The intellectual foundation of the field dates to the 1980s, when the oxidative modification hypothesis first proposed that the oxidation of low-density lipoprotein, LDL, is an essential step in the formation of foam cells and the initiation of atherosclerotic lesions. In the 1990s this idea fostered the expectation that broadly acting antioxidants such as vitamins E and C might slow the disease. That expectation collapsed in large clinical trials, and the review argues that this failure forced a fundamental scientific pivot. ROS, once viewed merely as deleterious metabolic byproducts, are now understood as critical signaling molecules. The problem, the authors contend, is not that oxidative stress is unimportant, but that global, non-specific antioxidant supplementation is the wrong tool for a process that is spatially compartmentalized, enzymatically specific, and physiologically indispensable.

The review maps the enzymatic machinery that generates ROS in the vascular wall in remarkable detail. The NADPH oxidase family, with seven isoforms designated Nox1 through Nox5 and Duox1 and Duox2, transfers electrons across membranes to reduce oxygen to superoxide. In the vasculature, Nox1, Nox2, Nox4, and Nox5 dominate. Nox2, originally identified in phagocytes, is expressed in endothelial cells, and experimental studies show that Nox2 deficiency attenuates atherosclerosis in animal models. Nox1 is highly expressed in vascular smooth muscle cells and is upregulated in atherosclerotic lesions, where the pro-atherogenic hormone angiotensin II activates it to drive proliferation, migration, and inflammatory gene expression. Nox4 is more ambiguous, generating ROS constitutively but possibly exerting protective effects through hydrogen peroxide-mediated signaling. Nox5, a calcium-dependent isoform with an N-terminal calcium-binding domain, is induced in human atherosclerotic plaques and contributes to endothelial dysfunction and vascular inflammation.

These enzymatic sources converge on a single pivotal interface: the endothelium. NOX-derived superoxide rapidly reacts with nitric oxide to form peroxynitrite, depleting a molecule that normally keeps the vessel lining relaxed, anti-inflammatory, and anti-thrombotic. The result is a shift toward a pro-inflammatory, pro-thrombotic endothelial phenotype. The review also highlights a vicious cycle at the level of endothelial nitric oxide synthase. Under normal conditions eNOS produces protective nitric oxide, but when it becomes uncoupled through deficiency of its cofactor tetrahydrobiopterin or its substrate L-arginine, it generates superoxide instead. Oxidative stress itself oxidizes BH4 and promotes this uncoupling, so the enzyme that should defend the vessel becomes a source of the very radicals that damage it. NOX-mediated ROS, together with mitochondrial dysfunction, also acts as an upstream signal that primes and activates the NLRP3 inflammasome in the vascular wall, triggering cleavage and release of the pro-inflammatory cytokines IL-1β and IL-18 and establishing a self-sustaining loop of endothelial damage and immune cell recruitment.

Mitochondria occupy a special place in this updated framework. Once regarded simply as ATP-generating powerhouses, vascular mitochondria are now recognized as primary drivers of redox signaling. Complexes I and III of the electron transport chain are the principal sites of superoxide leakage, particularly under metabolic stress. Recent work shows that macrophage-derived mitochondrial ROS drive foam cell formation through STAT5 signaling: mitochondrial superoxide facilitates the differentiation of macrophages into foam cells by modulating metabolic reprogramming and enhancing chromatin accessibility at STAT5 target genes. Excess mitochondrial ROS emission and defective mitophagy, the cellular housekeeping of damaged mitochondria, contribute to endothelial senescence and NLRP3 inflammasome activation across atherogenesis. Two other enzymatic sources round out the picture. Xanthine oxidase, which catalyzes the oxidation of hypoxanthine and xanthine to uric acid, generates superoxide and hydrogen peroxide and is elevated in cardiovascular disease. Myeloperoxidase, abundant in neutrophils and macrophages, produces hypochlorous acid, modifies LDL, consumes nitric oxide, activates metalloproteinases, and is enriched within human atheroma, where elevated plasma levels predict cardiovascular risk independently of traditional factors.

