Ambulatory Blood Pressure Is Associated With Polymorphic ...hyper.ahajournals.org/content/hypertensionaha/52/5/980.full.pdf · Ambulatory Blood Pressure Is Associated With Polymorphic
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Ambulatory Blood Pressure Is Associated With PolymorphicVariation in P2X Receptor Genes
Julian Palomino-Doza, Thahira J. Rahman, Peter J. Avery, Bongani M. Mayosi, Martin Farrall,Hugh Watkins, Christopher R.W. Edwards, Bernard Keavney
Abstract—The P2X receptor gene family encodes a series of proteins that function as ATP-gated nonselective ion channels.P2X receptor channels are involved in transducing aldosterone-mediated signaling in the distal renal tubule and arepotential candidate genes for blood pressure regulation. We performed a family based quantitative genetic associationstudy in 248 families ascertained through a proband with hypertension to investigate the relationship between commongenetic variation in the P2X4, P2X6, and P2X7 genes and ambulatory blood pressure. We genotyped 28 singlenucleotide polymorphisms, which together captured the common genetic variability in the 3 genes. We corrected ourresults for multiple comparisons specifying a false discovery rate of 5%. We found significant evidence of associationbetween the single nucleotide polymorphism rs591874 in the first intron of the P2X7 gene and blood pressure. Thestrongest association was found for nighttime diastolic blood pressure (P�0.0032), although association was present forboth systolic and diastolic blood pressures measured by an observer during the day and at night. Genotypes wereassociated with a 0.2 SD (�2.5 mm Hg) difference in night diastolic blood pressure per allele and accounted for �1%of the total variability in this measurement. Other suggestive associations were found, but these were nonsignificant aftercorrection for multiple testing. These genetic data suggest that drugs affecting P2X receptor signaling may have promiseas clinical antihypertensive agents. (Hypertension. 2008;52:980-985.)
Hypertension is a multifactorial complex disorder involv-ing interactions between genetic and environmental
determinants. The heritability of systolic and diastolic bloodpressures (BPs) has been estimated at between 30% and60%.1 Case-control studies of hypertension as a “diseasestate,” a widely used design, introduce an artificial dichotomyinto the analysis of a quantitative trait. Even very largecase-control genetic studies of hypertension have not, thusfar, been successful in identifying susceptibility genes.2
However, BP considered quantitatively presents specificdifficulties as a phenotype for study because of the uncer-tainty in estimating an individual’s “usual” BP from a limitednumber of measurements. Ambulatory BP measurement hasbeen shown to provide more reproducible estimates of usualBP than isolated clinical readings, and a quantitatively basedapproach using ambulatory BP measurement, therefore, hasthe potential to increase the power of genetic studies.3,4
Purinergic P2X receptors are nonselective ATP-gated ionchannels that are widely expressed in mammalian cells.Molecular studies have identified 7 P2X receptor genes that
form both homotrimeric and heterotrimeric complexes invivo.5 The P2X4 gene is abundantly expressed in vascularendothelial cells,6 where it is the major mediator of Ca2�
influx induced by ATP and blood flow. P2X4 knockout micehave marked suppression of normal endothelial cell responsesto blood flow, have deficient downstream adaptive vascularremodeling, and develop hypertension.7 P2X4 receptors arealso expressed in cardiac myocytes8 and in the renal collect-ing duct.9 Aldosterone stimulates the extracellular release ofATP from the basolateral membrane of renal tubular cells,which acts via P2X4 or P2X4-like receptors to mediatecontraction of the adjacent epithelial cells, resulting in trans-epithelial sodium transport via ENaC.10
The amino acid sequence of P2X7 is closely related toP2X4, and the 2 genes lie adjacent to each other separated bysome 24 kb on chromosome 12, suggesting that they mayhave evolved by a process of gene duplication. The P2X7gene is expressed in response to injury (eg, diabetic nephrop-athy) in the podocytes, mesangial cells, and glomerularendothelial cells of the kidney and in vascular endothelial
Received March 18, 2008; first decision April 12, 2008; revision accepted September 15, 2008.From the Institute of Human Genetics (J.P.-D., T.J.R., B.K.), School of Mathematics and Statistics (P.J.A.), and Office of the Vice Chancellor
(C.R.W.E.), Newcastle University, Newcastle upon Tyne, United Kingdom; Department of Medicine (B.M.M.), University of Cape Town, Cape Town,South Africa; and the Department of Cardiovascular Medicine (M.F., H.W.), University of Oxford, Oxford, United Kingdom.
