Abstract

A bimodal pattern of mortality in systemic lupus erythematosus (SLE) exists. Early-stage deaths are predominantly caused by infection, whereas later-stage deaths are mainly caused by atherosclerotic disease. Further, although SLE-related mortality has reduced considerably in recent years, cardiovascular (CV) events remain one of the leading causes of death in people with SLE. Accelerated atherosclerosis in SLE is attributed to both an increase in traditional CV risk factors and the inflammatory effects of SLE itself. Many of these changes occur within the microenvironment of the vascular-immune interface, the site of atherosclerotic plaque development. Here, an intimate interaction between endothelial cells, vascular smooth muscle cells, and immune cells dictates physiological vs pathological responses to a chronic type 1 interferon environment. Low-density neutrophils (LDNs) have also been implicated in eliciting vasculature-damaging effects at such lesion sites. These changes are thought to be governed by dysfunctional metabolism of immune cells in this niche due at least in part to the chronic induction of type 1 interferons. Understanding these novel pathophysiological mechanisms and metabolic pathways may unveil potential innovative pharmacological targets and therapeutic opportunities for atherosclerosis, as well as shed light on the development of premature atherosclerosis in patients with SLE who develop CV events.

Key Indexing Terms:

Systemic lupus erythematosus (SLE) is an autoimmune disease involving chronic systemic inflammation. Despite marked improvement in the 10-year survival of patients with SLE over the past 5 decades, SLE mortality rates remain high compared with those in the general population.1 SLE was also among the top 20 leading causes of death in female individuals 5 to 64 years of age in 2000 to 2015.2 Moreover, patients with SLE have a 1.8-fold increased mortality rate compared with the general population, with increased cause-specific mortality rates for cardiovascular (CV) disease (CVD) and infections.3 Thus, SLE is still one of the most noteworthy risk factors for CV events, particularly for female patients with SLE under 40 years of age, even after controlling for traditional Framingham risk score (FRS) factors4 such as diabetes, hypertension, dyslipidemia, obesity, and smoking status.5

Additionally, there is a concern that clinical risk scores such as the FRS, Systematic Coronary Risk Evaluation (SCORE), QRESEARCH risk estimator version 3 (QRISK3), and others underestimate the increased CV risk in patients with SLE that present with atherosclerotic plaque.4,6,7 Cardiac magnetic resonance imaging, another potential diagnostic tool, has been shown to be effective in detecting subclinical cardiac involvement in patients with SLE and has also been suggested as the preferred first-line noninvasive cardiac imaging modality when available.8 However, it is an expensive imaging tool and is often not readily available. Therefore, improved methods for predicting CVD risk are required.

The atherogenic lipid profile in SLE is characterized by dyslipidemia,9 but Petri et al observed that despite promising findings from short-term studies,10,11 statin therapy provided less CV protection in patients with SLE compared to the general population than was initially thought,12 suggesting that SLE-accelerated atherosclerosis may not be entirely lipid-dependent. So, although the overall mechanisms associated with the development of premature atherosclerosis and CVD in SLE still remain unclear, they are largely considered to be the result of an intricate interplay between profound immune dysregulation, inflammation, and traditional CVD risk factors. Various cell types with high heterogeneity are involved in this pathogenic process, notably, differentially activated monocytes and macrophages, dendritic cells, neutrophils, T and B lymphocytes, and endothelial cells (Figure 1). The foamy macrophage and its intricate link with endothelial damage has been discussed extensively13,14; however the role of cellular metabolism in macrophages and other cells of interest and how it relates to atherosclerotic lesion development in SLE is still unknown. Animal studies show that the local plaque in atherosclerosis is a unique metabolic microenvironment characterized by low glucose and high lactic acid,15-17 and offers a unique opportunity for therapeutic modulation.

Figure 1.

