Metabolic and Immune Crosstalk in Cardiovascular Disease

Abstract

Cardiovascular diseases (CVDs) including atherosclerosis and heart failure, arise from the intricate interplay of metabolic, immune, and neural dysregulation within vascular and cardiac tissues: This review focuses on integrating recent advances in metabolic and immune crosstalk of the cardiac vasculature that affects cardiometabolic health and disease progression. Coronary and lymphatic endothelial cells regulate cardiac metabolism, and their dysfunction is linked to CVDs. Lymphatics maintain tissue homeostasis, including by clearing metabolic waste, lipids, and immune cells, and their maladaptation in metabolic diseases worsens outcomes. Altered vascular endothelial metabolism in heart failure drives immune-mediated inflammation, fibrosis, and adverse cardiac remodelling. Concurrently, artery tertiary lymphoid organs (ATLOs) formed in the adventitia of advanced atherosclerotic arteries, serve as pivotal neuroimmune hubs, coordinating local immunity through T and B cell activation and neurovascular signalling via artery-brain circuits. T cells within plaques and ATLOs undergo clonal expansion as a result of peripheral tolerance breakdown, with pro-inflammatory CD4⁺ and CD8⁺ subsets amplifying atherosclerosis, effects further shaped by systemic immune activation. Therapeutic strategies targeting endothelial cell metabolism, lymphatic dysfunction, neuroimmune crosstalk and T cell plasticity hold promise for integrated CVD management.

Introduction:

The intricate interplay between metabolism, immunity, vascular and neuronal function lies at the heart of cardiovascular health and disease. Over the past decade, advances in transcriptomics, metabolomics, and imaging have revolutionized our understanding of how endothelial cells, lymphatic vessels, and immune cells orchestrate cardiac and vascular homeostasis. Blood vascular endothelial cells (BECs) and lymphatic endothelial cells (LECs) not only regulate nutrient exchange and waste clearance in tissues but also adapt their metabolic profiles — shifting between glycolysis and fatty acid oxidation — to support proliferation, barrier maintenance, and stress responses. This vascular metabolic flexibility, however, may be reduced under pathological conditions such as heart failure, atherosclerosis, and cardiometabolic disorders. This leads to vascular endothelial dysfunction and lymphatic insufficiency fueling chronic inflammation and adverse metabolic shifts in target organs. Concurrently, the innate and adaptive immune system plays a pivotal role, with T cells and artery tertiary lymphoid organs (ATLOs), emerging as critical mediators of atherosclerosis progression through antigen-driven responses and neuroimmune crosstalk. Adding further complexity, the adventitia of arteries serves as a neuroimmune cardiovascular interface, linking peripheral nerves, immune cells, and vascular components to the brain via artery-brain circuits. This review integrates recent findings on the metabolic regulation of coronary and lymphatic endothelium, the impact of endothelial dysfunction in heart failure, the formation and function of ATLOs, and the dynamic roles of T cells within atherosclerotic plaques and lymphoid niches in driving coronaropathy. We will not extensively cover progress in understanding the phenotypic plasticity and recruitment of vascular smooth muscle cells (VSMCs) and monocytes/macrophages, nor in features of metabolic reprogramming related to trained immunity, and their role in cardiovascular pathology, as these aspects have been recently reviewed and discussed in detail elsewhere^1–4. Together and viewed in context, these insights delineate the multifaceted mechanisms underlying cardiovascular pathology and highlight novel therapeutic opportunities in targeting these intertwined metabolic, immune, and neural pathways to restore cardiovascular health.

1. Metabolic Roles of Endothelial Cells in Cardiac Health and Disease

Endothelial cells (ECs) line blood vessels and are crucial for vascular homeostasis, regulating blood flow, inflammation, and barrier integrity^{5, 6}. Both blood vascular and lymphatic ECs (BECs, LECs) are central to cardiac health, managing metabolic substrate delivery, waste clearance, and tissue homeostasis⁷. Over the past decade, advances in metabolomics, lipidomics, and single-cell transcriptomics (scRNA-seq) have revealed how changes in EC metabolism influences vascular and cardiac health in cardiovascular diseases (CVDs)⁸. Under normal conditions, BECs regulate nutrient and oxygen supply to the energy-demanding heart⁷, while LECs clear metabolic byproducts and maintain fluid and immune balance. This homeostatic state can quickly adapt to changes in tissue needs. For istance, during angiogenesis, glycolysis ramps up to fuel proliferation, while β-oxidation supports biomass production, including nucleotide synthesis⁹. Under stress, BECs shift to anaerobic glycolysis—despite abundant oxygen and fatty acids—to maintain barrier function⁶, potentially shifting intermediate metabolites like lactate, acetate and ethanol, from BECs to interstitial cells. In the heart, the metabolic preference of coronary ECs for glucose under conditions of stress, ensures oxygen and fatty acid availability for cardiomyocytes, with BECs acting as metabolic gatekeepers, directing substrates and hormonal signals to cardiac cells, and shuttling carbohydrate intermediate metabolites to parenchymal cells, including immune cells, fibroblasts, and LECs. However, metabolic dysregulation in microvascular and arterial ECs, triggered by cardiometabolic risk factors, such as obesity, diabetes, oxidative stress, and inflammation, leads to EC dysfunction, contributing to cardiac metabolism dysregulation and disease progression in heart failure (HF) and atherosclerosis⁸. This section highlights current knowledge on the metabolic roles of BECs and LECs, their heterogeneity, and their function in cardiac health and disease.

1.1. Metabolic Profiles and Endothelial Cell Heterogeneity

Advances in metabolomics and scRNA-seq have unveiled metabolic heterogeneity among cardiac EC subpopulations—arterial, capillary, venous BECs, and LECs—highlighting substrate preferences that influence cardiac metabolism, validated by ongoing metabolomic studies⁸. Transcriptomic-based targeted investigations have enabled analyses in distinct cardiac EC subpopulations, allowing molecular comparisons in unprecedented detail of metabolic programs in distinct parts of the vascular tree. For example, previous scRNA-seq studies comparing murine ECs from different organs have reported that cardiac BECs express much lower levels of glucose transporters as compared to brain BECs, but higher levels of genes involved in fatty acid uptake, an aspect shared with skeletal muscle BECs^{10, 11}. The coupling of such transcriptomic data by in situ metabolomic or lipidomic analyses will be necessary to 1) ascertain the functional relevance of altered gene expression profiles; and 2) determine the impact of altered EC metabolic pathways on cardiac pathophysiology.

In most vascular beds, the primary energy source for BECs is glycolysis; however, they are able to adapt the rate of glycolysis to meet their energy demands in response to varying levels of shear stress, oxygenation and nutrients across different anatomical locations but also in response to local cues such as injury or ischemia¹². In the quiescent state, arterial BECs experience high oxygen-rich blood flow and mechanical stress, which requires efficient redox-homeostasis and ATP production^{7, 13}, mediated by oxidative phosphorylation, for cytoskeletal maintenance and nitric oxide (NO) production. Capillary BECs, in contrast, are essential for the maintenance of endothelial barrier integrity while ensuring efficient nutrient- and oxygen-exchange, and providing protection against oxidative stress. These cells primarily rely on glycolysis for ATP production/energy, and their lower mitochondrial density prevents excessive reactive oxygen species (ROS) production. Venous BECs may exhibit an intermediate metabolic phenotype, balancing glycolysis and oxidative phosphorylation to adapt to lower oxygen content and higher lactate levels. On the other hand, LECs that derived from venous BECs are key players in multiple homeostatic tissue functions, such as fluid drainage, immune cell trafficking, lipid transport, and inflammatory resolution¹⁴. LECs primarily utilise glycolysis for energy production and fatty acid β-oxidation /glutaminolysis for biosynthetic functions, and for clearing lipids and metabolic waste^15–17. In the activated (angiogenic / lymphangiogenic) state, BECs switch from quiescence to vessel sprouting, accompanied by increased glucose uptake and glycolysis to allow EC proliferation and capillary outgrowth, while LECs increase fatty acid and ketone body oxidation¹⁸ (see below).

