Trained immunity: new paradigm in the immunological memory of cardiovascular disease

Abstract

Cardiovascular disease (CVD) remains the leading cause of death worldwide, despite significant progress in identifying and managing traditional risk factors such as hyperlipidemia, hypertension, and diabetes. While targeted therapies addressing these factors reduce the risk of primary and secondary cardiac events, a substantial “residual risk” persists even after successful clinical intervention. This residual risk has prompted renewed interest in understanding the long-term biological effects of cardiovascular risk factors, particularly through the lens of chronic inflammation. Recent advances highlight a pivotal role for trained immunity—a form of innate immune memory driven by epigenetic and metabolic reprogramming—in driving this inflammation. Unlike adaptive immune memory, trained immunity occurs in innate immune cells and enhances their responsiveness to subsequent, unrelated stimuli. Emerging evidence suggests that various cardiovascular risk states, including hypercholesterolemia, obesity, and diabetes, can induce trained immunity, leading to heightened inflammatory tone that persists over time. Cardiac macrophages, as central mediators of tissue homeostasis and inflammation in the heart, are increasingly recognized as critical targets of this phenomenon. In this review, we explore how established cardiovascular risk factors can induce trained immunity on cardiac macrophages and examine the implications for disease progression, myocardial remodeling, and post-injury repair. Finally, we discuss emerging therapeutic strategies aimed at modulating trained immunity to reduce residual cardiovascular risk, offering a new frontier in the prevention and treatment of CVD.

Cardiovascular disease remains the leading cause of mortality worldwide, representing nearly 20.5 million deaths in 2021.¹ Central to the pathophysiology of cardiovascular diseases is immune activation, with systemic inflammation being implicated in atherosclerosis, heart failure, arrhythmias, and aortic aneurysms.^2–6 Inflammation is both a cause and consequence of these pathologies and therefore provides a mechanistic link connecting earlier conditions and future disease. While the immune system’s role in cardiovascular pathologies was recognized as early as the 19^th century, the translation of these fundamental concepts towards new therapeutics is still limited.⁷

Illustrative of this gap in clinical intervention is the observation that patients who receive optimal control of traditional risk factors (ie dyslipidemia, hypertension) remain at elevated risk of future cardiac pathologies—a phenomenon known as “residual risk.” For example, in the Pravastatin or Atorvastatin Evaluation and Infection Therapy–Thrombolysis in Myocardial Infarction 22 (PROVE IT-TIMI 22) trial, it was shown that despite intensive lipid lowering with statin treatment, 22.4% of patients experienced a clinical event 2 yr following the end point of the study.⁸ Both preclinical studies and large-scale clinical trials now unequivocally identify chronic inflammation as a central, independent driver of this residual risk. One of the landmark studies highlighting this was the Canakinumab Anti-inflammatory Thrombosis Outcome Study (CANTOS) trial, in which it was shown that blocking interleukin (IL)-1β signaling in high-risk patients with a prior myocardial infarction led to a reduction in recurrent clinical cardiovascular events.⁹ These results have been further supported by the LoDoCo2 and COLCOT trials which have tested the efficacy of colchicine, another anti-inflammatory medication.^10,11 Therefore, the problem of residual cardiovascular risk can be reframed as a question of persistent, unresolved immune activation: Why do these inflammatory pathways remain active even when the primary stimulus is pharmacologically controlled?

Both innate and adaptive immune cell subsets are critical mediators of the inflammatory component of residual risk and have been reviewed extensively in other recent articles.^5,12,13 Given the focus of this special edition on cardiac macrophages, rather than highlighting all immune mechanisms contributing to cardiovascular disease risk and outcomes, in this review, we highlight the contribution of trained immunity, an innate immune phenomenon. We explore the pathways and mediators identified in the growing field of trained immunity that represent the next frontier of therapeutic targets for resolving this persistent clinical challenge.