Contribution of dysregulated metabolism in monocytes/macrophages, LDNs, and CD4+ T cells toward the induction of inflammation-driven endothelial damage in individuals with SLE. (1) Chronic IFN-α secretion in patients with SLE drives the characteristic systemic inflammatory profile and contributes to metabolic rewiring in macrophages, LDNs, and CD4+ T cells. (2) This rewiring occurs due to mitochondrial dysregulation characterized by an increased glycolytic and oxidative phenotype combined with the release of mitochondrial nucleic acids. (3) The effects of this phenotype are seen particularly at the immune-vascular interface at plaque sites. These cells have complex interactions with the endothelium in a proinflammatory, glucose-rich, lactate-deficient, and chronic IFN-α secretion microenvironment. (4) IFN-α is chronically secreted by the cells in the plaque milieu; combined with other inflammatory factors, this causes endothelial toxicity. Over time, this vascular damage presents as premature atherosclerosis in individuals with SLE. Created with BioRender.com. CVD: cardiovascular disease; IFN: interferon; LDN: low-density neutrophil; mtDNA: mitochondrial DNA; mtRNA: mitochondrial RNA; OXPHOS: oxidative phosphorylation; SLE: systemic lupus erythematosus.

This review aims to explore recent evidence linking conserved metabolic pathways, such as glycolysis and oxidative phosphorylation (OXPHOS), to their role in inducing vascular damage in SLE, as well as the metabolic adaptations of specific cells at the vascular-immune interface that promote atherosclerosis development.

Chronic type I interferon production drives vascular damage in SLE

Over the last few years, short-term clinical trials with statins have shown improved outcomes for patients with SLE with CVD, such as reduced arterial stiffness10 and improved endothelial cell-mediated vasodilation.11 However, a longer trial investigating the vasculoprotective effects of statins failed to demonstrate this in SLE populations without CV events.12 At 2 years, there was no significant difference in the progression of coronary artery calcium, carotid intima-media thickness or carotid plaque, disease activity, measures of inflammation, or endothelial cell activation. Other studies have also shown similar results,18,19 suggesting that traditional CV risk factors do not completely explain the increased CVD risk in patients with SLE, and subsequently that inflammation plays a key role in driving this process.

Although the clinical manifestations of SLE vary, serum interferon (IFN)-α appears to be present in significantly higher quantities in patients with SLE than in healthy controls.20 Further, the type I IFN (IFN-I) transcriptional signature correlates with the clinical and serological markers of disease activity.21 Accumulating evidence from both human and murine studies supports the role of IFN-I in atherogenesis and its linked clinical manifestations. Experimental data show that systemic or intraplaque IFN-I drives atherogenesis by promoting foam cell formation, activating endothelium and immune cells, and enhancing proinflammatory leukocyte recruitment to arteries.22 Evidence of direct endothelial dysfunction associated with the activation of the IFN-I system in patients with SLE has been documented.23 Specifically, IFN-I plays a prominent role in decreased vascular repair through the upregulation of the interleukin 1 (IL-1) receptor antagonist, repression of IL-1 pathways, and downregulation of the proangiogenic molecule, vascular endothelial growth factor. Further, there is evidence that IFN-I induces premature atherosclerosis by increasing the number of smooth muscle progenitor cells in the bloodstream, which gives rise to macrophages and eventually foam cells, therefore promoting atherosclerotic lesions within the vasculature.24

Immune cell-mediated endothelial damage as a result of aberrant activation by IFN-I has also been noted. Specifically, IFN-I treatment has been shown to upregulate scavenger receptor class A (SR-A), one of the main receptors involved in foam cell formation, in human and mouse monocytes/macrophages, leading to increased foam cell formation.25 Elevated SR-A expression can also be found in human peripheral blood mononuclear cells (PBMC) from individuals with an elevated IFN signature, such as patients with SLE and those with HIV infection.26 As people with SLE or HIV are more prone to CVD, this suggests a causal link between IFN-I–induced foam cell formation and a heightened risk of CVD. In support of this hypothesis, treatment of atherosclerosis-prone mice with IFN-β shows increased lesion size, macrophage content, and plasma chemokine (C-C motif) ligand 5 (CCL5) levels.27

Thus, IFN-I directly and indirectly induces dysregulated cellular functions in immune cells that promote endothelial dysfunction and plaque formation, which could in part be underpinned by metabolic rewiring (Figure 1).