Endothelial heterogeneity is defined as the functional and molecular differences in ECs across the vascular tree in different tissues and organs^{5, 19–21}. In the heart, pre-arterial cells have been shown to drive arterialization during development. These cells include: 1) individual venous cells within the immature vascular plexus that abruptly upregulate mature arterial markers²² and later contribute to building coronary arteries, 2) cells originating from independent pre-arterial and pre-venous sources to contribute to an integrated coronary plexus²³, linking the ability of ECs to invade the myocardium with subsequent formation of arteries. In the adult heart, coronary BECs represent the largest non-myocyte and non-fibroblast cell population, encompassing a broad spectrum of vascular EC subtypes, each adapted to a specific function within distinct segments of the vasculature (Fig.1). Their functionality can substantially change with ageing, including gradual loss of angiogenic capacity. It has been hypothesised that coronary capillary BECs that are in close contact with cardiomyocytes and pericytes represent the primary BEC lineage, exhibit transcriptional changes in multiple disease states such as myocardial infarction, heart failure, or hypertension. This hypothesis is supported by findings from multi-organ single cell/nuclei RNA-seq, and single-cell multiomics atlases established in the past decade, highlighting the metabolic diversity within coronary BECs in the heart. This heterogeneity, driven by distinct metabolic profiles, affects how different BEC subpopulations respond to mechanical and metabolic stressors, with significant implications for cardiac dysfunction including fibrosis, ischemia, and diastolic dysfunction. Following the identification of specific coronary capillary clusters that exhibit gene expression indicative of stress-responses in CVDs, other BEC subclusters have been found to show enrichment in gene sets associated with local immune responses^24–30. In addition, proteomic analysis has revealed sex-specific differences in the underlying factors of coronary capillary BEC dysfunction, with inflammatory responses being predominant in male subjects, while ventricular remodelling and fibrosis responses were higher in female subjects³¹.

Fig 1. Coronary EC heterogeneity and the vascular response to cardiometabolic stressors.

Cardiometabolic diseases, including obesity, insulin resistance, type 2 diabetes, hypertension, dyslipidaemia, and chronic low-grade inflammation, induce coronary changes characterised by fluctuating flow, oxygenation, and oxidative/nutrient stress, and increase the risk of cardiovascular complications. A, intramyocardial arteries; C, intramyocardial capillaries; D, diseased coronary vessel; H, healthy coronary vessels; metab/infl, metabolic and inflammatory; NO, nitric oxide; PCs, pericytes; ROS, reactive oxygen species; TEM, Trans-endothelial migration; V, subepicardial veins. Created in https://BioRender.com.

The concept of EC heterogeneity is also supported by recent findings using scRNA-seq demonstrating that specific coronary BEC subtypes express pro-inflammatory genes in response to metabolic stress³². This vascular pro-inflammatory profile is conductive to immune cell recruitment to the myocardium, creating a vicious cycle of inflammation, oxidative stress, endothelial senescence and damage. Over time, this pro-inflammatory environment promotes vascular endothelial dysfunction and a pro-thrombotic state, further disrupting blood flow in the heart. Inflammation and oxidative stress aggravate fibrosis and diastolic dysfunction, contributing to HF progression³³. A link between endothelial senescence and the pathobiology of HF with preserved ejection fraction (HFpEF) has been established in mouse models of accelerated aging, and in metabolic and hypertensive comorbidities. Various triggers of endothelial senescence all converge on cell cycle arrest and senescence-associated secretory phenotype (SASP) causing vascular inflammation and oxidative stress in HF^{34, 35}. However, not all BECs may react the same way. Under athero-prone shear-stress for example, scRNAseq revealed distinct EC subtypes that undergo pro-inflammatory endothelial-to-mesenchymal transition (EndMT) involving the Notch1/p38 MAPK-NF-kB signalling axis (see below), promoting adverse vascular remodelling and barrier dysfunction³⁶. Such heterogeneous responses may also include unexpected immune cell-like EC phenotypes as a mechanism contributing to promote local inflammation³⁷.

Cardiometabolic diseases – including obesity, insulin resistance, type 2 diabetes, hypertension, and dyslipidaemia, – are associated with vascular changes characterised by fluctuations in flow rates, leading to reduced tissue perfusion resulting in oxidative and nutrient stress. These phenomena are associated with chronic, low-grade sterile inflammation, a key risk factor of cardiovascular complications. A comprehensive understanding of the various roles of BEC subpopulations within the cardiac vasculature, their intricate communication with neighbouring cells, and their reactions to changes in homeostasis, is paramount to improve therapeutic strategies in conditions including both ischemic and non-ischemic HF. Conditions such as diabetes and obesity exert distinctive effects on EC subpopulations: in type 2 diabetes, arterial BECs exhibit an increased susceptibility to oxidative stress, apoptosis, binucleation and NO uncoupling²⁹. Conversely, metabolic risk factors such as obesity or impaired endothelial glycolysis affect capillary BECs, leading to compromised vascular barrier function, microvascular rarefaction, and disrupted interactions with neighbouring cells³⁸. Chronic inflammation has been demonstrated to induce EC dysfunction, leading to impaired vasodilatory responses and increased leukocyte adhesion in BECs³³. Such vascular deficiencies can exacerbate cardiac remodelling and dysfunction, ultimately propelling HF. However, BEC subpopulation-specific studies are needed to delineate precisely where and how inflammation has the most detrimental impact on the coronary vasculature. Although inflammation can impair EC function, it may also be vital for tissue healing; the net effect likely depends on site-specific and temporal factors³⁹. Furthermore, it has been suggested that venous dilation may be a viable target for HF for reducing resting and exercise cardiac filling pressures⁴⁰. Recent advances in scRNA-seq have significantly expanded our understanding of EC heterogeneity, particularly highlighting distinct metabolic reprogramming across different EC clusters in various physiological and pathological contexts. Metabolic shifts are critical during EndMT in diabetic atherosclerosis. ECs undergoing EndMT show decreased fatty acid oxidation and increased glycolytic metabolism, facilitating mesenchymal transition and contributing to plaque progression⁴¹. Metabolic adaptations have also been identified in cardiac ECs after myocardial infarction⁴², in obesity-induced cardiac arteriole ECs⁴³ and in ECs in choroidal neovascularisation⁴⁴. Obesity induces distinct transcriptional shifts in cardiac arterial ECs, particularly in genes involved in fatty acid uptake (e.g. Meox2, Tcf15). In angiogenic conditions, such as choroidal neovascularisation and tumour angiogenesis, scRNA-seq data identified distinct EC subpopulations with unique metabolic signatures (SQLE, involved in cholesterol biosynthesis, and ALDH18A1, involved in proline biosynthesis), highlighting the intricate link between metabolism and angiogenesis⁴⁴. Endothelial cells exposed to disturbed blood flow, characteristic of atherosclerosis, show remarkable metabolic reprogramming. scRNA-seq and single-cell ATAC-seq analyses reveal subpopulations that acquire pro-inflammatory and mesenchymal features under these conditions, suggesting a robust metabolic shift directly linked to altered haemodynamics³⁷. In summary, the heterogeneity of coronary BECs, particularly in their metabolic and stress-response profiles, influences their vulnerability to dysfunction under metabolic stress. This diversity underlies region-specific patterns of fibrosis, ischemia, and diastolic dysfunction (Fig.1).

1.2. Lymphatic Contributions to Cardiovascular Lipid Homeostasis

Lymphatics play a pivotal role in cardiac homeostasis by clearing metabolic waste (e.g., lactate, ketone bodies) and excess lipids, protecting against lipotoxicity and inflammation. Indeed, a rapid and substantial (>40%) increase in lactate levels in cardiac lymph fluid has been reported in dogs following acute myocardial ischemia⁴⁵. Mature lymphatics are characterized by elevated β-oxidation, although they also have the capacity to use glycolysis for generation of energy¹⁸. Indeed, efficient lipid handling is part of the “molecular identity” of lymphatics encoded by their signature transcription factor, Prox1. This key factor, expressed in both LECs and cardiomyocytes, directly drives gene expression of several essential regulators of fatty acid uptake and use¹⁸, ensuring elevated levels of fatty acid transporters (e.g. Cd36, Fabp1-6) and mitochondrial β-oxidation enzymes (e.g. Cpt1a). Secondly, LECs are often positioned far from blood capillaries and thus reside in a microenvironment with lower glucose availability but elevated levels of free fatty acid or triglycerides, enriched in lymph fluid. This lipid-rich context also has the capacity to regulate LEC identity and fatty acid-dependent metabolism. For example, free fatty acid, such as oleic acid, increase gene expression in LECs of fatty acid transporters to stimulate further fatty acid uptake, likely both for use as fuel and as a source of biomass, and for enabling enhanced lymphatic clearance from tissues¹⁸. Moreover, intracellular fatty acid-derived bioactive metabolites, such as acetyl coenzyme A, have recently been found to contribute to epigenetic regulation of Prox1 target gene expression in LECs¹⁸. This includes positive regulation of the Flt4 gene, encoding the essential lymphatic growth factor receptor VEGFR3. This lipid-stimulated regulatory feedback ensures that lymphatic vessels, exposed to free fatty acid-rich fluids and expressing elevated levels of fatty acid transporters and lipases, maintain lymphatic identity, including cell metabolic programs favouring β-oxidation over glycolysis. However, during active lymphangiogenesis LECs may also upregulate pathways for anaerobic glycolysis^{10, 46}, with lactate or ketone bodies as potential intermediate substrates often found enriched in lymph fluid⁴⁵. Experimental studies in mice have demonstrated that loss of the key enzyme controlling ketone body oxidation, 3-oxoacid-CoA-transferase-1, leads to lymphatic dysfunction⁴⁷. Conversely, ketone body supplementation in vivo improved cardiac lymphangiogenesis after myocardial infarction in mice⁴⁷. Currently, a clinical trial [NCT03991897] is ongoing to evaluate the potential benefit of ketone body supplementation in patients with lymphatic dysfunction.