The oxidation of LDL itself is a chemically rich process. Free radical chain reactions abstract hydrogen atoms from polyunsaturated fatty acids in LDL phospholipids and cholesteryl esters, generating lipid peroxyl radicals and, ultimately, reactive aldehydes such as malondialdehyde and 4-hydroxynonenal. Enzymatic oxidation may be equally or more important in vivo: 12/15-lipoxygenase directly oxidizes LDL lipids, myeloperoxidase modifies both lipid and protein components, and secreted sphingomyelinase promotes LDL aggregation and oxidation. The resulting oxidized LDL is a heterogeneous population of particles. Minimally modified LDL retains recognition by the LDL receptor but acquires pro-inflammatory properties, while fully oxidized LDL has its apoB-100 lysine residues modified by aldehydes, preventing LDL receptor uptake and enabling scavenger receptor internalization. Among the bioactive products are oxidized phospholipids and oxysterols, cholesterol oxidation products such as 7-ketocholesterol, 7β-hydroxycholesterol, and 27-hydroxycholesterol. These oxysterols are markedly cytotoxic, triggering intracellular ROS generation, organelle dysfunction, and macrophage apoptosis, and biologically relevant oxysterol mixtures upregulate the scavenger receptor CD36 in monocytes and macrophages through PKCδ-, ERK1/2-, and PPARγ-dependent signaling, accelerating oxLDL internalization and foam cell formation.

Foam cell formation, the hallmark of early lesions, follows from a regulatory loophole. Native LDL is not efficiently internalized by macrophages because intracellular cholesterol downregulates LDL receptors, but scavenger receptors are not subject to this feedback. Macrophage uptake of oxLDL is principally mediated by scavenger receptor class A member 1, also called MSR1 or CD204, alongside CD36, which recognizes oxidized phospholipids, and LOX-1, which also mediates endothelial-macrophage cross-talk. Genetic knockout models in mice show that deficiency of either SR-A1 or CD36 significantly reduces macrophage oxLDL accumulation and attenuates lesion development. Once inside the cell, lysosomal acid lipase hydrolyzes cholesteryl esters, and acyl-CoA:cholesterol acyltransferase-1 re-esterifies free cholesterol for storage as lipid droplets. When cholesterol influx overwhelms the efflux transporters ABCA1 and ABCG1, unesterified cholesterol accumulates in organelle membranes, triggering prolonged endoplasmic reticulum stress, mitochondrial dysfunction, and inflammatory activation. Cholesterol-overloaded macrophages shift toward aerobic glycolysis, the Warburg effect, with the glycolytic enzyme PKM2 playing a pivotal role in sustaining pro-inflammatory responses. In early lesions, apoptotic foam cells are cleared by efferocytosis, but in advanced plaques persistent oxidative stress impairs this clearance, and uncleared cells undergo secondary necrosis, feeding the expanding necrotic core.

Oxidative stress also manufactures the vulnerable plaque. Acute cardiovascular events typically result from plaque rupture or erosion rather than gradual narrowing, and vulnerable plaques share specific features: a lipid-rich necrotic core exceeding 40 percent of plaque volume, a fibrous cap thinner than 65 micrometers, abundant macrophage inflammation, and neovascularization. ROS activate matrix metalloproteinases by oxidizing cysteine residues in their pro-domains, relieving autoinhibition, and also increase MMP expression through NF-κB and AP-1. Collagenases MMP-1, MMP-8, and MMP-13 cleave the fibrillar collagens of the fibrous cap, while MMP-9 degrades denatured collagen and basement membrane. Human coronary plaque studies have shown co-localization of MMP-1 with regions of high circumferential stress, suggesting mechanical forces and proteolytic degradation collaborate in rupture. Advanced plaques also sprout abnormal neovessels from the vasa vasorum whose leaky junctions permit intraplaque hemorrhage; heme and iron from trapped erythrocytes generate ROS through Fenton chemistry, and cholesterol crystals formed from excess free cholesterol potently activate the NLRP3 inflammasome. Under hypoxic conditions, HIF-1α target genes such as heme oxygenase-1 can drive ferroptosis in foam cells, further destabilizing the plaque, while vascular smooth muscle cells transdifferentiate into macrophage-like foam cells that compromise cap integrity.

