid species are another key component of the overall structure of HDL. Recent lipidome analysis from Kontush et al. revealed that more than half of the total HDL mass is accounted for by lipid components, the majority being phospholipids – accounting for 40–60% of the total lipid mass.¹⁰ Of these phospholipids, phosphatidylcholines are the largest population, making up 33 – 45% of total lipid mass, and play critical roles in particle stability, cholesterol efflux, and molecular interactions with HDL-associated enzymes.¹⁰

HDL can be classified into different sub-populations using various techniques. An overview of HDL classification is summarized in Table 1, based on the publication of Rosenson et al.¹¹ Briefly, the use of density gradient ultracentrifugation and nondenaturing gradient gel electrophoresis can distinguish HDL sub-populations on the basis of density and size, respectively; from smallest to largest, HDL3c, HDL3b, HDL3a, HDL2a, and HDL2b.¹¹ ApoA-I containing HDL sub-populations can also be defined on the basis of size and charge: pre-β-1 HDL (very small, discoidal HDL with apoA-I and phospholipid), α-4 HDL (very small, discoidal HDL with apoA-I, phospholipid, and free cholesterol), α-3 HDL (small, spherical HDL with apoA-I, apoA-II phospholipid, free cholesterol, cholesteryl ester, and triglyceride), α-2 HDL (medium-sized, spherical HDL with same constituents as α-3 HDL), and α-1 HDL (very large, spherical HDL with same constituents as α-3 HDL but nearly no apoA-II).¹¹ Using more sophisticated techniques, such as nuclear magnetic resonance (NMR), 26 different HDL sub-populations have been identified, but due to limited measurement precision they are simply described as small, medium, and large.¹¹ Recently, an NMR-based clinical analyzer called the Vantera® was developed to measure total HDL particle number in clinical laboratory settings.¹²

Table 1.

Classification of HDL

Separation	Analytical Method		Very Small	Small	Medium	Large	Very Large
Density	Ultracentrifugation	Classification	HDL3		HDL2
		Range (F1.2)	0 - 3.5		3.5 - 9
	Density gradient ultracentrifugation	Classification	HDL3c	HDL3b	HDL3a	HDL2a	HDL2b
		Range (g/mL)	1.154 - 1.170	1.129 - 1.154	1.110 - 1.129	1.088 - 1.110	1.063 - 1.087
Size	Nondenaturing gradient gel electrophoresis	Classification	HDL3c	HDL3b	HDL3a	HDL2a	HDL2b
		Range (nm)	7.2 - 7.8	7.8 - 8.2	8.2 - 8.8	8.8 - 9.7	9.7 - 12.9
	2-D gel electrophoresis	Classification	pre-β-1	α-4	α-3	α-2	α-1
		Range (nm)	5.0 - 6.0	7.0 −7.5	8.5 −8.5	9.0 - 9.4	10.8 - 11.2
	NMR	Classification	Small		Medium	Large
		Range (nm)	7.3 −8.2		8.2 −9.4	9.4 - 14

3. Changes in HDL Composition in SLE Patients

3.1. Dyslipoproteinemia

Many SLE patients have increased levels of VLDL and LDL and decreased levels of HDL, considered as the “lupus lipoprotein pattern”. Low HDL-cholesterol (HDL-C) is one of the most prevalent dyslipidemia indicators observed in SLE patients, including pediatric populations.^13–18 In a multiethnic U.S. cohort study containing 546 SLE patients, 81% had low HDL-C levels (< 35 mg/dL).¹⁷ A recent Egyptian study reported 45% of 221 SLE patients presented low HDL-C level (< 40 mg/dL).¹⁸ In some instances, no significant differences in HDL-C levels were found between SLE and controls, which may be explained by differences in patient population.⁸ Nevertheless, other HDL-related abnormalities were identified in those studies such as reduced PON-1 activies.⁸ It is noteworthy that HDL-C, which is routinely determined in clinical laboratories, is usually considered synonymous to HDL particle level. However, it has been increasingly recognized that the HDL-C may not be an approriate surrogate of HDL levels since the cholesterol content of HDL does not correlate perfectly with the number of HDL particles.¹² Alternatively, some studies determined HDL levels by measuring apoA-1 concentrations by immunochemistry methods, and lower apoA-1 levels were found in SLE patients relative to healthy subjects.¹³ Recently, direct measurement of HDL particle numbers has been made possible in clinical labs by nuclear magnetic resonance (NMR) methods.¹² Chung et al. found that SLE patients had lower numbers of HDL particles compared to healthy volunteers with NMR method.¹⁹ Patients with active disease have lower HDL levels^20–23, while use of prednisone and hydroxychloroquine correlates with higher HDL-C. ^22,24 In contrast, simvastatin and atorvastatin have not been found to modify HDL-C levels in SLE patients.^25,26

