Abstract
Blood pressure (BP) is a dynamic phenotype that varies rapidly to adjust to changing environmental conditions. Standing upright is a recent evolutionary trait, and genetic factors that influence postural adaptations may contribute to BP variability. We studied the effect of posture on the genetics of BP and intermediate BP phenotypes. We included 384 sib-pairs in 64 sib-ships from families ascertained by early-onset hypertension and dyslipidemia. Blood pressure, three hemodynamic and seven neuroendocrine intermediate BP phenotypes were measured with subjects lying supine and standing upright. The effect of posture on estimates of heritability and genetic covariance was investigated in full pedigrees. Linkage was conducted on 196 candidate genes by sib-pair analyses, and empirical estimates of significance were obtained. A permutation algorithm was implemented to study the postural effect on linkage. ADRA1A, APO, CAST, CORIN, CRHR1, EDNRB, FGF2, GC, GJA1, KCNB2, MMP3, NPY, NR3C2, PLN, TGFBR2, TNFRSF6, and TRHR showed evidence of linkage with any phenotype in the supine position and not upon standing, whereas AKR1B1, CD36, EDNRA, F5, MMP9, PKD2, PON1, PPARG, PPARGC1A, PRKCA, and RET were specifically linked to standing phenotypes. Genetic profiling was undertaken to show genetic interactions among intermediate BP phenotypes and genes specific to each posture. When investigators perform genetic studies exclusively on a single posture, important genetic components of BP are missed. Supine and standing BPs have distinct genetic signatures. Standardized maneuvers influence the results of genetic investigations into BP, thus reflecting its dynamic regulation.
Keywords: blood pressure, family studies, genetic linkage, postural adaptations
high blood pressure (BP) levels and variability are associated with a high risk of adverse outcomes (21, 32). It is estimated that up to 50% of BP variance is genetically determined, with the rest of it due to the effect of environment (24). However, individual quantitative trait loci (QTLs) contribute to only a small percentage of BP variation (7, 28). In fact, genetic variants identified to date in genome-wide association studies (GWAS) may explain only 1–2% of BP variance in the population (9, 19, 29). The effect of a QTL on BP may be overridden by compensatory mechanisms (genetic modifiers and/or physiological adaptations) or obscured by the acute influence of environment at the time of BP measurement (random variation/acute hemodynamic adjustments) as well as by antihypertensive therapy (26).
In contrast to stable or relatively stable quantitative traits, such as height or weight, BP is a highly dynamic phenotype designed to preserve tissue perfusion in continuously changing environmental conditions, such as emotional stress, time of the day, and posture. BP variability is clinically relevant since it is coupled with vascular complications, e.g., stroke (25, 32). Moreover, the genetic determinants of BP responses to the environment may contribute to BP levels (6, 13).
BP regulation requires the action of redundant systems (11), which may dilute the effect of single genetic variants, influencing certain regulatory pathways. Thus, their hypothetical outcome may only be observed when the involved pathway operates under certain stimulation. In fact, experiments on rats have shown BP QTLs that were only apparent after specific pharmacological blockade of the renin-angiotensin, sympathetic nervous, or nitric oxide systems (38). Likewise, family studies have pointed out the existence of BP QTLs specific to certain environmental stimuli, such as posture or sleep (15). Furthermore, genetic variants, such as vasoregulatory hormones [e.g., renin, atrial natriuretic peptide (ANP)], or hemodynamic traits [e.g., total peripheral resistance (TPR), heart rate (HR), etc.], may influence intermediate BP phenotypes without causing any change in the final BP phenotype (6, 28). However, few studies have conducted concurrent phenotyping of intermediate BP traits or assessed the genetic determinants of BP variability under standardized physiological testing.
In humans, lying supine causes blood redistribution from the lower extremities to the thoracic pool, resulting in increased ANP secretion, whereas standing upright evokes a relative decrease in blood above the heart with the activation of compensatory mechanisms to maintain brain perfusion (11). Postural adaptation to standing upright is a relatively recent evolutional trait (41) that may share many genetic determinants with BP. We hypothesized that postural adaptation as a phenotype will reveal novel polymorphisms associated with BP.
Previous work in hypertensive black sib-pairs (22) and family studies (16) found that posture is linked with changes in heritability (h2) estimates of BP. Moreover, preliminary research on hypertensive sib-pairs revealed that posture was coupled with changes in the genetic linkage of polymorphisms of angiotensin II type I receptor and BP (7). However, only two single nucleotide polymorphisms (SNPs) in one gene were tested in that investigation (7). Thus, we hypothesized that physiological adaptations to the change in posture will result in variation of genetic linkage of blood pressure and intermediate blood pressure phenotypes. This study was conducted in sib-pairs from a founder population of French-Canadian families ascertained by the presence of hypertension and dyslipidemia (15). In addition to systolic (SBP), diastolic (DBP), and mean arterial BP (MAP), nine intermediate BP phenotypes including hormones (renin, aldosterone, epinephrine, norepinephrine, ANP), second messengers (cGMP, cAMP), and hemodynamic traits [TPR, stroke volume (SV), HR] were measured at multiple times during supine or standing after careful withdrawal of antihypertensive medications. Individual traits, summary variables obtained by factor or component analyses as well as delta phenotypes (change in trait values from supine to standing), were tested for genetic linkage. Simulations were used to test for significance of postural differences on linkage.
METHODS
Study population.
We included 384 sib-pairs obtained from 64 sib-ships in the present study. Sib-ship size ranged from two to nine individuals (average number of sib-pairs in a sib-ship, 6; range, 1–36 sib-pairs). Sib-ships were obtained from a cohort of independent families and details of family recruitment and structure have been reported previously (15, 30, 34).
The number of sib-pairs with genotype-phenotype information was 380. In brief, families were ascertained by the presence of at least one sib-pair with early-onset (≤55 yr old) hypertension and dyslipidemia, and by the absence of secondary hypertension, DBP >110 mmHg, diabetes mellitus, body mass index >35 kg/m2, renal or liver dysfunction, degenerative central nervous system diseases, malignancies, pregnancy, and substance abuse. The study was approved by the Ethics Committees of Centre de santé et de services sociaux Chicoutimi, Université du Québec à Chicoutimi, and the Centre hospitalier de l'Université de Montréal.
Phenotyping.
Phenotyping was undertaken according to standardized procedures described elsewhere (22). Normotensive and hypertensive subjects were invited to undergo extensive phenotyping. Hypertensive subjects without any contraindications had their antihypertensive drugs withdrawn for 1 wk, and lipid-lowering agents were withdrawn for 1 mo in all subjects. In the present study, BP medications were discontinued in all participants. Orthostatic maneuvers were performed in 291 individuals (147 men, average age 42 yr old), 160 of them normotensive, and 131 hypertensive. In brief, after 30 min of equilibration, the subjects were monitored for 30 min in the supine position and then in the standing position for another 10 min. Hemodynamic variables, including HR (n = 237), TPR (n = 237), SV (n = 237), and BP (SBP, DBP, and MAP; n = 291) were measured every 5 and 2 min during recumbency and standing, respectively, by impedance plethysmography (SORBA Medical Systems), equipped with an automated BP monitor (Dynamap, Johnson & Johnson Medical). Renin (n = 221), aldosterone (n = 221), ANP (n = 184), catecholamines [epinephrine (EP) n = 202, norepinephrine (NE) n = 207], and cyclic nucleotides (cAMP n = 222, cGMP n = 222) were measured after 25 min in the supine and after 10 min in the upright position.
Genotyping.
Total genomic DNA was extracted from human blood, precipitated with ethanol and dissolved in sterile Tris-EDTA (TE) buffer. Affymetrix's GeneChip Mapping 50K XbaI arrays were used. Genomic DNA was diluted with reduced TE buffer at a concentration of 50 ng/μl. Genomic DNA (250 ng) was digested with XbaI enzyme and amplified by a single polymerase chain reaction (PCR) primer. Purified PCR products (40 μg) were digested with DNase I, end-labeled, injected into a XbaI array made up of 2.5 million 25-mer probes, and hybridized for 16 h at 48°C. The array was then washed and stained with streptavidin-phycoerythrin in an Affymetrix F-450 Fluidics station. After staining, the array was read by Affymetrix GeneChip Scanner 3000 7G. Automation of the fluidics station and scanner was supported by Affymetrix GeneChip operating software (GCOS 1.2). GeneChip genotyping analysis software (GTYPE) evaluated the array data, generated genotype calls, and exported array reports for biostatistical analysis.
Selection of candidate genes and SNPs.
Since the goal of the present study was to understand the dynamics of genetic linkage rather than conduct genome-wide scans, we adopted a candidate gene approach. We systematically searched for genes in the PUBMED database with the limit “human” and key words “vascular function,” “autonomic function,” “cardiac function,” “heart rate,” “kidney function,” “hypertension,” “insulin resistance,” and “metabolic syndrome,” as well as in the OMIM database with the key words “hypertension” and “hypotension.” We added the list of candidate genes for BP homeostasis and/or hypertension published previously by Halushka et al. (12), Okuda et al. (27), and Iwai et al. (20). After deleting duplicates and nongenes, we selected 312 candidate genes.
We then screened our database to find SNPs within candidate genes or neighboring SNPs in linkage disequilibrium (LD) (D′ = 1 and LOD ≥2; Haploview) with any SNP located in candidate genes in our database or in the HapMap database for Utah residents with Northern and Western European ancestry (CEU population) (36). Allele frequencies and Hardy-Weinberg equilibrium were calculated by exact test, as described by Wigginton et al. (40).
Statistical analysis.
Phenotypes were adjusted by age and sex. Trait values during supine or standing were compared by paired t-test (SPSS software). Differences between normotensives and hypertensives were compared by unpaired t-test. Polygenic h2, i.e., total additive genetic h2, and genetic covariance, i.e., polygenic bivariate analysis with age and sex as covariates, were estimated by SOLAR (2).