Since the majority of LECs originate from venous endothelium during development¹⁴, they share many molecular characteristics with venous BECs, including similar expression of key metabolic regulators. However, lymphatic capillaries—due to their specialized open-ended design and negative lumen pressures that facilitate absorption—are better suited than veins to drain macromolecules, including lipoproteins from tissues. Notably, when dietary fatty acid overload occurs, or cardiac fatty acid utilization drops (e.g. during ischemia), cardiomyocytes can generate lipoproteins to secrete excess lipids into the interstitium⁴⁸, where cardiac lymphatics subsequently handle their transport and clearance (Fig. 2A). Studies in other tissues, including large blood vessels, have demonstrated that lymphatics efficiently take up lipids from the interstitium, ranging from albumin-bound free fatty acids, such as oleic acid and palmitic acid; lipoprotein-bound triglycerides; as well as preβ- or discoidal-high density lipoprotein-bound cholesterol through the process of reverse-cholesterol transport⁴⁹ (Fig. 2c). In addition to protecting the heart from potential lipotoxicity and/or coronary atherosclerosis, these evacuated excess lipids returning to the blood circulation via lymphatic shunts at the level of the subclavian veins, may also be used directly as fuel in cardiac LECs, thus potentially coupling cardiomyocyte and LEC metabolic activities. Another key mechanism by which lymphatics may influence cardiac metabolic health relates to their essential role in maintaining cardiac homeostasis of both fluid and immune cells (Fig. 2a). Lymphatic clearance of excess extracellular macromolecules, solutes, and water is necessary to prevent formation of myocardial oedema, which severely impacts both local microvascular blood flow and cardiac diastolic function by acutely increasing myocardial stiffness⁵⁰. Similarly, insufficient lymphatic uptake in CVDs of cardiac-infiltrating immune cells leads to delayed resolution of inflammation with accumulation of proinflammatory mediators that may reduce cardiac systolic function⁵¹. It is well known that severe myocardial inflammation alters cardiac metabolism, with the most notable case being increased glucose uptake and use in the setting of myocarditis⁵². However, the role of cardiac lymphatics in limiting cardiac substrate-switching in inflammatory heart diseases remains to be further investigated. The question whether decreased cardiomyocyte use of free fatty acid and/or increased cardiac production of lactate or ketone bodies, such as following myocardial ischemia or development of HF, has the capacity to influence metabolic activity in cardiac lymphatics has not yet been addressed.

Fig. 2. Illustration of the functional role of lymphatics in cardiometabolic disorders.

In the heart, lymphatic dysfunction, induced by cardiovascular risk factors, may contribute to cardiac dysfunction and adverse remodelling by aggravating myocardial edema, inflammation, fibrosis and potentially steatosis. (a). Several metabolic risk factors, including obesity, diabetes and hyperlipidemia, are associated with lymphatic dysfunction. Insufficient lymphangiogenesis and/or lymphatic transport dysfunction slows resolution of inflammation and may promotesexpansion of adipose tissue (b). Insulin regulates several aspects of lymphatic biology (metabolism, inflammatory profile, transport function), and insulin resistance may thus contribute directly to lymphatic transport dysfunction (c). Through the process of reverse-cholesterol transport (RCT), insufficient lymphatic uptake and clearance of lipids, in part mediated by the transporter SR-B1, may contribute to local build-up of cholesterol and other lipids and progression of atherosclerosis (d). Created in https://BioRender.com

Peripheral lymphatic dysfunction predisposes individuals to obesity, while obesity exacerbates lymphatic dysfunction, creating a vicious cycle between metabolic disorders and low-grade inflammation due to lymphatic malfunction⁵³. In addition, lymphatic vessels play a crucial role in the clearance of lipids and lipoproteins from tissues, including the vessel walls of large arteries, through reverse-cholesterol transport. This process appears dependent on lymphatic expression of the scavenger receptor class B type I (SR-BI). Notably, reverse-cholesterol transport is impaired in obesity-prone models (ob/ob mice), while high-fat diet-fed mice display systemic lymphatic dysfunction^{54, 55}. Obesity-associated low-grade inflammation may also alter lymphatic structure and drainage capacities, thus contributing to inefficient lipid and immune cell clearance (Fig. 2a). For example, excessive adipose-tissue expansion promotes prostaglandin and cytokine release from adipocytes, leading to recruitment and activation of inducible NO synthase (iNOS)-expressing proinflammatory macrophages, which can overwhelm local NO˙) gradients and thus cause lymphatic transport dysfunction⁵⁵.

Systemic hyper-permeability in lymphatic collecting vessels has been described in a mouse model of type 2 diabetes⁵⁶. In vitro studies revealed that defective insulin signalling in LECs led to decreased NO˙ production, reduced metabolism, and increased expression of proinflammatory molecules contributing to lymphatic dysfunction. In agreement, blocking the insulin receptor signalling adapter Irs1 sufficed to suppress lymphangiogenesis in vivo⁵⁷. Furthermore, as lymph propulsion in collecting lymphatic vessels depends on LEC-derived NO˙ production to induce spontaneous contractions of lymphatic mural cells, lymphatic insulin resistance, by reducing NO˙ levels, may also lead to lymphatic transport dysfunction (Fig. 2b). However, it is still unknown whether peripheral and/or cardiac lymphatic dysfunction, induced in dysmetabolic states (e.g. hypercholesterolemia and hyperglycaemia), may contribute to cardiac complications associated with obesity and diabetes. Of note, a clinical study demonstrated reduced peripheral lymphatic function in patients with cardiometabolic HFpEF⁵⁸. In the clinic, lymphatic remodelling in arterial walls has been demonstrated in patients with atherosclerosis⁵⁹. In mice susceptible to atherosclerosis (ApoE^−/−), hypercholesterolemia leads to lymphatic leaks and loss of lymphatic valves, causing lymphatic dysfunction⁴⁹. More recently, defective lymphatic transport has been associated with impaired reverse cholesterol transport contributing to the development of atherosclerotic plaques in ApoE^−/− mice⁶⁰. Importantly, ezetimibe, a drug approved for reduction of hypercholesterolemia in patients, has been shown to depend on efficient lymphatic drainage to reduce the formation of atherosclerotic plaques⁶¹. This suggests that enhancing lymphatic function, alongside cholesterol-lowering treatments, could be a promising approach for managing atherosclerosis. Nevertheless, the molecular mechanism underlying lymphatic remodelling and dysfunction in response to hypercholesterolemia remains unclear. It is conceivable that the accumulation of oxidized LDL (oxLDL) within atherosclerotic plaques could directly cause lymphatic dysfunction (Fig. 2c), similar as observed in the liver⁶².

1.3. Endothelial Dysfunction in Heart Failure

BECs experience mechanical shear-stress from blood flow, which plays a fundamental role in maintaining vascular health. Under normal conditions, BECs respond to this shear stress by activating signalling pathways that promote anti-inflammatory, anti-thrombotic, and antioxidant responses. NO, produced by endothelial NO synthase (eNOS) is a key player in this response, which helps relax blood vessels and reduces oxidative stress. Proper eNOS activity and NO production are essential for maintaining vascular health and are modulated by the cellular metabolic state⁶³. However, when EC metabolism is altered, this mechanotransduction—the process by which cells convert shear stress into biochemical signals—can become impaired, contributing to vascular endothelial dysfunction⁶⁴. Under these altered metabolic conditions, BECs often experience decreased mitochondrial efficiency, reduced glycolytic capacity, and changes in β oxidation^{65, 66}. These metabolic shifts interfere with the production of ATP and other energy sources necessary to support eNOS activity and NO production, which are critical for an adequate shear-stress responses⁶⁷. As NO production declines, the protective effects against inflammation and oxidative stress diminish⁶⁸. Additionally, metabolic alterations increase ROS generation, which not only damages cellular components but also disrupts the cell’s response to shear-stress⁶⁹. ROS can further inhibit NO production, leading to a vicious cycle where BECs become increasingly less responsive to shear-stress, losing their ability to maintain vascular tone and protect against thrombosis and inflammation⁷⁰. Over time, the inability of the vasculature to adequately respond to blood flow changes under altered metabolic conditions exacerbates the inflammatory and oxidative tissue environment, promoting atherosclerosis, vascular stiffness, and ultimately HF⁷¹. Thus, addressing both metabolic and mechanotransductive dysfunctions is crucial in therapeutic approaches aiming to restore vascular function and prevent cardiovascular diseases.