Immune cell mitochondrial metabolic rewiring and atherosclerosis

The development of atherosclerosis in patients with mitochondrial diseases may result from a primary pathological mechanism or from a secondary mechanism associated with, but not confined to, arterial hypertension or hyperlipidemia. Patients with mitochondrial DNA (mtDNA) mutations develop primary mitochondrial atherosclerosis through mechanisms thought be associated with increased oxidative stress, energy deficiency, mitophagy alterations, or accumulation of toxic metabolites.28 Thus, atherosclerosis can develop in patients with mitochondrial disorders in the absence of traditional atherosclerosis risk factors, suggesting that atherosclerosis can be a primary consequence of a mitochondrial defect.

The above studies provide a rationale for the development of atherosclerosis due, at least in part, to dysfunctional mitochondria and/or reduced mitochondrial fitness in the context of existing disease. It is now well established that patients with SLE have impaired mitochondrial function and repair processes.29,30 Work done by Xing and colleagues highlighted mitochondrial rewiring in the development of atherosclerosis in patients with SLE.31 They conducted high-throughput RNA sequencing and differential expression analysis in PBMCs from 19 pairs of age-matched patients with SLE with or without atherosclerosis. Subsequent pathway analysis revealed that PBMCs from the SLE + atherosclerosis group had significantly upregulated oxidative phosphorylation. This oxygen-consuming phenotype promotes an exacerbation of mitochondrial-derived reactive oxygen species (mROS) levels and the concurrent accumulation of inflammatory metabolites. Such an imbalance of mitochondrial redox homeostasis when driven by heightened levels of mROS has been termed “mito-inflammation,” a new concept suggesting that the mitochondrion acts as central regulator, checkpoint, and arbitrator of the inflammatory process, particularly in inflammatory-related diseases such as autoimmune diseases, CVDs, neurodegenerative diseases, and cancer.

Future research on the development of atherosclerosis in the context of existing SLE disease should focus on how this oxygen-consuming phenotype induces endothelial damage at the vascular-immune interface, what metabolites are involved, and which immune cells are implicated.

Do metabolically dysregulated CD4+ T cells promote accelerated atherosclerosis in SLE?

CD4+ T cells rely on glycolysis for inflammatory effector functions, but previous studies have shown that mitochondrial metabolism supports their chronic activation. Both autoimmune and chronic inflammatory diseases support a key role for T lymphocytes of either dysmetabolic or autoimmune/autoinflammatory origin in vascular inflammation.32 CD4+ T cells exert multiple proinflammatory functions through the release of effector cytokines and participate in the vasculitic reaction, contributing to the recruitment of macrophages, neutrophils, dendritic cells, natural killer cells, B cells, and T cells. From a mechanistic standpoint, the pathogenic outcome of spontaneous and/or sustained IFN-I production is shown to be associated with metabolic and bioenergetic changes in several peripheral immune cell subsets.33 The IFN-I–induced inflammatory environment can significantly affect the metabolism of immune cells, particularly CD4+ T cells (Figure 2).34,35

Figure 2.

Monocytes/macrophages, CD4+ T cells, and LDNs undergo metabolic rewiring induced by IFN-α in the immune-vascular niche, leading to a hypermetabolic phenotype accompanied by the release of proinflammatory cytokines. (A) In macrophages, this causes GSK3β inhibition and subsequent release of cathepsin K, promoting endothelial toxicity. (A,C) In both LDNs and macrophages, enhanced metabolism leads to an aberrant TCA cycle driving extracellular release of succinate, triggering SUCNR1, and promoting endothelial toxicity. (B) CD4+ T cells downregulate antiinflammatory molecules such as IL-10R and upregulate CXCR3. Concurrent release of CXCR3-ligands CXCL9, CXCL10, CXCL11 by the endothelium leads to aberrant CD4+CXCR3+ T cell trafficking in this niche, further contributing to endothelial damage. Created with BioRender.com. CXCL: chemokine ligand; CXCR3: chemokine receptor 3; GSK3β: glycogen synthase kinase 3β; IFNAR: IFN-α receptor; IFN: interferon; IL-10R: interleukin 10 receptor; LDN: low-density neutrophil; MMP2: matrix metalloproteinase-2; OXPHOS: oxidative phosphorylation; SLE: systemic lupus erythematosus; SUCNR1: succinate receptor 1; TCA: tricarboxylic acid; TGF-β1: transforming growth factor β1.