Single point identity was calculated from genotypes for all family members (e.g., using all available data on extended pedigrees), by descent (IBD) probabilities with GENIBD (SAGE). Model-free two-point linkage analysis, according to the weighted Haseman-Elston regression method (W4) as implemented in SIBPAL (SAGE), investigated the presence of linkage. This method transforms sib-pair trait values to weighted combinations of the squared trait difference and squared mean corrected trait sum, adjusting for the nonindependence of sib-pair squared sums and differences in large sib-ships (35).
Linkage analysis was conducted on average phenotypes or on the delta change from supine to standing, with Bonferroni correction accounting for multiple testing, including number of SNPs and phenotypes. The alpha threshold (at P < 0.05) for linkage during recumbency or standing with average phenotypes was 2.4 × 10−6 [e.g., 0.05/803 SNPs × 2 analyses (supine or standing) × 13 phenotypes] and 4.8 × 10−6 for delta phenotypes. Moreover, for hemodynamic traits for which we obtained 12 measurements over time, principal component analysis (PCA) and factor analysis (FA) were used to find uncorrelated univariate traits containing shared variance (PCA) or correlation (FA) among all traits at all times during supine or standing.
Differences in nominal or corrected linkage P values ≥1 × 103 between recumbent and standing postures were considered to be of interest. However, only disappearance of the linkage signal in a given posture was taken as a significant change in nominal linkage.
We designed permutations to calculate empirical P values for hemodynamic variables at each posture, taking into account all available measurements and testing for differences in linkage between supine and standing. We adopted the gene-dropping approach, as implemented in MERLIN to simulate genotypes (1). In brief, it segregates random chromosomes through pedigrees retaining the family structure of the original pedigree, the recombination fraction, and the original pattern of missing values. The net result is a random chromosome that is unlinked to any trait, but the distribution of phenotypes is conserved intact.
Linkage analysis of 10,000 permutations for all hemodynamic traits through all time points during recumbency and standing conserving the time-series structure of the data was conducted. Then, the proportion of simulated replicates that was equal to or greater than the minimum t-value (lowest linkage statistic) observed at each period (supine or standing) was selected as the empirical P value for that period. Empirical P values for differences in linkage between supine and standing were calculated by estimating the proportion of simulated replicates with an equal to or greater difference of minimum t-values between supine and standing than the difference between both postures.
To further test the hypothesis that change in posture has a significant effect on the linkage results, a permutation test on correlations between time series (time points) of the “t-statistic” derived from linkage tests and a square signal, representing change in posture was done (3). The permutation test is part of a family of methods known collectively as “resampling procedures” (4). The algorithm used in this work is an adaptation of previous investigation by Belmonte and Yurgelun-Todd (3). The computation steps are described in detail in the appendix.
RESULTS
Screening of the 312 preselected genes for SNPs in our database yielded a total of 1,018 SNPs in 196 genes. After excluding SNPs with minor allele frequency <5% and deviations from Hardy-Weinberg equilibrium (P < 0.01), we finally selected 803 SNPs (Supplementary Table S1).1 SNPs in LD were not excluded from the analyses and were counted in the correction by multiple testing. However, the SNPs tested did not exhaustively cover all genes studied.
Effects of posture on phenotypic variance and heritability.
Table 1 reports the means for all traits during supine and standing positions, as well as the delta change. Hypertensive individuals had higher supine SBP, DBP, TPR, HR, aldosterone, cAMP, and ANP levels but lower SV compared with normotensive subjects (Table 1). Factors associated with supine SBP (P < 0.05) were SV, TPR, HR, renin, ANP, cGMP, and cAMP levels, whereas supine DBP was associated with HR, SV, TPR, and cAMP. Age and sex were used as covariates in all models.
Table 1.
Hemodynamic and neuroendocrine traits by hypertension status
| Without HBP (n = 160)† |
With HBP (n = 131)† |
All (n = 291)† |
||
|---|---|---|---|---|
| Trait | Mean ± SD | Mean ± SD | Mean ± SD | h2 |
| SBP; mmHg | ||||
| Supine SBP | 112 ± 11 | 136.7 ± 16.2b | 122.5 ± 19.7 | 0.4 |
| Standing SBP | 115 ± 13 | 140.4 ± 18.4b | 125.6 ± 20.8 | 0.58 |
| Delta SBP | 3.1 ± 7.22 | 3.0 ± 8.92 | 3.0 ± 8.02 | 0.37 |
| DBP; mmHg | ||||
| Supine DBP | 67.9 ± 7.7 | 81.0 ± 9.7b | 73.9 ± 10.8 | 0.58 |
| Standing DBP | 74.9 ± 9.0 | 87.2 ± 10.6b | 79.5 ± 11.2 | 0.74 |
| Delta DBP | 5.8 ± 4.62 | 5.3 ± 5.12 | 5.5 ± 4.82 | 0.19 |
| HR, beats/min | ||||
| Supine | 64.3 ± 9.4 | 66.7 ± 9.4a | 65.2 ± 9.5 | 0.56 |
| Standing | 75.4 ± 11.3 | 77.5 ± 12.8 | 74.5 ± 10.9 | 0.34 |
| Delta heart rate | 9.5 ± 6.22 | 9.0 ± 5.92 | 9.3 ± 6.12 | 0.15 |
| SV, ml | ||||
| Supine | 75 ± 19.5 | 67.3 ± 23.3a | 71.3 ± 21.4 | 0.13 |
| Standing | 50.1 ± 12.8 | 49.4 ± 15.9a | 49.2 ± 24.2 | 0.45 |
| Delta change | (−) 24.9 ± 14.42 | (−) 18.2 ± 302 | (−) 21.6 ± 15.22 | 0.06 |
| TPR, Sorba units | ||||
| Supine | 771.5 ± 246.3 | 1,007.3 ± 317.5b | 878.2 ± 308.8 | 0.19 |
| Standing | 1,138.8 ± 355.9 | 1,357.5 ± 468.6b | 1,235.9 ± 424.4 | 0.42 |
| Delta TPR | 365.4 ± 232.22 | 355.8 ± 312.32,b | 360.6 ± 272.22 | 0.31 |
| Renin, ng/ml/h | ||||
| Supine | 1.4 ± 1.1 | 1.3 ± 0.9 | 1.4 ± 1.0 | 0.44 |
| Standing | 2.2 ± 2 | 1.9 ± 1.3 | 2.0 ± 1.7 | 0.23 |
| Delta | 0.7 ± 1.22 | 0.5 ± 0.92 | 0.6 ± 1.12 | 0.03 |
| Aldo, pg/ml | ||||
| Supine | 4.9 ± 3.5 | 6.0 ± 3.6a | 5.4 ± 3.5 | 0.2 |
| Standing | 6.0 ± 3.9 | 7.7 ± 4.4a | 6.8 ± 4.2 | 0.12 |
| Delta | 1.1 ± 22 | 1.7 ± 2.42,a | 1.3 ± 2.32 | 0.15 |
| ANP, pg/ml | ||||
| Supine | 52.8 ± 23 | 64.1 ± 30.2a | 58.4 ± 27.2 | 0.29 |
| Standing | 43.8 ± 18.8 | 55.1 ± 24a | 49.3 ± 22.1 | 0.14 |
| Delta | (−) 9.0 ± 11.62 | (−) 9.1 ± 18.61 | (−) 9.1 ± 15.42 | 0.38 |
| cAMP, pg/ml | ||||
| Supine | 11.7 ± 2.9 | 12.6 ± 2.9a | 12.2 ± 2.9 | 0.28 |
| Standing | 13.4 ± 3.4 | 14.1 ± 3.9 | 13.8 ± 3.7 | 0.19 |
| Delta | 1.7 ± 3.51 | 1.5 ± 3.92 | 1.6 ± 3.72 | 0 |
| cGMP, pg/ml | ||||
| Supine | 4.7 ± 1.7 | 4.9 ± 1.9 | 4.8 ± 1.9 | 0.17 |
| Standing | 4.9 ± 1.7 | 5.0 ± 1.7 | 5.0 ± 1.7 | 0.19 |
| Delta | 0.2 ± 1.11 | 0.1 ± 1.2 | 0.16 ± 1.11 | 0.19 |
| EP, pg/ml | ||||
| Supine | 42.8 ± 17.2 | 42.9 ± 20.9 | 42.9 ± 18.7 | 0.24 |
| Standing | 52.9 ± 30.5 | 47.9 ± 21.6 | 51.0 ± 27.4 | 0.21 |
| Delta | 10.1 ± 30.61 | 4.9 ± 22.1 | 8.1 ± 27.71 | 0 |
| NE, pg/ml | ||||
| Supine | 169.4 ± 62.3 | 163.5 ± 70.7 | 166.7 ± 66.3 | 0.45 |
| Standing | 399.6 ± 146.1 | 394.5 ± 165.7 | 397.4 ± 155.8 | 0.2 |
| Delta | 230.2 ± 120.62 | 231.1 ± 124.12 | 230.6 ± 122.82 | 0.06 |
Superscript letters indicate comparison of hypertensives (HBP) vs. normotensives (without HBP):
P < 0.05; bP < 0.001 (unpaired t-test). Superscript numbers indicate comparison of supine vs. standing: 1P < 0.05; 2P < 0.001 (Paired t-test). h2: heritability (SOLAR).
Total number of individuals included; systolic (SBP) and diastolic blood pressure (DBP) (n = 291); heart rate (HR) (n = 237); stroke volume (SV) (n = 237); total peripheral resistance (TPR) (n = 237); renin (n = 221); aldosterone (Aldo) (n = 221); cyclic guanosine monophosphate (cGMP) (n = 222); cyclic adenosine monophosphate (cAMP) (n = 222); epinephrine (EP) (n = 202); norepinephrine (NE) (n = 207); atrial natriuretic peptide (ANP) (n = 184). BP, blood pressure.