The clinical translation story is where the review delivers its sharpest critique. Large randomized trials of antioxidant vitamins failed consistently: the HOPE and HOPE-TOO trials of 400 IU daily vitamin E in high-risk patients showed no reduction in cardiovascular events and hinted at increased heart failure risk, the Women’s Health Study found no effect of vitamin E on major cardiovascular events among 39,876 women, and meta-analyses concluded that vitamin E does not reduce cardiovascular mortality. The ALL-HEART trial found no reduction in cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke with allopurinol in ischemic heart disease. The NADPH oxidase inhibitor setanaxib, formerly GKT137831, produced dose-dependent atheroprotection in diabetic mice but has so far been evaluated clinically mainly in diabetic kidney disease, with cardiovascular efficacy unproven. Mitochondria-targeted antioxidants such as MitoQ, a ubiquinone derivative that accumulates in mitochondria based on membrane potential, and SS-31, a peptide that targets cardiolipin, showed protective effects in preclinical models, and MitoQ improved endothelial function and reduced aortic stiffness in a small randomized study of older adults, but cardiovascular outcome benefits remain unproven.

Against this backdrop, the most convincing clinical wins have come from targeting the inflammation that oxidative stress ignites rather than mopping up radicals indiscriminately. The CANTOS trial demonstrated that canakinumab, an IL-1β-neutralizing antibody, reduced cardiovascular events in patients with residual inflammatory risk independently of LDL lowering, establishing inflammation as a therapeutic target, though at the cost of increased serious infections. Low-dose colchicine, which inhibits microtubule polymerization and NLRP3 inflammasome activation, reduced cardiovascular events after myocardial infarction in the COLCOT trial and in chronic coronary disease in LoDoCo2. Statins, the cornerstone of prevention, exert pleiotropic antioxidant effects, reducing NADPH oxidase expression and activity and improving eNOS coupling beyond what LDL reduction alone explains. The authors argue that the future lies in precision redox medicine: integrating compartmentalized redox dynamics with immunometabolism, trained immunity, and single-cell multi-omics profiling, and matching targeted redox therapies to specific disease stages and patient biomarker profiles. Emerging biomarkers such as soluble LOX-1 offer mechanistic rationale for monitoring intraplaque oxidative dynamics but remain limited by assay standardization, and targeting LOX-1 with the humanized monoclonal antibody MEDI6570 has yielded inconclusive clinical outcomes. The failure of vitamin trials, the review concludes, does not invalidate the oxidative stress hypothesis; it underscores that only site-specific, stage-specific strategies are likely to convert three decades of mechanistic insight into clinical benefit.

Subject of Research: The role of oxidative stress in LDL oxidation, foam cell formation, and plaque instability in atherosclerosis

Article Title: Oxidative stress and atherosclerosis: from LDL oxidation and foam cell formation to plaque instability – three decades of evidence and remaining therapeutic challenges

Article References: Ji, L., & Du, H. (2026). Oxidative stress and atherosclerosis: from LDL oxidation and foam cell formation to plaque instability – three decades of evidence and remaining therapeutic challenges. Frontiers in Physiology, 17, Article 1968314. https://doi.org/10.3389/fphys.2026.1968314

Image Credits: AI Generated

DOI: 10.3389/fphys.2026.1968314

Keywords: atherosclerosis, oxidative stress, reactive oxygen species, oxidized LDL, foam cells, NADPH oxidase, NLRP3 inflammasome, mitochondria-targeted antioxidants, matrix metalloproteinases, plaque instability, antioxidant vitamin trials, precision redox medicine