Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Functional Epistatic Interaction between rs6046G>A in F7 and rs5355C>T in SELE Modifies Systolic Blood Pressure Levels

Functional Epistatic Interaction between rs6046G>A in F7 and rs5355C>T in SELE Modifies Systolic Blood Pressure Levels

  • Said El Shamieh, 
  • Ndeye Coumba Ndiaye, 
  • Maria G. Stathopoulou, 
  • Helena A. Murray, 
  • Christine Masson, 
  • John V. Lamont, 
  • Peter Fitzgerald, 
  • Athanase Benetos, 
  • Sophie Visvikis-Siest
PLOS
x

Abstract

Background

Although numerous genetic studies have been performed, only 0.9% of blood pressure phenotypic variance has been elucidated. This phenomenon could be partially due to epistatic interactions. Our aim was to identify epistatic interaction(s) associated with blood pressure levels in a pre-planned two-phase approach.

Methods and Results

In a discovery cohort composed of 3,600 French individuals, we found rs6046A allele in F7 associated with decreased blood pressure levels (P≤3.7×10−3) and rs5355T allele in SELE associated with decreased diastolic blood pressure levels (P = 5×10−3). Both variants interacted in order to influence blood pressure levels (P≤0.048). This interaction was replicated with systolic blood pressure in 4,620 additional European individuals (P = 0.03). Similarly, in this replication cohort, rs6046A was associated with decreased blood pressure levels (P≤8.5×10−4). Furthermore, in peripheral blood mononuclear cells of a subsample of 90 supposed healthy individuals, we found rs6046A positively associated with NAMPT mRNA levels (P≤9.1×10−5), suggesting an eventual involvement of NAMPT expression in blood pressure regulation. Confirming this hypothesis, further transcriptomic analyses showed that increased NAMPT mRNA levels were positively correlated with ICAM1, SELL, FPR1, DEFA1-3, and LL-37 genes expression (P≤5×10−3). The last two mRNA levels were positively associated with systolic blood pressure levels (P≤0.01) and explained 4% of its phenotypic variation.

Conclusion

These findings reveal the importance of epistatic interactions in blood pressure genetics and give new insights for the role of inflammation in its complex regulation.

Introduction

Blood pressure (BP) is a heritable trait with estimates indicating that 30–70% of its variance is attributed to genetics [1], [2]. In family studies its heritability varies, according to measurement processes, from ≈31% [single-measure of systolic blood pressure (SBP) and diastolic blood pressure (DBP)], to ≈57% (long-term average of SBP and DBP phenotypes) and to ≈68% (24-hour profile of SBP and DBP) [3]. Both BP and essential hypertension (HTN) are considered polygenic traits [4]. Inflammation, blood coagulation cascade, cellular adhesion molecules and lipid metabolism appear to have significant roles [5].

The largest Genome-wide association study (GWAS) on BP including ≈200,000 individuals [6] reported 29 loci to be associated with SBP, DBP and/or essential HTN [6]. However, their genetic risk score explained only 0.9% of BP phenotypic variation [6], this representing the so-called ‘dark matter’ of genetic risk [7]. Despite the very large sample-size studies used for gene discovery, many common variants with small effects on BP may remain unidentified [8]. A large ‘hidden heritability’ of unknown nature may be explained by rare variants, structural large variants, epistatic [gene*gene (G*G)] and gene*environment (G*E) interactions [7]. We pointed out that epistatic interactions might also play an important role in discovering new genes [7]. This statement has been extensively reviewed in the last years and multi-locus methods have been developed to detect such interactions [7].

Epistatic interactions have been documented for susceptibility to cancer [9], morphology [10] and autoimmune conditions [11]. However, to date they have not been extensively studied in BP regulation. We hypothesize that the research of epistatic interactions among candidate single nucleotide polymorphisms (SNPs) represents a challenge in the investigation of disease-risk variants, as their application to high-dimensional genome-wide data exhaustively including all SNPs combinations is not yet feasible [7]. In previous candidate gene studies, we showed interesting results concerning the identification of BP candidate SNPs [12][16]. However, these studies were conducted in limited-sized populations.

Therefore, in the present study, we investigated BP epistasis mechanisms in a pre-planned two-phase approach gathering 8,220 European individuals. The effect of 10 candidate SNPs and then, G*G interactions between significant SNPs were assessed in a discovery population of 3,600 individuals. Highlighted epistases were replicated in 4,620 additional European individuals. We further searched for association(s) with 10 inflammation-related genes in peripheral blood mononuclear cells (PBMCs) (LL37, DEFA1-3, FPR1, ICAM1, SELL, SELP, NAMPT (visfatin), LEP, TNF and IL-6) [17] of a subsample of 90 supposed healthy individuals. Finally, we sought to propose a possible molecular mechanism of action.

Materials and Methods

Ethics Statement

All participants involved in the present study were recruited in accordance with the latest version of the Declaration of Helsinki for Ethical Principles for Medical Research Involving Human Subjects and gave written informed consent. Genetic studies protocols were approved by the local ethics committees for the protection of subjects for biomedical research: 1) the Comité Consultatif de Protection des Personnes dans la Recherche Biomédicale de Lorraine, Nancy, France, for populations recruited in the Center of Preventive Medicine. 2) the Comité d’Ethique du Centre Hospitalier Universitaire de Cochin, Paris, France, for ERA population. 3) The ethic committee of Belfast, Ireland, for population recruited in Ireland.

Study Populations

Discovery population.

A sample of 2,971 unrelated individuals was recruited during free medical check-ups at the Center of Preventive Medicine of Vandœuvre-lès-Nancy in the East of France. They were Caucasians, born in France for three generations and their clinical and biological data were collected at baseline before any eventual drug prescription following consultation. They were selected on the basis of the following criteria: (1) no antihypertensive drug therapy at recruitment; (2) complete clinical and genotypic data available; (3) and BP levels ranging from normotensive to stage 2 HTN (for hypertensive individuals, data were gathered before the prescription of any medication).

