Pharmacogenetic Implications of eNOS Polymorphisms (Glu298Asp, T786C, 4b/4a) in Cardiovascular Drug Therapy

Abstract

Endothelial nitric oxide synthase (NOS3 or eNOS) is the enzyme responsible for the highest production of nitric oxide, with the greatest impact on the cardiovascular system, encoded by the eNOS gene, which presents various polymorphisms. ENOS gene polymorphisms play an important role in the response to drugs affecting nitric oxide (NO) signaling. This review discusses the pharmacogenetic impact of eNOS polymorphisms on the response to drugs affecting NO activity: angiotensin converting enzyme inhibitors, angiotensin II receptor antagonists, calcium blockers, beta-blockers, diuretics, phosphodiesterase inhibitors, and statins. The identification of biomarkers that accurately predict particular phenotypes is a challenge that needs additional large studies, in different populations. Efforts should be oriented towards a more accurate evaluation of the effects of eNOS genetic variants on biochemical parameters reflecting eNOS gene expression and enzymatic activity, in different diseases, as well as following drug treatment. This approach will allow for a better understanding of the role of eNOS genetic variants in cardiovascular disease progression and for cardiovascular drug therapy optimization.

Lack of a positive response to drugs or a toxic reaction to usual doses are responsible for high morbidity and mortality rates and for increasing health care costs (1,2). The variability of patients’ response can be explained by pharmacogenetics, an interdisciplinary field combining pharmacological and genetic information, which addresses how genetic polymorphisms may influence the response or lack of response to different classes of drugs, as well as the development of various degrees of toxicity (1,3). Previous studies have showed that 20-95% of the variation in individual drug responses can be explained by genetic polymorphisms (4).

In the cardiovascular field, the response to administrated medications is influenced by many factors, such as doses, patient’s disease and particularities (dyslipidemia, hypertension, atherosclerotic diseases, genetic polymorphism) (5-7).

Endothelial dysfunction represents one of the first steps of atherosclerosis, which develops even before any angiographic evidence of disease. One basic characteristic of this dysfunction involves the alteration of nitric oxide (NO) synthesis, release and activity (8).

The production of NO is mediated by the endothelial nitric oxide synthase (eNOS) (9,10) from L-arginine by NOS (neuronal, endothelial and inducible) (10). Decreased NO synthesis results in the development of cardiovascular diseases, responsible for the increased number of deaths worldwide (11). Various eNOS gene polymorphisms have been described (10), such as insertions/deletions, microsatellites, single nucleotide polymorphisms (SNPs), and variable number of tandem repeats (VNTRs).

NO signaling can be modulated by different drugs, and their effect may be influenced by eNOS gene polymorphism (12-18). Despite being constitutively expressed, eNOS is regulated by many stimuli at transcriptional, posttranscriptional and posttranslational level (19). In this regard, variations in the eNOS gene influence its activity, which consequently affects NO production (20).

NO plays an important role in the normal physiology of the cardiovascular system, and dysfunctional NO signaling has been associated with development and progression of cardiovascular diseases – myocardial infarction, coronary spasm, intra-stent thrombosis, heart failure (21-25,26-31).

The following eNOS gene polymorphisms, as the most extensively studied, will be described here: i) G894T (Glu 298Asp), exon 7 (rs1799983), ii) T786C, promoter SNP (rs2070744), and iii) Intron 4 (4b/4a), VNTR.

The single nucleotide polymorphism, rs1799983 is located in exon 7, position 894 in the NOS3 gene (G894T) and determines a glutamine to aspartate change at position 298 of the protein (Glu298Asp) (32). The variant allele for G894T polymorphism reduces eNOS binding to caveolin-1, leading to eNOS diminished availability and activity in endothelial cells (33). The presence of reduced NO formation in patients with this allelic variant supports these in vitro findings (34,35). At a transcriptional level, eNOS can be affected by a single nucleotide polymorphism, rs2070744 (T786C), which dramatically reduces ENOS transcriptional activity (36,37). This effect is linked to replication protein A1 (RPA1), which binds to the eNOS promoter with higher affinity in the presence of the mutation (38). In fact, inhibition of RPA1 expression can restore the transcriptional activity in the eNOS promoter with the C allele, whereas RPA1 over-expression has the contrary effect (38). Interestingly, these in vitro studies are consistent with in vivo changes, which have shown that the circulating levels of NO-related markers in subjects carrying the C allele tend to be lower (16) compared to that of T allele carriers, supporting a functional importance of this polymorphism (38).

