Genetic Markers of Endothelial Dysfunction

Abstract

The rate of endothelial dysfunction is influenced by genetic variation and thus inherited in families. Genetic disorders, such as familial hypercholesterolemia and homocystinuria, are at risk for premature atherosclerosis, and exhibit early endothelial dysfunction. The known spectrum of mutations in LDL receptor, APOB and PCSK9 gene represent the monogenic dominant hypercholesterolemia. An autosomal recessive form of hypercholesterolaemia in the caused by homozygous mutations in the LDL-R adaptor protein. The polygenic hypercholesterolaemia for patients with a clinical diagnosis of FH is based on the cumulative effect of LDL-C-raising alleles with a cumulative effect, in a complex interaction with the environment that leads to an increase in LDL-C, producing an FH-like phenotype and presenting this type of hypercholesterolaemia as a typical complex disease. The various causes of homocysteinaemia like genetic causes include mutations and enzyme deficiencies such as the most frequently mentioned 5, 10-methylenetetrahydrofolate reductase (MTHFR), but also methionine synthase (MS) and cystathionine β-synthase (CβS) but also by deficiencies of folate, vitamin B12 and, to a lesser extent, deficiencies of vitamin B6, which affects methionine metabolism, and leads also to endothelial disfunction in different mechanismms. Mutations in genes coding enzymes in homocysteine metabolism and also in nitric oxide (NO) synthesis, the main vasodilatator is also presented in this chapter. The crucial importance of microRNAs in endothelial physiology following EC-specific inactivation of the enzyme Dicer which is involved in altered expression of key regulators of endothelial function, including endothelial nitric oxide synthase (eNOS), vascular endothelial growth factor receptor 2 (VEGF), interleukin-8, Tie-1 and Tie-2. The new discoveries based on genome-wide screening (GWAS) complement the knowledge of the topic.

1. Introduction

A healthy vascular endothelium exerts atheroprotective effects through vasoactive mediators such as nitric oxide (NO), prostacyclin and endothelium-derived hyperpolarizing factor (EDHF). The endothelium plays an important role in regulating vasomotor tone and maintaining vascular integrity. Endothelial dysfunction, impaired endothelium-dependent dilation, is a fundamental element in the pathogenesis of cardiovascular disease. There is evidence that as we age, the endothelium is exposed to the deleterious effects of elevated blood pressure and increased levels of cholesterol, glucose, homocysteine, to products of the inflammatory response and to components of cigarette smoke, and these protective properties decrease, leading to a state of endothelial dysfunction [1]. Although endothelial dysfunction is associated with a number of risk factors for atherosclerosis, these risk factors are not the only determinants of endothelial dysfunction. The rate of endothelial dysfunction is influenced by a number of factors that are determined by genetic variation and thus inherited in families. These issues will be addressed in this chapter.

Vascular atherosclerosis, as the most common sign of endothelial dysfunction, usually manifests itself at a later age, although studies of twins and adoptions indicate that this more common form is also partly heritable, although the inheritance is complex, arising from shared environmental exposures (risk factors) and many common gene variants (polymorphisms) with small to moderate effects. Endothelial dysfunction manifested in atherosclerosis also results from single-gene diseases that strongly modify risk factors, such as hypercholesterolemia or hyperhomocysteinuria. Children with certain single-gene disorders, such as homocystinuria and familial hypercholesterolemia, are at risk for premature atherosclerosis, and also exhibit early endothelial dysfunction [2, 3, 4].

2. Aging endothelium—mechanisms of endothelial senescence

Endothelial cell senescence is a physiological process of irreversible cell cycle arrest to which various biological stress conditions such as, telomere shortening, DNA damage, reactive oxygen species (ROS) production and mitochondrial dysfunction contribute. Cellular senescence is a process in which vascular cells stop dividing and undergo characteristic phenotypic changes, such as profound changes in chromatin and secretome [5]. Vascular endothelial cell senescence has been found to play a key role in vascular aging, leading to the initiation, progression and development of vascular atherosclerosis [6]. Aging vascular endothelial cells typically become flatter and enlarged with increasingly polypoid nuclei. These changes are accompanied by modulation of cytoskeletal integrity, angiogenesis, cell proliferation and migration [7]. Aging endothelial cells exhibit decreased production of endothelial nitric oxide (NO), increased release of endothelin-1 (ET-1), increased inflammation and cell apoptosis [7]. Senescence of endothelial cells thus induces structural and functional changes in blood vessels, exacerbating thrombosis, inflammation and atherosclerosis with impaired vascular tone, angiogenesis and vascular integrity, which contributes to the development and progression of atherosclerosis [8]. However, the molecular mechanisms of vascular endothelial cell aging and their relationship to underlying pathophysiological changes are not yet fully understood. In this chapter, the role of genetic factors affecting the mechanisms of endothelial cell senescence in the process of vascular aging and the development of atherosclerosis will be discussed.