Trained immunity

Classically, immunological memory has been attributed solely to the adaptive immune system, which provides highly specific protection against invading pathogens. This specific response is mediated by a series of gene recombination and somatic hypermutation events in naive lymphocytes, leading to the production of mature lymphocytes that express receptors highly specific to a given epitope. The pool of memory lymphocytes generated in this process are responsible for mounting faster, more robust responses to future exposures to the same antigen. In contrast, the more evolutionarily conserved innate immune system relies on non-specific recognition of conserved motifs, expressing pattern recognition receptors that activate in response to microbe-associated molecular patterns (MAMPs), host damage-associated molecular patterns (DAMPs), or the absence self-signals. Given the non-specific nature of the innate immune response, its potential to develop memory responses has historically been overlooked. One of the first descriptions of innate immune memory was made in 1962, when it was shown that macrophages that had been sensitized to Listeria monocytogenes provided enhanced future protection.¹⁴ However, it wasn’t until the 2000s that this phenomenon was again described and coined “trained immunity.”¹⁵ Unlike adaptive memory, innate immune memory is mediated through durable metabolic and epigenetic changes that persist after withdrawal of the initial stimulus.^16,17 These metabolic and epigenetic changes alter the proinflammatory capacity of the trained cell, thus providing non-specific protection against a wide range of secondary immune challenges. Critically, these changes persist even after withdrawal of the initial stimulus and the cell returns to baseline, differentiating this phenomenon from immune priming. While trained immunity has been classically shown to permit faster, more robust secondary responses, a distinct, but related phenomenon known as immune tolerance has also been described.^18,19 Immune tolerance represents the opposite outcome in which the initial stimulus drives cells to an exhausted phenotype, reducing their ability to mount as strong of a subsequent response (Fig. 1).

Figure 1.

Comparison of innate immune responses to repeated or prior inflammatory stimuli.

Trained immunity can occur either to mature circulating or tissue myeloid cells, termed peripheral training, or to bone marrow hematopoietic stem and progenitor cells (HSPCs), called central trained immunity. Whereas peripheral trained immunity is limited by the short lifespan of circulating cells, central trained immunity impacts the HSPCs, a quiescent pool of long-lived cells. In central trained immunity, metabolic and epigenetic reprogramming is maintained across multiple cell divisions and hematopoietic differentiation, allowing trained progenitors to give rise to a continuous pool of new myeloid cells that also exhibit features of immune training.^20–23 As such, the innate immune memory implicated in chronic inflammatory conditions is often attributed to central immune training.

Initial studies in trained immunity focused on its ability to provide cross-protection after an initial pathogenic exposure, such as the Bacillus Calmette–Guérin tuberculosis vaccine, bacterial components such as lipoprotein polysaccharide (LPS), and fungal elements like beta-glucan.^{16,17,24–27} However, it has subsequently been shown that a surprising number of endogenous ligands—including oxidized low-density lipoprotein (oxLDL), glucose, and catecholamines—are also capable of reprogramming the body’s innate immune system.^28–32 Notably, many of these endogenous inducers of trained immunity are associated with known cardiovascular disease risk factors, including hyperlipidemia and diabetes (Table 1). Therefore, this agnostic memory introduces a potential link between temporally distal inflammatory events and future inflammation-driven disease. Accumulating evidence now identifies trained immunity as a key mechanism driving the long-lasting inflammation seen in cardiovascular diseases.^31,56,57

Table 1.

Mediators of central and peripheral trained immunity.