The 4b/4a VNTR polymorphism in intron 4 (variable number of tandem repeats on intron 4) of the eNOS gene regulates eNOS post transcriptionally, by altering the formation of a small interfering RNA (siRNA) (39). The types of this polymorphism with four (variant 4a) or five copies (variant 4b) of the 27 bp siRNA fragment (20) are the most commonly met alleles. In vitro studies have shown higher siRNA levels in endothelial cells containing five copies, determining lower eNOS mRNA levels, compared to cells containing only four copies (39,40).

Correlations between eNOS gene polymorphisms and differentiated responses to cardiovascular drugs have also been reported (41). In this review we discuss the pharmacogenetic impact of eNOS polymorphisms on drugs that affect eNOS activity, such as antihypertensive drugs with a role in NO bioavailability. These include: i) angiotensin converting enzyme inhibitors, ii) angiotensin II receptor antagonists, iii) calcium blockers, iv) beta-blockers, v) diuretics, and other drugs whose action is influenced by nitric oxide, such as phosphodiesterase inhibitors and statins.

Angiotensin-converting Enzyme Inhibitors

Antihypertensive drugs’ effects are influenced by ENOS polymorphism (14,18). Angiotensin-converting enzyme (ACE) inhibitors are among the most used antihypertensive drugs (42). ACE inhibition is responsible for vasodilation and improvement of endothelial function, through decreasing the levels of angiotensin II and increasing those of bradykinin (43). Bradykinin stimulates endothelial cell receptors, leading to eNOS activation, NO release and vasodilation (44).

Silva et al., showed in a study on hypertensive patients treated with enalapril (an ACE inhibitor), that the “C” alleles of the rs2070744 polymorphism were more frequently found in patients with a good response to Enalapril (18). On the other hand, the same group has also shown that there were no differences between responders and non-responders to enalapril when they beared the 4b/4a and Glu298Asp alleles of the eNOS polymorphism (18).

Sandrim et al., reported that the “C Glu b” haplotype is more frequently found in NT (normotensive healthy controls) compared to HT (hypertensive patients) or RHT (resistant hypertensive patients) (NT 21% versus HT 8% and RHT 7%, both p<0.00625) (45). In HT or the RHT groups, the “C Asp b” haplotype was more frequently found (HT 22% and RHT 20%, versus NT 8%, both p<0.00625) (45). According to this study, no significant difference was found between HT and RHT patients regarding distribution of eNOS haplotypes in the HT and RHT groups, suggesting an inconsequential influence of the eNOS gene on the antihypertensive therapy’s resistance.

Oliveira-Paula et al., (46) also reported that GG genotype for rs16960228 polymorphism of the PRKCA influenced the quality of enalapril response. In a recent review, Luizon et al. (47), showed that eNOS polymorphisms are associated with susceptibility of preeclampsia (PE) and affect the response to antihypertensive treatment in PE. In fact, eNOS haplotypes combining the “C, a, Glu” of the T-786C (rs2070744), 27 bp VNTR a/b and Glu298Asp (rs1799983) were more frequent in the responsive subgroup of PE to antihypertensive treatment (47).

Angiotensin II Receptor Blockers (ARBs)

Clinical studies have confirmed a significant amelioration of the endothelium-dependent vasodilation in patients with arterial hypertension treated with angiotensin II receptor blockers (ARBs) compared to placebo or other antihypertensive agents (48-51). The increased NO release from the ARB therapy can be pharmacologically explained by the reduced activity of angiotensin II, resulting in enhanced antioxidant protection and NO bioavailability (52,53). Even though bradykinin concentration is not increased by ARBs,the latter can induce eNOS expression (54-56). Mason et al., (55) have reported higher endothelial NO release induced by all the tested ARBs (losartan, olmesartan, telmisartan, valsartan) compared to untreated cells. Different ARBs affect the NO release in various ways, depending on eNOS polymorphisms. Olmesartan (55) can raise NO levels by approximately 30 % in different eNOS genotypic backgrounds, with the most evident difference in ECs from donors homozygous for the nucleotide mutations T-786C and G894T. This result shows the specific effects of ARBs on eNOS function, which is in turn influenced by single nucleotide substitution, followed by changes in NO metabolism and finally leading to augmented cardiovascular risk, including hypertension (57-62). The relative contribution of the different mutations to ARB responses requires a separate thorough study.

Beta-blockers

Liljedahl et al., have demonstrated that the G allele of the 2996A/G eNOS polymorphism is associated with a higher blood pressure (BP) response to a β-blocker, and the A allele of the 498G/A eNOS polymorphism is associated with higher BP responses to a β-blocker and an angiotensin II receptor blocker (63).