3.2. Changes in HDL Size Distribution in SLE patients

Lupus patients have been found to have a different distribution of HDL sub-fractions and HDL sizes compared to healthy controls (Table 2). However, it should be noticed that these studies are inconsistent and yet remained to be controversial due to the dynamic changes of HDL in inflammatory states which is further discussed in detail in a subsequent section. Hua et al. reported that small HDL is less prevalent in SLE patients, while the overall HDL size is increased in SLE patients compared to healthy controls.²⁷ Similarly, Chung et al. reported that SLE patients have a lesser proportion of large HDL; however, the overall sizes of HDL were no different between SLE and healthy individuals.¹⁹ Other studies have also reported no significant differences in HDL size between SLE patients and healthy controls.¹⁶ Juárez-Rojas et al. reported that SLE patients tend to have lower proportions of HDL2b and higher proportions of HDL3b and HDL3c.²⁸ Similar results were reported by Formiga et al., where SLE patients had higher levels of HDL3-C and lower levels of HDL2-C.²⁹ Since the negative correlation between large HDL levels, positive correlation between small HDL levels and cardiovascular risks have been frequently reported in non-SLE population,³⁰ low levels of large HDL may also be associated with enhanced proatherogenic responses in SLE patients.

Table 2.

Changes in lipoprotein profile and HDL composition in SLE patients

Author (Year)	Subjects	HDL changes	Ref.
Delgado Alves (2002)	32 lupus patients with 20 matched control	PON1 activity ↓, HDL2 ↓, HDL3 ↓	35
McMahon (2006)	154 women with SLE	Pro-inflammatory HDL ↑	38
Juárez-Rojas (2008)	30 women with uncomplicated SLE, and 18 matched controls	Cholesteryl ester ↓ Triglycerides ↑ ApoA-1 ↓ HDL size ↓ HDL2b ↓ HDL3b, HDL3 ↑	28
Batuca (2009)	77 lupus patients	PON1 activity ↓	34
Smith (2014)	SLE patients and healthy control	Oxidized ApoA-1 ↑	39
Marsillach (2015)	54 SLE patients and 25 matched healthy controls	PON3 ↓	62
Gaal (2016)	51 SLE patients with 49 matched healthy controls	SAA combination ↑ PON1 arylesterase activity ↓	16
Han (2016)	18 subjects with SLE	SAA combination ↑	33
Kiss (2017)	37 SLE patients	ApoA-1 ↓ PON1 activity ↓	8

3.3. Proteomic and Lipidomic Changes in SLE HDL Composition

HDL in SLE also has abnormal proteomic features, with changes in apoA-I being the most noteworthy. In the study of Machado et al., female adolescents with SLE had higher HDL-C/apoA-I ratio compared to healthy controls, indicating the HDL particles in those SLE patients had a lesser amount of apoA-I.³¹ Juárez-Rojas et al. reported less apoA-I and more apoE content in HDL particles isolated from women with SLE but without other known conditions affecting lipoprotein profiles such as diabetes, infection and lipid-regulating drugs.²⁸ However, in another study comparing 51 SLE patients with 49 healthy controls, SLE subjects were found to have higher apoA-I/HDL ratio, although the overall serum apoA-I level was lower.¹⁶ Similar to proteomic alterations, the lipidome of HDL is also compromised in the SLE setting. Juárez-Rojas et al. found that HDL isolated from SLE patients had less cholesteryl ester and more triglyceride compared to healthy controls.²⁸ Moreover, lipid peroxidation is enhanced in SLE patients due to enhanced oxidative stress that promotes a significant increase in the production of oxidized lipids.³² Further, Gaal et al. and Han et al. found that SLE patients display significantly increased levels of serum amyloid A (SAA) compared to healthy subjects.^16,33 Additionally, multiple studies confirmed a significant reduction in paraoxonase- (PON) 1 and PON-3 activity in SLE patients.^8,16,34,35 The abnormal lipoprotein profiles in SLE patients are summarized in Table 2.

