Genetic factors contributing to atherosclerosis

Abstract

Purpose of review

Atherosclerosis is a highly complex and heterogeneous disease characterized by inflammation and lipid accumulation in the arterial wall. Genetic studies to identify loci and genes predisposing to the disease have now been carried out in large human cohorts and the underlying pathways have been studied in animal models, primarily mice. This review summarizes recent findings relating to genetic contributions to the disease and implications for diagnosis and treatment.

Recent findings

Recent discoveries of genes contributing to all aspects of atherosclerosis and related diseases, including inflammation, lipid levels and oxidation, clonal hematopoiesis, hypertension, cardiac function, and aortic stenosis, are discussed. Altogether, several hundred contributing genes have been identified, and these are enabling a better understanding of the important pathways involved. These findings have also led to the development of improved diagnostic methods, including overall risk evaluation, and the identification of new therapeutic targets.

Summary

Genetics is proving to be a powerful approach to dissecting atherosclerosis. Genetic studies promise to revolutionize the diagnosis and treatment of the disease.

INTRODUCTION

It has been said that “genetics is to biology as mathematics is to physics”, and, indeed, genetic studies have proven key to understanding atherosclerosis and other cardiovascular disorders. It has been estimated that genetics explains about 40% of disease incidence of atherosclerosis, including rare Mendelian disorders, such as familial hypercholesteremia, and common forms of the disease that are due to the interactions of many genes, each with a modest impact. As of this writing, several hundred loci contributing to the common forms have been identified using very large cohorts for genome-wide association studies (GWAS) (see below). Studies with genetically engineered mouse models have been important in validating candidate genes at these loci. Such studies have revealed that genetic variation affects many different cell types and pathways involved in the institution and progression of atherosclerosis [1]. Here, separated by topic area, we review some recent papers dealing with the role of genetic variation in atherosclerosis.

GENETICS OF RISK FACTORS FOR ATHEROSCLEROSIS

Lipid levels

Circulating levels of total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides were among the first atherosclerosis risk factors to be investigated using both GWAS strategies and rare variant analyses. With respect to common variation, recent studies have sought to identify novel loci for lipids through interactions with environmental factors. For example, 17 previously unknown loci for lipid levels were identified through a genome-wide gene-sleep interaction study [2]. Some of the positional candidate genes at the identified loci, such as SLC25A31, SLC8A1, and ASPH, are known to be the targets of commonly prescribed drugs, suggesting potential therapeutic opportunities for individuals with sleep disturbances. A similar GWAS interaction approach with sex revealed genetic contributions to lipid metabolism that differed between males and females [3].

Rare variant analyses using either whole-exome sequencing (WES) or whole-genome sequencing (WGS) has also shed light on the genetic architecture of lipids. For example, a large meta-analysis of whole genome sequencing identified approximately 50 novel rare coding and noncoding variants for LDL cholesterol levels [4]. With respect to the latter type of genetic variation, rare variants in long noncoding RNAs (lncRNAs) have been identified at known loci for lipids, suggesting a broader role of lncRNAs in regulating lipid metabolism [5]. Beyond biological implications, genetic analyses with whole-genome sequence data in large numbers of subjects has also provided insight into the so-called “missing heritability” of lipid levels (and other complex traits). For example, heritability estimates for LDL cholesterol levels based on WGS analysis with both common and rare variants no longer differ from heritability estimates based on the conventional method using pedigree analyses [6]. Thus, these observations demonstrate that narrow sense heritability for at least some lipid traits appears to be fully explained by WGS data.

Beyond gene discovery, there has also been a shift towards leveraging the large number of identified genes and alleles for risk prediction or therapeutic development. Among familial hypercholesterolemia patients, genetic screening that identified a pathogenic mutation was associated with better clinical management and a greater reduction in LDL cholesterol levels [7]. There is also significant effort being devoted to utilizing genetic findings in the form of polygenic risk scores (PRS) to inform treatment and predict outcomes in patients. Notably, individuals with a higher burden of lipid-raising alleles had increased plaque rupture and mortality and increased elevations of lipid levels as a function of age [8,9].