The development of myocardial fibrosis represents a pivotal component in the progression of cardiac dysfunction, exhibiting a close association with a deficiency in cardiac contractility ⁷². Myocardial fibrosis may be categorised into the following: subepicardial fibrosis, perivascular fibrosis or diffuse fibrosis, which is most frequently observed in HFpEF. Focal (replacement) fibrosis is a typical consequence of ischemic injury (area of restricted blood flow) or scar tissue replacement of dead cardiomyocytes associated with HFrEF⁷³. In the context of pathophysiological conditions, activated fibroblasts and myofibroblasts have been observed to increase extracellular matrix protein deposition. However, research has also identified the acquisition of a fibrogenic phenotype in immune cells, cardiomyocytes and vascular cells⁷⁴. In response to pathological stimuli, such as Angiotensin II, ECs have been observed to upregulate glycolysis and exhibit increased susceptibility to EndMT, leading to fibrosis in specific heart regions^{75, 76}. ECs can experience both short-term and long-lasting changes in their fate and function, as exemplified by EndMT, where certain EC subsets undergo major transformations. Thus, EndMT-responsive ECs may respond to homeostatic disruptions in ways that either worsen or alleviate vascular dysfunction. The precise molecular pathways underlying these divergent effects, and their context-specific or broad applicability, remain to be fully elucidated and are the focus of ongoing research⁷⁷. It has been demonstrated that certain coronary ECs, particularly those located in the capillary and arterial regions, as well as the valvular endothelium, exhibit an increased propensity to undergo EndMT under disease conditions. Within these sites, specific BEC subpopulations, often exposed to unique hemodynamic forces or inflammatory stimuli, appear “primed” to adopt mesenchymal traits^78–80. The study of these subsets and the molecular cues governing their transition is an active and evolving field, with significant implications for therapeutic interventions aimed at preventing or reversing pathologic tissue remodelling. It has been recently suggested that under disease stimuli or pathological transforming growth factor-beta stimulation, cardiac microvascular ECs directly contribute to cardiac fibrosis by inducing the expression of Sox9 and activating neighbouring fibroblasts. Conversely, loss of Sox9 in BEC prevents cardiac fibrosis and dysfunction⁸¹.

Metabolically altered ECs in larger coronary arteries are critical for NO production, essential for vasodilation. High shear stress normally stimulates eNOS in these ECs, supporting NO production. In a murine model, the absence of functional eNOS has been demonstrated to be a significant predictor of congestive heart failure⁸², while the presence of eNOS has been shown to be protective. Under metabolic stress, eNOS becomes dysfunctional “uncoupled” contributing to oxidative stress and NO decline leading to vascular dysfunction and increased vascular stiffness. This hinders coronary blood flow during diastole, exacerbating diastolic dysfunction. Reduced NO bioavailability also leads to a reduction in coronary flow reserve, which contributes to heart workload under stress, worsening diastolic dysfunction⁶⁸. Notably, endothelium-derived hyperpolarizing factors (EHFs), such as epoxyeicosatrienoic acids, hydrogen peroxide, and potassium ions play a compensatory role in maintaining vasodilation when NO bioavailability is reduced. Although EHF-mediated responses are often preserved or enhanced in HF to regulate vascular tone, their function can be impaired in severe or prolonged HF, particularly in the presence of risk factors like hypercholesterolemia or diabetes, exacerbating vascular dysfunction⁸³. Lipid overload and free fatty acids in metabolic syndrome and obesity have long been recognized to drive vascular inflammation and dysfunction, in part through disrupting NO production⁸⁴. The blocking of eNOS activity under metabolic overload has been shown to force iNOS mediated nitrosative and unfolded protein response (UPR) stress and inflammation, leading to diastolic dysfunction particularly in male mice⁸⁵. Under stress conditions, mitochondrial dysfunction and lipid peroxidation in BECs can increase oxidative stress susceptibility. Under altered metabolic conditions, ECs produce excessive ROS, leading to oxidative damage and cell apoptosis. This damage results in microvascular rarefaction, reducing blood nutrient and oxygen supply and causing localized ischemia in the myocardium. Hypoxia and ischemic conditions in the myocardium trigger maladaptive remodelling and fibrosis, impairing both systolic and diastolic function and ultimately leading to heart failure^{69, 86–88}. One potential molecular mechanism driving microvascular rarefaction is through deregulation of the NAD- dependent histone deacetylase family of sirtuins. Sirtuin-3 in particular is downregulated in the endothelium of diabetic patients, causing metabolic reprogramming, deficient mitochondrial function, increased ROS production and reduced angiogenic potential in the heart⁸⁹. A further proposed mechanism that can lead to reduced or inhibited angiogenic potential and, consequently, microvascular rarefaction within the heart under conditions of angiotensin II-induced pressure overload, is the production and vesicular secretion of SEMA3A. This, in turn, competes with VEGFA binding to the pro-angiogenic co-receptor Nrp1⁹⁰. In addition, microvascular ECs communicate with fibroblasts through SEMA3A-containing extracellular vesicles to activate myofibroblasts and drive the onset and progression of cardiac fibrosis. A reverse communication route, between myofibroblasts and BECs may also contribute, with AngPTL4 upregulation in fibroblasts being suggested to inhibit angiogenesis and drive coronary EC apoptosis, causing rarefaction in the mouse model of metabolic HFpEF⁹¹. These changes impair diastolic function and exacerbate cardiac remodeling, highlighting EC metabolism as a key driver of heart failure progression. Indeed, the multi-centre international prospective study PROMIS-HFpEF confirmed a high prevalence of coronary microvascular dysfunction in HFpEF patients, linking microvascular rarefaction and systemic vascular dysfunction to HF severity⁹². Whether specific BEC subpopulations discovered in the current area of single cell studies are more prone to cause microvascular rarefaction remains to be explored.

Collectively, targeting these unique EC metabolic profiles may provide novel therapeutic avenues for HF. However, as is invariably the case, single-cell analyses are confronted with challenges in sample preparation and computation, as well as differences in the animal model systems³⁰. Consequently, rigorous evaluation is imperative when testing novel therapeutic targets.

2. Vascular Immune Metabolic Reprogramming in Atherosclerosis

Atherosclerosis is initiated by subendothelial accumulation of lipids in the areas of disturbed blood flow that trigger chronic inflammation^{71, 93}. This process involves functional alterations of ECs including active LDL transcytosis, monocyte recruitment, differentiation of monocytes to macrophages, lipid uptake by macrophages and formation of foam cells, as extensively reviewed elsewhere^{1, 71, 94}. Excessive cholesterol uptake can overwhelm macrophage cholesterol metabolism, driving pathological processes that sustain a complex inflammatory and structural environment. This persistent inflammation contributes to significant residual cardiovascular risk, even after acute coronary syndromes, largely due to unresolved inflammation. Increasingly, these events are categorized by their underlying plaque morphology, offering insights into distinct mechanisms of disease progression and therapeutic targets^{95, 96}.

Plaque macrophages exhibit considerable plasticity, adopting context-dependent phenotypes ranging from lipid-processing to pro-inflammatory states. Traditionally, macrophages in atherosclerosis have been broadly categorized into lipid-laden foam cells, IL1B-producing inflammatory macrophages, and tissue-resident macrophages expressing TIMD4, LYVE1, and FOLR2^{2, 97}. However, single-cell technologies have uncovered transitional states between inflammatory macrophages and lipid-laden foam cells, indicating a dynamic continuum of macrophage phenotypes rather than fixed, discrete subsets⁹⁸. These findings reinforce the notion that certain macrophage subsets contribute to plaque destabilization including lipid-associated pro-inflammatory PLIN2^hi TREM1^hi macrophages found in human plaques⁹⁹. Moreover, macrophage polarization and function are closely linked to their metabolic state: pro-inflammatory subsets rely on glycolysis, while anti-inflammatory macrophages depend on oxidative phosphorylation and fatty acid oxidation, as reviewed elsewhere¹⁰⁰. Emerging evidence indicates that monocytes and macrophages can undergo ‘trained immunity,’ a form of innate immune memory characterized by epigenetic and metabolic reprogramming, exacerbates atherosclerosis³. The intricate relationship between macrophage metabolism and trained immunity in atherosclerosis progression has been explored in greater detail in other comprehensive reviews^{3, 98, 100}.