3.4. Pro-inflammatory HDL

HDL exerts anti-inflammatory functions in healthy subjects. However, during inflammatory states, the composition of HDL is altered, and the function of HDL is compromised. This can lead to the transformation of HDL into a dysfunctional, pro-inflammatory particle with reduced cholesterol efflux capacity and unable to perform its normal anti-inflammatory and anti-oxidative functions (Figure 1).³⁶ Compared to normal HDL, pro-inflammatory HDL (piHDL) is characterized by increased SAA content, decreased apoA-I levels and increase apoA-I oxidation, and decreased PON activity.¹⁶ Increased SAA content of HDL results in its displacement of apoA-I, thus, reducing the atheroprotective properties of HDL.³⁷ McMahon et al. reported that 44.7% of SLE patients have increased levels of piHDL in comparison to only 4.1% of controls and 20.1% of rheumatoid arthritis patients.³⁸ Their subsequent study identified piHDL to be associated with increased carotid intima-media thickness (IMT) and plaque by carotid ultrasound, suggesting that dysfunctional piHDL significantly contributes to the development of subclinical atherosclerosis in SLE.⁶

Figure 1.

Formation of dysfunctional HDL or proinflammatory (piHDL) in SLE. In SLE, multiple HDL proteomic changes occur leading to the impairment of HDL’s function. Oxidation of apoA-I methionine and tyrosine residues by chemical and enzymatic pathways leads to reduced HDL’s ability to efflux cholesterol outside the cell and neutralize oxidized lipids. Increased levels of serum amyloid A (SAA) caused displacement of apoA-I on HDL and reduced cholesterol efflux and anti-inflammatory activity. Reduced levels and activity of HDL associated PON-1 lead to the reduced anti-oxidant activity of HDL. Abbreviations: ApoA-I, apolipoprotein A-I; Met, methionine; Tyr, tyrosine; PON, paraoxonase.

3.5. Oxidized HDL

HDL also undergoes structural changes in SLE due to oxidation, and it has been found that SLE patients have higher oxidized HDL (Figure 1).³⁹ Myeloperoxidase (MPO) is the main oxidant enzyme responsible for the oxidation of HDL. MPO-catalyzed oxidation converts normal tyrosine on apoA-I to 3-chlorotyrosine or 3-nitrotyrosine, leading to oxidized HDL.⁴⁰ MPO also oxidizes three methionine residues of apoA-I, methionine (Met)-86, Met-112, and Met-148, contributing significantly to enhancing levels of oxidized apoA-I.⁴¹ In SLE patients, a high concentration of serum MPO is reported, implying a link between increased HDL oxidation and increased MPO in this disease. Recent studies found associations between HDL oxidation and elevated formation of neutrophil extracellular traps (NETs).³⁹ Under normal conditions, NETs play an important antimicrobial role, however, in SLE, NETs are dysregulated. It was reported that the degradation of NETs is hindered in SLE patients, resulting in aberrant elevation of NETs and infiltration of netting neutrophils in tissues. ⁴² Carlucci et al. demonstrated that SLE patients present higher levels of low-density granulocytes (LDGs), which are one subset of neutrophils with enhanced NET formation capacity.⁴³ The increased LDGs levels are found to be associated with the impaired cholesterol efflux capacity of HDL isolated from lupus patients.⁴³ It has been proposed that enhanced levels of NETs enhances the oxidant potential in SLE and leads to the externalization to the extracellular space of oxidant enzymes such as MPO, nitric oxide synthase (NOS), and NADPH oxidase (NOX), ultimately promoting oxidation of HDL (Figure 2). ^43,44 Data from SLE patients showed that MPO and NOX from NETs promote 3-chlorotyrosine modification, and NOS and NOX promote 3-nitrotyrosine modification on HDL, while inhibition of NETs decreased the oxidation of HDL. These observations support the notion that NETs play a major role in the oxidation of HDL in SLE.³⁹

Figure 2.