Endogenous ligand	Associated disease state(s)	Evidence type	Model/System	Ref
oxLDL	Hypercholesterolemia, Atherosclerosis	Ex vivo	Human monocytes	31 , 32 , 33–37
LDL, VLDL, lipoprotein remnants	Hypercholesterolemia	Ex vivo	Human monocytes	38 , 39
Western Diet	Hypercholesterolemia	In vivo	Murine bone marrow and circulating myeloid cells (Ldlr−/− mice)	40–42
High-fat diet	Obesity	In vivo	Murine bone marrow progenitors and adipose tissue macrophages	43 , 44
Stearic acid and palmitic acid	Obesity	In vivo	Murine bone marrow and adipose tissue macrophages	45
		In vitro	Murine bone marrow derived macrophages
Unknown (Weight cycling)	Weight cycling	In vivo	Murine bone marrow progenitors	46 , 47
Palmitic acid or obese adipose tissue-conditioned media	Weight cycling	In vitro	Murine bone marrow-derived macrophages	46
Lipoprotein(a)	Genetic	Ex vivo	Human monocytes	38
		In vitro	Human primary monocytes
Aldosterone	Hypertension	Ex vivo	Human monocytes	48
High salt diet	Hypertension	In vivo	Murine bone marrow progenitors	49
High glucose	Diabetes	In vivo	Murine BM progenitors (STZ mouse model)	29 , 30
		In vitro	Murine bone marrow macrophages
Heme	Hemolysis, tissue damage	In vitro	Murine monocytes	50
		Ex vivo	Murine bone marrow derived macrophages	50
Norepinephrine	Myocardial Infarction	In vivo	Murine bone marrow progenitors	51 , 52
	Pheochromocytoma	In vivo	Human monocytes	23
		In vitro	Primary human monocytes
Unknown (sleep interruption)	Sleep restriction	In vivo	Murine bone marrow progenitors	53 , 54
			Human monocytes	53
IL1β	Stroke	In vivo	Murine bone marrow progenitors	55
		In vitro	Human primary bone marrow derived monocytes	23

Immune training of vascular monocytes/macrophages and atherosclerosis

Atherosclerosis arises in large arteries at sites of disturbed flow where endothelial dysfunction permits entry of low-density lipoprotein (LDL) cholesterol particles into the vessel wall. These particles undergo oxidative and enzymatic modifications to form more inflammatory lipoproteins, including oxidized LDL (oxLDL). This then stimulates chemokine release by immune cells within the microenvironment, upregulating adhesion molecule expression and recruiting monocytes. Inflammatory monocytes dominate this infiltrate and differentiate into macrophages. Macrophages internalize lipids to form foam cells, a hallmark of early lesions, and their interaction with modified lipoproteins amplifies local inflammation. Plaque evolution is shaped by the balance between growth and stability. Persistent leukocyte recruitment, macrophage proliferation, and unchecked apoptosis promote chronic inflammation, enlarge necrotic cores, and thin the fibrous cap, leaving lesions vulnerable to rupture or erosion and thereby precipitating clinical events.⁵⁸

Although lipid-lowering therapies mitigate atherosclerotic disease progression, they do not fully eliminate risk of future cardiac events. This observation has prompted a number of efforts to elucidate the mechanisms linking hyperlipidemic states to persistent, chronic inflammation.^59–61 One pivotal finding that emerged from these works was the discovery that oxLDL could induce peripheral trained immunity. In these foundational studies, human monocytes were transiently exposed to oxLDL in vitro, after which they exhibited persistent epigenetic and metabolic changes.³⁵ Subsequent studies have shown that oxLDL-trained monocytes have enhanced oxidative phosphorylation, glycolysis, and glutaminolysis.^31,34,62,63 Metabolites produced by these pathways activate various histone modifiers, including Set 7 methyltransferase and the demethylase KDM5, which drive persistent alterations in the epigenetic regulation of transcription.^62,64 Epigenetic changes induced by oxLDL-driven trained immunity include increased trimethylation of histone 3 lysine 4 (H3K4me3), an activating histone modification found at promoters, at pro-inflammatory gene loci.³¹ These changes permit oxLDL-trained monocytes to mount an augmented pro-inflammatory response when re-stimulated with TLR-2 or -4 agonists.^31,35

The in vivo relevance of these findings has been assessed by evaluating circulating immune cells from patients with Familial Hyperlipidemia (FH), a genetic form of severe high cholesterol. Similar to prior in vitro findings, mature circulating monocytes from these patients had increased H3K4me3 at proinflammatory gene promoters, corresponding to increased transcriptional activity at these loci. Furthermore, these patients exhibited features of central trained immunity, including increased numbers of peripheral HSPCs, a sign of increased hematopoiesis. When removed from the in vivo hyperlipidemic environment and cultured ex vivo, HSPCs from patients with FH exhibited persistent myeloid skewing and enhanced cytokine production.⁶⁵ Importantly, these changes persist even after initiation of statin therapy in these patients, highlighting the gap that currently exists in our treatment approaches to CVD risk states.