The first 2 authors contributed equally to this article.Correspondence to Bernard Keavney, Institute of Human Genetics, Central Parkway, Newcastle upon Tyne, NE1 3BZ United Kingdom. E-mail
cells. P2X7 is constitutively expressed in the hypothalamusand nucleus tractus solitarius, indicating a potential role forreceptors incorporating P2X7 subunits in the regulation ofsympathoadrenal tone.
P2X6 receptor subunits coassemble with P2X4 subunits,11
and the P2X6 gene is expressed in the distal renal tubule andthe collecting duct principal cells. Some previous genome-wide linkage scans for BP have detected linkage in the regionof chromosome 22 that includes the P2X6 gene.12–14 TheP2X4, P2X6, and P2X7 genes are, therefore, plausible can-didate genes for BP regulation. We have investigated theimpact of common variation in these genes on BP in a large,family based association study using ambulatory BPmonitoring.
MethodsThe collection strategy of this family study has been describedpreviously.15 Families were ascertained through a proband diagnosedwith essential hypertension. The following criteria defined eligibilityas a proband: a mean systolic BP �140 mm Hg and mean diastolicBP �90 mm Hg on daytime ambulatory BP monitoring, �3 officeBP readings �160 mm Hg systolic and 95 mm Hg diastolic, ortreatment with �2 antihypertensive drugs. These relatively stringentcriteria were applied to provide maximum security that probandswere indeed at the upper end of the population BP distribution.Secondary hypertension was excluded using the standard screeningprotocol applied in the hypertension clinic.
To be suitable for the study, families were required to consist of�3 siblings clinically assessable for BP if �1 parent of the sibshipwas available to give blood for DNA analysis and to consist of �4assessable siblings if no parent was available for DNA analysis.Quantitatively assessed sibships were recruited either in the gener-ation of the proband or his or her offspring. Where members of thesibship were found to be hypertensive, families were extended andthe spouses and offspring of hypertensive siblings collected. Themajority of individuals in the family collection, therefore, had BPswithin the conventionally accepted “normal range,” and the familycollection included some extended families, although most werenuclear families.15 A total of 1425 subjects from 248 families wererecruited. The median family size was 5 people, with 60% of familiescomposed of between 4 and 6 genotyped and phenotyped members.A total of 71% of families were 2 generation, and 29% were 3generation. Eighty-four percent of families had an assessable sibshipin the generation of the proband, whereas 16% of families consistedof a proband and their nuclear family (spouse and children �18years) only. The study received ethical clearance from the appropri-ate review committees and corresponded with the principles of theDeclaration of Helsinki. All of the participants gave informedconsent to participate in the study.
BP was measured using ambulatory monitoring for a period of 24hours in all of the subjects willing to undergo monitoring, using theA&D TM2421 monitor according to a protocol described previously16
(see data supplement available online at http://hyper.ahajournals.orgfor additional detail). A full clinical history was taken, whichincluded the subject’s medical history and lifestyle factors, includingconsumption of alcohol and tobacco, as well as habitual physicalexercise. Anthropometric measurements, including height, weight,and waist and hip circumferences, were performed. DNA wasextracted from blood samples using standard methods.