As our purpose was to assess BP as a continuous trait, and in order to have a proper inter-individual variability, we included individuals from the ERA cohort (Evolution de la Rigidité Artérielle) in the discovery population. ERA participants were selected from a Parisian cohort that had a health check-up at the ‘Investigations Préventives et Cliniques’ center. The details of this study have been previously presented [18]. Six hundred and twenty nine individuals randomly selected in ERA were incorporated in the discovery population. As no significant differences between minor allele frequencies (MAFs) of the investigated genetic variants and BP levels in these samples were found, we regrouped both discovery samples in order to perform our statistical analyses.

The corresponding samples were part of a human sample storage platform: the Biological Resources Bank ‘Interactions Gène-Environnement en Physiopathologie CardioVacsulaire’ (BRC IGE-PCV) in Nancy, East of France.

Replication population.

We used a non-overlapping sample extracted from the BRC IGE-PCV. Altogether, 4,620 individuals with (1) no antihypertensive drug therapy at recruitment; (2) complete clinical and genotypic data for rs5355C>T in SELE and rs6046G>A in F7 were available; (3) BP levels ranging from normotensive to stage 2 HTN; and (4) only European origins were analyzed (Ireland, French). Stage 3 HTN patients were also excluded in the replication population as they were treated with antihypertensive medication.

Clinical and Biological Data Collection

SBP and DBP were measured under constant temperature (19°C–21°C) and standardized conditions (supine position) using a manual sphygmomanometer (Colonne à mercure, Mercurius) by expert nurses [18]. The recorded values were the means of 3 readings with 20 min intervals. An adjustable BP cuff was used to correct errors due to variations in arm circumference [19]. HTN was defined as SBP≥140 mmHg or DBP≥90 mmHg as recommended in the Seventh Report of the Joint National Committee on the prevention, detection, evaluation, and treatment of high BP [20]. All individuals underwent complete medical examination including anthropometric and biochemical measurements collected with standardized methods as described elsewhere [17].

Genotyping Assays

We selected rs1799752Ins>del in ACE, rs5882A>G in CETP, rs1801133C>T in MTHFR rs662A>G in PON1 and rs1800629G>A in TNF from the “Cardio-Vascular Disease 35” assay, a multilocus genotyping assay developed in collaboration with Roche Molecular Systems [12]. These genetic variants were candidate markers for cardiovascular disease (CVD) risk factors, specifically involved in the predisposition to essential HTN (rs1799752Ins>del in ACE), in the development of atherosclerotic plaques and in the progression of atherosclerosis (rs5882A>G in CETP, rs1801133C>T in MTHFR rs662A>G in PON1 and rs1800629G>A in TNF) [12]. In addition, rs5355C>T in SELE [13], [21], rs1800790G>A in FGB [14], rs6046G>A in F7 [15], rs328C>G in LPL [16], [22] were chose based on our previous published studies that found these SNPs associated with BP levels and/or HTN in European populations [12][16], [21], [22]. Finally, rs3025058T>Ins in MMP3 was selected from an internal investigation showing a link between this genetic variant and BP levels.

A summary of investigated genetic variants (nearby gene, location, type and mutation) was shown in Supplementary Data S1.

Genomic DNA was extracted from peripheral blood samples using the salting out method [23]. Genotyping was performed using two methods in the discovery population. 1) A multilocus assay with an immobilized probe approach designed by Roche Molecular Systems, Pleasanton, California, USA [24]. After PCR amplification using pooled biotinylated primers and hybridization to sequence-specific oligonucleotide probes, two independent observers using proprietary Roche Molecular Systems image processing software performed genotype assignments. Among 2,971 individuals, discordant results (<3% of all scoring) were resolved by a third observer and if necessary, by a joint reading. 2) Evidence Investigator™ biochip designed by Randox Laboratories, Antrim, UK was used to genotype ERA participants. This genotyping assay is based on a combination of probe hybridization, ligation, PCR amplification and microarray hybridization. This unique design permits high assay multiplexing and ready discrimination between genotypes. For the validation of genotyping results, blinded replication analysis was performed on 50 common samples. Both genotyping methods gave matched results at 99% (data available on demand).

Replication population.

Only rs5355C>T in SELE and rs6046G>A in F7 were genotyped. Among all individuals; 2,059 were genotyped by Kbioscience company using the competitive allele specific PCR (KASP) chemistry coupled with a FRET-based genotyping system (http://www.kbioscience.co.uk/reagents/KASP/KASP.html). The remaining 2,561 individuals were genotyped by Roche multilocus assay as described previously.

PolyPhen Analysis of Nonsynonymous SNPs

The prediction of nonsynonymous SNPs possible impacts on their protein structures was performed using PolyPhen [25].

Peripheral Blood Mononuclear Cells Collection

Freshly drawn peripheral venous blood (10 ml) was collected into tubes containing EDTA (Vacutainer, Becton Dickinson) under fasting conditions. PBMCs were isolated by centrifuging on a density gradient of Ficoll as described previously and stored at -80°C until RNA extraction [26]. PBMCs bank with high recovery of lymphocytes (97.5%) was constituted as described elsewhere [26].

RNA Extraction and qRT-PCR Analysis

Using a microarray analysis [5]; we selected the top 10 inflammation-related genes (from a total of 182 genes) having a higher expression in PBMCs of hypertensive individuals when compared with normotensives. Total RNA was isolated from PBMCs by an automated isolation procedure (MagNa Pure LC instrument). mRNA quality and stability were carefully tested [26] and reverse transcribed as previously described [26]. Quantitative real-time PCR (qRT-PCR) was performed using LightCycler instrument (Roche Diagnostics, Mannheim, Germany) with Master Plus SYBR Green I kit for all gene transcripts. SELE and F7 were not quantified, as they were not expressed. Specific primers were designed using Primer Premier 3.0 software (Supplementary Data S1). All experiments were carried out in duplicates in a total reaction volume of 20 µl containing 0.5 mM of each specific primer. Negative and internal controls were included. All mRNA levels were normalized to the mRNA levels of POL2RA. The specificity of all PCR products was further verified by electrophoresis on 10% polyacrylamide gel (data available on demand). The clinical characteristics of the studied subsample were presented in Supplementary Data S1.