Hypertension

Hypertension is one of the strongest risk factors for atherosclerosis and has been a major area for growth in cardiovascular genetics in recent years. A GWAS of over one million European-ancestry adults reported more than 2000 blood pressure (BP) trait loci, including 113 novel loci, increasing the SNP-based herit-ability of BP traits to more than 60% [10]. There have also been a number of sub-population GWAS analyses of BP traits. An ancestry-specific approach revealed loci specific for BP traits in Han Taiwanese individuals, and using Mendelian Randomization confirmed hypertension as a causal factor in unstable angina [11]. A sex-specific approach revealed that nearly half of GWAS BP trait-associated loci were female-specific [12]. The COL4A1/COL4A2 locus was one such female-specific locus and also associated with fibromuscular dysplasia and spontaneous coronary artery dissection, both of which disproportionately affect female patients. Genetics also affects the beneficial effects of diet on hypertension. The impact of a DASH (Dietary Approaches to Stop Hypertension) on SBP is modified by SNPs in MTHFS and WWOX, among other genes [13]. A FinnGen and UKB study examined only resistant HTN (requiring >3 medications for treatment), and found a subset of five loci, perhaps highlighting optimal targets for this difficult-to-treat subpopulation [14].

In addition to genome-wide analyses, genetics has been used to tease out specific effects of NDST3 and STRN on salt-sensitivity [15,16]. Salt-sensitive BP elevations represent a subtype of hypertension pathophysiology which may benefit from different medication strategies, but assessing these responses is labor-intensive in clinical settings. Risk alleles of both NDST3 (a heparan sulfate modifying enzyme) and STRN (a scaffolding molecule) were shown to help differentiate these patients and could be useful to alleviate such a burden.

Finally, the clinical utility of genetics in hypertension treatment has been explored. A subset of Black patients from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) were genotyped and BP outcomes assessed [17]. An increased PRS was associated with a reduced BP response to chlorthalidone. Incorporation of a PRS also improved prediction of incident hypertension in a cohort of more than 2500 patients [18]. Thus, genetic testing has shown diagnostic utility in both preventive and chronic hypertension settings.

Inflammation

Inflammation is a key driver of atherosclerosis initiation and progression [19,20]. Indeed, many of the loci identified in GWAS have been involved in inflammation. Genetic factors have been associated with many aspects of inflammation and multiple cell types, including macrophages and lymphocytes [1].

A body of evidence has suggested that the oxidation of lipids in lipoproteins trapped in the vessel wall produce pro-inflammatory species, leading to leukocyte recruitment and extravasation. Transcriptomic analyses have identified a subpopulation of macrophages that express the transcription factor GATA2. GATA2 is upregulated by oxidized LDL and such upregulation is associated with increased proliferation and apoptosis [21]. Octanol, a byproduct of lipid oxidation, that is widely found in plant oils and food additives, was also recently shown by Wang et al. [22], to increase atherosclerosis in mouse models by upregulating the expression of sodium-hydrogen exchanger 1 (NHE1). The role of lipid oxidation is supported by the identification of NRF2, a transcription factor regulating many genes involved in antioxidant mechanisms [23]. NRF2 resides with mitochondria, where it likely protects the organelle from oxidative stress-mediated decay. NRF2 expression also affects autophagy, a process involved in the turnover of subcellular components [24]. A candidate gene study in mice also showed that the nuclear receptor DAX1 promotes atherosclerosis by autophagy suppression as well as lipid transport inhibition in macrophages [25]. DAX1 is a co-repressor of several protective nuclear receptors to suppress autophagy and result in lipid accumulation and inflammation in macrophages.