Pacanowski et al., have shown that patients with eNOS polymorphisms have relatively lower BP during treatment and enhanced rates of BP control (64). Although this correlation was important only for patients assigned to the verapamil SR strategy, a similar tendency appeared when alternative treatment using atenolol was administered. Thus, aggressive multi-drug regimens used in INVEST (International Verapamil SR/Trandolapril Study) (65), irrespective of the drug types included, may induce a superior therapeutic response in patients with the -786T>C polymorphism. INVEST was a prospective, randomized trial comparing antihypertensive therapy with β-blockers versus calcium antagonists in 22,576 CAD (coronary artery disease) patients with hypertension (65). The accurate mechanism of the stronger association observed in subjects treated with verapamil is not known.

Interestingly, in agreement with prior evidence, this result supports the fact that atenolol does not influence endothelial function or alteration of NO biodisponibility (66-67). A difference of 4-7 mmHg in BP across the different genotype groups in the verapamil SR arm is a clinically significant variation, knowing the fact that a constant reduction of 12 mmHg in systolic BP prevents one death for every 11 patients treated (68). There is no consistent association of eNOS polymorphisms with BP phenotypes (69-71). These results suggest that eNOS genotype might contribute to heterogeneity of responses to several antihypertensive drugs, and pharmacogenetic studies should pursue a more complete characterization of this biologically plausible polymorphism drug relationship.

Calcium Blockers

Several calcium blockers, such as nifedipine, have shown to improve both endothelial function and NO bioavailability (72-73). According to previous studies, amlodipine (a calcium blocker) may provoke vasodilation through eNOS activation (74). At least one of the pathways in eNOS activation is known to be calcium-dependent (72-73). Amlodipine blocks Ca efflux, deactivating eNOS through regulation of Ca. In a randomized clinical trial, Zhang et al., reported that for rs1799983 (Glu298Asp G>T), lower all-cause mortality was observed in minor allele carriers treated with amlodipine versus lisinopril [for GG HR=1.01 (95%CI=0.91-1.13), GT+TT=0.85 (95%CI=0.75-0.97), p=0.04] (75). The authors also showed significant associations between eNOS variants with CHD and heart failure as well as significant pharmacogenetic effects for stroke and all-cause mortality. This suggests the fact that eNOS3 polymorphisms might give practical information with regard to treatment choosing decisions in the future.

Diuretics

Thiazide diuretics are extensively used for hypertension control, with hydrochlorothiazide (HCTZ) being the most used in the clinical practice (76). It has been established that the Glu298Asp polymorphism in the eNOS gene is capable to modulate the response to hydrochlorothiazide, thus subjects homozygous for the Glu allele have a significantly greater reduction in BP levels compared to individuals with the Asp allele (76). Despite having a significant effect, it is rather is small, demonstrating that, for a single polymorphism, it is difficult to explain the entire genetic background underlying variations in the pharmacological responses of a medication class.

Phosphodiesterase Type 5 (PDE-5) Inhibitors

eNOS polymorphisms also influence the treatment of erectile dysfunction (ED) (15,77). Phosphodiesterase type 5 (PDE-5) inhibitors are commonly used for the treatment of ED since inhibition of the PDE-5 enzyme increases the tissue cGMP concentration, mainly in the absence of NO signaling (77). Eisenhardt et al., have shown that homozygous subjects for the “Asp” allele of the Glu298Asp eNOS polymorphism are less responsive to sildenafil (an PDE-5 inhibitor) (13), however, this association is still subject to controversy (13,15). Patients with ED carrying the “4a” and “C” alleles show enhanced responses to sildenafil versus the “4b”and “T” alleles for the 4b/4a VNTR and g.-786T>C eNOS polymorphisms, respectively (13,15). Muniz et al., have reported that reduced plasma nitrite concentration predicts better responses to sildenafil. In addition, response to sildenafil is influenced by eNOS haplotypes including g.-786T>C, Glu298Asp and 4b/4a VNTR (15).

Lacchini et al., have demonstrated that the “T” allele of the Glu298Asp eNOS polymorphism, the “4b” allele of the 4b/4a VNTR and the “C” allele (T786>C) are associated with a poor response to therapy with PDE-5 inhibitors (77).

In a recent study by Di Salvatore et al., the role of eNOS polymorphism in the development of hypertension and proteinuria was presented, by associating the eNOS c.-894T genotype with a significantly higher rate of grade 3-4 hypertension and proteinuria (p=0.0002) (78).