Characterizing SLE CD4+ T cell metabolism in humans has been challenging. Very few real-time metabolic studies on SLE CD4+ T cells have been conducted to date, with results having contradictory outcomes. In 1 study, it was noted that CD4+ T cells isolated from the peripheral blood of patients with SLE had increased oxygen consumption rate (OCR), extracellular acidification rate, and spare respiratory capacity in comparison to CD4+ T cells from healthy controls with and without anti-CD3 and anti-CD28 activation.36 Conversely, when CD4+T cells from patients with SLE were assessed in another study in a similar manner, no significant changes in basal OCR, maximal OCR, or spare respiratory capacity were observed.33 Despite these contradictory results, Yang and colleagues suggest that targeting CD4+ T cell metabolism may be a promising approach for the prevention and treatment of atherosclerosis and CVDs in SLE.37 In support of this, mouse models indicate that transferring CD4+ T cells from SLE-prone mice into dyslipidemic mice accelerates the development of atherosclerosis,38 and suggests that CD4+ T cells represent a functional link in a reciprocal enhancement between atherosclerosis and SLE. Lü et al previously demonstrated the therapeutic potential of targeted regulation of glycolytic metabolic reprogramming of CD4+ T cells in delaying the progression of atherosclerosis.39 Additionally, it has been suggested that the normalization of CD4+ T cell metabolism reverses SLE.36

Thus, although murine studies have attempted to answer the question of whether aberrant CD4+ T cell metabolism drives atherosclerosis, human studies are required to confirm and strengthen these findings. We propose that atherosclerosis driven by CD4+ T cells is underpinned by CD4+ T cell metabolic rewiring; however, studies targeting OXPHOS and glycolysis in CD4+ T cells from patients with SLE are required to unravel how this metabolic rewiring contributes to vascular inflammation.

Monocytes and macrophages from patients with CVD, rheumatoid arthritis, and SLE show similar metabolic profiles

The vast majority of macrophages present in the atherosclerotic plaque derive from circulating monocytes (Figure 2). Aberrant inflammatory activation of these macrophages potentiates inflammation and tissue injuries in SLE. These macrophages are characterized by increased glycolytic flux to sustain the rapid demand of energy for adenosine triphosphate (ATP) production. Interestingly, monocyte-derived macrophages from patients with coronary artery disease display elevated glucose utilization compared to control subjects,40 and macrophages from patients with rheumatoid arthritis (RA) and CVD share a hypermetabolic phenotype that is both oxidative and glycolytic. Mechanistically, the enhancement of metabolism leads to the deactivation of glycogen synthase kinase-3β, which induces the excessive production of cathepsin K, thereby promoting epithelial toxicity.41

Signaling through IFN-β–IFN-α receptor subunit–signal transducer and activator of transcription 1 triggers CCL5 production by macrophages, increasing CCR5-mediated monocyte recruitment.27 This explains a correlation between upregulated IFN-I signaling and CCL5 expression in advanced human lesions. IFN-α–induced monocytes acquire a proinflammatory state that is probably a consequence of epigenetic priming occurring in myeloid precursors in the bone marrow. This epigenetic priming also extends to genes associated with cellular metabolism, resulting in the preferential use of aerobic glycolysis over oxidative phosphorylation. This subsequently contributes to the maintenance of a prolonged state of hyperactivation, increasing the ability of monocytes to reach the atherosclerotic plaque and become macrophages.