Plaque rupture, marked by fibrous cap disruption overlying a necrotic core, has long been recognized as a key trigger of acute coronary syndromes. However, advanced imaging has identified plaque erosion as a distinct mechanism, occurring without cap rupture. Eroded plaques typically show intact fibrous caps with endothelial injury, fewer immune cells, and lower markers of inflammation, pointing to distinct underlying mechanisms. A recent study further demonstrated that patients with plaque erosion have reduced inflammatory protein expression and a lower rate of recurrent cardiovascular events compared to those with plaque rupture supporting the need for tailored anti-inflammatory strategies¹⁰¹. In addition to differences in inflammation, the composition of the plaque and resulting thrombus also varies by phenotype. While both plaque rupture and erosion trigger platelet activation, the mechanisms and extent of thrombus formation differ. Emerging evidence suggests that platelets may not only drive thrombus formation but also influence plaque healing - a process that can occur even without full vessel occlusion¹⁰². Understanding how lesion type, haemodynamics, and matrix composition shape platelet responses is therefore critical critical for developing effective antithrombotic strategies, improving risk stratification, and optimizing secondary prevention. However, multi-omics data at single-cell resolution from well-characterized cohorts are still lacking and may be crucial to uncover the distinct mechanisms driving plaque erosion versus rupture, although innate immune cell activation may promote plaque regression¹⁰³.

In addition to immune cells, VSMCs play a critical role in atherosclerotic plaque development and stability. VSMCs originate from the medial wall and clonally expand to form the fibrous cap⁴. They transdifferentiate into macrophage-like, osteogenic, or fibrochondrocytic states^104–106. These transitions are driven by the inflammation. e.g., IL-1b suppresses the contractile VSMC phenotype¹⁰⁷, whereas VSMC can adapt to inflammatory macrophage like phenotype¹⁰⁸, that may influence transdifferentiation of VSMCs into osteogenic-like cells via IL-1β signaling^{109, 110}. Overall, these data indicate that these mechanisms may contribute to the development of new therapeutics to treat atherosclerosis. However, most insights into VSMC plasticity come from mouse studies, as lineage tracing in humans is challenging. Recent single-cell multimodal profiling has identified CD200 as a reliable marker for tracking VSMCs and their derivatives in humans, enabling better analysis of phenotypic shifts in this lineage across disease states in humans¹¹¹. Complementary single-cell and epigenetic studies have further revealed regulatory networks driving VSMC proliferation and phenotypic switching¹¹². Disease-relevant VSMC subsets exhibit enhancer activation and transcriptional programs linked to key regulators such as RUNX1 and TIMP1, the latter promoting proliferation through CD74-mediated STAT3 signaling¹¹². These findings not only reveal new therapeutic targets but also underscore the heterogeneity of VSMC responses in vascular pathology.

Recent studies showed a significant enrichment of T cells in mouse and human atherosclerosis, underscoring the importance of the adaptive immune system in disease progression and T cell heterogeneity within plaques^113–118. Importantly, immune cell aggregates termed artery tertiary lymphoid organs (ATLOs) that specifically form in the adventitia, as specialized hubs orchestrating local and systemic responses to plaque progression in mice and humans^119–124 (Fig 3). Although it is established that both arteries and the peripheral nervous system uses the adventitia as their conduits, interaction between the peripheral nervous system and the arterial tree have not been studied with the exception of the resistance arterioles where it is known that sympathetic nervous system axons reach the media smooth muscle cell controlling blood pressure¹²⁵. ATLOs have been shown to form intricate interactions among the immune system, nervous system, and cardiovascular system¹²⁶. These structures integrate immune activation, neural signaling, and vascular remodeling, shaping disease outcomes in context-dependent ways. These structures significantly influence atherosclerotic plaque progression and stability.

Fig. 3. Neurovascular Communication in Atherosclerosis.

NICI is formed in the adventitia of diseased artery segments. Inflammatory mediators may activate nociceptive receptors at sensory axon terminals to convey peripheral inflammatory signals to the brain via dorsal root ganglia and nodose ganglia via sensory and vagal afferents and the spinal cord. Neuronal input from the brain and spinal cord are projected to the vascular adventitia, via sympathetic and vagal efferents, to modulate vascular functions. Unlike the aortic arch, the abdominal aorta with ATLOs contains sensory and sympathetic innervations, whereas sensory, sympathetic and vagal nerves innervate the heart and its coronaries.

2.1. ATLOs act as Communication Hubs in Advanced Atherosclerosis

ATLOs are disease-specific immune cell aggregates that form in the adventitia adjacent to atherosclerotic plaques, distinguishing them as a subset of tertiary lymphoid organs (TLOs) observed in cancer, chronic infection and autoimmune diseases^{120, 121, 127, 128}. Unlike stereotyped secondary lymphoid organs (e.g., lymph nodes, spleen among others), TLOs arise in diseased adult tissues in response to localized chronic inflammation triggered by diverse exogenous or endogenous stimuli such as bacteria in chronic infection, autoantigens in autoimmune diseases, mismatch of major histocompatibility complex proteins in transplant rejection and cancer among many others^{120, 127, 128}. ATLOs show a high degree of territoriality and age-related development because they arise in arterial wall segments burdened by atherosclerotic plaques in hyperlipidemic mice and human arteries, but not in plaque-free artery segments^{119–122, 126, 128–131} (Fig 3). They evolve through distinct stages: from scattered T and B cell clusters (Stage I), to segregated T and B cell zones (Stage II), to complex structures with B cell follicles and germinal centers supported by follicular dendritic cells (Stage III)¹²². This progression mirrors lymph node architecture, suggesting a purposeful immune adaptation. In mice, they primarily develop in atherosclerotic abdominal aorta of aged ApoE^−/− mice featuring all developmental stages and are less frequent in innominate arteries with stage I features¹²². ATLO formation is accompanied by neoangiogenesis, abnormal lymphatics - resembling those found in some forms of cancer^{127, 132}, high endothelial venules (HEVs) and conduit networks^{122, 130}. These connective tissue constituents, like their counterparts in lymph nodes, are known to promote the recruitment and accumulation of both innate and adaptive immune cells and to organize their movement into and out of chronically inflamed tissues¹²¹. Adoptive transfer experiments showed that HEVs can efficiently recruit lymphocytes into the diseased arterial wall, indicating that plaques instruct ATLO development to amplify local immunity¹²². This local response contrasts with reliance on distant secondary lymphoid organs, highlighting ATLOs as specialized “special purpose vehicles” of the immune system tailored to atherosclerosis. ATLOs exhibit functional variability, which are under intensive investigation to identify both their mechanisms of formation and impacts on disease progression^{120–122, 128, 129, 132–135}. Interestingly, meta-analysis of 28 single-cell RNA sequencing data sets from 5 vascular diseases including atherosclerosis, abdominal aortic aneurysm, intimal hyperplasia, isograft, and allograft revealed heterogeneous mechanisms of adventitial TLO formations¹²⁸.

Considering the functional impact of various ATLO immune cell subsets, we have suggested that ATLOs may be dichotomic in nature: Some immune cells may be atherosclerosis-protective such as T regulatory cells while others may promote the disease such as CD8⁺ lymphocytes^{130, 136}. Indeed, lymphotoxin beta receptor signalling may have pro- or anti-atherogenic properties in mice under specific conditions^{130, 137, 138}. This dichotomy suggests that ATLOs’ impact evolves with plaque development, aging, and immune senescence and factors yet to be identified, challenging the notion of a singular role. Recent discovery demonstrated peripheral nerves and their axons as another integral component of the adventitia and ATLOs^139–141, where they interact with immune cells and vascular cells at multiple levels^140–143. Structurally, the proximity of nerves, immune cells, and vascular components in tissues like the adventitia and ATLOs facilitates their crosstalk. Peripheral axon endings directly access to ATLO components including smooth muscle cells and immune cells, but not the plaque and there they form neuro-immune connections with immune cells and neuro-vascular connections with smooth muscle cells^{126, 144}. These interactions play a significant role in regulating cardiovascular function in both physiological and diseased conditions, such as atherosclerosis, hypertension, and HF^144–147

2.2. Neuroimmune Cardiovascular Interfaces (NICIs): Gateways for Neurovascular Communication

The adventitia is a complex and metabolically active tissue containing vasa vasora, lymphatic vessels, resident immune-/stromal-/progenitor- cells and nerves, all of which have the ability to affect vascular health and disease. Beyond hosting ATLOs, it is hardwired to the peripheral and central nervous systems via sensory and sympathetic nerve branches. Considering the anatomy of the adventitia, it can be viewed as a tissue to maintain and control the structure of arteries fundamentally involved in the maintenance of blood vessel homeostasis and serves as a conduit for neural signals to distant targets—a role noted since Andreas Vesalius in the 16^th century^{140, 148}. However, the interaction between the peripheral nervous system and the arterial tree have not been studied with the exception of the resistance arterioles where it is known that sympathetic nervous system axons reach the media smooth muscle cell controlling blood pressure¹²⁵. In atherosclerosis, adventitial restructuring parallels plaque progression by integrating neural, immune, and vascular responses in mice and humans^119–123.