Cellular senescence is a physiological or pathological process that occurs throughout life [9]. Under physiological conditions, cellular senescence is involved in embryonic tissue development, tissue repair, and tumor suppression responses [9]. However, the accumulation of senescent cells can lead to loss of replicative capacity, cell apoptosis, unfavorable structural changes and the associated development of atherosclerosis [9]. Cellular senescence is usually associated with aging and age-related disorders. In human coronary arteries, endothelial cells with increased β-galactosidase activity associated with enhanced senescence are observed during aging, suggesting that aging is also associated with decreased endothelial cell regeneration and endothelial cell senescence, which is associated with decreased endothelium-dependent arterial relaxation [10] and the development of arterial stiffness [9]. Several studies have found that NO donors reduce arterial stiffness in healthy subjects and in patients with hypertension and hypercholesterolemia [10]. These data support a role for aging vascular endothelial cells in the pathogenesis of arteriosclerosis. However, while few clinical studies have examined the relationship between endothelial aging, arterial stiffness and hypertension, those that have been conducted have shown that aging is closely associated with arterial stiffness and atherosclerosis. For example, data from the Framingham Heart Study showed that older age was significantly associated with higher carotid-femoral pulse wave velocity and mean arterial pressure [11]. Arterial stiffness has been shown to be an independent biomarker of atherosclerotic morbidity and mortality in the general population, in aging individuals, in patients with hypertension and in patients with end-stage renal disease [12]. With aging and the associated arterial stiffness, systolic blood pressure tends to increase while diastolic blood pressure tends to decrease, and this pathophysiological change results in an increase in pulse pressure and pulse wave velocity in the aorta. Indeed, it has also been observed that the prevalence of hypertension, especially isolated systolic hypertension, is increased in the aging population [13]. Increased systolic pressure increases left ventricular afterload with an associated increase in myocardial oxygen demand. Declining diastolic pressure reduces perfusion of the coronary circulation during diastole. These consequences of arterial stiffness, increased systolic pressure and decreased diastolic pressure further induce left ventricular hypertrophy and subsequent myocardial ischaemia, remodeling and other cardiovascular complications in aging individuals [8].

3. Cholesterol—factor of endothelial senescence and endothelial dysfunction

Senescence of endothelial cells is known to mediate the endothelial damage that occurs during the initial phase of atherosclerosis. Aging cells of the vascular wall lead to endothelial dysfunction, resulting in the synthesis of inflammatory cytokines and promoting the progression of atherosclerosis. The second stage of developing atherosclerosis, fibrous plaque formation, is characterized by increased lipid accumulation in the intima, resulting in fibrous tissue proliferation and vitreous degeneration, forming characteristic plaques in the intima. Also, macrophages accumulate in the subendothelial space, where they induce pathology by increasing the expression of key atherogenic and inflammatory cytokines and chemokines [14]. In the third stage, atherosclerotic plaque formation, the fibrous tissue is large and necrotic, enriched in lipids, while the lesion surface is thinner and few foam cells are present at the base and margin. In atherosclerotic lesions, smooth muscle cells of the vascular wall migrate from the media to the intima, accumulate around the lipid core formed by dead foam cells and switch from a contractile to a synthetic phenotype. Macrophages, on the other hand, which phagocytized lipids, display an abnormal or activated phenotype, which promotes pathological vascular proliferation [15]. At this stage, proliferation dominates the smooth muscle cells of the vascular wall, but aging does not occur and a typical atherosclerotic plaque is formed. The fourth stage involves changes secondary to atherosclerotic plaques, in which aging macrophages promote plaque instability, degradation of elastic fibers and thinning of the fibrous cap, as well as increased expression of metalloproteases and formation of ulcers and thrombi [14]. At this stage, foam cells induce senescence of human vascular endothelial cells by releasing 4-hydroxynonenal (4-HNE) [16], which exacerbates senescence and induces atherosclerosis. Senescent human vascular wall smooth muscle cells differentiate into an osteogenic phenotype and undergo expression of calcifying factors, which eventually leads to calcification of the atherosclerotic plaque. It is noteworthy that human vascular smooth muscle cells proliferate in the early phase of atherosclerotic plaque formation. However, the proliferation rate of these cells is lower in advanced plaques than in early lesions, indicating that cell senescence may occur [17]. In addition, vascular injury and phenotypic transformation of senescent human vascular wall smooth muscle cells also play a role in mediating vascular calcification [18]. Cellular aging is not a consequence of a single cause, but there are many factors that can induce cellular aging. Premature cellular aging, can be caused by factors such as miRNAs, homocysteine, hyperglycaemia, hypertension, hyperlipidaemia, hyperphosphataemia and oxidative stress, by reducing telomerase activity, increasing ROS production and promoting vascular calcification, mitochondrial dysfunction and DNA damage.