Since these seminal studies, animal studies have provided compelling in vivo support for the suggested role of central trained immunity in cardiovascular disease. For example, Western diet feeding, which drives high cholesterol levels in genetically prone mice, has been shown to cause persistent transcriptional and epigenetic reprogramming of HSPCs, leading to production of mature myeloid progeny that are skewed toward a hyperinflammatory state.^31,40 These changes persist even after returning these mice to a standard chow diet to normalize their cholesterol levels. Indeed, when bone marrow progenitors isolated from these “trained” mice are transplanted to hypercholesterolemia-naive mice, the mature myeloid cells that reconstitute recipient mice retain their proinflammatory phenotype and promote accelerated atherosclerosis.⁶⁶ In vivo studies have further advanced our understanding of innate immune memory, demonstrating that it is not only dependent on exposure to a given stimulus but also on the duration and interval of that exposure. For example, the effects of high fat diet on innate immune memory are dependent on whether this exposure is continuous or intermittent. Indeed, modeling weight cycling with intermittent periods of high fat diet led to increased metabolic activity and inflammation of adipose macrophages compared to a standard continuous high fat diet obesity model.⁴⁶ Another study showed that early intermittent exposure to a high-cholesterol Western diet (WD) accelerated atherosclerosis as compared to late continuous WD exposure by altering the number and phenotype of resident-like arterial macrophages.⁴²

Additional studies have shown the potential for ASCVD risk factors other than hypercholesterolemia to induce immune training. Hyperglycemia, a state associated with diabetes mellitus, has been shown in a number of studies to induce peripheral immune training. Brief culturing of human or murine monocytes in high glucose media conditions led to increased proinflammatory cytokine production upon subsequent LPS stimulation.³⁰ In addition to peripheral training, hyperglycemia has also been shown to cause central trained immunity, with transient exposure of human cord blood HSPCs to high glucose resulting in persistent increases in mitochondrial reactive oxygen species, proinflammatory cytokine release, and NFkB-p65 activity. These changes persist even after return to normoglycemic conditions in both the undifferentiated HSPCs as well as the HSPC-derived monocytes.⁶⁷ While the metabolic derangements in diabetes mellitus extend beyond hyperglycemia, there is evidence that exposure to diabetes in vivo may also induce trained immunity. Bone marrow derived macrophages differentiated from the STZ diabetic mouse model, for example, exhibited augmented cytokine release with stimulation and enhanced modified LDL uptake and foam cell formation.^29,30 HSPCs collected from sternal bone marrow of patients with type 2 diabetes mellitus also exhibited enhanced NF-κB activation upon ex vivo stimulation, leading to augmented pro-inflammatory cytokine gene expression. Consistent with established mechanisms of innate immune memory, these changes were driven through stable, epigenetic changes, including increased recruitment of poised enhancers at pro-inflammatory gene loci.⁶⁸

Trained immunity is believed to contribute to atherosclerosis not only as a result of exposure to cardiovascular risk factors but also in response to general inflammation. For example, exposure of atheroprone Apolipoprotein E knockout (Apoe^−/−) mice to subclinical endotoxemia (low-dose LPS) led to aggravated atherosclerosis. The monocytes isolated from these mice demonstrated persistent pro-inflammatory skewing even one month after cessation of LPS exposure. Further, the authors found that adoptive transfer of low-dose LPS-exposed monocytes to naïve Apoe^−/− mice was sufficient to drive the increase in atherosclerosis.⁶⁹ While this study highlighted a peripheral trained immune mechanism, it has been shown that exposure of human HSPCs to interleukin-1β itself can induce features of trained immunity.²³ These findings raise the intriguing question of how broadly these mechanisms generalize to common inflammatory signals and conversely, what biological mechanisms may restrain the induction of trained immunity. Addressing these questions will be critical for understanding the cumulative impact of chronic inflammatory exposures on atherogenic risk.