Chronic inflammation and metabolic syndrome are strongly associated with atherosclerosis [19].*********** Human GWAS studies recently showed that genetic variants in the cis-regulation element of the histone deacetylase (HDAC9) gene are associated with atherosclerosis, plaque destabilization, and inflammasome activation in myeloid cells. Furthermore, therapeutic inhibition of HDAC in myeloid cells reduced plaque size and improved stability in mouse models [26].

Efferocytosis, a process by which apoptotic cells are rapidly cleared by macrophages, reduces inflammation by averting secondary necrosis [27]. Genetic studies in mice linked impaired efferocytosis and atherosclerosis to endoplasmic reticulum (ER) stress and identified TRIB33 as a mediator of ER stress in lipid-loaded macrophages [28]. Efferocytosis offers a therapeutic node through which to target atherosclerosis, and the authors suggest TRIB33 as a molecular target to increase this process.

Insulin-like growth factor binding protein 6 (IGFBP6) was identified as a candidate gene for atherosclerosis by examining differential gene expression responses in cultured endothelial cells as well as human atherosclerosis databases. Mouse knockout and overexpression studies then revealed that IGFBP6 protects against atherosclerosis by inhibiting endothelial inflammation, including reducing expression of adhesion molecules [29].

Mendelian randomization and other modeling approaches using human genetic and transcriptomic profiling data, have implicated inflammatory pathways in atherosclerosis. Zhang and Chen [30] focused on regulatory network approaches to implicate numerous inflammatory genes and pathways in atherosclerosis. For example, heme oxygenase 1 (HMOX1) was found to be a critical player associated with lipid oxidation and ferroptosis. Villaplana- Velasco et al. [31] examined several large cohorts to identify inflammatory markers associated with cardiovascular disease, stroke and inflammation. Similarly, Wu et al. [32] used Mendelian randomization to demonstrate associations between atherosclerosis and eight common immune-mediated inflammatory diseases, including Crohn's disease, systemic lupus erythematosus, psoriasis, and multiple sclerosis. Jin et al. [33] used network modeling to identify a T lymphocyte network regulating plaque inflammation.

It is clear that cellular senescence, a process in which cells stop dividing and secrete high levels of inflammatory cytokines, growth factors and proteases, plays an important role in the disease, particularly in aging individuals [34,35]. Recent studies integrating GWAS data and expression data in smooth muscle cell and endothelial cells have linked senescence to atherosclerosis and identified genes contributing to the process, which may serve as pharmacologic targets [36■,37]. For example, Lipscomb et al. [38] showed in a mouse model that resolvin D2 limits senescent cell accumulation in atherosclerosis plaques.

Recently, a GWAS of the Hybrid Mouse Diversity Panel identified a locus on mouse chromosome 9 significantly associated with atherosclerosis in both sexes [39,40]. A candidate gene, encoding N-nicoti- namide N-methyltransferase (NNMT), was charactered. Knockdown of NNMT resulted in a dramatic reduction in atherosclerosis and significant reduction in macrophage replication in the lesions. Based on this, bone marrow transplantation studies with bone marrow from NNMT deficient mice were performed and these revealed a large reduction in atherosclerosis. Nicotinamide (NAM), the substrate of NNMT, can be salvaged to produce NAD, and. studies with heterozygous NNMT knockout mice revealed more than a 50% increase in NAD/NADH ratio in cultured macrophages. To further examine the role of NAD levels in macrophage proliferation and apoptosis in atherosclerosis, the authors examined CD38 deficient mice. CD38 catalyzes the conversion of NAD to ADP ribose and NAM. As with the NNMT knockout studies, transplantation of bone marrow from CD38 knockout mice into wild-type mice resulted in reduced macrophage proliferation and increased apoptosis in the recipient mice. The authors concluded that NAD levels are critical in macrophage proliferation and survival. Reduced NAD levels have been associated with cellular senescence and increased expression of SIRT-1, which through FOXO3 increases the likelihood of cell cycle arrest and resistance to oxidative stress [39].