The P2X4 and P2X7 genes are located adjacent to each other onthe long arm of chromosome 12 (12q24.32 and 12q24.2, respec-tively), and the P2X6 gene is located on chromosome 22 (22q11.21).Tag single nucleotide polymorphisms (SNPs) within each gene and15 kb in either direction (to incorporate close-range upstream anddownstream regulatory sequences) were identified by reference tothe SNP data from the HapMap CEU samples of Northern andWestern European ancestry (http://www.hapmap.org).17 The taggingstrategy was implemented with the Tagger utility in the Haploview
software package, with the parameters r2�0.8 and minor allelefrequency �0.05.18 14 tag SNPs were required in P2X7, 5 tag SNPsin P2X4, and 6 tag SNPs in P2X6. Selected SNPs, their location inthe genes, and linkage disequilibrium between the SNPs (calculatedusing Haploview from our data) are shown in the Figure.
Twenty-five SNPs were genotyped by the MassEXTEND methodon a Sequenom MassArray matrix assisted laser desorptionionization-time of flight platform. The primers were designed usingthe MassArray assay designer software (version 3.0.2.0, Sequenom).Details of primers for these assays are shown in Table S1. ThreeSNPs for which satisfactory Sequenom assays could not be devel-oped were genotyped using restriction fragment length polymor-phism assays. DNA sequence for these SNPs was downloaded fromthe Ensembl database (http://www.ensembl.org)19 and manuallyanalyzed for the presence of restriction sites and to determine primersequences. PCR primers, annealing temperatures, and restrictionenzymes used are shown in Table S2. Control individuals of knowngenotypes were included in every plate, and 100 randomly selectedsamples were genotyped twice for each polymorphism. Genotypingwas carried out blinded to phenotypic information.
Mendelian inheritance of all of the genotypes and Hardy-Weinberg equilibrium for each marker were checked using PED-STATS.20 Additional checks based on unlikely recombination pat-terns within families were carried out using the error-checking optionin MERLIN version 1.1.1.21 Errors were corrected when possible byreference to the raw genotyping data, and when this was not possible,genotypes were excluded from analysis. We used MINITAB version14 to examine the BP phenotypes for normality. All of the BPvariables required log-transformation to adequately conform to anormal distribution (see Figure S1). We used MINITAB to adjust theBP phenotypes for the significant covariates age, sex, and habitualphysical activity using linear regression. The main analyses includedonly those individuals in whom BP measurements free from antihy-pertensive treatment were available, so no adjustment for drugtreatment was required. Supplementary analyses including bothuntreated participants and participants taking antihypertensive med-ications were performed; in these analyses, the effects of each of themain classes of drugs (diuretics and �-blockers) on BP wereestimated from the data by regression and the appropriate adjustmentmade to the on-treatment BP values. The log-transformed, covariate-adjusted residuals were entered into the quantitative trait geneticassociation analyses, which were performed (for both untreated-onlyand all-subject analyses) using a variance-components approach,which takes account of shared polygenic effects in members of thesame pedigree, implemented in the QTDT program.22 We first testedfor the presence of population stratification for each SNP bycomparing the between- and within-family components of variancein QTDT. Because there was no evidence of stratification (P�0.05),we used the total association model, which incorporates bothcomponents, in the principal analyses. We also conducted subsidiaryanalyses using the “orthogonal” or transmission/disequilibrium testmodel specifiable in QTDT, which tests only transmission fromheterozygote parents and, thus, uses only a fraction of the availablegenotype/phenotype information; despite the considerable loss ofpower, if such transmission/disequilibrium test analyses are positive,they potentially provide an additional level of security in the result.To take some account of the risk of false-positive findings becauseof multiple comparisons, we interpreted the results based on the falsediscovery rate.23 We used the program QVALUE running on top ofthe statistical package R to determine q values24 (see data supple-ment). We adopted an arbitrary false discovery rate threshold of 0.05(ie, 1 in 20 of the associations passing this criterion were expected tobe false).