Standing was associated with changes in hemodynamic variables and neuroendocrine activity (Table 1). Both DBP and SBP rose with TPR and HR and declined in left ventricular SV (Table 1). These hemodynamic adaptations were accompanied by elevated renin and aldosterone levels, as well as increased circulating catecholamines and cyclic nucleotides, but decreased ANP (Table 1). The increase in TPR and aldosterone levels during standing was greater in hypertensives than in normotensives (Table 1). Hypertensives had larger phenotypic variance for most traits compared with normotensives (Table 1). Near one-third (n = 88) of participants showed no change or decreased SBP upon standing, whereas 12% of participants (n = 35) had blunted or reduced DBP upon standing.
Standing was associated with a larger mean increase of DBP compared with SBP. However, SBP had larger SD. Age, baseline SBP, and delta change in EP were linked with postural change in SBP. Prior use of BP medications (including diuretics or β-blockers) was not associated with delta SBP. Delta TPR was weakly correlated with delta SBP (r = 0.2, P = 0.05), whereas delta SV, delta HR, and composite delta cardiac output showed no correlation with change in SBP upon standing. These findings indicate that supine SBP in part determines the delta change of SBP upon standing. Delta change in DBP was associated with supine DBP, delta HR (P < 0.001, r = 0.28), and delta NE levels (P = 0.002, r = 0.26) but weakly correlated with delta TPR (r = 0.13) and SV (r = −0.13). h2 estimates for traits during supine or standing are listed in Table 1. They were higher for hemodynamic factors (excluding HR) upon standing but decreased or did not change for neuroendocrine factors (Table 1).
Effects of posture on genetic correlation among traits.
Polygenic bivariate analysis was conducted to investigate genetic covariance (RhoG) among BP and intermediate BP phenotypes during when supine and standing (Table 2). Genetic covariance between SBP and DBP did not change from supine to standing (0.84 and 0.88, respectively). However, differences in genetic correlation were seen between postures (Table 2). For instance, TPR was the factor with the greatest genetic covariance with BP during supine (Table 2). However, during standing, cAMP had the highest genetic covariance with BP (Table 2, see also Fig. 3). Plasma renin activity and aldosterone were the best genetic correlates with TPR during supine, while NE level was the best genetic correlate of aldosterone in the standing position (Table 2).
Table 2.
Genetic covariance (RhoG) among BP phenotypes
| Factor | SBP | DBP | SV | HR | TPR | Renin | Aldo | EP | NE | cGMP | cAMP | ANP |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Supine | ||||||||||||
| SBP | 1.00 | 0.84 | 0.22 | 0.07 | 0.81 | 0.25 | 0.17 | 0.30 | 0.47 | 0.14 | 0.34 | nc |
| DBP | 0.84 | 1.00 | 0.38 | 0.09 | 0.93 | 0.06 | 0.30 | 0.21 | 0.39 | 0.03 | 0.35 | nc |
| SV | 0.22 | 0.38 | 1.00 | 0.90 | 0.72 | 0.21 | 0.77 | nc | 0.08 | 0.40 | 0.32 | 0.68 |
| HR | 0.07 | 0.09 | 0.90 | 1.00 | 0.10 | 0.16 | 0.48 | nc | 0.46 | 0.12 | 0.64 | 0.16 |
| TPR | 0.81 | 0.93 | 0.72 | 0.10 | 1.00 | 0.55 | 0.55 | 0.08 | 0.16 | 0.48 | 0.27 | nc |
| Renin | 0.25 | 0.06 | 0.21 | 0.16 | 0.55 | 1.00 | 0.37 | 0.10 | 0.29 | nc | 0.13 | 0.31 |
| Aldo | 0.17 | 0.30 | 0.77 | 0.48 | 0.55 | 0.37 | 1.00 | 0.39 | 0.06 | 0.30 | 0.21 | 0.11 |
| EP | 0.30 | 0.21 | nc | nc | 0.08 | 0.10 | 0.39 | 1.00 | 0.06 | 0.04 | nc | 0.74 |
| NE | 0.47 | 0.39 | 0.08 | 0.46 | 0.16 | 0.29 | 0.06 | 0.06 | 1.00 | 0.27 | 0.04 | nc |
| cGMP | 0.14 | 0.03 | 0.40 | 0.12 | 0.48 | nc | 0.30 | 0.04 | 0.27 | 1.00 | 0.40 | 0.20 |
| cAMP | 0.34 | 0.35 | 0.32 | 0.64 | 0.27 | 0.13 | 0.21 | nc | 0.04 | 0.40 | 1.00 | 0.55 |
| ANP | nc | nc | 0.68 | 0.16 | nc | 0.31 | 0.11 | 0.74 | nc | 0.20 | 0.55 | 1.00 |
| Standing | ||||||||||||
| SBP | 1.00 | 0.88 | 0.31 | 0.01 | 0.60 | 0.38 | nc | 0.06 | 0.14 | 0.03 | 0.78 | 0.13 |
| DBP | 0.88 | 1.00 | 0.13 | 0.23 | 0.63 | 0.15 | nc | 0.07 | 0.59 | 0.06 | 0.72 | 0.06 |
| SV | 0.26 | 0.01 | 1.00 | 0.13 | 0.66 | 0.22 | 0.15 | 0.10 | 0.15 | 0.37 | 0.45 | 0.01 |
| HR | 0.01 | 0.23 | 0.13 | 1.00 | 0.04 | 0.02 | 0.18 | 0.12 | 0.57 | 0.59 | 0.11 | 0.59 |
| TPR | 0.60 | 0.63 | 0.66 | 0.04 | 1.00 | 0.40 | 0.41 | 0.02 | 0.17 | 0.43 | 0.86 | 0.05 |
| Renin | 0.38 | 0.15 | 0.22 | 0.02 | 0.40 | 1.00 | 0.04 | 0.09 | 0.22 | 0.22 | 0.21 | 0.13 |
| Aldo | nc | nc | 0.15 | 0.18 | 0.41 | 0.04 | 1.00 | 0.46 | 0.87 | 0.52 | nc | 0.21 |
| EP | 0.06 | 0.07 | 0.10 | 0.12 | 0.02 | 0.09 | 0.46 | 1.00 | 0.18 | 0.70 | 0.20 | 0.57 |
| NE | 0.14 | 0.59 | 0.15 | 0.57 | 0.17 | 0.22 | 0.87 | 0.18 | 1.00 | 0.15 | 0.25 | 0.04 |
| cGMP | 0.03 | 0.06 | 0.37 | 0.59 | 0.43 | 0.22 | 0.52 | 0.70 | 0.15 | 1.00 | 0.23 | 0.85 |
| cAMP | 0.78 | 0.72 | 0.45 | 0.11 | 0.86 | 0.21 | nc | 0.20 | 0.25 | 0.23 | 1.00 | 0.39 |
| ANP | 0.13 | 0.06 | 0.01 | 0.59 | 0.05 | 0.13 | 0.21 | 0.57 | 0.04 | 0.85 | 0.39 | 1.00 |
Results of bivariate polygenic analysis with SOLAR. Age and sex were covariates. Numbers represent genetic correlations (RhoG) between factors during recumbency or standing. nc, RhoG SE not computable by SOLAR.
Fig. 3.
Dynamic genetic architecture of blood pressure (BP) and intermediate BP phenotypes during orthostatic stress. Genetic interactions of supine (A) hemodynamic and neuroendocrine traits or after adopting the standing posture (B). C: interactions for delta change phenotypes. Hemodynamic factors are represented as open rings (SBP, systolic blood pressure; DBP, diastolic blood pressure; TPR, total peripheral resistance; HR, heart rate); neuroendocrine factors appear as ellipses (ANP, atrial natriuretic peptide; Aldo, aldosterone; EP, epinephrine; NE, norepinephrine); second messengers (cyclic nucleotides) are symbolized as dashed rings, and genes, as gray-shaded ellipses. Traits and genes are arranged on the basis of their genetic proximity (as with the degree of genetic covariance or linkage). Blue and red lines represent genetic correlations (RhoG), positive and negative respectively, among factors after adjustment for other interrelated factors (see Table 2: Genetic covariance among BP phenotypes). Unbroken lines represent RhoG >0.7; dashed lines, RhoG = 0.5–0.7; and dotted lines, RhoG 0.4. Only genetic correlations >0.40 are displayed and used for genetic adjustments. Gray lines linking genes and traits represent genetic linkage. SV and EP are not represented in C, since genetic correlations for these factors were not computable.
Genetic linkage.
Linkage analyses were conducted using the average of trait values during each posture using Bonferroni's correction. Significant linkage results after Bonferroni correction (cP value, only significant results at cP < 0.05 are listed) are shown in Table 3. Top P values were obtained for rs10509540 (renalase gene) and rs3798573 (estrogen receptor gene) linked to cGMP levels (cP < 1.0 × 10−10) and renin levels (cP < 1.0 × 10−10), respectively. Most of significant linkage results in this analysis were obtained for neuroendocrine factors with only six genes linked to hemodynamic traits. No SNP linked simultaneously to both hemodynamic and hormonal traits.
Table 3.