Statistical Analyses

Statistical analyses were performed using the SPSS® statistical software version 19.0 (SPSS, Inc, Chicago, Illinois). Polymorphisms with MAF deviating from Hardy-Weinberg equilibrium (HWE) were excluded from individual analyses. In order to determine the effect of the 10 selected genetic variants on SBP and DBP assuming additive models using the common wild type as the reference group; age, gender and body mass index (BMI)-adjusted linear regressions were performed for individual association analyses. Due to multiple testing, the significance level was set at P≤5×10−3 in the discovery and replication populations.

G*G interactions.

Two-locus additive epistasis was defined as significant statistical interaction between two SNPs [27] and was determined when significant interaction existed on a linear additive model adjusted for age, gender and BMI. Epistatic interactions were only tested between individually significant associated SNPs. In both populations, Bonferroni correction for multiple testing was applied. The significance level was set at P≤0.05.

SNP-mRNA association analysis.

Linear regressions were performed to assess the effect of SNPs previously associated to SBP and/or DBP in the first stage of our analyses on mRNA levels. In epistatic conditions, an interaction term was introduced in the model. The significance level was set at P≤5×10−3 due to multiple testing.

Pearson’s correlation analyses.

Pearson’s correlation was used to test the correlation between all genes expression and NAMPT levels (values log-transformed). The significance level was set at P≤5×10−3 due to multiple testing.

Linear regression analysis between genes expression and BP levels.

Linear regression models were used to further assess the association of SELL, FPR1, ICAM1, DEFA1-3 and LL-37 with mRNA levels with SBP and DBP after adjustment for age and gender. The significance level was set at P≤0.01 due to multiple testing.

URLs

Primer Premier 3.0 is available at: http://frodo.wi.mit.edu/primer3/.

Polyphen is available at: http://genetics.bwh.harvard.edu/pph2/.

Results

Table 1 presents the clinical characteristics of the studied populations. According to the Seventh Report of the Joint National Committee [28], 21.8% of participants had normal BP, 32% were pre-hypertensive and 46.2% had HTN stage 1 and 2 in the discovery population (Table 1). In the replication set, 34% had normal BP, 39.8% were pre-hypertensive and 26.2% were stage 1 and 2 hypertensive (Table 1). A higher frequency of HTN was observed in the discovery compared to the replication population (46.2% vs. 26.2% respectively), which is partly due to the presence of older individuals in the discovery set.

Table 2 shows genetic variants associated with BP traits. We found two SNPs, rs5355C>T in SELE and rs6046G>A in F7 showing associations with SBP and/or DBP respectively in the discovery population (Pdiscovery≤5×10−3, Table 2). rs5355T allele in SELE was associated with decreased DBP levels (P = 5×10−3, β = −0.04, Table 2), whereas rs6046A allele in F7 was associated with decreased SBP and DBP levels respectively (P = 3.7×10−3 and P = 8.2×10−4 respectively, Table 2). Both SNPs are nonsynonymous, introducing amino acid substitutions (Leu575Phe and Arg353Gln respectively). According to Polyphen, they were predicted to have a null effect on their corresponding protein structures. full individual association results with BP in the discovery and the replication population were shown in Supplementary Data S1.

In order to examine whether rs5355C>T in SELE and rs6046G>A in F7 may also indirectly influence BP levels, we tested their G*G interaction (Table 3). Both SNPs interacted in order to influence SBP and DBP in the discovery population (P = 0.048 and P = 0.047 respectively, Table 3A). Table 3A shows BP variations according to rs5355T allele in SELE and rs6046G/A genotypes in F7 when compared to rs5355C allele in SELE. We found that individuals carrying rs5355T allele in SELE and rs6046GG in F7 had 6.5 mmHg and 8 mmHg decrease in SBP and DBP respectively when compared with carriers of rs5355C allele in SELE and rs6046GG genotype in F7 (Table 3A). In contrast, individuals carrying rs5355T allele in SELE and one minor allele of rs6046G>A (rs6046GA) had 6.1 mmHg and 1.2 mmHg increase in SBP and DBP respectively when compared with carriers of rs5355C allele in SELE, rs6046GA genotype in F7 (Table 3A). Furthermore, carriers of rs5355T allele in SELE and two minor alleles of rs6046G>A (rs6046AA) had higher BP levels when compared with those carrying rs5355C allele in SELE and rs6046AA genotype in F7 (5.1 mmHg and 3.8 mmHg increase in SBP and DBP respectively) (Table 3A). We concluded that rs6046A might invert the BP-lowering effect of rs5355T on DBP and SBP.

thumbnail
Table 3. Blood pressure variations according to rs5355T allele in SELE and rs6046G/A genotypes in F7 when compared to rs5355C allele in SELE.

https://doi.org/10.1371/journal.pone.0040777.t003

In the replication population, rs6046G>A in F7 was also associated with decreased SBP (Preplication = 8.45×10−4 and Pmeta = 2.03×10−4) and DBP (Preplication = 2.58×10−7and Pmeta = 9.16×10−4). In contrast, rs5355C>T was not associated with DBP (Preplication = 0.86). Most importantly, we found rs5355C>T in SELE and rs6046G>A in F7 interacting in order to influence the SBP (Preplication = 0.03, Table 3B). Similar SBP variations according to rs5355T allele in SELE and rs6046G/A genotypes in F7 were successfully found (Table 3B).

