Atrial natriuretic peptide (ANP) is a hormone synthesized in cardiac atria and released into the circulation exerting relevant natriuretic, diuretic, and vasodilatory effects (1). Elevated ANP levels are associated with increased cardiovascular risk in the general population (2- 4) and with increased mortality in patients with heart failure, ischemic heart disease, and stroke (5- 6). The rs5065 ANP gene variant, corresponding to a T to C substitution at position 2238 of the pre-proANP gene, leads to an alteration of ANP primary sequence and translates into an altered ANP peptide. The latter has been shown to reduce endothelial cell viability, proliferation and endothelial cell tube formation in vitro and to modulate common mechanisms underlying the transition from stable to unstable plaque (7). The frequency of the rs5065 variant in the general population ranges from 14% to 23%, and it has been associated with increased risk of stroke and myocardial infarction (MI) in previous studies (8- 11). Nevertheless, the impact of rs5065 variant on susceptibility to develop cardiovascular events has never been investigated through a prospective type of approach. In addition, it is not clear whether this ANP variant, apart from predisposing to increased incidence of acute cardiovascular events due to its actions on mechanisms of plaque vulnerability, might also affect development of atherosclerotic disease. In this regard, emerging biomarkers of vascular atherosclerosis and plaque instability (i.e., lipoprotein-associated phospholipase A2 [Lp-PLA2], oxidized low-density lipoprotein [OxLDL], and myeloperoxidase [MPO]) (12- 14) might help our understanding of the real impact of rs5065 on atherosclerotic plaque progression/vulnerability.

In the present study we aimed at: 1) investigating the role of rs5065 variant on susceptibility to acute coronary syndrome (ACS) through the analysis of 2 independent ACS populations and 2 independent cohorts of stable angina (SA) patients; 2) assessing its effect on coronary artery disease (CAD)-related outcomes through a prospective follow-up approach in SA patients; and 3) exploring the impact of rs5065 on atherosclerotic plaque progression/instability within the context of ischemic heart disease through the determination of vascular biomarkers like Lp-PLA2, OxLDL, and MPO.

The study population consisted of 1,004 consecutive Caucasian patients referred to the Cardiovascular Center OLV Aalst (Belgium) for cardiac catheterization from January 2000 to December 2002 because of: 1) SA with at least 1 significant coronary stenosis (diameter >50% as assessed by quantitative coronary angiography) and positive functional stress test; or 2) ACS (e.g., unstable angina, non–ST-segment elevation MI or ST-segment elevation MI) (15). Control population consisted of 393 healthy Caucasian volunteers. To replicate the findings observed in SA patients, genotype analysis was also performed in an independent population of 482 consecutive stable angina (rSA) Caucasian patients with at least 1 significant coronary stenosis (diameter >50% as assessed by quantitative coronary angiography) and positive functional stress test referred for cardiac catheterization from January 2007 to December 2009. To replicate the findings observed in ACS patients, genotype analysis was also performed in an independent population of 675 Caucasian patients with ACS, originating from the Belgian cohort of the GRACE registry (Global Registry of Acute Coronary Events) (16). Caucasian ethnicity of the subjects included in the study was established on the basis of a questionnaire to exclude non-Caucasian origin up to 3 generations of ancestors. Only white European subjects and patients from 3 generations of ancestors were included in the study.

Systemic arterial hypertension was defined as systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg on at least 2 separate occasions (17) or ongoing antihypertensive treatment. Diabetes mellitus was defined as the presence of an active treatment with insulin or an oral antidiabetic agent (18). Patients were defined as hypercholesterolemic when they had total cholesterol plasma concentrations ≥220 mg/dl or they were receiving lipid-lowering drugs. Patients reporting regular smoking in the previous 6 months were considered current smokers.

Clinical endpoints were evaluated at clinic visits alternated with phone contacts for up to 6 years in the SA population and 4 years in the rSA population. The primary endpoint was major adverse cardiovascular event (MACE), defined as the composite of: 1) mortality from all causes (Death); 2) MI; and 3) revascularization by percutaneous coronary intervention or bypass surgery (Revasc). Secondary endpoints were death, MI, revascularization, cerebrovascular accidents (CVAs), and major adverse cardiac and cerebrovascular event (MACCE) (composite of death, MI, revascularization, and CVA).

The study was approved by the Ethical Committees of the participating institutions, and informed consent was obtained from all subjects.