Linkage of hemodynamic and neuroendocrine phenotypes in supine or standing position
| Chromosome | Gene | SNP | MAF | Phenotype | P Value | cP Value |
|---|---|---|---|---|---|---|
| Supine | ||||||
| 2 | KCNG3 | rs6716767 | 0.09 | Aldo | 1.2 × 10−6 | 0.02 |
| SLC8A1 | rs10490208 | 0.43 | Aldo | 5.0 × 10−7 | 0.01 | |
| APOB | rs10495711 | 0.23 | Aldo | 8.5 × 10−14 | 1.9 × 10−9 | |
| 3 | TGFBR2 | rs1367610 | 0.15 | cAMP | 3.0 × 10−7 | 6.8 × 10−3 |
| GHRL | rs715827 | 0.09 | Renin | 1.6 × 10−9 | 3.3 × 10−5 | |
| 4 | GC | rs1453459 | 0.14 | ANP | 3.0 × 10−7 | 6.8 × 10−3 |
| rs222016 | 0.14 | ANP | 2.2 × 10−7 | 2.2 × 10−3 | ||
| rs842873 | 0.46 | SV | 2.0 × 10−7 | 4.5 × 10−3 | ||
| 6 | ESR1 | rs1884052 | 0.17 | Renin | 2.6 × 10−9 | 6.0 × 10−5 |
| rs1884054 | 0.32 | Renin | 2.4 × 10−9 | 5.4 × 10−5 | ||
| rs726281 | 0.2 | Renin | 2.5 × 10−8 | 5.2 × 10−4 | ||
| rs726282 | 0.07 | Renin | 1.1 × 10−10 | 2.4 × 10−6 | ||
| rs3798573 | 0.07 | Renin | 4.2 × 10−13 | 8.8 × 10−9 | ||
| 8 | ADRA1A | rs520180 | 0.27 | ANP | 1.0 × 10−7 | 2.2 × 10−3 |
| KCNB2 | rs349337 | 0.05 | ANP | 1.1 × 10−11 | 2.4 × 10−7 | |
| rs2253667 | 0.12 | ANP | 8.8 × 10−13 | 1.9 × 10−8 | ||
| 10 | RNLS | rs10509540 | 0.24 | cGMP | 1.0 × 10−14 | 1.0 × 10−10 |
| rs2162361 | 0.24 | cGMP | 7.9 × 10−8 | 1.8 × 10−3 | ||
| rs4934391 | 0.35 | cGMP | 1.9 × 10−11 | 4.3 × 10−7 | ||
| rs10509547 | 0.27 | cGMP | 1.2 × 10−8 | 2.7 × 10−4 | ||
| rs10509548 | 0.26 | cGMP | 1.7 × 10−8 | 3.8 × 10−4 | ||
| rs10509549 | 0.26 | cGMP | 3.5 × 10−8 | 7.9 × 10−4 | ||
| rs10509550 | 0.25 | cGMP | 1.6 × 10−5 | 3.6 × 10−5 | ||
| rs2437871 | 0.48 | cGMP | 5.9 × 10−8 | 1.3 × 10−3 | ||
| rs413898 | 0.19 | cGMP | 1.0 × 10−14 | 1.0 × 10−10 | ||
| TNFRSF6 | rs4406737 | 0.4 | cGMP | 1.2 × 10−8 | 2.8 × 10−4 | |
| rs2296600 | 0.4 | cGMP | 6.2 × 10−8 | 1.4 × 10−3 | ||
| rs1977389 | 0.4 | cGMP | 6.2 × 10−8 | 1.4 × 10−3 | ||
| 17 | CRHR1 | rs436667 | 0.3 | ANP | 7.0 × 10−7 | 0.01 |
| rs647483 | 0.3 | ANP | 8.0 × 10−7 | 0.02 | ||
| rs241031 | 0.3 | ANP | 7.0 × 10−7 | 0.01 | ||
| rs17688434 | 0.3 | ANP | 6.0 × 10−7 | 0.13 | ||
| rs17688452 | 0.29 | ANP | 9.0 × 10−7 | 0.02 | ||
| rs10491144 | 0.3 | ANP | 7.0 × 10−7 | 0.01 | ||
| rs17563501 | 0.29 | ANP | 1.6 × 10−6 | 0.03 | ||
| rs1981997 | 0.29 | ANP | 7.0 × 10−7 | 0.01 | ||
| Standing | ||||||
| 1 | F5 | rs2420134 | 0.06 | Aldo | 1.8 × 10−9 | 4.1 × 10−5 |
| KCNH1 | rs1934614 | 0.1 | TPR | 2.0 × 10−7 | 4.5 × 10−3 | |
| 2 | KCNG3 | rs6716767 | 0.09 | Aldo | 1.0 × 10−7 | 2.2 × 10−3 |
| SLC8A1 | rs10490208 | 0.43 | Aldo | 3.0 × 10−7 | 6.8 × 10−3 | |
| rs10495711 | 0.23 | Aldo | 3.4 × 10−9 | 7.0 × 10−5 | ||
| 3 | PPARG | rs4135268 | 0.08 | Renin | 3.0 × 10−7 | 6.8 × 10−3 |
| GHRL | rs715827 | 0.09 | Renin | 4.4 × 10−12 | 1.0 × 10−7 | |
| 4 | PPARGC1A | rs2324240 | 0.12 | Renin | 2.1 × 10−6 | 0.04 |
| PKD2 | rs4128340 | 0.33 | cGMP | 1.5 × 10−6 | 0.03 | |
| EDNRA | rs1429119 | 0.25 | DBP | 3.9 × 10−8 | 8.8 × 10−4 | |
| rs1429119 | 0.25 | MAP | 6.2 × 10−9 | 1.4 × 10−3 | ||
| rs2115560 | 0.33 | MAP | 8.0 × 10−7 | 0.01 | ||
| 6 | ESR1 | rs1884052 | 0.17 | Renin | 2.0 × 10−7 | 4.1 × 10−3 |
| rs1884054 | 0.32 | Renin | 6.0 × 10−7 | 0.01 | ||
| rs726281 | 0.2 | Renin | 8.1 × 10−12 | 1.8 × 10−7 | ||
| rs726282 | 0.07 | Renin | 2.5 × 10−13 | 5.7 × 10−9 | ||
| rs3798573 | 0.07 | Renin | 1.0 × 10−14 | 1.0 × 10−10 | ||
| 10 | RET | rs788271 | 0.17 | NE | 1.5 × 10−6 | 0.03 |
| RNLS | rs10509540 | 0.24 | cGMP | 1.0 × 10−13 | 1.0 × 10−10 | |
| rs2162361 | 0.24 | Aldo | 1.0 × 10−7 | 2.2 × 10−3 | ||
| rs413898 | 0.19 | cGMP | 2.0 × 10−7 | 4.1 × 10−3 | ||
| 17 | ITGB3 | rs10514919 | 0.19 | DBP | 1.9 × 10−6 | 0.04 |
| PRKCA | rs10512510 | 0.06 | TPR | 4.4 × 10−14 | 1.0 × 10−9 | |
| rs10512510 | 0.06 | SV | 2.0 × 10−9 | 4.6 × 10−5 | ||
| 20 | MMP9 | rs4812989 | 0.05 | MAP | 6.0 × 10−7 | 0.01 |
Nominal and Bonferroni (cP) P values (Alpha threshold = 2.4 × 10−6) for single nucleotide polymorphisms (SNPs) in candidate genes for BP homeostasis and hypertension. Hemodynamic and neuroendocrine factors involved in BP regulation were measured with subject in a supine or upright, standing position.
Table 3 shows that most linkage signals (72%) varied with posture. Nearly one-third of SNPs remained linked (corrected P < 0.05) both during supine and standing positions. For the SNPs that remained significant during both postures, one-third showed a large change (P ≥ 1 × 103) in linkage scores with posture (e.g., rs413898, rs10495711, rs726281, rs726282, rs10512510). For instance, rs413898 (in RNLS gene) was linked to cGMP levels much more when supine (cP = 1.0 × 10−10) than standing (cP = 4.1 × 10−3). Figures 1 and 2 provide examples of the effect of posture on linkage signals. In Fig. 1, the level of linkage of SNP rs842873 and SV is highest during supine but disappears during standing. In contrast, linkage signal between SNP rs10494478 and MAP (Fig. 2) becomes evident in standing position.
Fig. 1.
Linkage of rs842873 and stroke volume (SV) when supine and standing upright. A: SV (cardiac impedance units, means ± SE) during supine (0–30 min) and standing (30–40 min) in sib-pairs from hypertensive families (n = 237). B: corresponding P values for 2-point linkage between rs842873 located on chromosome 4q13 and SV throughout the experimental periods (n = 258 sib-pairs).
Fig. 2.
Linkage of rs10494478 and mean arterial blood pressure (MAP). A: MAP (mmHg, means ± SE) during supine (0–30 min) and standing (30–40 min) in sib-pairs from hypertensive families (n = 291). B: corresponding P values for 2-point linkage between rs10494478 located on chromosome 1q23 and MAP throughout the experimental periods (n = 366 sib-pairs).
Empirical linkage.
Permutations were designed to obtain empirical estimates of linkage for each posture and to test for differences in linkage signals between supine and standing. For this experiment, all time point estimates for each hemodynamic trait (SBP, DBP, HR, SV, TPR) during supine or standing (e.g., every 5 min during supine, every 2 min during standing) were used, and 10,000 permutations using the gene dropping approach were conducted, conserving intact the time-series structure of the data (only randomly changing the genetic structure). There was evidence of linkage for 63 SNPs: 52 SNPs were linked when supine, 15 SNPs during standing, and 26 SNPs remained significant for both supine and standing (Table 4). For those significant SNPs, additional 10,000 simulations were done to test for linkage differences between supine and standing position (Table 4, difference in linkage P value). The proportion of simulated replicates with an equal to or greater difference of minimum t-values (measure of linkage) between supine and standing was the empirical P value for this experiment. Linkage signal was significantly different between supine and standing for 43 SNP-phenotype couples.
Table 4.