Formation of dysfunctional HDL and autoantibodies against HDL and apoA-I in SLE. In SLE, abnormal elevation of NETs is observed leading to endothelial cell damage. In presence of NETs increased levels of MPO, NOS, NOX, and ROS are observed causing oxidation of HDL and apoA-I. Oxidation of apoA-I and HDL induces the formation of anti-HDL and anti-apoA-I autoantibodies. Furthermore, oxidation of apoA-I at Met148 leads to conformational changes of apoA-I promoting protein misfolding, dissociation of misfolded apoA-I from HDL and formation of apoA-I amyloid fibrils. This aggregated apoA-I in more immunogenic leading to further increase in anti-apoA-I autoantibody titers. Abbreviations: ApoA-I, apolipoprotein A-I; EC, endothelial cell; SM, smooth muscle cell; NETs, neutrophil extracellular traps; MPO, myeloperoxidase; NOS, nitric oxide synthase; NOX, NADPH oxidase; Met, methionine; Tyr, tyrosine.

In addition to MPO, lipid peroxidation products also mediate the oxidation of HDL. Studies showed that oxidized phospholipids could covalently react with apoA-I at various sites, forming lipid-protein adducts.⁴⁵ It was also found that lipid hydroperoxide could chemically react with apoA-I and apoA-II, leading to the oxidation of Met residues to methionine sulfoxide.⁴⁶ Since clinical studies have shown that lipid peroxidation is enhanced in SLE patients, this may also contribute to the increased oxidation of apoA-I and HDL in this disease.⁴⁷

4. Functional Changes in HDL in SLE Patients

4.1. Impaired Cholesterol Efflux Capacity

Removal of excess cholesterol from macrophages in the artery wall is recognized to be the key process of HDL for protection against atherosclerosis and improvement of cardiovascular outcomes. This process, known as reverse cholesterol transport (RCT), allows translocating excess cholesterol and other lipids from lipid-laden macrophages in atherosclerotic lesions to the liver for elimination. The first and most crucial step in RCT is to efflux cholesterol from macrophages to HDL. HDL isolated from SLE patients displays a 15% decrease in cholesterol efflux ability compared to healthy control HDL.³⁹Decreased cholesterol efflux capacity of HDL purified from the SLE patients has been found to signfiicantly correlate with increased noncalcified coronary plaque burden.⁴³ Multiple studies propose that the reason for decreased cholesterol efflux ability of HDL in SLE is due to increased levels of SAA. As mentioned earlier, increased levels of SAA promote dysfunctional piHDL with diminished ability to remove cholesterol from macrophages and traffic cholesteryl ester to liver.⁴⁸

Oxidation of HDL can also contribute to its reduced cholesterol efflux capacity in SLE. HDL containing oxidized apoA-I has reduced the ability to efflux cholesterol.^49,50 Furthermore, when Met-148 is oxidized in apoA-I, HDL loses the ability to interact with lecithin-cholesterol acyltransferase (LCAT), an enzyme responsible for the conversion to cholesterol ester, which is a key step in RCT.⁵¹ Thus, the oxidation of apoA-I could be a key reason for the impaired process of cholesterol efflux in SLE.