A blood sample collection was performed from a peripheral vein for deoxyribonucleic acid analysis. Deoxyribonucleic acid was extracted from whole blood with a commercially available kit (Qiagen). The ANP gene single nucleotide polymorphism rs5065 was characterized as previously described (8,19).

Blood sampling for biomarker assessment was performed at the time of study enrolment in rSA patients. The N-terminal (NT)-proANP levels were assessed by a commercially available enzyme-linked immunoadsorbent assay (ELISA) kit (Gruppe Biomedica, Vienna, Austria). The MPO levels were measured by Mercodia MPO ELISA kit (Mercodia, Uppsala, Sweden); Lp-PLA2 levels were assessed by Lp-PLA2 PLAC test Diadexus ELISA kit (Diadexus, South San Francisco, California); OxLDL levels were determined by a sandwich ELISA kit with the monoclonal antibody 4E6 (Mercodia). The antibody 4E6 is especially against an epitope in the apoB-100 moiety of OxLDL that is formed from substitution of lysine residues of apoB-100 with aldehydes (20).

All data analyses were performed with SPSS software package (version 13.0, SPSS, Inc., Chicago, Illinois). Normal distribution was assessed by Kolmogorov-Smirnov test. Continuous variables are expressed as mean ± SD or median [interquartile ranges]; categorical variables are expressed as frequencies and percentages. Student t test or Mann-Whitney test was used to compare continuous variables, as appropriate. Comparisons between categorical variables were evaluated with 2-tailed Fisher exact test or Pearson's chi-square test, as appropriate. The NT-proANP levels within the different genotypes were compared with Kruskal-Wallis test. Predictors of biomarkers elevation were assessed by linear regression analysis for continuous variables adjusted for age, sex, body mass index (BMI), hypertension, hypercholesterolemia, diabetes, and smoking. Genotype frequencies and Hardy-Weinberg equilibrium were estimated with chi-square test.

The risk associated with ANP variant in the occurrence of ACS and risk of significant coronary stenosis at angiography were estimated by logistic regression analysis, computing the odds ratio with the respective 95% confidence interval under the assumption of a dominant model (score of 0 for WT; 1 for heterozygous and double mutant). Due to the low number of homozygotes for the rs5065 minor allele (MA), neither additive nor recessive models of inheritance were included. A multivariate logistic regression analysis model was built including the following variables as covariates: age, sex, BMI, hypertension, hypercholesterolemia, diabetes, and smoking.

Survival curves for MACE were constructed with the Kaplan–Meier method. Clinical endpoints were evaluated with the Cox proportional hazard analysis. Cox proportional hazard analysis for cumulative MACE at follow-up was performed with 4 multivariate models: 1) adjusted for age, sex, BMI, hypertension, diabetes, hypercholesterolemia, smoking habit, multivessel disease; 2) adjusted for age, hypertension, diabetes, hypercholesterolemia, smoking habit, therapy with angiotensin-converting enzyme inhibitors (ACE-I), diuretic agents, aspirin; 3) adjusted for sex, hypertension, diabetes, hypercholesterolemia, smoking habit, therapy with ACE-I, diuretic agents, aspirin; and 4) adjusted for multivessel disease, hypertension, diabetes, hypercholesterolemia, smoking habit, therapy with ACE-I, diuretic agents, aspirin. Results are expressed as hazard ratios (diuretic agents) and 95% confidence intervals. For all tests, a 2-tailed p < 0.05 was considered significant.

Clinical characteristics of the patients are shown in (Table 1). With the exception of BMI and therapy with diuretic agents, clinical features and medical therapy of patients with SA and ACS were significantly different as compared with the control population. The SA patients were characterized with higher frequency of multivessel disease as compared with ACS.

The rSA patients were younger, with higher incidence of hypertension and hypercholesterolemia, as compared with the SA group. By contrast, smoking habit, multivessel disease, and treatment with ACE-I and aspirin at discharge were lower as compared with the SA group.

The rACS patients were slightly older and more frequently smokers, whereas they presented with lower frequency of male sex, hypertension, diabetes, and hypercholesterolemia, compared with ACS patients. Medical therapy also differed at discharge, as compared with ACS patients (Table 1).

Genotype and allele frequency of the patients is reported in (Table 1). The observed genotype distribution for rs5065 in the overall population was in Hardy-Weinberg equilibrium (p = 0.48). Frequency of the rs5065 MA in control subjects was similar to that previously published in other control populations (8,10,16). As shown in (Table 1), both SA and rSA patients presented an rs5065 MA frequency similar to that observed in control subjects. In contrast, both ACS and rACS patients presented higher frequency of rs5065 MA, as compared with control and SA patients.