Empirical linkage of hemodynamic traits during recumbency and standing positions
| Chromosome | Gene | SNP | MAF | Phenotype | Supine P | Standing P | P Diff* |
|---|---|---|---|---|---|---|---|
| 1 | KCND3 | rs10494136 | 0.07 | SV | 4.4 × 10−3 | 0.24 | 0.06 |
| rs10494136 | 0.07 | TPR | 7.8 × 10−3 | 0.33 | 0.05 | ||
| F5 | rs10494478 | 0.08 | MAP | 0.04 | 4.0 × 10−4 | 0.03 | |
| rs10494478 | 0.08 | SBP | 0.16 | 5.5 × 10−3 | 0.04 | ||
| rs10489184 | 0.08 | SBP | 8.8 × 10−3 | 1.3 × 10−3 | 0.49 | ||
| rs10489184 | 0.08 | DBP | 0.02 | 2.1 × 10−3 | 0.56 | ||
| rs10489184 | 0.08 | MAP | 4.0 × 10−3 | 2.8 × 10−3 | 0.83 | ||
| HF1 | rs3753396 | 0.21 | SBP | 0.02 | 7.0 × 10−4 | 0.20 | |
| rs3753396 | 0.21 | MAP | 7.0 × 10−3 | 2.7 × 10−3 | 0.89 | ||
| rs3753396 | 0.21 | DBP | 9.9 × 10−3 | 3.6 × 10−3 | 0.93 | ||
| KCNH1 | rs1934613 | 0.11 | MAP | 9.0 × 10−4 | 6.7 × 10−3 | 0.06 | |
| rs1934613 | 0.11 | DBP | 4.0 × 10−4 | 0.02 | 0.01 | ||
| rs1934613 | 0.11 | SBP | 9.3 × 10−3 | 0.04 | 0.1 | ||
| rs1934614 | 0.17 | SV | 0.90 | 9.0 × 10−4 | <0.0001 | ||
| rs1934614 | 0.17 | TPR | 0.25 | 1.8 × 10−3 | 0.01 | ||
| rs1934614 | 0.17 | DBP | 0.01 | 0.06 | 0.07 | ||
| 2 | SRD5A2 | rs206847 | 0.39 | PR | 6.0 × 10−3 | 0.02 | 0.29 |
| SLC8A1 | rs395404 | 0.13 | SV | 0.22 | 1.2 × 10−3 | 0.08 | |
| rs395404 | 0.13 | TPR | 0.05 | 7.4 × 10−3 | 0.26 | ||
| 3 | CAST | rs724333 | 0.32 | DBP | 9.9 × 10−3 | 0.04 | 0.10 |
| rs724333 | 0.32 | MAP | 9.0 × 10−4 | 0.22 | 0.0005 | ||
| rs724333 | 0.32 | SBP | 4.5 × 10−3 | 0.43 | 0.001 | ||
| rs9311603 | 0.10 | DBP | 3.7 × 10−3 | 8.7 × 10−3 | 0.17 | ||
| rs9311603 | 0.10 | MAP | 0.01 | 0.07 | 0.09 | ||
| KCNAB1 | rs3773714 | 0.14 | DBP | 3.2 × 10−3 | 0.02 | 0.05 | |
| rs10513494 | 0.12 | DBP | 1.7 × 10−3 | 5.0 × 10−3 | 0.18 | ||
| rs10513494 | 0.12 | MAP | 2.2 × 10−3 | 7.6 × 10−3 | 0.16 | ||
| rs10513494 | 0.12 | SBP | 0.01 | 0.02 | 0.32 | ||
| SLC2A2 | rs10513688 | 0.21 | DBP | 0.05 | 1.4 × 10−3 | 0.13 | |
| rs10513688 | 0.21 | MAP | 0.07 | 5.0 × 10−3 | 0.21 | ||
| rs10513689 | 0.22 | DBP | 0.04 | 2.8 × 10−3 | 0.26 | ||
| rs10513689 | 0.22 | MAP | 0.05 | 4.9 × 10−3 | 0.32 | ||
| 4 | CORIN | rs7661217 | 0.32 | TPR | 2.7 × 10−3 | 0.89 | 0.0005 |
| GC | rs1453459 | 0.14 | MAP | 0.02 | 0.21 | 0.2 | |
| rs842873 | 0.46 | SV | 3.5 × 10−3 | 0.22 | 0.04 | ||
| FGF2 | rs308388 | 0.37 | MAP | 0.02 | 0.89 | 0.0003 | |
| EDNRA | rs1429119 | 0.25 | MAP | 0.18 | 7.8 × 10−3 | 0.07 | |
| NR3C2 | rs3846329 | 0.20 | DBP | 3.9 × 10−3 | 0.03 | 0.04 | |
| rs10519959 | 0.39 | SBP | 0.04 | 4.0 × 10−4 | 0.12 | ||
| rs10519959 | 0.39 | MAP | 1.3 × 10−3 | 6.0 × 10−4 | 0.54 | ||
| rs10519959 | 0.39 | DBP | 6.4 × 10−3 | 0.02 | 0.19 | ||
| rs4835136 | 0.41 | MAP | 3.2 × 10−3 | 1.3 × 10−3 | 0.81 | ||
| rs4835136 | 0.41 | SBP | 0.05 | 1.6 × 10−3 | 0.12 | ||
| rs4835136 | 0.41 | DBP | 6.6 × 10−3 | 0.02 | 0.24 | ||
| rs2358469 | 0.38 | MAP | 0.02 | 5.5 × 10−3 | 0.78 | ||
| 5 | PCSK1 | rs4077817 | 0.06 | DBP | 8.4 × 10−3 | 0.04 | 0.08 |
| 6 | PLN | rs3951042 | 0.50 | DBP | 1.5 × 10−3 | 0.42 | 0.0002 |
| rs3951042 | 0.50 | MAP | 4.0 × 10−4 | 0.56 | 0.0001 | ||
| rs4027875 | 0.30 | SBP | 7.2 × 10−3 | 0.35 | 0.004 | ||
| rs4027875 | 0.30 | MAP | 0.01 | 0.66 | 0.0005 | ||
| rs10484287 | 0.22 | MAP | 6.4 × 10−3 | 0.76 | <0.0001 | ||
| rs7759088 | 0.22 | SBP | 0.01 | 0.61 | 0.002 | ||
| rs7759088 | 0.22 | MAP | 5.1 × 10−3 | 0.73 | 0.0002 | ||
| GJA1 | rs9320815 | 0.20 | DBP | 2.0 × 10−4 | 0.01 | 0.005 | |
| rs9320815 | 0.20 | MAP | <1.0 × 10−4 | 0.04 | 0.0004 | ||
| rs9320815 | 0.20 | SBP | 1.6 × 10−3 | 0.13 | 0.003 | ||
| 7 | NPY | rs16109 | 0.20 | HR | 7.0 × 10−4 | 0.40 | 0.001 |
| rs16089 | 0.25 | PR | 3.0 × 10−4 | 0.81 | <0.0001 | ||
| CD36 | rs6961069 | 0.41 | TPR | 0.06 | 2.2 × 10−3 | 0.20 | |
| rs7789369 | 0.38 | TPR | 0.18 | 9.0 × 10−4 | 0.04 | ||
| PON1 | rs10485996 | 0.18 | MAP | 0.01 | 0.01 | 0.37 | |
| rs10485996 | 0.18 | SBP | 2.0 × 10−3 | 0.03 | 0.02 | ||
| PON2 | rs854523 | 0.47 | SBP | 4.1 × 10−3 | 0.03 | 0.06 | |
| CYP3A5 | rs1357319 | 0.06 | DBP | 3.1 × 10−3 | 4.9 × 10−3 | 0.27 | |
| rs1357319 | 0.06 | MAP | 7.2 × 10−3 | 7.9 × 10−3 | 0.62 | ||
| rs2037498 | 0.06 | DBP | 5.3 × 10−3 | 5.2 × 10−3 | 0.33 | ||
| rs2037498 | 0.06 | MAP | 9.7 × 10−3 | 6.3 × 10−3 | 0.71 | ||
| rs2037499 | 0.06 | DBP | 4.2 × 10−3 | 5.0 × 10−3 | 0.32 | ||
| rs2037499 | 0.06 | MAP | 8.9 × 10−3 | 7.5 × 10−3 | 0.71 | ||
| AKR1B1 | rs1732045 | 0.43 | TPR | 0.60 | 1.0 × 10−4 | 0.007 | |
| 8 | WRN | rs4733224 | 0.28 | DBP | 0.01 | 0.03 | 0.1 |
| KCNB2 | rs349337 | 0.05 | SV | 8.0 × 10−4 | 0.67 | 0.009 | |
| rs349337 | 0.05 | TPR | 2.8 × 10−3 | 0.93 | 0.003 | ||
| rs2383869 | 0.49 | SBP | 0.02 | 0.20 | 0.03 | ||
| rs2253667 | 0.12 | TPR | 5.6 × 10−3 | 0.12 | 0.11 | ||
| TRHR | rs10505127 | 0.41 | TPR | 3.0 × 10−4 | 0.90 | <0.0001 | |
| rs3110037 | 0.43 | TPR | 1.0 × 10−4 | 0.90 | <0.0001 | ||
| rs10505126 | 0.47 | TPR | 5.1 × 10−3 | 0.90 | <0.0001 | ||
| 11 | TRPC6 | rs7945727 | 0.13 | SBP | 0.08 | 2.1 × 10−3 | 0.08 |
| rs7945727 | 0.13 | MAP | 0.07 | 6.4 × 10−3 | 0.17 | ||
| rs4129255 | 0.36 | MAP | 0.02 | 1.8 × 10−3 | 0.39 | ||
| rs4129255 | 0.36 | DBP | 0.01 | 0.01 | 0.43 | ||
| rs10501981 | 0.41 | MAP | 0.06 | 2.5 × 10−3 | 0.15 | ||
| rs7925012 | 0.49 | MAP | 0.07 | 5.7 × 10−3 | 0.26 | ||
| MMP3 | rs678815 | 0.37 | MAP | 3.3 × 10−3 | 0.08 | 0.01 | |
| rs678815 | 0.37 | SBP | 9.7 × 10−3 | 0.12 | 0.03 | ||
| 13 | EDNRB | rs1937388 | 0.08 | TPR | 2.2 × 10−3 | 0.61 | 0.01 |
| rs1924921 | 0.10 | TPR | 1.0 × 10−3 | 0.73 | 0.002 | ||
| 17 | CRHR1 | rs17651134 | 0.28 | TPR | 7.0 × 10−3 | 0.04 | 0.21 |
| ITGB3 | rs10514919 | 0.19 | DBP | 0.14 | 4.2 × 10−3 | 0.03 | |
| rs10514919 | 0.19 | MAP | 0.10 | 7.5 × 10−3 | 0.18 | ||
| PRKCA | rs10512510 | 0.06 | SV | 0.22 | 2.0 × 10−3 | 0.10 | |
| 19 | LDLR | rs1799898 | 0.14 | TPR | 3.6 × 10−3 | 0.03 | 0.25 |
Empirical P values for SNPs in candidate genes for BP. Phenotypes (hemodynamic factors) were measured several times during the recumbent position and upon adopting the standing posture. The gene-dropping approach (MERLIN) was used to simulate genotypes, and 10,000 permutations were conducted through all time points during supine or standing position, conserving the time-series structure of the data. For significant SNPs, empirical P values for the difference in linkage between supine and standing (
P Diff) were calculated. MAF, minor allele frequency.