ATLOs integrates vascular, lymphatic, immune and neural elements forming NICIs in atherosclerotic adventitia segments through peripheral sensory and sympathetic nerves that connect the vascular system to the brain to establish a hardwired artery-brain circuit (ABC)¹²⁶. Thus, NICIs control complex, cross-organ tripartite interactions among 3 major biological systems, i.e. the nervous system, the immune system, and the cardiovascular system (Fig. 3). Within ABCs, the sensory nerves - through nociceptors and mechanoceptors - detect changes in endothelial, parenchymal and immune cells to relay this information to the brain. Distinct brain regions in the brainstem, the hypothalamus and the cortex integrate these sensory signals and project autonomic outputs by sympathetic and parasympathetic efferent arms to arteries via the peripheral parasympathetic and sympathetic ganglia modulating inflammation and plaque dynamics¹²⁶ (Fig. 3). Moreover, adventitial neuronal restructuring involves newly formed axon networks of sensory, sympathetic, and vagal parasympathetic axons, was only observed in diseased segments of arteries innervation^{126, 149, 150} (Fig. 3), and resembles the innervation pattern of normal lymph nodes¹⁵¹. Future studies are expected to uncover tissue-specific innervation patterns in other vascular segments tailored to their functions and anatomical microdomains, to regulate cardiovascular metabolism and homeostasis.

Beyond atherosclerosis, NICIs are implicated in other cardiovascular conditions: in hypertension, the sympathetic nervous system drives immune activation in the spleen, leading to inflammation that exacerbates blood pressure elevation^{145, 146, 152}; while in HF, neuroimmune crosstalk involving the stellate and superior cervical ganglia - key sympathetic hubs - can promote arrhythmias through inflammatory remodeling^{147, 152}. These examples illustrate how NICIs integrate systemic signals to fine-tune cardiovascular responses.

NICIs facilitate their tripartite interactions through diverse mechanisms: i) Sympathetic nerves release norepinephrine to modulate it’s activity through interaction with β2 adrenergic receptors in immune cells and smooth muscle cells^{126, 144}. Activation of these receptors (especially α- and β-adrenergic receptors) by neurotransmitters like norepinephrine can lead to vasoconstriction and an increase in vascular permeability. This response facilitates immune cell extravasation, allowing immune cells to move from blood vessels to inflamed tissues. Chronic activation of the SNS and persistent norepinephrine release have been associated with prolonged inflammation, as seen in diseases like hypertension and atherosclerosis^{126, 144–146}. Additionally, chronic stress can amplify adrenergic signalling, further promoting vascular inflammation^{153, 154}. Sympathetic nerves also co-release neuropeptide Y to regulate cholinergic signaling in heart and possibly arteries as well as modulate the heart’s blood flow and influence glucose sensing^{155, 156}. Vascular ECs can respond to neural signals by upregulating adhesion molecules (e.g., ICAM-1, VCAM-1) that enable immune cells to attach and migrate across the endothelium¹⁵⁷. Thus, neuronal signalling creates a microenvironment that either promotes or inhibits immune cell recruitment and activation in a context- dependent manner. ii) Sensory nerves, when activated, release bioactive neuropeptides like substance P and calcitonin gene-related peptide, which interact with immune cells including T cells, VSMCs and ECs expressing specific neuropeptide receptors^{143, 151, 158–161}. Importantly, vascular smooth muscle cells express calcitonin gene related peptide receptors such as transient receptor potential vanilloid 1 (TRPV1) and receptor activity modifying protein 1 that are implicated in atherosclerosis and host defense: TRPV1 activation in VSMCs reduces foam cell formation and inflammation ¹⁶², but its global depletion may worsen atherosclerosis by increasing inflammatory cytokines^{163, 164}, whereas SMC-specific receptor activity modifying protein 1 depletion influences the host defense¹⁶⁰. iii) Parasympathetic vagal pathways, conversely, dampen inflammation via the cholinergic anti-inflammatory pathway by reducing pro-inflammatory cytokines such as tumour necrosis factor-α, potentially slowing down atherosclerosis progression^{146, 165}. Multiple T subtypes also release acetylcholine to inhibit pro-inflammatory cytokine release from macrophages^166–168. iv) Peripheral nerve-derived CXCL12 signalling through its receptor CXCR4 expressed in LECs guides lymphatic sprouting and patterning in embryos and in adult mice¹⁶⁹, while perineuronal lymphatics regulates local T cell inflammation, linking neural signaling to immune and vascular outcomes in disease ^{170, 171}. Moreover, lymphatic vessels present in epi-and peri-neurium of peripheral nerves, including the sciatic nerve of mice and humans, include two subpopulations of LECs (Rnd1^Hi precollector-like LECs and Ccl21^Hi capillary LECs) in sciatic nerves ¹⁷². In line with this, sympathetic denervation after sciatic nerve injury causes expansion of draining lymph node¹⁷³.

Mapping distinct NICIs by scRNA-seq and spatial transcriptomics of the adventitia or the peripheral ganglia may identify the ication of molecular cues, such as inflammatory mediators, neurotransmitters and their congnate, receptors in immune cells, vascular cells and neurons that actively participate in the ABC. For example, neuroimmune guidance cue netrin-1, and semaphorins are expressed by macrophages and ECs in human and mouse atherosclerotic plaques, where they exhibit pro- and anti-atherosclerotic functions depending on cell types involved and ligand-receptor interactions^174–176. Likewise, neuron-derived guidance cues (e.g. semaphorin 3A), calcitonin gene-related peptide, chemokine (e.g., CCL2) and neurotrophic factors (e.g., neuronal growth regulator 1) have shown to impact vascular and immune functions^{158, 177, 178}. Conversely, immune mediators like interleukin-1β or tumor necrosis factor can alter neural activity, creating a feedback loop¹⁶⁵. These advancements will enhance our understanding of their roles in the cardiovascular system and support translational studies in human PNS and the brain. We expect that novel unbiased experimental techniques and ABC-related circulatory biomarkers will be identified in the coming years.

2.2. T Cell Dynamics and Tolerance Breakdown

Recent scRNA-seq studies highlights T cells as key players in atherosclerosis^{113–118, 124}, shifting focus from macrophage dominance^{99, 179–182}. T cells originate from bone marrow-derived common lymphoid progenitors, and migrate to the thymus for differentiation and selection to ensure self-tolerance. Central tolerance is enforced through two key mechanisms: negative selection, which eliminates thymocytes with high-affinity T cell receptors (TCRs) for self-antigens to prevent autoimmunity; and clonal diversion, which redirects persistent self-reactive T cells into regulatory phenotypes to maintain immune homeostasis¹⁸³. As thymic selection shapes the initial T cell repertoire, peripheral lymph nodes play a pivotal role in T cell responses in atherosclerosis by acting as central hubs for antigen presentation and immune activation^{93, 183}. With aging, thymic involution exacerbates T cell dysregulation¹⁸⁴, reduces thymopoiesis, increases cellular senescence and shifts T cell composition toward pro-inflammatory subsets like effector memory and exhausted T cells^{185, 186}. These changes contribute to chronic inflammation, including atherosclerosis, by disrupting T cell homeostasis and promoting immune-mediated vascular pathology. Beyond thymic selection, peripheral lymphoid organs, particularly lymph nodes regulate T cell responses in atherosclerosis by acting as central hubs for antigen presentation and immune activation^{93, 183} Apolipoprotein B (ApoB), a key component of low-density lipoprotein (LDL) lipoprotein, is not typically an auto-antigen in its native state. However, when modified through oxidation, it can become immunogenic and acts as an auto-antigen^{93, 187}, since the body may recognize the modified protein as foreign. It is a prominent antigen detected in regional lymph nodes draining mid- to large-sized arteries in mice and humans¹⁸⁸. Plaque-derived dendritic cells (DCs) migrate via lymphatic vessels to present (auto)antigens, such as ApoB^{183, 187–190} to naive T cells in an MHC I or II dependent manner, alongside co-stimulatory signals (e.g., CD28-B7, ICOS-ICOSL, CD40-CD40L)^{183, 191, 192}. This triggers clonal expansion of antigen-specific cytotoxic or helper T cells, which differentiate into effector subsets, enter circulation through efferent lymphatics, and potentially homing back to atherosclerotic plaques. In line with this, lymph nodes draining aortic and carotid arches (e.g., axillary and cervical) enlarge as well as adventitial and subepidermal lymphatics expand in atherosclerotic mouse models^{188, 189}, and in carotid atherosclerosis patients¹⁹³, supporting a broader role for lymphoid structures in disease progression. Lymphatics in the arterial adventitia further facilitate reverse cholesterol transport¹⁹⁴ and DCs’ migration from plaques to the draining lymph nodes, connecting local inflammation to systemic immune responses^{188, 190}. However, whether cholesterol-laden macrophages exit plaques via lymphatics or transendothelial migration into the vessel lumen^{190, 195} remains unresolved.