Clonal hematopoiesis of indeterminate potential

The discovery of clonal hematopoiesis of indeterminate potential (CHIP) has led to the development of a subfield of atherosclerosis genetics in the last decade since Jaiswal et al. [41] first reported an association between CHIP and atherosclerosis. Throughout the aging process, hematopoietic stem cells (HSCs) accumulate somatic mutations. When such a somatic mutation occurs in a leukemia-causing locus, thereby conferring a proliferative advantage, this leads to clonal expansion, and is deemed clonal hematopoiesis [42■]. While this is a precursor state in the development of leukemia, the risk of developing leukemia is low (<1% annual risk) [43,44]. The major driver of increased early mortality in these patients (1.4x relative risk) is cardiovascular events [41]. In recent work, CHIP has now been associated with more severe coronary artery disease (CAD), which underlies the risk of cardiac events, and calcific aortic valve disease, which is a common comorbidity in patients with atherosclerosis [45,46].

Recent studies have also improved our understanding of how this genetic condition leads to atherosclerosis and ASCVD events. Some have suggested that CHIP is an artifact of hyperlipidemic states; in other words, that the inflammation and hyperlipidemia characteristic of atherosclerosis may actually have a reverse causality leading to HSC proliferation and mutation accumulation [47]. The use of deep sequencing in a longitudinal cohort by Diez-Diez et al. [48] allowed for early detection of CHIP mutants and confirmed that CHIP precedes atherosclerosis while atherosclerosis does not affect CHIP expansion. Indeed, CHIP and hyperlipidemia contribute synergistically to the development of atherosclerosis [49]. Recent studies have also expanded on the established pathways of inflammation in CHIP, implicating IL-17 signaling in cardiovascular disease related to CHIP with JAK2 mutations [50■]. Previous literature has reported mixed results on the role of IL-17 in cardiovascular disease, and this may be due to heterogeneity in CHIP populations [51].

Ultimately, the identification of CHIP in clinical practice may allow for genetically guided therapeutics. A secondary analysis of the Canakinumab Antiinflammatory Thrombosis Outcomes Trial (CANTOS) showed more benefit among those with TET2 mutant CHIP [52], while a secondary analysis of the Low-Dose Colchicine 2 (LoDoCo2) trial showed that colchicine attenuated clonal growth among those with TET2 mutants [53■]. Human and murine data support the potential use of both of these agents in atherosclerosis [54,55], but no trials in CHIP of any agents have reported results at the time of writing. Early phase trials of an oral NLRP3 inhibitor [56] and combination NLRP3 inhibitor and bispecific IL-1β/IL-18 monoclonal antibody [57] have been completed but have not reported efficacy results.

GUT microbes

The gut microbiome contributes to a variety of immune and metabolic traits involved in numerous diseases, including atherosclerosis. Host genetic and environmental factors interact with the gut microbiome by affecting the gut microbiome composition [58]. Host genetic factors also mediate interactions with the microbiome, such as the metabolism of microbiome-derived products. For example, trimethylamine N-oxide, isoproterenol and imidazole propionate (ImP), all produced by certain bacterial species, have major impacts on cardiovascular disorders, including atherosclerosis [59]. A recent review examines host gene-by-gut microbe interactions and discusses potential applications to cardiovascular diagnosis and therapy [60].

GENETICS OF CLINICAL EVENTS RELATED TO ATHEROSCLEROSIS

Clinical atherosclerotic cardiovascular events and heart failure

Prior large-scale GWAS have identified more than 300 loci for clinical atherosclerosis outcomes, including CAD and myocardial infarction [61–63]. As with other risk factors, the field has now moved more towards sequencing approaches, which have provided a better understanding of the genetic basis of atherosclerosis. Integration of whole exome sequencing (WES) and whole genome sequencing (WGS) data with machine learning methods has revealed ultrarare variants at known genes for CAD or related risk factors that have large effect sizes and heritability estimates for CAD [64,65]. WES analysis of patients with CAD, stroke, and peripheral artery disease analyzed collectively have also identified rare, deleterious variants of HTRA1, SGTB, and RBM12 that were associated with atherosclerosis independent of traditional risk factors [66].