High cholesterol and triglyceride levels have also been found to be associated with an increased risk of atherosclerosis and shorter life expectancy. In fact, the vascular endothelial dysfunction that occurs during human aging is the factor, and the accumulation of lipids in the vascular endothelium activates leukocytes to produce cytokines and chemokines that recruit macrophages. On the other hand, macrophages enhance the inflammatory response and secrete vascular endothelial growth factor, a key cytokine that mediates angiogenesis and the inflammatory response. And hyperlipidaemia itself is a major risk factor for aging, hypertension and diabetes.

The relationship between hypercholesterolemia, atherosclerosis and aging is still poorly understood. Low-density lipoprotein (LDL) (cholesterol) in general is an important physiological compound for cellular function, but in high concentrations can lead to atherosclerosis. It is generally accepted that the oxidized form of cholesterol leads to endothelial dysfunction, which is the initial step in the formation of atherosclerotic plaques. Oxidized low-density lipoprotein acts by binding to multiple scavenger receptors (SRs), such as SR-AI, SR-A2, and can also increase the expression of endothelial cells’ own LOX-Ion receptor and activate these cells [19, 20, 21]. Under physiological conditions, endothelial cells secrete many factors, monitor the transport of plasma molecules and regulate vascular tone. In addition, endothelial cells are involved in the regulation of cholesterol and lipid homeostasis, signal transduction, immunity and inflammation [22]. And, in addition, oxidized low-density lipoprotein promotes the growth and migration of smooth muscle cells, fibroblasts and macrophages. Vascular lesions are most often caused by hypercholesterolaemia, which can be induced by dietary supplementation, overproduction of lipoproteins by the liver or genetic mutations of lipid receptors and other proteins that regulate lipid homeostatic pathways.

3.1 Mutations of genes regulating the cholesterol level

Hypercholesterolaemia is a common, and still underdiagnosed, autosomal dominantly inherited disorder that is estimated to occur at a prevalence of ≈1 in 220 people worldwide. Familial hypercholesterolaemia (FH) is characterized by a persistent lifelong elevation of low-density lipoprotein cholesterol (LDL-C) and, if untreated, leads to early onset atherosclerosis and an increased risk of cardiovascular events. Untreated hypercholesterolaemia in men and women is associated with a very high risk ranging from 30–50% of having a fatal or non-fatal cardiac event at 50 and 60 years of age, respectively [23]. The most common cause of single-gene familial hypercholesterolaemia is pathogenic variants in the LDL receptor gene (LDL-R), which account for 85–90% of genetically confirmed cases of familial FH. Pathogenic variants in the gene for apolipoprotein (ApoB), a ligand for the LDL receptor, a component of LDL resulting in reduced binding of LDL to LDL-R, or gain-of-function mutations in the gene for proprotein convertase subtilisin/kexin 9 (PCSK9), resulting in increased destruction of LDL-R, account for 5–15% and 1% of cases of monogenic hypercholesterolaemia, respectively [24]. There is also an autosomal recessive form of hypercholesterolaemia in the human population, caused by homozygous mutations in the LDL-R adaptor protein which, is associated with the mild phenotype of homozygous hypercholesterolaemia found in Sardinian residents [25].