Effects of random resampling on linkage signals (permutation test).
To further test whether differences in linkage P values between supine or standing were real or due to chance, we applied a permutation test to the full linkage results from all time points. In this analysis, resampling of all time points was conducted and compared with a random distribution. The permutation test showed that ∼35% of all experiments (SNP-phenotype combinations) yielded significant variation in linkage signals (asymptotic correlation test) from supine to standing. After 10,000 permutations, the number of significant results (variation in linkage signal between supine and standing) was 27%. Thus, one-third of all linkage tests, including previously identified SNP-phenotype pairs, disclosed consistent differences in linkage signal between supine and standing.
Effect of posture on linkage of summary traits.
Summary traits obtained by PCA or FA represent the combined phenotypic variance or covariance among BP and intermediate BP phenotypes. PCA or FA results in different summary traits; PCA traits are uncorrelated linear combinations of the original variables that explain common variance, whereas FA results in underlying factors that explain correlation among variables. PCA resulted in four principal components that explained 89 or 80% of the phenotypic variance of supine or standing traits; the first two components during supine or standing reflect common variance of SBP, DBP, TPR, and SV. The third component was mostly influenced by HR, whereas the last component mainly reflects SBP and DBP variance (Supplementary Table S2).
On the other hand, FA resulted in three main factors explaining 81 or 71% of covariance among traits during supine or standing. The first factor mainly reflects covariance between SBP, DBP, and TPR; the second factor mostly correlates with SV and TPR, whereas the third factor is mostly influenced by HR. (Supplementary Table S2).
Table 5 reports significant linkage results obtained for PCA and FA traits. After Bonferroni correction, 11 SNPs were linked. Most linkage signals were obtained from standing position, and no linkage signal remained significant during both postures. The best linkage results were attained for the second factor during standing (rs10512510, PRKCA gene; FA2; cP = 8.71 × 10−9; Table 5). This SNP (rs10512510) was the only marker linked to both PCA and FA traits (Table 5).
Table 5.
Linkage of principal components and factor analysis traits
| Chromosome | Gene | SNP | MAF | Phenotype | P Value | cP Value |
|---|---|---|---|---|---|---|
| Supine | ||||||
| 8 | KCNB2 | rs349337 | 0.05 | FA2 | 4.0 × 10−6 | 4.0 × 10−6 |
| 16 | SCNN1G | rs8045190 | 0.16 | FA3 | 3.0 × 10−7 | 3.0 × 10−7 |
| rs10492791 | 0.25 | FA3 | 7.0 × 10−7 | 7.0 × 10−7 | ||
| Standing | ||||||
| 1 | KCNH1 | rs1934614 | 0.17 | FA2 | 5.0 × 10−7 | 5.0 × 10−7 |
| 2 | SLC8A1 | rs395404 | 0.13 | FA2 | 3.5 × 10−8 | 3.5 × 10−8 |
| 7 | CFTR | rs2237725 | 0.10 | PC1 | 3.0 × 10−6 | 3.0 × 10−6 |
| 9 | IKBKAP | rs10512384 | 0.07 | FA2 | 8.1 × 10−12 | 8.1 × 10−12 |
| FXN | rs953588 | 0.45 | FA2 | 4.1 × 10−6 | 4.1 × 10−6 | |
| 17 | PRKCA | rs10512510 | 0.06 | FA2 | 7.7 × 10−13 | 7.7 × 10−13 |
| rs10512510 | 0.06 | PC1 | 4.2 × 10−6 | 4.2 × 10−6 | ||
| rs172939 | 0.34 | PC3 | 3.0 × 10−7 | 3.0 × 10−7 | ||
| 20 | MMP9 | rs4812989 | 0.05 | FA1 | 4.0 × 10−6 | 4.0 × 10−6 |
Nominal and Bonferroni: (cP) P values (Alpha threshold = 4.46×10−6) for principal components of hemodynamic traits in supine or standing position. PC1, PC2, PC3: principal components explaining the variance among all traits during supine or standing. FA1, FA2, FA3: factors explaining intercorrelations among all traits during recumbency or standing.
Linkage of delta change phenotypes.
Table 6 reports the linkage results for delta change from supine to standing. After Bonferroni correction, 16 SNPs located in 12 genes were significantly linked to any phenotype. Top linkage (Bonferroni's cP < 1.0 × 10−12) was found for delta change cGMP on chromosome 1 in the NPHS2 gene. A top linkage result for hemodynamic traits was obtained for rs1435010 located in KCNH7 on chromosome 2. This SNP was linked to delta MAP (cP = 1.92 × 10−9), delta DBP (cP = 1.04 × 10−7), and delta SBP (cP = 4.1 × 10−3, Table 6).
Table 6.
Linkage of delta change phenotypes from supine to standing
| Chromosome | Gene | SNP | MAF | Phenotype | P Value | cP Value |
|---|---|---|---|---|---|---|
| 1 | NPHS2 | rs928016 | 0.05 | cGMP | 1.0 × 10−16 | 1.0 × 10−12 |
| rs1410590 | 0.05 | cGMP | 1.0 × 10−16 | 1.0 × 10−12 | ||
| KCND3 | rs10494136 | 0.07 | SV | 2.8 × 10−6 | 0.02 | |
| 2 | KCNH7 | rs1435013 | 0.09 | MAP | 2.0 × 10−7 | 2.0 × 10−3 |
| rs1435013 | 0.09 | DBP | 2.9 × 10−6 | 0.03 | ||
| rs9287822 | 0.10 | MAP | 8.3 × 10−7 | 8.3 × 10−3 | ||
| rs1435010 | 0.06 | MAP | 1.9 × 10−13 | 1.9 × 10−9 | ||
| rs1435010 | 0.06 | DBP | 9.9 × 10−12 | 1.0 × 10−7 | ||
| rs1435010 | 0.06 | SBP | 4.0 × 10−7 | 4.1 × 10−3 | ||
| SLC8A1 | rs430314 | 0.10 | EP | 5.1 × 10−7 | 5.1 × 10−3 | |
| 3 | TGFBR2 | rs1367610 | 0.15 | Aldo | 2.0 × 10−7 | 2.0 × 10−3 |
| rs10501982 | 0.29 | Renin | 1.4 × 10−6 | 0.01 | ||
| 5 | LMNB1 | rs1051644 | 0.45 | Renin | 1.8 × 10−6 | 0.01 |
| 6 | PLN | rs724868 | 0.23 | HR | 3.1 × 10−7 | 3.1 × 10−3 |
| 7 | PIK3CG | rs1724262 | 0.36 | NE | 3.1 × 10−6 | 0.03 |
| 10 | RET | rs2505535 | 0.22 | Renin | 1.2 × 10−6 | 0.01 |
| 12 | KCNA6 | rs4765780 | 0.37 | Renin | 1.0 × 10−7 | 1.0 × 10−3 |
| 14 | CMA1 | rs1956917 | 0.40 | MAP | 1.0 × 10−7 | 1.0 × 10−3 |
| rs1956917 | 0.40 | SBP | 3.8 × 10−6 | 3.9 × 10−2 | ||
| 16 | SLC6A2 | rs9928481 | 0.06 | Aldo | 2.0 × 10−7 | 2.0 × 10−3 |
Nominal and Bonferroni's (cP) P values for SNPs in candidate genes for BP homeostasis and hypertension. The delta change from supine to standing was used as a trait. *Alpha threshold = 4.81 × 10−6.
Figure 3 summarizes the findings of the present study. It discloses genetic correlations among hemodynamic and neuroendocrine factors during supine, standing, and delta change phenotypes as well as the genes linked during each condition. The genetic proximity of BP intermediate phenotypes shifted with change in posture (Fig. 3, A and B). Moreover, a different genetic picture emerged from delta phenotypes (Fig. 3C), with aldosterone having a central role and being strongly genetically correlated with NE as well as with renin and ANP (Fig. 3C). Furthermore, overall linkage architecture varied among the three conditions (Fig. 3, A–C).
DISCUSSION
The main findings of this study are that genetic linkage for BP and intermediate BP phenotypes is influenced by posture and that postural adaptations to standing measured as a delta phenotypes revealed novel polymorphisms not seen when only supine or standing traits were used. Postural differences were found consistently for nominal and empirical linkage; changes in genetic linkage were seen in either neuroendocrine or hemodynamic traits and included greater variations in significance (>1 × 103). Similarly, postural differences in linkage were also obtained using PCA or FA summary variables. These findings indicate that a dynamic genetic architecture underlies BP regulation, which was underscored by a simple physiological maneuver, such as standing upright.