The NT-proANP plasma levels were assessed in 428 rSA patients. No differences were detected among the 3 genotypes: 4,138 (2,988 to 6,578) fmol/ml in rs5065 WT, 4,266 (2,833 to 6,373) fmol/ml in rs5065 heterozygotes, 4,677 (2,531 to 8,137) fmol/ml in rs5065 mutant homozygotes (p = 0.92).

The MPO, Lp-PLA2, and OxLDL were assessed in 212 rSA patients (Figure 1). The MPO levels were significantly higher in rs5065 MA carriers as compared with WT (597 [345 to 832] μg/l vs. 488 [353 to 612] μg/l, p = 0.038). At the linear regression analysis, rs5065 MA was the only independent predictor of MPO elevation (unadjusted: beta = 0.21, t = 3.04, p = 0.003; adjusted: beta = 0.18, t = 2.66, p = 0.008). No significant difference was observed between rs5065 WT and rs5065 MA carriers with regard to Lp-PLA2 (251 [204 to 311] μg/l vs. 252 [215 to 315] μg/l, p = 0.59) and OxLDL (60 [43 to 79] U/l vs. 55 [45 to 75] U/l, p = 0.78) levels.

Predictors of ACS are shown in (Table 2). The multivariate adjusted analysis confirmed the rs5065 MA to be an independent predictor of ACS.

The rs5065 variant neither showed an effect on the risk of significant coronary artery stenosis at the angiography (p = 0.15) nor was associated with the number of diseased vessels (p = 0.32).

In SA patients, median clinical follow-up was 37 months (13 to 70 months), and it was obtained in 376 patients (87%). A total of 145 (39%) patients experienced MACE during the follow-up period (Table 3). At the univariate analysis, MA carriers showed a significantly higher rate of MI, the combined endpoint of death/MI, revascularization, MACCE, and MACE as compared with WT (Table 3). No difference was observed in the rate of death and CVA. Survival analysis revealed a significantly lower MACE-free survival in MA carriers (Figure 2A). At Cox proportional hazard analysis, irrespective of the multivariate model, rs5065 MA carriers were independently associated with significantly higher MACE rate at follow-up (Table 4). Medical treatment at follow-up did not significantly differ between rs5065 WT and MA carriers (data not shown), with the exception of diuretic agents. In particular, 123 (24%) WT patients and 156 (30%) rs5065 MA carriers were taking diuretic agents (p = 0.02).

In rSA patients, median clinical follow-up was 31 months (23 to 41 months), and it was obtained in 462 patients (96%). A total of 86 (19%) patients experienced MACE during the follow-up period (Table 3). At the univariate analysis, MA carriers showed a significantly higher rate of MI, the combined endpoint of death/MI, CVA, MACCE, and MACE as compared with WT (Table 3). No difference was observed in the rate of death and Revasc. Survival analysis revealed a significantly lower MACE-free survival in MA carriers (Figure 2B). At Cox proportional hazard analysis, irrespective of the multivariate model, rs5065 MA carriers were independently associated with significantly higher MACE rate at follow-up (Table 4). Medical treatment at follow-up did not significantly differ between WT and MA carriers (data not shown).

Our findings demonstrate that patients carrying the rs5065 MA are at higher risk of ACS and show lower MACE-free survival rate.

The ANP variant resulting from a T to C substitution at position 2238 of the pre-proANP gene is present in 14% to 23% of individuals from the general population (8,21). It leads to a mutant peptide able to induce endothelial cell damage by increasing reactive oxygen species production and reducing antioxidative stress response (7). This mutant peptide was also shown to up-regulate genes encoding enzymes involved in extracellular matrix degradation that might favor the transition from stable to unstable plaque (7).

The rs5065 ANP variant has been previously associated with increased risk of stroke and MI (8- 11). In the present study, we demonstrate increased frequency of rs5065 MA in ACS patients, with a nearly 2-fold independent risk of ACS. Most importantly, the direct contributory role to cardiovascular events of rs5065 MA is supported by the observation of lower event free survival with higher MI and higher combined death/MI, revascularization, MACCE, and MACE rates in SA patients carrying rs5065 MA. Yet, no association of rs5065 MA with significant coronary artery stenosis or with the number of diseased vessels was observed. This apparent discrepancy might be related to the reported in vitro effects of this ANP variant favoring plaque instability (7).