With the exception of the homozygous form of familial hypercholesterolemia (HoFH), FH is generally a silent disease. HoFH usually manifests with pathognomonic physical symptoms in childhood, such as cantelosis, tendon xanthoma and corneal arching. FH is diagnosed clinically based on a weighted combination of physical findings, personal or family history of hypercholesterolemia, early ischemic disease in the family and circulating LDL-C levels. The genetic cause is highly heterogeneous. Mutations in the LDL receptor genome are very common and occur at different sites disrupting receptor function in different ways. They therefore have different pathological significance. The spectrum of functional alterations in APOB outside the fragments routinely screened is growing. The ClinVar database at NCBI shows all the mutations in this gene described to date. There are about 3000 of them, and of these mutations that are labeled as pathological there are about 1000. They are mainly missense, nonsense frameshift mutations including about 500 deletions and 170 duplications. The largest number of known mutations are single nucleotide mutations mainly in coding regions of the gene, about 2000.

The known spectrum of mutations in APOB has been increasing in recent years thanks to next-generation sequencing (NGS) techniques, which allow all 29 exons of APOB to be studied without increasing laboratory workload [26, 27, 28, 29]. However, as APOB is a highly polymorphic gene, these variants require functional assessment before a clear diagnosis can be made [27]. It is also known that mutations in the APOB gene do not have 100 per cent penetrance, and the phenotype of patients is usually milder than in patients with FH caused by LDLR mutations [30].

The ClinVar database from NCBI is being updated with known pathological mutations in the APOB gene. There are currently 84 of them, most of which are located in the hydrophilic part of the apoB protein, the part that can bind to the LDL receptor. Mutations of the nonsense, missense and reading frame shift types dominate among the pathologies leading to familial hypercholesterolemia.

Familial hypercholesterolemia (FH), a major risk factor for coronary artery disease (CAD), is typically caused by mutations in genes that code for proteins responsible for removing low density lipoprotein (LDL) from the circulation. Only 17 pathogenic mutations in the PCSK9 gene are currently known and presented in the ClinVar database from NCBI. PCSK9 was discovered in 2003 when gain-of-function (GOF) mutations in this gene were identified as causative of FH in an autosomal dominant manner [31]. These GOF mutations are associated with hypercholesterolemia and a higher risk of CAD [32, 33, 34, 35, 36]. For example, a mutation in the apoB gene p.S127R is specifically associated with overproduction of this protein, resulting in greater synthesis of very low-density lipoprotein (VLDL), intermediate-density lipoprotein (IDL) and, consequently, LDL [32]. Another mutation of this gene p.E670G is associated more with serum lipid parameters, including total cholesterol (TC), high-density lipoprotein (HDL) and Apo B [33], as well as with an increased risk of stroke due to large vessel atherosclerosis and ischaemic stroke [36]. Serum PCSK9 levels, has been identified as a major predictor of carotid atherosclerosis independent of other risk factors in asymptomatic patients [37]. Furthermore, the contribution of PCSK9 concentrations to FH severity appears to be independent of LDL receptor genotype [38]. Recently, a homozygous gain-of-function mutation of the PCSK9 gene was characterized that is associated with the phenotype of a patient whose cholesterol is 316 mg/dl and LDL was 234 mg/dl at the age of 11 years [39]. This patient has no mutations in the LDL receptor or Apo B genes [39].

Loss-of-function mutations of the PCSK9 gene are associated with hypocholesterolaemia and significant protection against CAD [40, 41, 42, 43]. Notably, the p.Y142X mutation is found only in 0.4 per cent of African Americans, but not in other ethnic groups [40]. The p.C679X mutation is more common in African Americans and Zimbabwean Africans, but very rare in European Americans [41]. One individual has been described who is homozygous for the p.R46L mutation and has a total cholesterol level of 11 mg/dl [42]. In one family, six of the eight members who carry the p.R46L mutation have LDL levels below the bottom 10% percentile of LDL [42]. Another study reported that two healthy women with ‘loss of function’ mutations affecting both alleles of the PCSK9 gene have extremely low LDL cholesterol levels (14 mg/dL) [41, 42, 43].

The concept of polygenic hypercholesterolaemia for patients with a clinical diagnosis of FH but no monogenic cause was presented in 2013 by Talmud et al. [44]. This concept is based on the cumulative effect of LDL-C-raising alleles with a cumulative effect, perhaps in a complex interaction with the environment that leads to an increase in LDL-C, producing an FH-like phenotype and presenting this type of hypercholesterolaemia as a typical complex disease.