4.2. Reduced Anti-inflammatory Function

In recent years, it has become widely accepted that HDL can directly inhibit the inflammation processes that lead to the development of atherosclerosis.⁵² Although the complex anti-inflammatory mechanisms of HDL have not been fully elucidated, it is suggested that several signaling pathways play a role in this process. First, HDL can inhibit toll-like receptor (TLR) pathways through activation of the transcriptional repressor, activating transcription factor 3 (ATF3). Activated ATF3 is translocated into the nucleus and inhibits the promotion of TLR-induced inflammatory cytokines.⁵³ HDL can also inhibit nuclear factor-κB (NF-κB)-activated cell adhesion molecule expression, thus preventing vascular inflammation.⁵⁴

In contrast to the anti-inflammatory response promoted by healthy HDL, HDL purified from SLE patients induces a pro-inflammatory response. Smith et al. reported that HDL in SLE fails to inhibit cytokine induction driven by TLR pathways.⁵⁵ In comparison to macrophages exposed to healthy HDL, macrophages treated with SLE HDL induced activation of NF-κB and increased expression of tumor necrosis factor-α (TNF-α) and IL-6. In addition, macrophages treated with SLE HDL had significantly repressed ATF3 activation compared to control HDL or untreated group, suggesting that HDL from SLE cannot inhibit TLR pathways via ATF3 activation.

The oxidation of HDL in SLE promotes the binding of HDL to lectin-like oxidized low-density lipoprotein receptor 1 (LOX1R), preventing ATF3 nuclear translocation and leading to increased synthesis of inflammatory cytokines.⁵⁶ Lupus HDL was also found to be able to directly upregulate monocyte platelet-derived growth factor receptor β (PDGFRβ) and increase chemotaxis and TNF-α release.⁵⁷ Another study associated the loss of anti-inflammatory function of HDL in SLE to the increased SAA content in HDL particles. In normal conditions, HDL can inhibit inflammatory responses by disrupting lipid rafts and sequestering plasma membrane cholesterol, however, it is suggested that the SAA binding on the surface of HDL impedes the interaction between HDL and cell membrane, decreasing the anti-inflammatory role of HDL.³³

4.3. Reduced Antioxidant Capacity

In both the general population and in individuals with SLE, increased levels of oxidized LDL (oxLDL) is a well-recognized risk factor for CVD.⁵⁸ Under normal conditions, HDL prevents LDL oxidation by scavenging reactive oxygen species (ROS). The increased levels of oxidized HDL and oxidized apoA-I in SLE reduce the ability of HDL particles to scavenge ROS. In addition, piHDL in SLE may promote LDL oxidation. Elevated levels of oxLDL increase recruitment and adherence of monocytes to activated endothelial cells by increasing expression of adhesion molecules and proinflammatory cytokines.⁵⁹ These monocytes then transmigrate to the arterial intima, taking up oxLDL and eventually maturing to form foam cells.

PON enzymes also play a critical role in the antioxidant functions of HDL. Among the PON enzymes, PON-1 is the major antioxidant in HDL and prevents LDL from oxidation, thereby eliminating biologically active oxLDL.⁶⁰ SLE patients have reduced PON-1 activity^{8,16,34,35,61} and this correlates with the loss in antioxidative function of HDL.¹⁶ While the reasons for the loss in PON-1 activity are not fully understood, levels of various autoantibodies inversely correlate with PON-1 activity. These include IgG against apoA-I^7,16,34, HDL^34,35, and β2-glycoprotein I³⁵, suggesting that autoantibodies may contribute to the decreased activity of PON-1. In addition to PON-1, PON-3, another member of the PON enzymes, is decreased in SLE patients with subclinical atherosclerosis, potentially promoting the loss of antioxidant ability of HDL in SLE (Figure 1).⁶²

5. Formation of Autoantibodies in SLE

The production of autoantibodies is the key manifestation of SLE. In patients affected by autoimmune disorders, highly reactive IgG antibodies against human apoA-I are detected and can bind to both lipid-free apoA-I as well as apoA-I on HDL particles.³⁴ It was reported that 32.5% of SLE individuals tested positively for the presence of anti-apoA-I autoantibodies in association with decreased levels of HDL.⁶³ Similar studies have confirmed the presence of anti-apoA-I and its association with higher disease activity in SLE patients.^34,64,65 In comparison, only 1% of healthy individuals and 20% of acute coronary syndrome patients without an autoimmune disorder have detectable levels of anti-apoA-I.⁶⁶ Presence of anti-apoA-I is described to reduce the activity of PON, leading to increased LDL oxidation.^34,35 Likewise, in lupus-prone murine models, anti-apoA-I was associated with decreased levels of HDL-C and PON-1 activity.³⁴