The tricarboxylic acid (TCA) cycle occurs in the mitochondria and is the main source of cellular energy. Recent reports indicate that dysfunction of TCA cycle–related enzymes and metabolites causes human diseases, such as neurometabolic disorders and tumors; the unexplained roles of these enzymes and metabolites in disease have attracted increasing interest.42 One such metabolite, succinate, can be sensed extracellularly by succinate receptor 1 (SUCNR1) on other cells, including endothelial cells. The binding of extracellular succinate to SUCNR1 has been observed to promote epithelial toxicity in patients with CVD,43 suggesting that the succinate-SUCNR1 pathway and the TCA cycle44 could be implicated in epithelial toxicity. We have shown in vivo that chronic IFN-α treatment leads to increased levels of plasma succinate, indicating a potential role for this metabolite in atherogenesis in patients with SLE with high plasma IFN-α.45 Macrophages from patients with SLE also show similar pathogenic modifications that both induce endothelial damage as well as disrupt endothelial repair (Figure 2).46 In RA, monocytes exhibit metabolic dysfunction, which precedes the clinical manifestation of the disease.47 Similarly, assessing the monocyte metabolic profile may also be relevant in predicting CVD in patients with SLE.

Metabolic characterization of monocytes and macrophages from patients with SLE with and without CVD is required to determine the involvement of the succinate-SUCNR1 pathway and other pathways in inducing epithelial damage through a transcriptomic/metabolomic approach. This may lead to the discovery of biomarkers and/or therapeutic targets in patients with SLE with atherosclerosis and CVD.

LDNs from patients with SLE are hypermetabolic and contribute to epithelial toxicity

A specific group of immune cells termed “low buoyancy granulocytes” were first reported in PBMC preparations obtained from adult patients with SLE over 3 decades ago. A few years later, microarray analysis of PBMCs from pediatric patients with SLE had identified a high expression of neutrophil-specific genes, which eventually was attributed to an increase of LDNs in the PBMC layer.48 Further, this signature was observed in several newly diagnosed, untreated pediatric patients with SLE and was therefore not caused by steroid treatment. SLE LDNs are a heterogenous population consisting of mature LDNs (SSChiCD15+CD14−CD16+) and immature LDNs (SSChiCD15+CD14−CD16–).49 Most of these cells express the transcriptional profile of mature inflammatory cells, whereas a minority display an immature phenotype with a granule protein gene signature that is similar to the transcriptional signature found in early stages of neutrophil development.50

A small population of mature LDNs are known to occur in healthy individuals51; however, immature LDNs are not. Limited data show that LDNs possess a vasculature-damaging phenotype52 that is linked to IFN-α production, as well as a decreased ability to circulate in the microvasculature,53 which contributes to damaging the endothelium.54 Further, in a study that looked at patients with SLE with and without high vascular inflammation or noncalcified plaque burden together with indicators of early vascular disease, plaque vulnerability, and rupture risks, it was found that gene expression profiles pertaining to IFN signaling were significantly associated with low-density granulocytes.55 Our in vivo model of chronic IFN-α treatment in male Wistar rats induced a significant increase in the frequency of these LDNs in the blood following 3 weeks of treatment when compared to the saline control rats,45 which could suggest that patients with SLE with high levels of circulating IFN-α and LDNs may have a higher risk for premature atherosclerosis. Thus, the involvement of high levels of IFN-α and LDNs in atherogenesis and the mechanisms driving this pathogenic phenotype should be investigated further.

Previous cancer studies indicate that LDNs exhibit enhanced global bioenergetic capacity through their ability to engage mitochondrial dependent ATP production and rely on the catabolism of glutamate and proline to support mitochondrial dependent metabolism in the absence of glucose.56 These results were mirrored in a recent study on SLE LDNs that showed that they possess high basal oxidative phosphorylation and glycolysis, and that they were highly proinflammatory. Proteomics analysis identified significantly higher levels of the glutaminase enzyme, which is responsible for catalyzing the breakdown of glutamine to form glutamate as a part of the glutaminolysis pathway to feed mitochondrial metabolism.57 Thus, it could be plausible that metabolic rewiring of LDNs in the presence of chronic levels of IFN-α exacerbate this pathogenic phenotype. It is yet to be determined whether this aberrant metabolic phenotype in LDNs may drive the vasculature-damaging phenotype, and whether their presence may predict early-onset atherosclerosis in patients with SLE. Last, future studies should include whether inhibition of the glutamate pathway may present as an attractive target for future therapies.