The more often publishing genes with polymorphisms contributing to the high cholesterol phenotype include cadherin EGF LAG 7-pass G-type receptor 2, ATP-binding cassette subfamily G members 5 & 8 (ABCG5/8), sterol regulatory element binding protein-2 (SREBP-2), signal transducing adaptor family member 1 (STAP1), and Apo E. Talmud’s group developed a genetic risk score (GRS) based on scoring 12 SNPs where individuals above the top decile of the distribution of LDL-C scores were described as having a higher probability of polygenic hypercholesterolaemia [44]. Then, by removing SNPs with smaller effects/lower frequencies, they showed that a weighted score of six SNPs performed as well as a score of 12-SNPs. The top three quartiles of the distribution also indicated a greater likelihood of a polygenic explanation for their elevated LDL-C [45]. Another study established the 10-SNP GRS, which showed a strong association with high LDL cholesterol, confirming the validity of this score as a genetic risk marker for elevated LDL cholesterol [46]. In this cohort, individuals with an extreme weighted GRS ≥1.96 (≥90th percentile) were defined as having polygenic severe hypercholesterolaemia. Research has gone further and a study of patients with severe hypercholesterolaemia found that a high polygenic score for 2 million-SNP LDL-C (upper 5th percentile) could explain hypercholesterolaemia in up to 23% of patients, while only 2% carried a monogenic mutation [47].

With the development of genetic testing in recent years, a mutation in any of the three known autosomal dominant genes causing familial hypercholesterolaemia is found in the majority of cases with a clinical diagnosis of familial hypercholesterolaemia. SituationBecause individuals with polygenic background hypercholesterolaemia do not have the same inheritance pattern observed in monogenic familial hypercholesterolaemia, familial cascade screening is not recommended for individuals with polygenic background, as only 30% of relatives have elevated LDL-C levels compared to 50% in monogenic families. The presence of a causative monogenic mutation is associated with the highest cardiovascular risk vs. no mutation or polygenic ancestry, providing prognostic information independent of LDL-C. This may also help to assess the intensity of intervention. Treatment adherence also appears to be higher after monogenic confirmation of hypercholesterolaemia.

4. Homocystein—factor of endothelial senescence and endothelial dysfunction

In addition to cholesterol, the second important factor whose high blood concentration causes vascular endothelial damage is homocysteine. Homocysteine (Hcy) is a sulfur-containing non-proteinogenic amino acid formed during the metabolism of the essential amino acid methionine. Hcy is considered an independent risk factor for atherosclerosis and cardiovascular disease, but the molecular basis of these compounds remains incompletely elucidated to date. There is a causal link, as studies have observed that impaired endothelial function, a key initial event in the development of atherosclerosis and CVD, is observed in hyperhomocysteinemia (HHcy). Various phenomena may explain the vascular toxicity associated with high homocysteine concentrations. For example, Hcy is an inhibitor of nitric oxide (NO) synthesis, a gaseous master regulator of endothelial homeostasis. In addition, Hcy is responsible for deregulating the signaling pathways associated with hydrogen sulphide another important endothelial gasotransmitter. Hcy is also involved in the loss of critical endothelial antioxidant systems and thus increases the intracellular concentration of reactive oxygen species (ROS) causing oxidative stress. ROS interfere with lipoprotein metabolism, forming oxidized forms of lipids that are removed by vascular wall macrophages contributing to the development of atherosclerotic vascular lesions. In addition, excess Hcy can be indirectly incorporated into proteins, a process referred to as N-homocysteinylation of proteins, inducing vascular damage. The inability to metabolize homocysteine and excess homocysteine decreases the synthesis of the universal methyl group donor, so necessary for epigenetic processes occurring in cells, and the hypomethylation of cellular DNA caused by the accumulation of S-adenosylhomocysteine (AdoHcy) also contributes to the molecular basis of Hcy-induced vascular toxicity and endothelial cell aging. A negative regulator of cellular methyltransferases, AdoHcy is a metabolic precursor of Hcy that accumulates under HHcy conditions [48].