By comparison, gene discovery efforts have been less fruitful for downstream complications of CAD and MI, such as heart failure. This is likely due to etiological heterogeneity of HF (e.g., ischemic, hypertensive, and genetic cardiomyopathies) and the lack of distinction in many studies between heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF) [67]. However, studies with sample sizes approaching approximately 2 million individuals or stratified by heart failure subtypes have made significant progress in this area by identifying numerous loci for heart failure, including distinct genetic risk factors for HFrEF vs. HFpEF, and ancestry-specific variants [68,69–71■].

Aortic stenosis

Aortic stenosis and atherosclerosis are distinct entities that have clear overlap, including endothelial injury, activated mesenchymal cells, and macrophage-driven inflammation [72]. A recent study attempted to interrogate this overlap and performed GWAS for aortic stenosis adjusted for CAD in over 450 000 individuals [73]. They found 17 aortic stenosis risk loci, of which only six were overlapping loci with CAD, and the genetic correlation of CAD and aortic stenosis was only 0.15. More recently, a GWAS of over 2.8 million individuals and over 80 000 AS cases, identified 261 independent loci, 225 of which had not previously been reported [74]. Separately, Kany et al. [75■] incorporated cardiac MRI metrics of aortic stenosis into a GWAS of nearly 60 000 individuals in the UK Biobank. Using this approach, they are able to assess the genetic architecture underlying early aortic stenosis prior to diagnosis. Both of the recent GWAS’s have reaffirmed the genetic role of lipoprotein(a) [Lp(a)], which is currently being studied as a therapeutic target to limit aortic stenosis progression (NCT05646381) [76]. These recent studies have advanced the genetic understanding of aortic stenosis and its interaction with atherosclerosis.

GENETICS OF ATHEROSCLEROSIS IN CLINICAL CARE

Therapeutic applications of genetics

A number of therapeutic targets in atherosclerosis have been either derived from or bolstered by studies in genetics. Over two decades ago, PCSK9 was identified through genetic study as a possible target in atherosclerosis [77,78]. Indeed, targeting PCSK9 with either evolocumab or alirocumab decreased cardiovascular events in studies of high-risk populations [79,80]. A small-interfering RNA (siRNA) targeting PCSK9, inclisiran, has shown efficacy in reducing LDL and Lp(a) [81], but cardiovascular outcomes have not yet been reported from the ORION-4 trial [82]. At the time of writing, an oral PCSK9 inhibitor, enlicitide, has also just recently proven to lower LDL and Lp(a), and a cardiovascular outcomes trial is ongoing with projected completion in 2029 [83,84■].

A combination of genetic and epidemiologic data has established Lp(a) as a target in atherosclerosis. Levels of Lp(a) are genetically determined, primarily through variation at the LPA locus, and due to its strong association with cardiovascular events, some societies such as the European Atherosclerosis Society now recommend measuring Lp(a) at least once in all adults [85,86]. While Lp(a) is modestly lowered through PCSK9 inhibition, specific targeting of LPA with antisense oligonucleotides (e.g., pelacarsen) or siRNAs (e.g., lepodisiran, olpasiran, and zerlasiran) can reduce levels by more than 90% [87–90]. Again, impacts on cardiovascular events are currently under study [91].

Genetic data has also underscored the IL-6 pathway as a therapeutic target [92]. Completed trials with the IL-6 monoclonal antibody ziltivekimab have shown reduced inflammatory markers in patients with high ASCVD risk [93]. The ongoing ZEUS and ARTEMIS trials will report impact of the IL-6 antibody ziltivekimab on incident major adverse cardiovascular events in chronic coronary syndrome and acute coronary syndrome, respectively [94,95].