TCR clonal expansion following antigen recognition is central to adaptive immunity in atherosclerosis. Upon antigen presented by dendritic cells in lymph nodes, naive T cells undergo rapid clonal expansion, driven by signals from the TCR and co-stimulatory molecules such as CD28¹⁹². This proliferation promotes differentiation of antigen-specific T cells into effector and memory subsets¹⁹². scRNA-seq and TCR sequencing of human carotid plaques and matched peripheral blood samples reveal significant plaque-specific clonal expansion of effector CD4⁺ T cells, expressing activation markers such as CD69, FOS, FOSB), highlighting their potential role in local immune activation¹⁹⁶. Moreover, clonally expanded T cells displayed transcriptomic signatures indicative of antigen-specific stimulation and active interaction with foam cells, exacerbating local inflammation. In contrast, clonal expansion of CD8⁺ T cells in plaques compared to peripheral blood is less pronounced, though a specific subset of CD8⁺ CD127^low T cells subsets show clonal expansion in a previous work using TCR bulk sequencing¹¹³. In parallel, peripheral T cell tolerance weakens in advanced atherosclerosis, with the most pronounced dysfunction observed in plaques, followed by ATLOs, lymph nodes, and blood¹⁹⁶. This antigen-driven immune dysregulation in atherosclerosis, marked by disrupted tolerance checkpoints, aberrant T cell transcripts in clonally expanded CD4⁺ and CD8⁺ T cells, exhaustion, and impaired antigen presentation, is confirmed by scRNA-seq in human coronary and carotid plaques^{183, 196}. Notably, aging exacerbates this dysregulation through the accumulation of clonally expanded memory CD8⁺ GZMK⁺ T cells in plaques, contributing to disease progression^{197, 198}, particularly in aged low density lipoprotein-deficient male mice on chow diet^{199, 200}. Experimental depletion of CD8⁺ T cells in aged mice reduces plaque progression, while their adoptive transfer from aged donors exacerbated plaque formation, highlighting pathogenic role and therapeutic potential of age-associated memory CD8⁺ T cells in atherosclerosis¹⁹⁷.

In atherosclerotic plaques, CD4+ T cells are the dominant leukocyte fraction and serve key adaptive immunity regulators. They exhibit both pro-atherogenic and regulatory phenotypes, including Th subtypes (Th1, Th2, Th9, Th17, Th22, T follicular helper, CD28^null) and Treg subtypes (FOXP3⁺ Tregs, type 1-Tregs, Ex-Tregs) -each with distinct functions have been extensively reviewed elsewhere^{115, 183, 201}, specifically Th1 cells primarily driving inflammation²⁰¹. Single-cell proteomics and scRNA-seq reveal Th1 and Th2 enrichment in plaques compared to peripheral blood mononuclear cells^{113, 115}, though traditional Th subtype clustering is less distinct in transcriptomics data. Recently scRNA-seq analysis of murine and human plaques circumvented this limitation by applying established Th subset-defining module scores to assess CD4⁺ lineage commitment¹¹⁵. This refined approach revealed that a substantial fraction of plaque T cells (~ 41%) exhibit multi-lineage commitment, challenging the traditional model of distinct Th subtypes in atherosclerosis. Both species exhibited mixed Th1 and Th17 signatures: murine plaque-resident T effector memory cells were Th1-enriched, whereas T follicular helper cells predominated in adventitia. In contrast, human plaques showed higher Th17 and Treg cells, with increased T cell activation, dysfunction, and reduced plasticity. While Th1 cells remained the most abundant subset in human plaques compared to paired blood, a significant proportion of plaque T cells display multi-lineage commitment. These CD4⁺ T cell dynamics are influenced by multiple regulatory mechanisms. For instance, multi-lineage T cells may originate from FoxP3⁺ Tregs under inflammatory conditions¹⁸⁷. Conventional DCs regulate Treg abundance via non-canonical chemokine signaling: CCL17 suppresses Tregs through CCR8, while CCL3 ablation boosts Tregs and limits atherosclerosis, although FoxP3⁺ Tregs are reduced in symptomatic human carotid atheroma²⁰². Additionally, the epigenetic regulator EZH2 (enhancer of zeste homolog 2 mediating H3K27me3) in CD4+ T cells promotes type 2 immune responses, leading to an accumulation of iNKT2 and Th2 cells, memory T cells, and anti-inflammatory macrophages, ultimately restraining atherosclerosis progression²⁰³. These processes underscore a dynamic interplay between regulatory and effector functions, contributing to immune dysregulation in atherosclerosis after prolonged antigen exposure.

ATLOs (as described above) sustain local T cell activation and differentiation by recruiting circulating T cells via HEVs to regulate local immune responses, potentially modulating the progression of atherosclerosis^{120, 121, 144}. scRNA-seq analyses of isolated adventitial specimens have identified major immune populations such as monocyte-macrophages, B cells, and T cells, as well as innate lymphoid cells and natural killer cells, revealing the immunological heterogeneity of adventitia and ATLOs²⁰⁴. Recent studies using paired scRNA-seq and TCR sequencing revealed that T cell tolerance breakdown in advanced atherosclerosis involves plaques and ATLOs^{114, 136, 183}. Immune tolerance is critical for preventing self-reactive responses and relies on five main principles: naive T cell quiescence, effector/memory T cell support, Treg immunosuppression, myeloid antigen presentation, and tissue T cell homeostasis. Compromised tolerance checkpoints such as clonal expansion of CD4⁺-, CD8⁺-T cells and Treg cells drive atherosclerosis^{114, 136}. Interestingly, CD8 tolerance dysfunction observed in the animal model and human plaques parallels CD4 dysregulation, suggesting an autoimmune component to atherosclerosis. This opens avenues for immunotherapeutics targeting antigen-specific T cell responses. However, current scRNA-seq studies of whole mouse aortas often include adventitial cells, complicating plaque-specific immune profiling. Integrated analysis of plaques and adventitia reveals distinct populations: in murine samples, adventitia is enriched in T cells, B cells and neutrophils compared to plaques^{115, 122, 123, 130, 131}, while human plaques contain 2.6- to 4.1-fold more CD4⁺ and CD8⁺ memory T cells, although adventitial T cells show stronger proliferation signatures^{115, 130}. This highlights the need for spatially resolved studies to distinguish plaque-specific immunity.

Collectively, the antigen-driven expansion of CD4⁺ and CD8⁺ T cells drives immune activation within plaques, while progressive TCR expansion and tolerance breakdown exacerbate disease pathology. Lymph nodes, particularly those in proximity to affected arteries, represent an important reservoir in this process by facilitating antigen presentation, T cell priming, and clonal expansion, thereby fueling chronic inflammation in atherosclerosis. These findings reinforce the possibility that atherosclerosis may have an autoimmune component and suggest that targeting antigen-specific T cell responses could offer novel therapeutic strategies.

3. Therapeutic Implications

Endothelial dysfunction underpins many cardiovascular pathologies by driving oxidative stress, inflammation, fibrosis and reducing nitric oxide bioavailability. By selectively modulating endothelial metabolism, reducing oxidative stress, or enhancing NO availability, these treatments provide promising avenues to combat endothelial dysfunction and its consequences in heart. In arteries, interactions between immune cells, vascular cells and nerve axons in plaque and ATLOs regulate vascular health and pathologies. Advances in pharmacology, lymphatic modulation, neuro-immune targeting, and immunology offer a variety of therapeutic strategies to enhance EC function, improve cardiometabolic pathways, and reduce vascular damage. This section outlines the clinical and research implications of these approaches, highlighting their potential in treating conditions like heart failure (HF), diastolic dysfunction, and atherosclerosis.

Pharmacological interventions modulating EC metabolism, oxidative stress and mitochondrial functions show promise in HF: AMPK activators like metformin have been shown to enhance glycolysis and NO production, and reduce ROS production to improve vasodilation and mitigate vascular inflammation^205–207. Inhibition of transforming growth factor-beta signaling also reduced fibrosis^{208, 209} that offers benefits in diastolic dysfunction and HF. Metformin’s dual role in endothelial protection and fibrosis reduction supports its broader use beyond diabetes, particularly in patients with early diastolic impairment. Sodium-glucose cotransporter 2 (SGLT2) inhibitors such as empagliflozin, originally developed for diabetes management, have shown cardiovascular benefits, particularly in HF patients²¹⁰. These agents reduce ROS, boost NO bioavailability, and enhance mitochondrial function to improve EC function and myocardial health^{211, 212}. Their success in HF patients underscores a shift toward metabolic therapies, especially in patients with endothelial-driven diastolic dysfunction. Soluble guanylate cyclase stimulator (e.g. vericiguat) and NAD⁺ boosters (e.g. nicotinamide riboside and nicotinamide mononucleotide) enhance vascular function by enhancing NO bioavailability and reducing oxidative stress²¹³, which have been implicated in HF with reduced ejection fraction (HFrEF)²¹⁴, and in HF with preserved ejection fraction (HFpEF)²¹⁵ respectively. However, PPARγ agonists (e.g., Pioglitazone) and dipeptidyl peptidase-4 inhibitors (e.g., Sitagliptin) require further trials to clarify their role in HF prevention versus risk exacerbation^{216, 217}, although they reduce inflammation, improve NO bioavailability and EC functions. Therapies targeting mitochondrial efficiency such as mitochondria-targeted antioxidants (e.g., MitoQ) and NAD⁺ boosters (e.g., Nicotinamide Riboside) directly address oxidative stress and position them as potential candidates for microvascular protection in HF and complementary therapeutic avenues^{215, 218}. These therapies illustrate how metabolic and targeted interventions can improve coronary EC function and thus cardiac health. By selectively modulating EC metabolism, reducing oxidative stress, or enhancing NO availability, these treatments provide promising avenues to combat endothelial dysfunction and its consequences in heart disease.