While the discussion thus far has focused on innate immune reprogramming in chronic inflammatory states, the consequences of trained immunity extend beyond atherogenesis to the acute immune activation that accompanies plaque rupture and ischemic injury. These acute ischemic events represent a setting in which trained immune programs may be newly induced or further reinforced.

Immune training of cardiac macrophages after myocardial infarction

Myocardial infarction (MI), commonly known as a “heart attack,” occurs when coronary blood flow to the heart is acutely reduced or blocked, usually from plaque rupture or thrombotic occlusion. The resulting ischemia causes hypoxia, cell stress, and death, which activate immune responses. Endothelial cells, cardiomyocytes, and macrophages release inflammatory mediators that recruit neutrophils and monocytes, which quickly differentiate into inflammatory macrophages. Within 24 h, these cells are replaced, and over time they transition into reparative subtypes that promote healing through factors like VEGF, IL-10, and TGF-β. Excessive inflammation or large infarcts can impair repair, leading to maladaptive remodeling and heart failure. High leukocyte counts predict worse outcomes and greater risk of recurrent events. Resident cardiac macrophages, a heterogeneous and plastic population essential for tissue homeostasis and repair, are at the epicenter of a cardiovascular event. Following an acute, ischemic injury such as a myocardial infarction, resident macrophages, along with infiltrating monocytes, are exposed to a powerful cocktail of DAMPs and inflammatory cytokines.^70–72 This exposure can induce stable epigenetic changes that prime these cells for more aggressive responses in the future, consistent with peripheral trained immunity.^51,73 More recently, evidence shows that MI also induces profound changes in bone marrow activity, reshaping immune regulation and contributing to long-term cardiovascular disease risk.^73–77

One such study assessed the effects of experimentally induced MI on peripheral and central trained immunity in atheroprone Apoe^−/− mice. Compared to mice receiving a sham surgery, mice that underwent a left anterior descending coronary artery ligation developed greater atherosclerosis compared to mice that received a sham surgery.⁵¹ Importantly, these effects were mediated through bone marrow progenitor cells, as transplantation of bone marrow from MI-experienced donors accelerated atherosclerosis in MI-naive recipients. These effects were mediated through epigenetic reprogramming of the bone marrow progenitors by the histone methyltransferase KMT5a.⁵¹

The systemic effect of immune training following an ischemic event was highlighted through a study that assessed ischemic stroke as a driver of immune training. A stroke is caused by ischemia to the brain, driven by occlusion of cerebrovascular vessels. In mice, experimental ischemic stroke causes long term alterations in bone marrow and circulating myeloid populations, leading to a selective expansion of Ly6c^high cardiac monocytes.⁵⁵ As a result, repopulation of cardiac tissue with these Ly6c^high monocytes and macrophages drives cardiac fibrosis and chronic diastolic dysfunction through increased MMP9 production. Importantly, transplantation of stroke-reprogrammed HSPCs to a stroke-naive recipient mouse similarly promoted recruitment of Ly6c^high monocytes to the heart, driving ECM remodeling and fibrosis. These findings support a model by which immune training of progenitor cells can lead to long-lasting alterations in mature myeloid cells, contributing to future disease pathology in sites distal from the initial insult.⁵⁵