We hypothesize that increased levels of anti-apoA-I in SLE patients correlates with increased lipid-free apoA-I and oxidized free apoA-I. Lipid-free apoA-I can exist in plasma via several pathways, either due to de novo synthesis or dissociation from HDL due to displacement by elevated levels of SAA. In addition, oxidation of apoA-I at the Met-148 position leads to conformational changes in the protein.⁶⁷ Oxidation of apoA-I favors protein misfolding from the native α-helical structure to β-sheets, facilitating dissociation of apoA-I from HDL particles and initiating apoA-I amyloid fibril formation.⁴¹ Thus, in SLE plasma and in the oxidative microenvironment associated with NETs and atherosclerosis, the levels of structurally modified apoA-I are likely to be elevated, leading to higher levels of lipid-free misfolded protein (Figure 2). Indeed, HDL isolated from SLE patients, the median levels of 3-nitrotyrosine and 3-chlorotyrosine were 1.9- and 120.9-fold higher than those from healthy controls.³⁹ The misfolded and oxidized apoA-I protein is likely more immunogenic, thus, leading to higher titers of anti-apoA-I in SLE.

Anti-HDL antibodies have been recently reported in SLE patients. The differences between anti-HDL antibodies and anti-apoA-I antibodies remain unclear.^34,63 To measure antibody titers in serum, ELISA plates are most commonly coated with lipid-free apoA-I. However, in some instances, the entire HDL particle is used in the assay. It is possible that HDL used in the assay is partially oxidized and antibodies present in SLE patients serum recognize either oxidized or partially misfolded apoA-I that remains to be lipid-bound in HDL particles.⁶⁸ A few studies have reported significantly elevated levels of anti-HDL in SLE compared to both healthy subjects or patients with the primary antiphospholipid syndrome (APS).^69,70 High anti-HDL titers were associated with increased SLE disease activity markers and decreased PON activity, which could lead to loss of antioxidant and atheroprotective functions of HDL and promotion of atherosclerosis development.^8,34,62

6. Applications of HDL therapeutics

Reconstituted HDL (rHDL), nanoparticles prepared from apoA-I or apoA-I mimetic peptides following reconstitution with phospholipids, have been extensively studied as an anti-atherosclerosis therapeutics since 1984.⁷¹ Infusions of rHDL have shown to increase levels of circulating HDL, improve plasma cholesterol efflux capacity, inhibit synthesis of proinflammatory mediators, and improve endothelial function leading to increased overall atheroprotection in animal models and in clinical trials.⁷² As of today, six different rHDL products have been tested in clinical trials, including SRC-rHDL, CSL-111, ETC-216, ETC-642, CER-001 and CSL-112. Two products, ETC-216 and CSL-111, were shown to reduce plaque burden in CVD patients as assessed by intravascular ultrasound.^73,74 CSL-112, a newer formulation of CSL-111, exhibited improved clinical safety and cholesterol efflux capacity in healthy volunteers compared to CSL-111.⁷⁵ A 17,400 patients phase 3 study to explore the ability of CSL-112 to reduce major adverse cardiovascular events in CVD patients is currently ongoing.⁷⁶

HDL therapy has also presented beneficial effects on inflammatory disorders such as sepsis. It can neutralize endotoxin from bacteria, regulate the inflammatory response in macrophages, and inhibit endothelial cell activation in sepsis.⁷⁷ As a result, infusion of CSL-111 suppressed proinflammatory cytokine production, sepsis-induced hypotension, and reduced the severity of clinical symptoms.⁷⁸ Similarly, L-4F, an apoA-I mimetic peptide, demonstrated a beneficial effect in studies of various animal sepsis models by inhibiting proinflammatory cytokine production, reducing sepsis-induced hypotension, protecting against organ damage, and increasing survival rate.^79,80 Charles-Schoeman et al. have proposed a possible therapy with apoA-I mimetic peptides in collagen-induced arthritis, a rodent model of rheumatoid arthritis.⁸¹ Rats treated with combination therapy of apoA-I mimetic peptides, D-4F, and pravastatin showed significantly improved clinical severity score and less erosive disease compared to both non-combination and control groups. Inflammatory cytokines and chemokines were notably reduced with combination therapy as well as improving anti-inflammatory properties of HDL.