In conclusion, rs5355C>T in SELE interacted with rs6046G>A in F7 in order to influence SBP in a total of 8,220 European individuals.

We investigated the eventual relation(s) between the epistatic interaction and the inflammation-related genes in a PBMCs model. rs5355C>T in SELE was not associated with any of the investigated transcripts. In contrast, rs6046A allele in F7 was positively associated with NAMPT mRNA levels in both models (individual association and epistatic interaction models) (P = 9.2×10−5, β = 0.489 and P = 1.1×10−5, β = 0.552 respectively).

Increased NAMPT mRNA levels were positively correlated with ICAM1 (P<1×10−4 and β = 0.576, Table 4), SELL (P = 5×10−3 and r = 0.308, Table 4), FPR1 (P = 2×10−4 and r = 0.394, Table 4), LL-37 (P = 4×10−3 and r = 0.452, Table 4) and DEFA1-3 (P = 5×10−3 and r = 0.28, Table 4) genes expression. In addition ICAM1, SELL, FPR1 and DEFA1-3 expressions were also correlated (P≤5×10−3, Table 4). Only DEFA1-3 and LL-37 mRNA levels were positively associated with SBP. We found that both mRNAs explained 4% of SBP phenotypic variation (P = 3×10−3, β = 0.04 and P = 0.01, β = 0.03 respectively).

thumbnail
Table 4. Pearson’s correlations between NAMPT, ICAM1, SELL, FPR1, DEFA1-3 and LL-37 genes expression.

https://doi.org/10.1371/journal.pone.0040777.t004

Discussion

In the current study, we found rs6046A allele in F7 associated with decreased BP levels (P≤3.7×10−3 and Pmeta≤2.03×10−4). In the discovery cohort, rs5355T allele in SELE was also associated with decreased DBP (P = 5×10−3).

rs6046G>A in F7 was shown to be associated with increased F7 plasmatic levels [15]. More interestingly, this SNP was reported to have a role in protection against myocardial infarction in two different studies performed on Italian populations [29], [30]. rs5355C>T in SELE is located in chr.1q, a genomic region linked to BP related phenotypes in two independent linkage studies [31], [32]. These findings were supported by observation of mouse and rat BP-related quantitative trait loci in regions homologous to the human 1q chromosomal locus [33].

Herein, we showed that in a total of 8,220 European individuals, rs5355C>T in SELE interacted with rs6046G>A in F7 and the latter SNP in order to alter SBP (Pdiscovery = 0.047 and Preplication = 0.03 respectively, Table 3). The above interaction was differently associated with SBP variations according to rs6046G>A genotypes (Table 3). In fact, epistatic interactions are phenomena where the effect of a gene is modified by another one [34], [35], thus although rs6046A allele in F7 was associated with decreased BP levels, it interacted with rs5355T allele in SELE in order to influence SBP levels, resulting an increase in SBP mean values.

The non-replication of the association between rs5355C>T in SELE and DBP is not surprising as insignificant interaction effect on DBP between these two variants was found in the replication cohort. It is important to point out that, it has been postulated that epistatic interactions may identify genetic markers that are not captured by individual marker analysis and/or revealed by the combinatory effect of loci in other pathways [34], [35]. This postulate might explain why the two variants investigated here (and many others) were not reported among the top GWAS SNPs.

It was proven that blood coagulation factors enhance the inflammatory response leading to endothelial dysfunction accounting in part, for the vascular complications occurring in CVDs and their risk factors [36]. Thus, we searched for eventual relation(s) between the epistatic interaction and the inflammation-related genes in a PBMCs model. Numerous studies have revealed the importance of studying PBMCs in a strategy targeting the metabolic pathways of cardiovascular risk factors, such as HTN [5], [37], [38]. It has been recently demonstrated that PBMCs mRNA expression closely mimic the in vivo state and generate more physiologically relevant data concerning many health related traits [39]. The role of multiple metabolic pathways in HTN makes the study of PBMCs transcriptome important for the possible developing of diagnostic and prognostic tests [40], we assessed associations between rs5355C>T in SELE and rs6046G>A in F7 with the inflammation-related genes expression. rs6046A allele in F7 was associated with increased NAMPT mRNA levels (P≤9.2×10−5). NAMPT levels were also positively correlated with ICAM1, SELL, FPR1, DEFA1-3 and LL-37 genes expression (P≤5×10−3, Table 4). In addition ICAM1, SELL, FPR1 and DEFA1-3 expressions were also correlated (P≤5×10−3, Table 4). Only DEFA1-3 and LL-37 expressions were associated with SBP (P = 3×10−3 and P = 0.01 respectively) and explained 4% of its variation. Therefore, we suggest that the associations of DEFA1-3 and LL-37 mRNAs and SBP reflect the epistatic interaction and not the main effect of rs6046G>A in F7. Visfatin is a multifunctional protein that has been reported to be involved in innate immune system [41] and several other biological processes such as the cardiovascular system [42]. However, its role in BP was unclear. Supporting our results, three different in vitro studies have demonstrated that visfatin induced an endothelial dysfunction by increasing inflammatory and adhesion molecules expression such as ICAM1 [43][45]. In addition, in a previous study we have reported that gene expression of an antimicrobial peptide LL-37 in PBMCs was associated with altered BP levels [46]. The above findings support our epistatic and the in vivo results revealing an indirect link between NAMPT gene expression and BP through the expression of adhesion and innate immune system molecules.

thumbnail
Figure 1. Summary of the study and hypothesis for rs5355C>T in SELE and rs6046G>A in F7 interaction.

rs6046A allele in F7 was associated with decreased BP levels. rs5355C>T in SELE and rs6046G>A in F7 interacted in order to alter SBP levels, rs6046A inverted the BP-lowering effect of rs5355T. rs6046A allele in F7 was positively associated with increased NAMPT gene expression. NAMPT levels were positively correlated with ICAM1, SELL, FPR1 and DEFA1-3 genes expression. Only DEFA1-3 and LL-37 expressions were correlated and associated with SBP levels and explained 4% of its variation.

https://doi.org/10.1371/journal.pone.0040777.g001

Strengths and Limitations

The genetics of BP has never been easy [47]. For many years, it has been dominated by the stark contrast between its high heritability and the frustrating reality that no clearly reproducible and functional genetic variant could be discovered [3], with epistatic interactions accepted as cause of discrepancies across the studies.