Opportunities for therapeutic intervention

These studies have identified trained immunity as a key contributor to cardiac pathologies, and one that is currently not addressed by standardized therapy options. This presents a promising avenue to leverage for the development of better prognostic tools as well as new therapeutics. The idea of being able to risk-stratify patients with immune training-promoting risk factors based on predicted risk of developing future CVD is an appealing one. Indeed, many studies have successfully applied the system immune-inflammation index (SIII), a clinical algorithm originally designed for malignancies, to show that peripheral immune cell differences are predictive of future risk.⁷⁸ However, factoring in a biomarker of trained immunity could further increase the accuracy of such tools and factor into clinical decision making. Indication that a patient’s immune system has been trained may indicate that they may experience greater benefit from starting anti-inflammatory therapy such as colchicine. Finding such a biomarker, however, has thus far been elusive. In one murine study examining the relationship between myocardial infarction, immune training, and atherosclerosis, it was shown that MI-driven immune training led to persistent upregulation of Syk gene expression. The authors proposed that SYK expression in circulating monocytes could therefore be leveraged clinically to determine patient risk of atherosclerosis following a MI.⁵¹ While these authors highlighted SYK expression for its potential as a biomarker of trained immunity-associated risk, SYK activation has been shown to be necessary for the induction of immune training by various ligands including β-glucan and heme.^50,79,80 This is of particular interest given that a variety of SYK inhibitors have already been developed and several are currently in clinical trials for the treatment of cancer and autoimmune diseases.⁸¹

From a therapeutic perspective, trained immunity may be targeted by preventing the development of immune training in the setting of an acute inciting stimulus. While prolonged anti-inflammatory therapy poses a significant degree of risk from its immunosuppressive effects, proof-of-concept studies show that short-term use of an IL-1β neutralizing antibody to prevent induction of training is effective at preventing future cardiac events. Treating mice with an IL-1β neutralizing antibody immediately after stroke prevented immune training-associated development of long-term cardiac dysfunction.⁵⁵ Other studies have targeted methyltransferase enzymes to block trained immunity, either with the methyltransferase inhibitor methylthioadenosine or by a siRNA-mediated knockdown of KMT5a.^31,51 Alternatively, if cells have already undergone training, it may be possible to intervene to reverse or modulate trained immune responses. For example, cenicriviroc (CVC), an experimental human immunodeficiency virus (HIV) and steatohepatitis drug, is a dual CC-chemokine receptors type 2 and 5 inhibitor. In a mouse model of immune training by stroke, daily treatment of mice with cenicriviroc was able to reduce monocyte trafficking to the heart.⁵⁵ These findings suggest that therapeutic modulation of trained immunity is possible, both at the level of induction and post-establishment. However, despite the existence of compounds that target the metabolic and epigenetic pathways underlying trained immunity, a major challenge remains: how to selectively deliver these therapies to the trained cell populations without broadly suppressing beneficial immune functions.

Rather than attempting to reverse trained immunity, one group instead leveraged innate immune memory by reprogramming bone marrow-derived monocytes to a tolerized, anti-inflammatory phenotype, using 4-phenylbutyric acid (4-PBA). Treating monocytes with 4-PBA reprogrammed them to reduce expression of ICAM-1 and CCL5, while increasing expression of the pro-resolving mediator CD24. Functionally, 4-PBA-treated monocytes took up less oxLDL in vitro and promoted resolution of oxLDL-treated monocytes in a co-culture system. In vivo, infusion of 4-PBA-treated monocytes into atheroprone mice limited the development of atherosclerosis, presenting a novel immune cell-based therapeutic avenue.⁸²

Taken together, these studies highlight multiple opportunities for intervention, including pharmacologic inhibition of trained immunity induction and modulation of established trained programs. Moving forward, translational efforts must identify strategies to deliver these interventions with precise spatial and temporal control in order to maximize efficacy without causing global immunosuppression.

Conclusions

Collectively, these advances establish trained immunity as a critical contributor to cardiovascular disease, establishing a framework in which inflammatory states bias myeloid cells towards pathogenic responses to subsequent cardiac or vascular insults. In parallel, secondary inflammation arising from acute cardiovascular events may result in peripheral and central immune training, acting in a positive feedback loop to drive worsened cardiovascular outcomes (Fig. 2).

Figure 2.

Schematic representation of how trained immunity links systemic cardiovascular risk states with primary and secondary cardiac events.