4.1 Genetics of homocysteinaemia

There are various causes of homocysteinaemia. Genetic causes include mutations and enzyme deficiencies such as the most frequently mentioned 5, 10-methylenetetrahydrofolate reductase (MTHFR), but also methionine synthase (MS) and cystathionine β-synthase (CβS). In addition, HHcy can be caused by a diet rich in folate, but also by deficiencies of folate, vitamin B12 and, to a lesser extent, deficiencies of vitamin B6, which affects methionine metabolism, and also by impaired renal function.

MTHFR catalyzes the conversion reaction of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate, which is an intermediate in the conversion of Hcy to methionine. Mutations in MTHFR occur frequently in the population and are common inborn errors of folate metabolism that result in phenotypes ranging from asymptomatic to severe neurological deterioration and even early death in the classic form of MTHFR deficiency [49].

Homocystinuria is also an autosomal recessive error of metabolism resulting from defects in the cobalamin (vitamin B12)-dependent pathway that converts Hcy to methionine and is catalyzed by the enzyme methionine synthase.

Hcy in the blood is generally found 70–80% as a disulfide bound to plasma proteins, 20–30% as a homodimer with itself and about 1% as a free thiol, or a heterodimer with other thiols [50]. Levels of the Hcy are usually controlled by 2 biochemical processes: (1) roughly ~50% of the Hcy goes to transsulfuration pathway for producing the glutathione and the remaining and (2) ~50% can be remethylated back to methionine [51, 52]. Normally, the synthesis and elimination of Hcy stay pretty much in balance, but in diseased conditions, i.e., in HHcy, the overall plasma Hcy levels tend to increase due to the errors in the Hcy metabolism [53].

Causes of homocystinaemia include regular consumption of an excessively methionine-rich protein diet, or B12/folate deficiency, or ‘loss-of-function type’ mutations of the CBS gene as heterozygous or homozygous, and finally insufficient Hcy clearance from the kidney. Several other factors are influential among which are gender, age, smoking, alcohol consumption, certain medications, and medical conditions that can potentially modulate the methionine cycle can increase Hcy levels. Furthermore, there are additional genetic factors that are key in promoting HHcy status, such as genetic defects in enzyme proteins involved in ‘1-carbon metabolism’ [54, 55, 56]. As this cycle is the only pathway that gives methyl group in both biosynthesis of cellular compounds such as creatine, epinephrine, carnitine, phospholipids, proteins, and polyamines and in epigenetic changes (like methylation of DNA, RNA, and histones) [57]. Nevertheless, HHcy mediated metabolic malfunctioning because of the higher circulating Hcy levels promote oxidant stress-induced vascular inflammation and vessel dysfunction leading to atherosclerosis, myocardial infarction, stroke, multiple sclerosis, cognitive impairment, epilepsy, dementia, Parkinson’s disease, and ocular disorders [58, 59].

An interesting scientific discussion is being conducted in the context of the importance of the common MTHFR gene polymorphism and its significance in endothelial diseases. Heterozygous polymorphisms of the MTHFR gene reduce enzyme activity by 40% (CT variant, MTHFR c. [665C > T];[665C =]) and up to 70% in the homozygous form (TT variant, MTHFR c. [665C > T], [665C > T]). The CT variant is very common as it occurs in up to 20–40% of the Caucasian population and 1–4% of most other ethnic groups. The homozygous TT variant occurs in about 10% of the general population in Europe.

Retrospective studies conducted in the 1980s showed an increased prevalence of homocysteine concentrations in the 15–30 μmol/l range dependent on the MTHFR 677C > T polymorphism (new nomenclature, c.665 C > T) in the presence of concomitant folate deficiency in patients with atherosclerosis after myocardial infarction, stroke and coronary artery disease, and with a history of venous thromboembolism (VTE), i.e. deep vein thrombosis and/or pulmonary embolism. Quite different results were published from a prospective study published in 2002 in which these correlations were shown to be weak or even non-existent. In 2010, the American College of Cardiology and the American Heart Association unequivocally spoke out against homocysteine determination in cardiovascular risk assessment, considering hyperhomocysteinaemia to be a non-significant risk factor at the public health level [60]. In contrast, during protein biosynthesis, Hcy can be misused by methionyl-tRNA synthase to produce homocysteine thiolactone (HTL), a cyclic thioester that reacts rapidly with proteins to form amide bonds with the amino groups of lysine residues [61]. The resulting N-homocysteinylated proteins with altered structure and biochemical properties contribute to the vascular pathology associated with HHcy [62]. In fact, studies on cell cultures confirmed that Hcy supplemented in the medium was converted to HTL, and the extent of this conversion was proportional to Hcy concentration.