The current study shows the first replicated epistatic interaction in the BP genetics field. This interaction between a coagulation factor gene (F7) and an adhesion molecule gene (SELE) is putatively functional through its link with five inflammation-related gene expression. Going in the same direction; it has been demonstrated that some blood coagulation factors can induce an endothelial dysfunction (SELE is a marker of endothelial dysfunction) through an inflammatory response accounting for the vascular complications occurring in CVDs and their risk factors [36]. Similarly, NAMPT expression was shown to increase the expression of inflammatory and adhesion molecules such as ICAM1 [43][45]. Based on these findings we speculate a biological plausibility for the reported epistatic interaction. Supporting this statement, Tomaszewski et al [48] showed that a genetic risk score including SNPs from the fibroblast growth factor signaling pathway was able to explain a larger proportion of variation in HTN as compared with a genetic risk score including a similar number of SNPs based on the previous top SNPs from the GWAS [48]. This suggests that biological knowledge might support the reported epistatic interactions.

However, our study also had several limitations. Whereas focusing on European populations, our findings cannot be generalized to other ethnic groups. We also were unable to further investigate SNPs associations with plasmatic levels of the inflammation-related genes as the availability of biological materials was unfortunately limiting. Similarly, further studies looking at the SNPs association with SELE and F7 expression in endothelial cells, would be of great value.

Conclusion

Our findings are summarized in Figure 1. In European populations, we confirmed that rs6046A in F7 is associated with decreased BP. Furthermore, we found that rs5355C>T in SELE and rs6046G>A in F7 interacted in order to alter SBP levels. In addition rs6046A allele in F7 was positively associated with increased NAMPT gene expression, which was linked with BP through inflammatory mechanisms via the expression of adhesion and innate immune system molecules.

Perspectives

Even if additional investigations are needed, the present study highlighted the importance of taking into account candidate genes, GWAS and epistatic interactions in order to in deep investigate BP genetic regulation. One must also consider the functionality of relationships and G*E interactions that might be at the origin of the low until now predictive values of results in HTN. This integrative approach could better explain the missing heritability of this complex trait.

Author Contributions

Conceived and designed the experiments: SVS SES. Performed the experiments: SES HAM CM. Analyzed the data: NCN MGS SES SVS. Contributed reagents/materials/analysis tools: PF JVL. Wrote the paper: SES SVS. Biological Resources Center (BRC): “Interactions Gène-Environnement en Physiopathologie Cardio-Vasculaire” (IGE-PCV) populations: SVS. ERA population: AB.