While there has been significant progress in understanding this evolutionarily conserved mechanism of innate immune memory and how it intersects with chronic disease states, many foundational questions remain unresolved. For example, current models of trained immunity likely oversimplify the in vivo reality and do not account for complex crosstalk between different cell types. While in vitro models clearly show that innate immune cells retain an epigenetic imprint of prior insults, it is more challenging in vivo to disentangle the direct effects of these stimuli on innate cells from indirect effects mediated through other cell types. For example, there is evidence suggesting that in some contexts, the ability to develop a trained immune response is dependent on the presence of T cells.^83–86 Reciprocally, innate immune cells that have been “trained” are likely to interact with the adaptive immune system differently than untrained cells.⁸⁷ Trained immunity causes changes in markers associated with antigen presentation and cytokine production, likely impacting T and B cell licensing, memory, and polarization.^83,88,89 While direct evidence of trained immunity has been shown to last for up to one year both in mice and humans, heterologous protection of vaccination against unrelated pathogens has been shown to last for up to five years.⁸⁹ It is possible that these longer changes in inflammation or host pathogen defense are due to innate-adaptive immune crosstalk.

Another major gap lies in definitively demonstrating how trained immunity functionally alters cardiac macrophages. While in vitro models are useful for evaluating functions such as lipid uptake or efferocytosis, macrophage identity and behavior are strongly influenced by the tissue environment, limiting the translatability of any in vitro findings. On the other hand, assessing cardiac macrophages in vivo is largely limited to transcriptomic or epigenomic characterization. Indeed, the majority of evidence supporting the existence of trained cardiac macrophages has been largely inferred from end-stage pathological outcomes and assessment of the circulating or bone marrow immune cells.

Furthermore, significant gaps remain in our understanding of the mechanisms of trained immunity. A diverse set of epigenetic and metabolic changes in response to trained immunity have been documented. However, it remains to be determined whether certain epigenetic and metabolic shifts represent universal features of innate immune memory, or if the mechanisms driving trained immunity are context specific. Some studies have demonstrated that different training stimuli cause distinct transcriptional responses to restimulation.⁸⁶ However, it is less well understood how macrophage subtype or environment may impact the trained immunity program. This is especially important given the heterogeneity of cardiac macrophages. Whether embryonically derived artery wall macrophages retain distinct memory programs compared with monocyte-derived lesional macrophages warrants further investigation. Clarifying these mechanisms will be essential for translating insights from basic research into therapeutic strategies that target trained immunity in cardiovascular disease.

Author contributions

Emma Hope—conceptualization (lead), writing—original draft preparation (lead), writing—review & editing (supporting). Azuah L. Gonzalez—writing—original draft preparation (equal). Lola S. Norman—writing—original draft preparation (equal). Hunter Smith—writing—original draft preparation (equal). Jean W. Wassenaar—writing—review & editing (supporting). Kasey C. Vickers—writing—review & editing (supporting). Jonathan D. Brown—writing—review & editing (supporting). Amanda C. Doran—conceptualization (supporting), writing—review & editing (lead), supervision (lead), funding acquisition (lead).

Emma Hope (Conceptualization [Lead], Writing—original draft [Lead], Writing—review & editing [Supporting]), Azuah L. Gonzalez (Writing—original draft [Equal]), Lola S. Norman (Writing—original draft [Equal]), Hunter Smith (Writing—original draft [Equal]), Jean Wassenaar (Writing—review & editing [Supporting]), Kasey C Vickers (Writing—review & editing [Supporting]), Jonathan D. Brown (Writing—review & editing [Supporting]), and Amanda Doran (Conceptualization [Supporting], Funding acquisition [Lead], Supervision [Lead], Writing—review & editing [Lead])

Funding

This work was supported by National Institutes of Health grants: T32 GM152284 (E.H.), F31 HL172670 (A.L.G.), R01 HL146654 (J.D.B.), R01 HL159487 (A.C.D.), and R01 HL174961 (A.C.D. & J.D.B.). Work was also supported by an American Heart Association Career Development Award 24CDA1278132 (J.W.W.).

Conflicts of interest

None declared.

Disclosures

The authors have nothing to disclose.