Bivariate variance component analyses revealed that changes in genetic covariance among BP and intermediate BP phenotypes vary with posture. For instance, common genetic factors explained 65% (r = 0.8) of the phenotypic variance of supine SBP and TPR, but only 36% (r = 0.6) of standing SBP and TPR. Similarly, common genetic factors accounted for 15% of supine variation of NE and DBP, whereas during standing common genetic factors influenced 36% of NE and DBP variation. However, genetic correlation between supine and standing was high; for instance, bivariate variance component analyses showed high genetic correlation for supine and standing SBP (r > 0.9) as well as for supine and standing DBP (r > 0.9). Hence, similar genetic factors contribute to SBP or DBP variation during supine and standing but their relative contribution to trait variance changes with posture. Since IBD probabilities do not vary with posture, change in linkage scores results from alterations in trait similarity among sib-pairs, which may reflect relative postural variation in the effect of polygenes.
Previous family studies found that genetic factors explain part of BP variability with posture (15, 16). However, in these investigations, intermediate phenotypes were not available and/or genetic testing was conducted with microsatellite markers, and no formal statistical comparison of linkage between supine or standing traits was done. The present study is the first, to our knowledge, to present linkage information for nine intermediate BP phenotypes during orthostatic maneuvers in a founder population. It also shows that the use of delta phenotypes discloses additional genetic linkage, which suggests the existence of QTLs that modulate changes in BP and intermediate BP phenotypes during the change in posture from supine to standing.
Indeed, h2 estimates of delta change suggest that genetic factors may account for up to 38% of variability of delta phenotypes studied in this work; higher h2 estimates were obtained for delta change in SBP, TPR, or ANP levels. Responses to orthostatism (e.g., supine to standing) involve the integration of cardiopulmonary and arterial baroreflexes that sense changes in blood volume in central compartments and modulate sympathetic activity and vascular resistance. Our findings suggest that genetic factors may influence some of these physiological loops. However, for other traits, such as change in SV or renin and catecholamine levels, low h2 estimates were obtained, indicating that they may be reflex responses. In fact, decreased blood volume of the body's central compartment upon standing reduces left ventricular filling pressures, causing reflex increased brain stem sympathetic outflow that stimulates catecholamines and renin release.
It has been reported previously that combination of highly correlated traits with PCA or linear combination of phenotypes increases multipoint linkage scores (15) and the power to detect linkage (17). In the present study, five genes not previously identified by average traits were identified by PCA. However, only one SNP was linked to both PCA and FA traits. This indicates that summary traits obtained by either method are distinct. In fact, there is a more clear separation of traits with FA than with PCA. For instance, the first principal component (PCA1) reflects the common variance of DBP and SBP, SV and TPR. In contrast, the first factor (FA1) mainly reflects the interrelation between SBP and DBP. Simulation studies have found that FA produces factors with stronger correlations of underlying traits than PCA (39). However, PCA linkage data may reflect a “deeper” underlying genetic effect of BP regulation since it reflects the common variance of BP traits.
Overall, top linkage scores were obtained for cGMP and three SNPs (rs10509540, rs413898, rs4934391) inside or flanking the Renalase (RNLS) gene. This gene was also linked to aldosterone and SV. RNLS encodes a secreted amine oxidase that metabolizes catecholamines and is involved in the regulation of cardiac function and BP (42). Moreover, RNLS polymorphisms have been associated with essential hypertension (43). In the present study, ANP levels were higher in hypertensives compared with normotensives. It has been shown that binding of ANP to NPR-A, the main receptor for the biological known functions of ANP, results in intracellular formation of cGMP (13), which in this study had the greatest genetic correlation with ANP, as shown in our previous studies in genetically hypertensive rodents (37).
Most BP GWAS published to date obtained BP measurements in a single posture in a nonstandardized environment. Cuff BP values vary considerably between office visits, which increases random error. The fact that linkage results of the present study were not replicated in recent large BP GWAS does not invalidate the findings of this work, whose main objective was to investigate the effect of posture on linkage. The present study was conducted in a founder population and included 10 intermediate BP phenotypes analyzed under standardized conditions in a controlled environment including removal of medications. Measurements were repeated with participants lying supine or standing upright compared with office sitting BP readings in most investigations. Moreover, stringent significance levels correcting for multiple testing and permutations were used to evaluate these findings. We also adopted a model free linkage statistical method that makes no assumptions about distribution of the data. Furthermore, postural changes in h2, genetic covariance, and linkage signals in genes were seen across different traits.
In addition, to further assess the effect of chance on postural linkage variation, we implemented a permutation test that evaluated changes in linkage scores (signal) over time. For this test, all trait linkage scores, obtained during 30 min of lying supine and 10 min of standing upright, with a particular SNP, were rearranged randomly in time sequence (e.g., the first 5-min linkage score for SBP and a particular SNP were exchanged for the linkage score obtained at 4 min of standing and so on) 10,000 times for each SNP-phenotype couple. The test found that one-third of linkage signals were not due to chance but caused by the change in posture. Thus, it is unlikely that our results are merely the effect of random variation in linkage signals.
It is well known that important regulatory mechanisms of vascular function, such as nonprostanoid, nonnitric oxide, endothelium-dependent vasorelaxation, can only be demonstrated after pharmacological or physiological blockade of other vasoactive pathways (10), as we have demonstrated in a rodent model of hypertension (438). The present study strongly indicates a shift in the genetic determinants of BP and intermediate BP phenotypes influenced by posture. Upright posture is evolutionary speaking a recent event, thus, it could be in less genetic equilibrium than supine posture. Thus, that the “selection pressure” induced by postural adaptation may have resulted in changes of allele frequencies and polymorphic gene variation (23). It is unclear if the “new” genetic architecture designed to tolerate the standing position also resulted in a predisposition to develop hypertensive disorders.
In conclusion, physiological adaptations to postural change from lying supine to standing upright are accompanied by a distinct genetic architecture of BP and intermediate BP phenotypes. Discovery of genes linked to BP will be incomplete if only a single posture is included in genetic studies.
GRANTS
The Quebec Hypertensive Family Study was initially funded by NIH/NCBLI SCORE and is currently funded by the Canadian Institutes of Health Research (P. Hamet, J. Tremblay, and O. Šeda), the Canadian Foundation for Innovation (P. Hamet and J. Tremblay), and the Canada Research Chair (P. Hamet). I. A. Arenas was supported by a postdoctoral fellowship from the Canadian Institutes of Health Research.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
AUTHOR CONTRIBUTIONS
Author contributions: I.A.A., J.T., B.D., O.S., D.G., E.M., T.A.K., A.W.C., and P.H. conception and design of research; I.A.A., J.T., B.D., and J.S. analyzed data; I.A.A., J.T., O.S., E.M., T.A.K., and P.H. interpreted results of experiments; I.A.A. prepared figures; I.A.A. and J.S. drafted manuscript; I.A.A., J.T., J.S., O.S., D.G., E.M., T.A.K., A.W.C., and P.H. edited and revised manuscript; I.A.A., J.T., B.D., J.S., O.S., D.G., E.M., T.A.K., A.W.C., and P.H. approved final version of manuscript; J.S. and P.H. performed experiments.
Supplementary Material
ACKNOWLEDGMENTS
Thanks go to the instructors of the Wellcome Trust Consortium Genetics Advanced Course for Analysis of Complex Traits for helpful suggestions and project discussion. We are indebted to Professor A. Ziegler and Dr. Mark E. Samuels for critical analysis of the results, and to Ovid M. Da Silva for editing the manuscript.
Footnotes
The online version of this article contains supplemental material.