5. Gene polymorphisms influencing vasomotor endothelial function

Nitric oxide (NO), is a key vasodilator. It is formed in the vascular endothelium by the oxidation of arginine through the catalytic activity of nitric oxide synthase (NOS). This reaction requires NADPH and O2 as co-substrate and yields NO and citrulline as end products. Importantly, the enzymatic activity of NOS is inhibited by methylated analogues of arginine, namely N-monomethylarginine (L-NMMA) and asymmetric dimethylarginine (ADMA) [63], which are synthesized in vivo by a family of enzymes known as protein arginine methyltransferases. Proteolysis of proteins containing L-NMMA and ADMA releases them into the endothelial cell cytosol, from where they are removed into the blood. Elevated serum ADMA levels are associated with atherosclerotic vascular disease [64].

More than 15 polymorphisms exist in the NOS3 promoter that might influence mRNA transcription and reduce gene expression. Two polymorphisms in NOS3, 786 T > C and 894G > T, are the most studied. 786 T > C resides in the promoter region of NOS3 and regulates transcriptional initiation [65]. However, the –786 T > C polymorphism has shown inconsistent associations with functional measures, and with clinical disease end points. The CC genotype at 786 T > C is associated with blunted forearm blood flow responses to Ach in hypertensive subjects [66] and no increases in NOS3 mRNA and endothelial nitric oxide synthase (eNOS) protein expression in response to laminar shear stress in endothelial cells from coronary heart patients [67]. Polymorphisms within the coding region of the NOS 3 gene could alter NOS enzymatic activity. The 894G > T polymorphism in exon 7 of NOS3 results in substitution of glutamate with aspartate at codon 298 (also denoted as Glu298Asp) [68]. There is currently a debate, with controversial studies on whether this polymorphism is indeed functional. Two studies have shown that eNOS Asp298 undergoes selective proteolytic cleavage in endothelial cells and vascular tissues, which may account for reduced vascular NO production [69]. However, other studies have suggested that this finding may be the result of an artifact [70].

ADMA is removed from the circulation by metabolism primarily by isoform two of the DDAH2 dimethylarginine dimethylaminohydrolase, which predominates in tissues that express eNOS, such as the endothelium. The main cause of elevated ADMA levels in patients at risk for vascular disease is not fully understood, but one potential explanation could be loss-of-function mutations in the DDAH enzyme gene that alter gene expression or enzyme activity. Six potentially pathological polymorphisms have been identified in the DDAH2 gene. Five of them are upstream of the translation start site and may affect gene transcription. An insertion-deletion polymorphism (6G/7G) at position 2871, which lies in the core promoter region, affected DDAH2 promoter activity in the promoter/reporter assay [71].

The realization that common gene variants can, at best, have little to moderate impact on physiology and disease susceptibility has led to the understanding that future studies of susceptibility to complex diseases, whether they address clinical endpoints or intermediate phenotypes such as endothelial function, will need to be much larger and include more variables simultaneously. Because many GWAS identify SNPs outside protein coding regions or in non-coding intervals, the contribution of small non-coding RNA (e.g., lncRNA, microRNA) in modulating endothelial function should be addressed.

6. Micro-RNA and its epigenetic role in endothelial pathophysiology

Indisputably, microRNAs are fundamental regulators of many biological processes. Regulation of basic vascular endothelial functions by microRNAs and its disruption can lead to endothelial dysfunction. MicroRNAs are small, generally non-coding RNAs that regulate the expression of many genes through post-transcriptional degradation or translational repression. The crucial importance of microRNAs in endothelial physiology has been demonstrated following EC-specific inactivation of the enzyme Dicer which is involved in the biogenesis and processing of microRNAs, which cleaves microRNA precursors into mature forms [72, 73]. The absence of Dicer in the endothelium leads to altered expression of key regulators of endothelial function, including endothelial nitric oxide synthase (eNOS), vascular endothelial growth factor receptor 2 (VEGF), interleukin-8, Tie-1 and Tie-2.