Of note, elevated MPO levels were measured in MA carriers, as compared with WT subjects, in rSA patients, consistent with a major effect on plaque vulnerability rather than on progression of vascular atherosclerosis. Myeloperoxidase seems to play a causal role in plaque instability (22) by exerting its deleterious effects through: 1) the initiation of lipid oxidation in the subendothelial space of the vessel wall, where it generates several reactive oxygen intermediates able to induce oxidative cellular damage; and 2) the activation of matrix metalloproteinases (23) and deactivation of matrix metalloproteinase inhibitors (24), thereby promoting the weakening of the fibrous cap, potentially leading to destabilized atherosclerotic plaque. Myeloperoxidase has been associated with higher risk of developing first acute coronary event as well as subsequent cardiovascular events in patients previously admitted for ACS (25- 26). Although we cannot clarify the causal interaction between mutant ANP and MPO levels, we suggest that MPO might act synergistically and exacerbate the effects of mutant ANP in CAD patients.

By contrast, we did not observe any association between rs5065 MA and levels of either OxLDL or Lp-PLA2. Lack of association with these 2 biomarkers could be partly explained by their involvement mostly in the progression, severity, and extent of atherosclerotic disease rather than on plaque vulnerability (27- 30). However, with regard to our findings on OxLDL, it should also be mentioned that the 4E6 antibody from Mercodia used in our assay does not have a high specificity for OxLDL, being especially prepared against an epitope in the apoB-100 moiety of OxLDL (31- 32), and consequently, it might have limited, at least in part, the results of our analysis.

Data in SA patients were replicated in an independent control patient population (rSA). It should be acknowledged that rSA patients showed lower MACEs as compared with SA patients (19% vs. 39%, p < 0.001), most likely due to different duration of follow-up, lower incidence of multivessel disease, and adoption of drug-eluting stents, unavailable at the time SA patients were recruited. Nevertheless, rSA patients carrying rs5065 MA showed significantly higher MI, CVA, MACCE, and MACE rates as well as lower MACE-free survival at follow-up, as compared with WT patients.

Data in ACS patients were also replicated in an independent patient population (rACS). It will be interesting in the future to assess whether rs5065 ANP variant might influence cardiovascular prognosis after acute coronary events.

A recent pharmacogenetic study has shown beneficial effects of treatment with diuretic agents in rs5065 MA carriers in terms of improved blood pressure control and reduced cardiovascular risk in hypertensive subjects (21). Thus, it is possible that a pharmacogenetic interaction might have slightly confounded our findings at least in SA patients, because rs5965 MA carriers took diuretic agents more frequently, as compared with WT patients. Therefore, the outcome of MA carriers might have been slightly improved, although it remained significantly worse as compared with that of WT patients.

We also tested the hypothesis that rs5065 ANP variant might produce its effects through an alteration of the circulating peptide levels. In fact, a substantial amount of evidence has shown that elevated ANP levels are associated with increased cardiovascular risk and with worse prognosis in patients affected by cardiovascular diseases (1). In particular, elevated NT-proANP levels were found to parallel the progression of coronary atherosclerosis (33). High ANP levels were reported in stroke (6,34), and they have been associated with increased post-stroke and post-MI mortality (6) as well as with mortality in heart failure patients (5). However, we were unable to identify significant differences among the 3 rs5065 genotypes in terms of circulating peptide concentrations. Although we cannot exclude a role of the ongoing medical therapy, the latter finding mainly suggests that the structural peptide alteration dependent from the T to C transition within exon 3 of pre-proANP gene is the predominant cause of the observed increased predisposition to ACS, consistent with our in vitro observations (7).

The rs5065 ANP variant leading to altered and dysfunctional ANP peptide is associated with higher risk of ACS and translates into worse long-term clinical follow-up in SA patients. Our prospective study provides the first demonstration of the important predictive role played by rs5065 ANP gene variant in CAD. We measured higher MPO levels, a biomarker associated with plaque vulnerability, in rs5065 MA carriers, corroborating previous in vitro evidence of a dysfunctional peptide.

More studies are needed to define whether systematic screening for ANP gene variants in CAD patients will help to better define the individual cardiovascular risk prediction. In particular, more aggressive and specifically tailored pharmacological treatment might be required to improve clinical outcome in patients carrying rs5065 MA.