References

  1. 1. El Shamieh S, Herbeth B, Azimi-Nezhad M, Benachour H, Masson C, et al. (2012) Human formyl peptide receptor 1 C32T SNP interacts with age and is associated with blood pressure levels. Clin Chim Acta 413(1–2): 34–38.S. El ShamiehB. HerbethM. Azimi-NezhadH. BenachourC. Masson2012Human formyl peptide receptor 1 C32T SNP interacts with age and is associated with blood pressure levels.Clin Chim Acta413(1–2)3438
  2. 2. Levy D, Ehret GB, Rice K, Verwoert GC, Launer LJ, et al. (2009) Genome-wide association study of blood pressure and hypertension. Nat Genet 41(6): 677–687.D. LevyGB EhretK. RiceGC VerwoertLJ Launer2009Genome-wide association study of blood pressure and hypertension.Nat Genet41(6)677687
  3. 3. Ehret GB (2010) Genome-wide association studies: contribution of genomics to understanding blood pressure and essential hypertension. Curr Hypertens Rep 12: 17–25.GB Ehret2010Genome-wide association studies: contribution of genomics to understanding blood pressure and essential hypertension.Curr Hypertens Rep121725
  4. 4. Levy D, DeStefano AL, Larson MG, O’Donnell CJ, Lifton RP, et al. (2000) Evidence for a gene influencing blood pressure on chromosome 17. Genome scan linkage results for longitudinal blood pressure phenotypes in subjects from the framingham heart study. Hypertension 36: 477–483.D. LevyAL DeStefanoMG LarsonCJ O’DonnellRP Lifton2000Evidence for a gene influencing blood pressure on chromosome 17. Genome scan linkage results for longitudinal blood pressure phenotypes in subjects from the framingham heart study.Hypertension36477483
  5. 5. Visvikis-Siest S, Marteau JB, Samara A, Berrahmoune H, Marie B, et al. (2007) Peripheral blood mononuclear cells (PBMCs): a possible model for studying cardiovascular biology systems. Clin Chem Lab Med 45: 1154–1168.S. Visvikis-SiestJB MarteauA. SamaraH. BerrahmouneB. Marie2007Peripheral blood mononuclear cells (PBMCs): a possible model for studying cardiovascular biology systems.Clin Chem Lab Med4511541168
  6. 6. Ehret GB, Munroe PB, Rice KM, Bochud M, Johnson AD, et al. (2011) Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk. Nature 478: 103–109.GB EhretPB MunroeKM RiceM. BochudAD Johnson2011Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk.Nature478103109
  7. 7. Ndiaye NC, Azimi NM, El Shamieh S, Stathopoulou MG, Visvikis-Siest S (2011) Cardiovascular diseases and genome-wide association studies. Clin Chim Acta 412: 1697–1701.NC NdiayeNM AzimiS. El ShamiehMG StathopoulouS. Visvikis-Siest2011Cardiovascular diseases and genome-wide association studies.Clin Chim Acta41216971701
  8. 8. Taal HR, Verwoert GC, Demirkan A, Janssens AC, Rice K, et al. (2011) Genome-Wide Profiling of Blood Pressure in Adults and Children. Hypertension. HYPERTENSIONAHA.111.179481 p.HR TaalGC VerwoertA. DemirkanAC JanssensK. Rice2011Genome-Wide Profiling of Blood Pressure in Adults and Children.HypertensionHYPERTENSIONAHA.111.179481
  9. 9. Fijneman RJ, de Vries SS, Jansen RC, Demant P (1996) Complex interactions of new quantitative trait loci, Sluc1, Sluc2, Sluc3, and Sluc4, that influence the susceptibility to lung cancer in the mouse. Nat Genet 14: 465–467.RJ FijnemanSS de VriesRC JansenP. Demant1996Complex interactions of new quantitative trait loci, Sluc1, Sluc2, Sluc3, and Sluc4, that influence the susceptibility to lung cancer in the mouse.Nat Genet14465467
  10. 10. Leamy LJ, Routman EJ, Cheverud JM (2002) An epistatic genetic basis for fluctuating asymmetry of mandible size in mice. Evolution 56: 642–653.LJ LeamyEJ RoutmanJM Cheverud2002An epistatic genetic basis for fluctuating asymmetry of mandible size in mice.Evolution56642653
  11. 11. Wandstrat A, Wakeland E (2001) The genetics of complex autoimmune diseases: non-MHC susceptibility genes. Nat Immunol 2: 802–809.A. WandstratE. Wakeland2001The genetics of complex autoimmune diseases: non-MHC susceptibility genes.Nat Immunol2802809
  12. 12. Cheng S, Grow MA, Pallaud C, Klitz W, Erlich HA, et al. (1999) A multilocus genotyping assay for candidate markers of cardiovascular disease risk. Genome Res 9: 936–949.S. ChengMA GrowC. PallaudW. KlitzHA Erlich1999A multilocus genotyping assay for candidate markers of cardiovascular disease risk.Genome Res9936949
  13. 13. Marteau JB, Sass C, Pfister M, Lambert D, Noyer-Weidner M, et al. (2004) The Leu554Phe polymorphism in the E-selectin gene is associated with blood pressure in overweight people. J Hypertens 22: 305–311.JB MarteauC. SassM. PfisterD. LambertM. Noyer-Weidner2004The Leu554Phe polymorphism in the E-selectin gene is associated with blood pressure in overweight people.J Hypertens22305311
  14. 14. Maumus S, Marie B, Vincent-Viry M, Siest G, Visvikis-Siest S (2007) Analysis of the effect of multiple genetic variants of cardiovascular disease risk on insulin concentration variability in healthy adults of the STANISLAS cohort. The role of FGB-455 G/A polymorphism. Atherosclerosis 191: 369–376.S. MaumusB. MarieM. Vincent-ViryG. SiestS. Visvikis-Siest2007Analysis of the effect of multiple genetic variants of cardiovascular disease risk on insulin concentration variability in healthy adults of the STANISLAS cohort. The role of FGB-455 G/A polymorphism.Atherosclerosis191369376
  15. 15. Sass C, Blanquart C, Morange PE, Pfister M, Visvikis-Siest S (2004) Association between factor VII polymorphisms and blood pressure: the Stanislas Cohort. Hypertension 44: 674–680.C. SassC. BlanquartPE MorangeM. PfisterS. Visvikis-Siest2004Association between factor VII polymorphisms and blood pressure: the Stanislas Cohort.Hypertension44674680
  16. 16. Sass C, Cheng S, Siest G, Visvikis S (2004) Genetic influences on blood pressure within the Stanislas Cohort. J Hypertens 22: 297–304.C. SassS. ChengG. SiestS. Visvikis2004Genetic influences on blood pressure within the Stanislas Cohort.J Hypertens22297304