6.1. Use of apoA-I mimetic peptides and rHDL in SLE animal models

To date, very few studies have tested the putative benefit of HDL treatment in SLE animal models. A study by Woo et al. showed significant improvements in SLE manifestations in a murine lupus model associated with accelerated atherosclerosis via treatment of apoA-I mimetic peptide, L-4F, alone or with pravastatin.⁸² Notably, treatment with L-4F alone or with pravastatin significantly reduced IgG anti-dsDNA, IgG anti-oxidized phospholipids, proteinuria, glomerulonephritis, and osteopenia. L-4F also improved the anti-inflammatory functions of plasma HDL while reducing pro-inflammatory effects of LDL. In a more recent study from Black et al., increased levels of apoA-I resulted in suppression of lymphocyte activation, IgG anti-dsDNA autoantibodies, interferon- γ (IFNγ)-secreting CD4⁺ Th1 cells, and follicular T helper cells, along with improved glomerulonephritis in normocholesterolemic murine SLE.⁸³ Smith et al. investigated whether rHDL can reverse the proinflammatory effects of lupus HDL by using ETC-642, an HDL mimetic composed of apoA-I mimetic peptide (ESP24218) and phospholipid complex administered in vivo to the NZM2328 lupus mouse model.⁵⁶ Indeed, macrophages exposed to a 1:4 SLE HDL: ETC-642 ratio significantly suppressed TNF-α, IL-6, and NF-kB activation while promoting ATF3 nuclear translocation, suggesting that rHDL can successfully mimic the effects of healthy HDL. The therapy led to significant increases in Atf-3 expression and lowered IL-6 levels in serum, indicating that rHDL can also decrease cytokine synthesis.

6.2. Future directions for rHDL therapy in SLE with CVD

Underscoring the fact that dysfunctional HDL may promote CVD in SLE, HDL therapy may serve as the potential therapeutic because it can restore both the quantity and quality of HDL. It is worth noting that rHDL and naked apoA-I mimetic peptides used in previous studies have not been optimized for CVD prevention and treatment trials in SLE, as they primarily focused on maximizing the efficacy of cholesterol efflux. Therefore, the potential benefit of rHDL therapy for CVD prevention and treatment in SLE patients remains unclear. An advantage of rHDL therapy is that it can be further customized to be more disease-specific. We have recently discovered that the lipid component of HDL can significantly alter cholesterol efflux capacity and anti-inflammatory properties of HDL.⁸⁴ Therefore, further investigations to understand the protective roles of rHDL in SLE with CVD are required and will advise optimization of the rHDL composition. Administration of optimized rHDL in SLE models would likely increase the level of pre-β-HDL and may restore the anti-inflammatory functions of HDL, reduce piHDL and oxLDL. Improved and restored HDL may then exert its diverse protective mechanisms including the reduction of inflammatory mediators and activation of endothelial cells, all whilst improving cholesterol efflux capacity (Figure 3). In addition, prevention of autoantibody recognition may be possible in rHDL therapy via altering protein/peptide composition. These observations suggest that HDL mimetics may serve as an effective strategy with low toxicity potential for reducing CVD risk and, potentially, disease activity in SLE.

Figure 3.

Reconstituted HDL (rHDL) as a putative therapeutic strategy in SLE patients at risk for CVD. Infusion of rHDL in SLE may increase the level of preβ-HDL, reduce the presence of inflammatory mediators and activation of endothelial cells, enhance anti-inflammatory properties, ATF-3 activation, and facilitate cholesterol efflux capacity.