Recent studies have identified miR-19a as an important driver of upregulation of important factors implicated in endothelial dysfunction, hyperlipidemia, inflammation and atherosclerosis, revealing a vicious cycle involving endothelial Hif-1a activation, hyperlipidemia and upregulation of miR-19a, promoting CXCR2 (C-X-C Motif Chemokine Receptor 2 which mediates neutrophil migration to sites of inflammation) dependent monocyte adhesion by increasing endothelial expression of its ligand CXCL1 [74]. It is also worth noting that microRNAs are also involved in switching the phenotype of VSMCs between a quiescent (pro-contractile, differentiated) and proliferative (pro-synthetic, differentiated) state [75], a critical step in the pathophysiology of atherosclerosis.

The endothelium has a critical role in maintaining vascular integrity and protecting against cardiovascular disease. Accumulated data indicate endothelial function is a heritable trait regulated by polygenic factors; however, these genetic factors have not been fully elucidated until now.

7. New discoveries based on genome-wide screening (GWAS)

Genome-wide association studies (GWAS) have been widely used in recent years to identify new genetic loci underlying chronic diseases. GWAS for endothelial function have been relatively limited due to the different phenotypes associated with it. The first such study was conducted by Vasan and colleagues for several cardiovascular traits, including FMD (%) and hyperemic flow velocity in 1345 individuals from the Framingham Heart Study, using a set of 100kSNPs [76]. They identified several SNPs associated with each trait in this study, including chloride channel (CFTR) and phosphodiesterase 5A (PDE5A) SNPs. Although these results have not been replicated, this was the first GWAS to directly examine endothelial function on a large population.

Yoshino and colleagues conducted an association study on the coronary vascular response to acetylcholine (ACh), a common index of coronary endothelial function, in 643 female and male subjects [77]. They used a set of 1536 SNPs located in genes related to cardiovascular physiology and pathology. The results showed that variants in adenosine A1 receptor (ADORA1) were associated with endothelial dysfunction in the entire cohort, while variants in adenosine A3 receptor (ADORA3) and lipoprotein(a) (LPA) had the strongest associations with increased risk of endothelial dysfunction in women only.

In recent genome-wide association study (GWAS) studies in European population, three novel sites related to endothelial dysfunction were found [78, 79]: Vascular endothelial growth factor A (VEGFA) rs9472135, Faciogenital dysplasia 5 (FGD5) rs11128722, Zinc Finger C3HC-type Containing 1 (ZC3HC1) rs11556924.

Because many GWAS identify SNPs outside protein coding regions or in non-coding intervals, the contribution of small non-coding RNA (e.g., lncRNA, microRNA) in modulating endothelial function should be addressed. In 2011, genome-wide association studies (GWAS) identified ANRIL as a biomarker closely associated with coronary heart disease (CHD) [80]. These studies identified, locus 9p21 which contains many single nucleotide polymorphisms (SNPs) that are located in a “gene desert” without any protein-coding genes. A key portion of the SNPs at the 9p21 locus overlap with six exons in the ANRIL gene also known as CDKN2B-AS or CDKN2B-AS1, which is transcribed in the antisense direction in the INK4b-ARF-INK4a gene cluster. ANRIL is expressed in vascular endothelial cells, vascular smooth muscle cells, mononuclear phagocytes and atherosclerotic plaques and its variation is associated with vascular endothelial malfunction, vascular smooth muscle cell (VSMC) including proliferation, migration, senescence, apoptosis, mononuclear cell adhesion and proliferation, glycolipid metabolism disorders and DNA damage [81].

Heritable changes in gene activity and expression also can be the result of epigenetic changes. Recent evidence suggests epigenetic changes such as those induced by histonemethyltransferase Set7 are associated with endothelial dysfunction, including impaired FMD in diabetics [82].

The problem with the paucity of GWAS studies is that most disease-relevant single nucleotide polymorphisms (SNPs) cannot be assigned to a specific gene, and even demonstrating that a single SNP affects gene expression is not possible for most SNPs. This is a consequence of the complex architecture of the genome, in which enhancers are often located far from their target gene in a two-dimensional sequence-based projection. The second aspect is a consequence of the heterocellularity of the atherosclerotic lesion, such that a specific SNP is relevant in only one of the many different cell types expressed in the lesion.

In the future, thanks to the already initiated GWAS studies in single cells of the atherosclerotic lesion, this second problem may be solved.