  17. 17. Visvikis-Siest S, Siest G (2008) The STANISLAS Cohort: a 10-year follow-up of supposed healthy families. Gene-environment interactions, reference values and evaluation of biomarkers in prevention of cardiovascular diseases. Clin Chem Lab Med 46: 733–747.S. Visvikis-SiestG. Siest2008The STANISLAS Cohort: a 10-year follow-up of supposed healthy families. Gene-environment interactions, reference values and evaluation of biomarkers in prevention of cardiovascular diseases.Clin Chem Lab Med46733747
  18. 18. Benetos A, Adamopoulos C, Bureau JM, Temmar M, Labat C, et al. (2002) Determinants of accelerated progression of arterial stiffness in normotensive subjects and in treated hypertensive subjects over a 6-year period. Circulation 105: 1202–1207.A. BenetosC. AdamopoulosJM BureauM. TemmarC. Labat2002Determinants of accelerated progression of arterial stiffness in normotensive subjects and in treated hypertensive subjects over a 6-year period.Circulation10512021207
  19. 19. Dasberg H, Blondheim SH, Sadovsky E (1962) An adjustable blood pressure cuff to correct errors due to variations in arm circumference. Br Heart J 24: 214–220.H. DasbergSH BlondheimE. Sadovsky1962An adjustable blood pressure cuff to correct errors due to variations in arm circumference.Br Heart J24214220
  20. 20. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, et al. (2003) Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 42: 1206–1252.AV ChobanianGL BakrisHR BlackWC CushmanLA Green2003Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure.Hypertension4212061252
  21. 21. Sass C, Pallaud C, Zannad F, Visvikis S (2000) Relationship between E-selectin L/F554 polymorphism and blood pressure in the Stanislas cohort. Hum Genet 107: 58–61.C. SassC. PallaudF. ZannadS. Visvikis2000Relationship between E-selectin L/F554 polymorphism and blood pressure in the Stanislas cohort.Hum Genet1075861
  22. 22. Sass C, Herbeth B, Siest G, Visvikis S (2000) Lipoprotein lipase (C/G)447 polymorphism and blood pressure in the Stanislas Cohort. J Hypertens 18: 1775–1781.C. SassB. HerbethG. SiestS. Visvikis2000Lipoprotein lipase (C/G)447 polymorphism and blood pressure in the Stanislas Cohort.J Hypertens1817751781
  23. 23. Miller SA, Dykes DD, Polesky HF (1988) A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 16: 1215.SA MillerDD DykesHF Polesky1988A simple salting out procedure for extracting DNA from human nucleated cells.Nucleic Acids Res161215
  24. 24. Zee RY, Cook NR, Cheng S, Erlich HA, Lindpaintner K, et al. (2006) Polymorphism in the beta2-adrenergic receptor and lipoprotein lipase genes as risk determinants for idiopathic venous thromboembolism: a multilocus, population-based, prospective genetic analysis. Circulation 113: 2193–2200.RY ZeeNR CookS. ChengHA ErlichK. Lindpaintner2006Polymorphism in the beta2-adrenergic receptor and lipoprotein lipase genes as risk determinants for idiopathic venous thromboembolism: a multilocus, population-based, prospective genetic analysis.Circulation11321932200
  25. 25. Ramensky V, Bork P, Sunyaev S (2002) Human non-synonymous SNPs: server and survey. Nucleic Acids Res 30: 3894–3900.V. RamenskyP. BorkS. Sunyaev2002Human non-synonymous SNPs: server and survey.Nucleic Acids Res3038943900
  26. 26. Marteau JB, Mohr S, Pfister M, Visvikis-Siest S (2005) Collection and storage of human blood cells for mRNA expression profiling: a 15-month stability study. Clin Chem 51: 1250–1252.JB MarteauS. MohrM. PfisterS. Visvikis-Siest2005Collection and storage of human blood cells for mRNA expression profiling: a 15-month stability study.Clin Chem5112501252
  27. 27. VanderWeele TJ (2010) Epistatic interactions. Stat Appl Genet Mol Biol 9. TJ VanderWeele2010Epistatic interactions.Stat Appl Genet Mol Biol 9
  28. 28. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, et al. (2003) Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 42: 1206–1252.AV ChobanianGL BakrisHR BlackWC CushmanLA Green2003Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure.Hypertension4212061252
  29. 29. Girelli D, Russo C, Ferraresi P, Olivieri O, Pinotti M, et al. (2000) Polymorphisms in the factor VII gene and the risk of myocardial infarction in patients with coronary artery disease. N Engl J Med 343: 774–780.D. GirelliC. RussoP. FerraresiO. OlivieriM. Pinotti2000Polymorphisms in the factor VII gene and the risk of myocardial infarction in patients with coronary artery disease.N Engl J Med343774780
  30. 30. Iacoviello L, Di Castelnuovo A, De Knijff P, D’Orazio A, Amore C, et al. (1998) Polymorphisms in the coagulation factor VII gene and the risk of myocardial infarction. N Engl J Med 338: 79–85.L. IacovielloA. Di CastelnuovoP. De KnijffA. D’OrazioC. Amore1998Polymorphisms in the coagulation factor VII gene and the risk of myocardial infarction.N Engl J Med3387985
  31. 31. Ehret GB, O’Connor AA, Weder A, Cooper RS, Chakravarti A (2009) Follow-up of a major linkage peak on chromosome 1 reveals suggestive QTLs associated with essential hypertension: GenNet study. Eur J Hum Genet 17: 1650–1657.GB EhretAA O’ConnorA. WederRS CooperA. Chakravarti2009Follow-up of a major linkage peak on chromosome 1 reveals suggestive QTLs associated with essential hypertension: GenNet study.Eur J Hum Genet1716501657
  32. 32. Hunt SC, Ellison RC, Atwood LD, Pankow JS, Province MA, et al. (2002) Genome scans for blood pressure and hypertension: the National Heart, Lung, and Blood Institute Family Heart Study. Hypertension 40: 1–6.SC HuntRC EllisonLD AtwoodJS PankowMA Province2002Genome scans for blood pressure and hypertension: the National Heart, Lung, and Blood Institute Family Heart Study.Hypertension4016
  33. 33. Faruque MU, Chen G, Doumatey A, Huang H, Zhou J, et al. (2011) Association of ATP1B1, RGS5 and SELE polymorphisms with hypertension and blood pressure in African-Americans. J Hypertens 29: 1906–1912.MU FaruqueG. ChenA. DoumateyH. HuangJ. Zhou2011Association of ATP1B1, RGS5 and SELE polymorphisms with hypertension and blood pressure in African-Americans.J Hypertens2919061912