REFERENCES
- 1. Abecasis GR, Cherny SS, Cookson WO, Cardon LR. Merlin - rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet 30: 97–101, 2002 [DOI] [PubMed] [Google Scholar]
- 2. Almasy L, Blangero J. Multipoint quantitative-trait linkage analysis in general pedigrees. Am J Hum Genet 62: 1198–1211, 1998 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Belmonte M, Yurgelun-Todd D. Permutation testing made practical for functional magnetic resonance image analysis. IEEE Trans Med Imaging 20: 243–248, 2001 [DOI] [PubMed] [Google Scholar]
- 4. Blair RC, Karniski W. Distribution-free statistical analyses of surface and volumetric maps. Functional Neuroimaging: Technical Foundations. Academic, 1994, p. 19–28 [Google Scholar]
- 5. Carey G, Williamson J. Linkage analysis of quantitative traits: increased power by using selected samples. Am J Hum Genet 49: 786–796, 1991 [PMC free article] [PubMed] [Google Scholar]
- 6. Cowley AW., Jr The genetic dissection of essential hypertension. Nat Rev Genet 7: 829–840, 2006 [DOI] [PubMed] [Google Scholar]
- 7. Deslauriers B, Antoniol G, Merlo E, Gossard F, Pausova Z, Gaudet D, Tremblay J, Kotchen TA, Cowley AW, Hamet P. Dynamic linkage of the angiotensin II type I receptor gene locus in essential hypertension. Abstract. Am J Hypertens 16: 16A, 2003 [Google Scholar]
- 8. Dudding T, Heron J, Thakkinstian A, Nurk E, Golding J, Pembrey M, Ring SM, Attia J, Scott RJ. Factor V Leiden is associated with pre-eclampsia but not with fetal growth restriction: a genetic association study and meta-analysis. J Thromb Haemost 6: 1869–1875, 2008 [DOI] [PubMed] [Google Scholar]
- 9. Ehret GB. Genome-wide association studies: contribution of genomics to understanding blood pressure and essential hypertension. Curr Hypertens Rep 12: 17–25, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Feletou M, Vanhoutte PM. The third pathway: endothelium-dependent hyperpolarization. J Physiol Pharmacol 50: 525–534, 1999 [PubMed] [Google Scholar]
- 11. Guyton AC, Coleman TG, Granger HJ. Circulation: overall regulation. Annu Rev Physiol 34: 13–46, 1972 [DOI] [PubMed] [Google Scholar]
- 12. Halushka MK, Fan JB, Bentley K, Hsie L, Shen N, Weder A, Cooper R, Lipshutz R, Chakravarti A. Patterns of single-nucleotide polymorphisms in candidate genes for blood-pressure homeostasis. Nat Genet 22: 239–247, 1999 [DOI] [PubMed] [Google Scholar]
- 13. Hamet P, Tremblay J, Pang SC, Garcia R, Thibault G, Gutkowska J, Cantin M, Genest J. Effect of native and synthetic atrial natriuretic factor on cyclic GMP. Biochem Biophys Res Commun 123: 515–527, 1984 [DOI] [PubMed] [Google Scholar]
- 14. Hamet P. Environmentally-regulated genes of hypertension. Clin Exp Hypertens 18: 267–278, 1996 [DOI] [PubMed] [Google Scholar]
- 15. Hamet P, Merlo E, Seda O, Broeckel U, Tremblay J, Kaldunski M, Gaudet D, Bouchard G, Deslauriers B, Gagnon F, Antoniol G, Pausova Z, Labuda M, Jomphe M, Gossard F, Tremblay G, Kirova R, Tonellato P, Orlov SN, Pintos J, Platko J, Hudson TJ, Rioux JD, Kotchen TA, Cowley AW., Jr Quantitative founder-effect analysis of French Canadian families identifies specific loci contributing to metabolic phenotypes of hypertension. Am J Hum Genet 76: 815–832, 2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Harrap SB, Cui JS, Wong ZY, Hopper JL. Familial and genomic analyses of postural changes in systolic and diastolic blood pressure. Hypertension 43: 586–591, 2004 [DOI] [PubMed] [Google Scholar]
- 17. He LN, Liu YJ, Xiao P, Zhang L, Guo Y, Yang TL, Zhao LJ, Drees B, Hamilton J, Deng HY, Recker RR, Deng HW. Genomewide linkage scan for combined obesity phenotypes using principal component analysis. Ann Hum Genet 72: 319–326, 2008 [DOI] [PubMed] [Google Scholar]
- 18. Head GA. Baroreflexes and cardiovascular regulation in hypertension. J Cardiovasc Pharmacol 26, Suppl 2: S7–S16, 1995 [PubMed] [Google Scholar]
- 19. International Consortium for Blood Pressure Genome-wide Association Studies Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk. Nature 478: 103–109, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Iwai N, Tago N, Yasui N, Kokubo Y, Inamoto N, Tomoike H, Shioji K. Genetic analysis of 22 candidate genes for hypertension in the Japanese population. J Hypertens 22: 1119–1126, 2004 [DOI] [PubMed] [Google Scholar]
- 21. Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden of hypertension: analysis of worldwide data. Lancet 365: 217–223, 2005 [DOI] [PubMed] [Google Scholar]
- 22. Kotchen TA, Kotchen JM, Grim CE, George V, Kaldunski ML, Cowley AW, Hamet P, Chelius TH. Genetic determinants of hypertension: identification of candidate phenotypes. Hypertension 36: 7–13, 2000 [DOI] [PubMed] [Google Scholar]
- 23. Kroymann J, Mitchell-Olds T. Epistasis, and balanced polymorphism influencing complex trait variation. Nature 435: 95–98, 2005 [DOI] [PubMed] [Google Scholar]
- 24. Kupper N, Willemsen G, Riese H, Posthuma D, Boomsma DI, de Geus EJ. Heritability of daytime ambulatory blood pressure in an extended twin design. Hypertension 45: 80–85, 2005 [DOI] [PubMed] [Google Scholar]
- 25. Nardo CJ, Chambless LE, Light KC, Rosamond WD, Sharrett AR, Tell GS, Heiss G. Descriptive epidemiology of blood pressure response to change in body position: the ARIC study. Hypertension 33: 1123–1129, 1999 [DOI] [PubMed] [Google Scholar]
- 26. Noël A, Seda O, Tremblay J, Gaudet D, Kotchen TA, Cowley AW, Hamet P. The impact of pharmacological environment on capacity to identify the genetic components of hypertension. Abstract. 23rd Meeting of the International Society of Hypertension Vancouver, Canada, 2010 [Google Scholar]
- 27. Okuda T, Fujioka Y, Kamide K, Kawano Y, Goto Y, Yoshimasa Y, Tomoike H, Iwai N, Hanai S, Miyata T. Verification of 525 coding snps in 179 hypertension candidate genes in the Japanese population: identification of 159 snps in 93 genes. J Hum Genet 47: 387–394, 2002 [DOI] [PubMed] [Google Scholar]
- 28. O'Shaughnessy KM. Dissecting complex traits: recent advances in hypertension genomics. Genome Med 1: 43, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Padmanabhan S, Melander O, Johnson T, Di Blasio AM, Lee WK, Gentilini D, Hastie CE, Menni C, Monti MC, Delles C, Laing S, Corso B, Navis G, Kwakernaak AJ, van der Harst P, Bochud M, Maillard M, Burnier M, Hedner T, Kjeldsen S, Wahlstrand B, Sjogren M, Fava C, Montagnana M, Danese E, Torffvit O, Hedblad B, Snieder H, Connell JM, Brown M, Samani NJ, Farrall M, Cesana G, Mancia G, Signorini S, Grassi G, Eyheramendy S, Wichmann HE, Laan M, Strachan DP, Sever P, Shields DC, Stanton A, Vollenweider P, Teumer A, Volzke H, Rettig R, Newton-Cheh C, Arora P, Zhang F, Soranzo N, Spector TD, Lucas G, Kathiresan S, Siscovick DS, Luan J, Loos RJ, Wareham NJ, Penninx BW, Nolte IM, McBride M, Miller WH, Nicklin SA, Baker AH, Graham D, McDonald RA, Pell JP, Sattar N, Welsh P, Munroe P, Caulfield MJ, Zanchetti A, Dominiczak AF. Genome-wide association study of blood pressure extremes identifies variant near umod associated with hypertension. PLoS Genet 6: e1001177, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Pausova Z, Gaudet D, Gossard F, Bernard M, Kaldunski ML, Jomphe M, Tremblay J, Hudson TJ, Bouchard G, Kotchen TA, Cowley AW, Hamet P. Genome-wide scan for linkage to obesity-associated hypertension in French Canadians. Hypertension 46: 1280–1285, 2005 [DOI] [PubMed] [Google Scholar]
- 31. Rice T, Rankinen T, Province MA, Chagnon YC, Perusse L, Borecki IB, Bouchard C, Rao DC. Genome-wide linkage analysis of systolic and diastolic blood pressure: the Quebec Family Study. Circulation 102: 1956–1963, 2000 [DOI] [PubMed] [Google Scholar]
- 32. Rothwell PM, Howard SC, Dolan E, O'Brien E, Dobson JE, Dahlöf B, Sever PS, Poulter NR. Prognostic significance of visit-to-visit variability, maximum systolic blood pressure, and episodic hypertension. Lancet 375: 895–905, 2010 [DOI] [PubMed] [Google Scholar]
- 33. Seda O, Liska F, Krenova D, Kazdova L, Sedova L, Zima T, Peng J, Pelinkova K, Tremblay J, Hamet P, Kren V. Dynamic genetic architecture of metabolic syndrome attributes in the rat. Physiol Genomics 21: 243–252, 2005 [DOI] [PubMed] [Google Scholar]
- 34. Seda O, Tremblay J, Gaudet D, Brunelle PL, Gurau A, Merlo E, Pilote L, Orlov SN, Boulva F, Petrovich M, Kotchen TA, Cowley AW, Jr, Hamet P. Systematic, genome-wide, sex-specific linkage of cardiovascular traits in French Canadians. Hypertension 51: 1156–1162, 2008 [DOI] [PubMed] [Google Scholar]
- 35. Shete S, Jacobs KB, Elston RC. Adding further power to the Haseman and Elston method for detecting linkage in larger sibships: weighting sums and differences. Hum Hered 55: 79–85, 2003 [DOI] [PubMed] [Google Scholar]
- 36. The International HapMap Consortium A haplotype map of the human genome. Nature 437: 1299–1320, 2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Tremblay J, Huot C, Willenbrock RC, Bayard F, Gossard F, Fujio N, Koch C, Kuchel O, Debinski W, Hamet P. Increased cyclic guanosine monophosphate production and overexpression of atrial natriuretic peptide A-receptor mRNA in spontaneously hypertensive rats. J Clin Invest 92: 2499–2508, 1993 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Ueno T, Tremblay J, Kunes J, Zicha J, Dobesova Z, Pausova Z, Deng AY, Sun YL, Jacob HJ, Hamet P. Resolving the composite trait of hypertension into its pharmacogenetic determinants by acute pharmacological modulation of blood pressure regulatory systems. J Mol Med 81: 51–60, 2003 [DOI] [PubMed] [Google Scholar]
- 39. Wang X, Kammerer CM, Anderson S, Lu J, Feingold E. A comparison of principal component analysis and factor analysis strategies for uncovering pleiotropic factors. Genet Epidemiol 33: 325–331, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Wigginton JE, Cutler DJ, Abecasis GR. A note on exact tests of Hardy-Weinberg equilibrium. Am J Hum Genet 76: 887–893, 2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Wood B. Origin and evolution of the genus Homo. Nature 355: 783–790, 1992 [DOI] [PubMed] [Google Scholar]
- 42. Xu J, Desir GV. Renalase, a new renal hormone: its role in health and disease. Curr Opin Nephrol Hypertens 16: 373–378, 2007 [DOI] [PubMed] [Google Scholar]
- 43. Zhao Q, Fan Z, He J, Chen S, Li H, Zhang P, Wang L, Hu D, Huang J, Qiang B, Gu D. Renalase gene is a novel susceptibility gene for essential hypertension: a two-stage association study in northern Han Chinese population. J Mol Med 85: 877–885, 2007 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.