Beyond blood endothelial-targeted therapies, cardiac lymphatic drainage plays a vital role in cardiovascular health²¹⁹. Experimental studies in animals reveal that lymphatic transport insufficiency in post-myocardial infarction (MI) or pressure overload exacerbates edema and and inflammation²²⁰, there by aggravates cardiac dysfunction. Therapeutic lymphangiogenesis, using growth factors like VEGF-C, has been shown experimentally to accelerate resolution of oedema and inflammation after MI, potentially restoring cardiac metabolism leading to partial prevention of HF development⁵¹. While its impact on cardiomyocyte metabolism or cardiac steatosis remains unclear, improved lymphatic function could mitigate lactate and triglyceride buildup during ischemic stress, potentially enhancing cardiac recovery. Moreover, enhancing lymphatic drainage alongside cholesterol-lowering drugs like ezetimibe could mitigate atherosclerosis by improving reverse cholesterol transport in plaques⁶¹, though impacts on cardiac steatosis require further study⁶².

Targeting the interplay between the nervous, immune, and cardiovascular systems emerge as novel therapeutic strategies. Understanding these communication interfaces and mechanisms of interactions opens potential therapeutic possibilities for treating cardiovascular diseases. Targeting these interfaces, through neuromodulation, pharmacological agents, or bioelectronic approaches, could provide new ways to manage cardiovascular diseases. For example, modulating the cholinergic anti-inflammatory pathway through vagal nerve stimulation (VNS) has shown potential in reducing systemic and vascular inflammation^{145, 146, 165, 167}. VNS has been explored as a way to leverage the body’s natural anti-inflammatory pathways, potentially offering therapeutic benefits in controlling atherosclerosis. Additionally, targeting neuropeptide pathways or modulating adrenergic receptor activity could provide new approaches for controlling inflammatory vascular diseases, such as atherosclerosis. Studies suggest that modulating the sympathetic input to the adventitia or bone marrow could reduce vascular inflammation and leukocyte supply, key drivers of atherosclerosis. These approaches highlight NICIs as active regulatory networks with profound therapeutic implications for health and disease, specifically for atherosclerosis and HF, with ongoing research needed to optimize neuromodulation protocols and identify responsive patient subsets.

Recent advances in scRNA-seq and TCR sequencing revealed T cell dynamics and T-cell clonal expansion in atherosclerotic plaques, ATLOs, lymph nodes and circulation^{113–115, 124}, suggesting chronic antigen-driven responses in coronary artery diseases^{114, 183, 196}. Clonally expanded T cells in plaques and lymphoid niches present targets for personalized therapies. Key research priorities including conducting human studies with multi-omics and scTCR-seq to dissect T-cell roles in plaque, adventitia, and lymphoid tissues; validating T-cell subsets and TCR clones in larger cohorts for clinical relevance; identifying disease-specific antigens (e.g., ApoB or others) responsible for T-cell clonal expansion will enable precision immunotherapies. Identifying these (auto)antigens and modulating pathogenic T cell responses could refine risk stratification and treatment in coronary artery disease, bridging adaptive immunity with clinical outcomes.

These therapeutic approaches—ranging from metabolic modulators to lymphatic and neuro-immune-targeted strategies— offer a multifaceted approach to treat cardiovascular disease. By improving EC function, reducing oxidative stress, and enhancing NO bioavailability, these interventions offer promising avenues for managing HF, diastolic dysfunction, and atherosclerosis. Emerging fields like therapeutic lymphangiogenesis and NICI modulation, alongside precision immunotherapies targeting T-cell clonality, signal a future of integrated, personalized cardiovascular care. Further research is needed to optimize patient selection, clarify mechanisms, and translate preclinical findings into clinical benefits.

Limitations:

This review explored the interplay of endothelial cell (EC) metabolism, lymphatic function, immune responses, and neuroimmune interactions CVDs. While these studies provide groundbreaking insights, they are subject to several limitations: 1) Limited Human Data: Human studies are underrepresented compared to preclinical models. For example, while scRNA-seq has revealed T cell clonal expansion in human plaques, these findings are based on limited cohorts (e.g., carotid plaques), and larger, well-characterized human cohorts with multi-omics data are needed to validate findings. Moreover, lineage tracing of VSMCs or ECs is challenging in humans, restricting insights into phenotypic switching or EndMT to mouse studies. Additionally, mouse models often involve specific genetic knockouts or controlled conditions that may not reflect the heterogeneity of human CVDs, particularly in complex diseases like HFpEF or atherosclerosis influenced by multiple comorbidities. 2) Incomplete mechanistic understanding: The molecular mechanisms underlying key processes, such as lymphatic remodeling in atherosclerosis or the role of NICIs in modulating inflammation, remain incompletely elucidated. The interplay between EC metabolism, immune responses, and neural signaling is complex, and causal relationships are often inferred rather than directly demonstrated due to the limitations of current experimental tools. 3) Translational and therapeutic gaps: While therapeutic strategies like VEGF-C for lymphangiogenesis or vagal nerve stimulation show promise in preclinical models, their efficacy in humans remains uncertain due to limited clinical trials. For instance, the clinical trial on ketone body supplementation (NCT03991897) is ongoing, and its outcomes are not yet available. The applicability of metabolic modulators (e.g., metformin, SGLT2 inhibitors) or neuroimmune-targeted therapies requires further validation in diverse patient populations, particularly those with comorbidities like diabetes or obesity. These limitations highlight the need for larger human studies, improved spatial and temporal resolution, and integrative approaches to validate findings and translate them into effective therapies.

Conclusions:

The intricate interplay between EC metabolism, lymphatic function, neuroimmune interactions, and T cell dynamics highlights the multifaceted nature of CVDs, particularly atherosclerosis and HF. Advances in metabolomics, single-cell transcriptomics, and spatial analyses have elucidated the critical roles of coronary and lymphatic ECs in orchestrating cardiac metabolism and maintaining homeostasis. BECs and LECs exhibit distinct metabolic profiles—with capillary BECs favoring glycolysis to support barrier function and angiogenesis during stress, while LECs mainly rely on β-oxidation to manage clearance of metabolic waste—highlighting their complementary contributions to cardiac health. Dysregulation of these pathways, driven by cardiometabolic risk factors, disrupts EC functions, reducing both metabolic-perfusion coupling in coronaries and lymphatic drainage in the heart, which exacerbates inflammation, fibrosis, and cardiomyocyte metabolic dysregulation, all of which trigger HF progression. Simultaneously, ATLOs emerge as a dynamic communication hub in advanced atherosclerosis forming NICIs to integrate immune, neural, and vascular responses during atherosclerosis. ATLOs amplify local immune responses through T and B cell activation, while NICIs facilitate cross-talk between the immune, nervous, and the cardiovascular system, with neural signals modulating inflammation and plaque progression. The presence of clonally-expanded T cells in plaques, ATLOs, and draining lymph nodes further emphasizes the adaptive immune system’s role, with multi-lineage CD4⁺ T cells and pro-atherogenic CD8⁺ subsets driving chronic inflammation and tolerance breakdown in coronary artery disease. These findings suggest that atherosclerosis may harbor a pathogenic autoimmune component, fueled by antigen-specific responses yet to be fully defined. Therapeutically, targeting BEC metabolism with agents like metformin or empagliflozin, enhancing lymphatic function with VEGF-C, modulating neuroimmune crosstalk through vagal nerve stimulation, and oligoclonal T cell with their cognate antigens present novel avenues to mitigate CVD progression. Collectively, these insights reveal a complex cardiometabolic landscape where coronary dysfunction, lymphatic insufficiency, neuroimmune interactions, and adaptive immunity converge to shape disease progression. However, considerable challenges remain, including the need for further human studies to validate preclinical findings, precise delineation of subpopulation-specific cellular responses, and clarification of molecular mechanisms underlying lymphatic remodeling and T cell clonality. Future research and clinical trials hold promise to refine these therapeutic strategies, paving the way for integrated approaches to cardiovascular health.