  34. 34. Harrap SB (2009) Blood pressure genetics: time to focus. J Am Soc Hypertens 3: 231–237.SB Harrap2009Blood pressure genetics: time to focus.J Am Soc Hypertens3231237
  35. 35. Misono M, Maeda S, Iemitsu M, Nakata Y, Otsuki T, et al. (2009) Combination of polymorphisms in the beta2-adrenergic receptor and nitric oxide synthase 3 genes increases the risk for hypertension. J Hypertens 27: 1377–1383.M. MisonoS. MaedaM. IemitsuY. NakataT. Otsuki2009Combination of polymorphisms in the beta2-adrenergic receptor and nitric oxide synthase 3 genes increases the risk for hypertension.J Hypertens2713771383
  36. 36. Hezi-Yamit A, Wong PW, Bien-Ly N, Komuves LG, Prasad KS, et al. (2005) Synergistic induction of tissue factor by coagulation factor Xa and TNF: evidence for involvement of negative regulatory signaling cascades. Proc Natl Acad Sci U S A 102: 12077–12082.A. Hezi-YamitPW WongN. Bien-LyLG KomuvesKS Prasad2005Synergistic induction of tissue factor by coagulation factor Xa and TNF: evidence for involvement of negative regulatory signaling cascades.Proc Natl Acad Sci U S A1021207712082
  37. 37. Kontaraki JE, Marketou ME, Zacharis EA, Parthenakis FI, Vardas PE (2011) Early cardiac gene transcript levels in peripheral blood mononuclear cells in patients with untreated essential hypertension. J Hypertens 29: 791–797.JE KontarakiME MarketouEA ZacharisFI ParthenakisPE Vardas2011Early cardiac gene transcript levels in peripheral blood mononuclear cells in patients with untreated essential hypertension.J Hypertens29791797
  38. 38. Pendergrass SA, Hayes E, Farina G, Lemaire R, Farber HW, et al. (2010) Limited systemic sclerosis patients with pulmonary arterial hypertension show biomarkers of inflammation and vascular injury. PLoS One 5: e12106.SA PendergrassE. HayesG. FarinaR. LemaireHW Farber2010Limited systemic sclerosis patients with pulmonary arterial hypertension show biomarkers of inflammation and vascular injury.PLoS One5e12106
  39. 39. Joehanes R, Johnson AD, Barb JJ, Raghavachari N, Liu P, et al. (2011) Gene Expression Analysis of Whole Blood, Peripheral Blood Mononuclear Cells, and Lymphoblastoid Cell Lines from the Framingham Heart Study. Physiol Genomics. R. JoehanesAD JohnsonJJ BarbN. RaghavachariP. Liu2011Gene Expression Analysis of Whole Blood, Peripheral Blood Mononuclear Cells, and Lymphoblastoid Cell Lines from the Framingham Heart Study.Physiol Genomics
  40. 40. Aziz H, Zaas A, Ginsburg GS (2007) Peripheral blood gene expression profiling for cardiovascular disease assessment. Genomic Med 1: 105–112.H. AzizA. ZaasGS Ginsburg2007Peripheral blood gene expression profiling for cardiovascular disease assessment.Genomic Med1105112
  41. 41. Luk T, Malam Z, Marshall JC (2008) Pre-B cell colony-enhancing factor (PBEF)/visfatin: a novel mediator of innate immunity. J Leukoc Biol 83: 804–816.T. LukZ. MalamJC Marshall2008Pre-B cell colony-enhancing factor (PBEF)/visfatin: a novel mediator of innate immunity.J Leukoc Biol83804816
  42. 42. Wang P, Du H, Zhang RY, Guan YF, Xu TY, et al. (2010) Circulating and local visfatin/Nampt/PBEF levels in spontaneously hypertensive rats, stroke-prone spontaneously hypertensive rats and Wistar-Kyoto rats. J Physiol Sci 60: 317–324.P. WangH. DuRY ZhangYF GuanTY Xu2010Circulating and local visfatin/Nampt/PBEF levels in spontaneously hypertensive rats, stroke-prone spontaneously hypertensive rats and Wistar-Kyoto rats.J Physiol Sci60317324
  43. 43. Kim I, Moon SO, Kim SH, Kim HJ, Koh YS, et al. (2001) Vascular endothelial growth factor expression of intercellular adhesion molecule 1 (ICAM-1), vascular cell adhesion molecule 1 (VCAM-1), and E-selectin through nuclear factor-kappa B activation in endothelial cells. J Biol Chem 276: 7614–7620.I. KimSO MoonSH KimHJ KimYS Koh2001Vascular endothelial growth factor expression of intercellular adhesion molecule 1 (ICAM-1), vascular cell adhesion molecule 1 (VCAM-1), and E-selectin through nuclear factor-kappa B activation in endothelial cells.J Biol Chem27676147620
  44. 44. Kim SR, Bae YH, Bae SK, Choi KS, Yoon KH, et al. (2008) Visfatin enhances ICAM-1 and VCAM-1 expression through ROS-dependent NF-kappaB activation in endothelial cells. Biochim Biophys Acta 1783: 886–895.SR KimYH BaeSK BaeKS ChoiKH Yoon2008Visfatin enhances ICAM-1 and VCAM-1 expression through ROS-dependent NF-kappaB activation in endothelial cells.Biochim Biophys Acta1783886895
  45. 45. Lee WJ, Wu CS, Lin H, Lee IT, Wu CM, et al. (2009) Visfatin-induced expression of inflammatory mediators in human endothelial cells through the NF-kappaB pathway. Int J Obes (Lond) 33: 465–472.WJ LeeCS WuH. LinIT LeeCM Wu2009Visfatin-induced expression of inflammatory mediators in human endothelial cells through the NF-kappaB pathway.Int J Obes (Lond)33465472
  46. 46. Benachour H, Zaiou M, Samara A, Herbeth B, Pfister M, et al. (2009) Association of human cathelicidin (hCAP-18/LL-37) gene expression with cardiovascular disease risk factors. Nutr Metab Cardiovasc Dis 19: 720–728.H. BenachourM. ZaiouA. SamaraB. HerbethM. Pfister2009Association of human cathelicidin (hCAP-18/LL-37) gene expression with cardiovascular disease risk factors.Nutr Metab Cardiovasc Dis19720728
  47. 47. Harrap SB (2003) Where are all the blood-pressure genes? Lancet 361: 2149–2151.SB Harrap2003Where are all the blood-pressure genes?Lancet36121492151
  48. 48. Tomaszewski M, Charchar FJ, Nelson CP, Barnes T, Denniff M, et al. (2011) Pathway analysis shows association between FGFBP1 and hypertension. J Am Soc Nephrol 22: 947–955.M. TomaszewskiFJ CharcharCP NelsonT. BarnesM. Denniff2011Pathway analysis shows association between FGFBP1 and hypertension.J Am Soc Nephrol22947955