In this context, carriers of the “T” allele of c.-894T eNOS polymorphism that show lower basal eNOS levels may have increased risk of hypertension induced by VEGF blockade (79-81).

Statins

The main effects of statins are due to the decrease in cholesterol synthesis (8), through blocking of the conversion of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) to mevalonate (8,10,82-83), independent of eNOS polymorphisms.

At the same time, as it has previously been suggested (10,41) that statins’ effects and eNOS polymorphisms modulate each other in a reciprocal way. The decrease in serum cholesterol can also cause an increase in endothelial nitric oxide production (8,82,84) and improve endothelial function (8), even before a significant decrease in serum cholesterol occurs (85-89).

Statins can affect eNOS regulation at the transcriptional and posttranscriptional level (90). Studies have shown that fluvastatin and atorvastatin can increase the transcriptional activity of the eNOS gene. Statins can also inhibit the replication protein A1 (RPA1), a repressor of the transcriptional activity of eNOS (1,16,41), particularly in subjects with the CC genotype (1,12,38,91). This has been demonstrated by increased circulating levels of nitric oxide biomarkers (nitrite and nitrate) following administration of statins, regardless of the genotype, but being higher in C-allele carriers (1,12,16,41,82,83). Such effects have been reported in hypertensive (92), dyslipidaemic subjects (93) and obese women (83) so far.

Statins can also increase eNOS expression and activity – compensating the genetic disadvantage of subjects with the CC genotype for the T786C polymorphism - (1,10,16,41,94-98), effect on eNOS expression not being lowered when LDL cholesterol values normalized (94).

Statins play an additional role in eNOS function by decreasing the level and inhibitory activity of caveolin (99,100). In patients with the eNOS Glu298Asp mutation (guanine to thymine conversion at position 894 of the gene), the mutant endothelial cells seem to produce less NO, reducing the availability of eNOS in the caveolae of these cells (41).

Statins inhibit isoprenoids, small GTP-binding proteins, Rho (which induces an increase in vascular smooth muscle sensitivity to calcium), Ras, and Rac (its activation leads to the formation of lamellipodia and membrane ruffles) (8,89). Therefore, statins cause an accumulation of inactive Ras and Rho in the cytoplasm (8,89) and increase eNOS expression (101) and eNOS mRNA stability (8,55,94), thus prolonging the eNOS mRNA half-life (101).

Statins also bear antioxidant properties, particularly in the CC genotype (1,8,16,83). They decrease different markers of atherosclerosis (8,82), the anti-inflammatory effect being modulated by eNOS polymorphism (1,16,41,82-83). This idea was demonstrated by a study using atorvastatin, which significantly reduced sCD40L, sVCAM-1, sP-selectin and MMP-9 concentrations only in subjects with the CC genotype (41,82).

Statins are capable of reducing the fluidity of the red blood cell plasma membrane, but only in the case of the CC genotype for the T786C polymorphism (41,82).

Stains’ modulation on the eNOS system is influenced by the polymorphism in intron 4. Adenosine-induced coronary vasodilation was assessed by Kunnas in healthy individuals (treated with pravastatin). Individuals with the “a” allele of the 4b/4a VNTR show a significant enhancement in vasodilation compared to individuals with the “bb” genotype, possibly due to a greater raise in endothelial NO production (102).

All these mechanisms seem to support the idea that statins are more useful in preventing cardiovascular diseases, by countering the mechanisms underlying the development of cardiovascular diseases, particularly in subjects with the CC genotype (1,41).

Conclusion

Further studies are needed to elucidate the functional and clinical implications of eNOS3 polymorphism. The finding of biomarkers able of predicting a particular phenotype with accuracy is a challenge that will probably involve large scale studies, with results applicable in different populations.

Efforts should be oriented towards an evaluation of the effects of NOS3 genetic variants on biochemical parameters reflecting eNOS gene expression and enzymatic activity (in patients developing different pathologies). This approach can permit progress in our understanding of how eNOS genetics may contribute to cardiovascular disease and help us optimize the relevant drug therapy.

We propose a change in treatment paradigm, from an approach based on average effects (reported in large cohorts) to individualized treatment to revolutionize both patients’ therapy and manage adverse health effects, especially for cardiovascular disease patients.

Conflicts of Interest

The Authors declare that there are no conflicts of interest.

Authors’ Contributions

All the Authors contributed to the conception and design of the article and approved the final version submitted for publication.