One exciting recent development based in the genetics of atherosclerosis is the first-in-human application of CRISPR-Cas9 gene-editing to lower cholesterol. Laffin et al. [96■■] used a lipid-nanoparticle-encapsulated CRISPR-Cas9 endonuclease mRNA and guide RNA targeting ANGPTL3 for hepatic deletion. Natural genetic variation in ANGPTL3 initially helped confirm its role as a therapeutic target in dyslipidemia [97]. Following efforts with antibody targeting and antisense oligonucleotides to oppose its expression [97–102]. Laffin et al. [96■■] now show that CRISPR-Cas9 editing of ANGPTL3 is well tolerated in the short-term and has a durable effect on cholesterol profiles [96■■]. Moving forward, gene therapy will likely have increased impact in atherosclerosis given the rate of progress in this area [103].

Genetic testing in clinical care

Translation of genetic testing into the clinical setting has been limited due to the lack of clear impact on clinical decision making (Fig. 1). The one exception is familial hypercholesterolemia. Professional societies including the AHA and ACC have published “Scientific Statements” supporting cascade testing for variants in LDLR, APOB, and PCSK9 in the last 15 years [104,105], and the European Atherosclerosis Society “Consensus Statement” supports testing for the above three genes in addition to LDLRAP1, ABCG5/G8, LIPA, and APOE to confirm diagnosis of familial hypercholesterolemia [106]. However, these recommendations are not universal (in part limited by concerns regarding cost) and have not been formally endorsed in AHA/ACC guidelines.

FIGURE 1.

Genetics in human atherosclerosis. The genetic study of atherosclerosis (left) can be separated into risk factors, disease onset, and sequelae and related diseases. Genetics is impacting clinical care of atherosclerosis (right) as genetic sequencing becomes more available and therapies leveraging CRISPR-cas9, antisense oligonucleotides (ASOs), and small interfering RNA (siRNA) are tested in humans.

An additional obvious role for genetic testing in cardiovascular clinics is in the identification of CHIP. Indeed, recent work has been done to make a CHIP diagnostic test cheaper and more accessible [107], but this remains in the research setting for the time-being. Currently, clinical genetic testing for CHIP mutations is nearly entirely in the setting of a new cancer diagnosis or in the work-up of cytopenia.

Outside of familial hypercholesterolemia and CHIP, specific mutation-driven testing is rare, but, as noted above, PRS hold promise for incorporation into risk prediction. Many studies have shown an improvement in risk discrimination when adding PRS on top of models based on traditional risk factors, and a 2025 consensus statement from the ESC Council on Cardiovascular Genomics, the ESC Cardiovascular Risk Collaboration, and the European Association of Preventive Cardiology summarizes this work in greater depth [108■■].

CONCLUSION

Numerous novel genes and pathways for atherosclerosis have been identified and characterized in the last few years. These have come primarily from human GWAS but also studies of natural variation in mice. Statistical methods for the identification of genes, such as Mendelian randomization, are proving increasingly useful and machine learning approaches based on DNA sequence are beginning to be investigated [109]. In combination with biochemical and cell biology studies, genetic studies are leading to the development of improved diagnostic methods and novel treatment strategies.

KEY POINTS.

• Atherosclerosis is a highly complex disease process due to the interactions of hundreds of genetic variations and environmental factors.

• Multiple pathways affecting inflammation, plasma lipids, hypertension, and the gut microbiome mediate the impact of genetics on atherosclerosis.

• Clonal hematopoiesis is a novel risk factor driven by somatic mutations which may allow for personalized approaches to atherosclerosis risk optimization.

• Polygenic risk scores and CRISPR-Cas9 offer opportunities to integrate genetics into the clinical care of atherosclerosis.