Conclusion and perspectives

Evidence in the literature shows the significant association of SLE with atherosclerosis and CVD. Indeed, the chance of myocardial infarctions in female individuals aged 35 to 44 years with SLE is reportedly 50-fold higher than matched controls without SLE.58 So far, SLE-CVD–specific risk calculators have been used to predict the risk of CVD in non-SLE individuals. However, they do not always consider the heterogeneity of the disease and the inflammatory variables that are increased in SLE. For example, in a cross-sectional study of 276 patients with SLE, 6% of patients fulfilled the definitions for high or very high cardiac risk according to the SCORE risk calculation. However, following carotid ultrasound assessment, 32% of the patients were reclassified as very high risk.4,7 Some improvements have been made over time, improving upon SCORE to yield SCORE2; however, this calculator is relatively new and has so far only been tested on White patients with SLE, thus limiting its generalizability at a global level.59

The use of traditional risk factors to diagnose CVD in non-SLE individuals is also limiting, as the presence of undiagnosed SLE could potentially challenge the diagnosis of CVD in certain groups, such as perimenopausal women, who historically do not form the “traditional” CVD demographic. Additionally, early events that govern the development of atherosclerosis still remain unclear. There is also little clarity on whether premature atherosclerosis is a blanket feature of SLE or if it affects only a subsection of patients.60

These issues highlight the need for understanding the immunometabolism of cells involved in inducing endothelial toxicity at the vascular-immune interface and how IFN-I potentially drives this damage within this microenvironment. Moreover, as IFN-I transcriptional signatures largely correlate with clinical and serological markers of disease activity,21 it follows that defining an IFN-I–specific metabolic signature would have translational potential. Correlating the metabolism of immune cells to IFN-I could be used to target the involved pathways/metabolites and ameliorate, if not reverse, the deleterious effects they cause in SLE pathophysiology. Such a metabolic characterization could also aid in supplementing existing predictive biomarkers of disease activity, treatment outcomes, and even disease stratification.

Further, comparing the IFN-I–specific metabolic signatures of individual peripheral immune cells in patients with SLE with premature atherosclerosis, with advanced CVD, and without any atherogenic manifestations could shed light on the temporal progress of the events that define and follow the former. Similar single-cell level studies on malignant tumors have aided in better understanding patterns of molecular evolution and disease-wide heterogeneity.61 IFN-I has also been shown to elicit variable effects on different immune cell types and have implications in treatment strategies.62 These studies underline the importance of understanding disease landscapes and the associated immune cell composition at the single-cell level. Indeed, Libby et al demonstrate sizable heterogeneity in monocyte subsets, and therefore macrophages, in CVD, which goes beyond historically established CD14/CD16-based monocyte classifications.63

On a broader note, linking conserved metabolic pathways to their role in inducing vascular damage in SLE and the involved immunometabolic adaptations of cells at the vascular-immune interface during atherosclerosis could be applied in better stratification of patients with SLE based on CVD status. This has direct application in the administration of differential pharmacological interventions to patients with and without risk of CVD. For instance, some studies have found that even the use of low-dose (5 mg daily) glucocorticoids and corticosteroids such as prednisolone, a common treatment regimen for SLE, augments the risk for the development of CV events in people with inflammatory conditions.64-67

Thus, delineating the complex interactions of immune cell subsets with the vascular endothelium, and the immunometabolic adaptations underpinning them, at lesion sites in patients with SLE with atherosclerosis is of vital importance in order to inform and update existing healthcare indices as well as clinical interventions.

Accepted for publication November 30, 2023.Copyright © 2024 by the Journal of Rheumatology

This is an Open Access article, which permits use, distribution, and reproduction, without modification, provided the original article is correctly cited and is not used for commercial purposes.