ResultsThe ambulatory BP measurements in the participants aresummarized in Table 1. Fifty-two percent of the participantswere women, and 36% were classified as hypertensive. Asexpected given the selection of families through a hyperten-sive proband, ambulatory BPs tended to be higher than would
Palomino-Doza et al Hypertension and P2X Genes 981
by guest on June 4, 2018http://hyper.ahajournals.org/
Figure. SNPs typed in the P2X4, P2X7 (A), and P2X6 (B) genes, and linkage disequilibrium relationships between them. Exons are rep-resented by horizontal bars. The linkage disequilibrium relationships between SNPs are represented by Haploview triangle plots; darkersmall squares denote higher linkage disequilibrium (represented by r2), and the numbers in the small squares show the significant pair-wise D� values derived from the present study.
982 Hypertension November 2008
by guest on June 4, 2018http://hyper.ahajournals.org/
be expected in a nonselected population. Quantitative day-time recordings of ambulatory BP were available in 958people (of whom 126 were pedigree founders), with on-treatment recordings available in a further 212. Because ofparticipants electing to switch off the monitor at night, fewernight recordings were available: quantitative night recordingswere available in 770 people (of whom 100 were pedigreefounders), with on-treatment recordings available in a further158. The sibling correlations for the traits ranged from 22% to36% for the different ambulatory measures and were compa-rable with previously published family data.25,26
Genotyping was successful for �98% of the samples forall of the SNPs. The estimated genotype miscall rate was�1% for all of the SNPs. All 25 of the markers were inHardy-Weinberg equilibrium at the 5% significance level.Table S3 shows genotype frequencies and counts for thepopulation. The minor allele frequencies ranged from 1.9% to49.0%, yielding marker heterozygosities from 3.6% to 51.0%.
Allele frequencies were very similar to the HapMap data forthe CEU population.17 As expected from the tagging strategythat we used, the correlation between SNPs was generallymodest (Figure) and corresponded well with previous pub-licly available data. Haplotypes with frequencies of �1% inthe population are shown in Tables S4 and S5. Twenty-fivesuch haplotypes were present in the region of chromosome 12spanning P2X4 and P2X7, accounting for �85% of thegenetic variation present. Seventeen such haplotypes werepresent in the region of chromosome 22 spanning P2X6,accounting for 96% of variation. The composition and fre-quency of the common haplotypes were in close agreementwith those described in the HapMap CEU population.
Using data from participants who were not taking antihy-pertensive medications, we found significant associationbetween genotypes at the SNP rs591874 in the P2X7 geneand diastolic BP, whether measured by the “clinic readings,”during the day or at night (P�0.015, 0.0072, and 0.0032,respectively, among 887, 915, and 742 genotyped and quan-titatively phenotyped individuals; QTDT “total” associationmodel; Table 2). Somewhat less significant associations werefound between the marker and systolic BP in all 3 of thesettings (P�0.028, 0.049, and 0.035, respectively; Table 2).Each rs591874 minor allele was associated with a higherlog-transformed, covariate-adjusted diastolic BP as measuredby “clinic” readings, daytime and nighttime recordings by0.126, 0.138, and 0.173 SDs, respectively. In our sample, thiswould correspond with an increment of between 2 and3 mm Hg in diastolic BP per allele. Genotype at rs591874accounted for �1% of the variability in the diastolic BPmeasurements. Association was also found between themarker rs656612 in the P2X7 gene and nighttime diastolic BP(P�0.0051; among 397 common homozygotes, 295 het-erozygotes, and 53 rare homozygotes). At this SNP, eachminor allele was associated with a 0.172-SD increase of BP.Genotype at rs656612 accounted for 1.15% of the variabilityof nighttime diastolic BP. The rs656612 SNP is situated closeto rs591874 in intron 1 of the P2X7 gene, and linkagedisequilibrium between these 2 SNPs is strong (D��0.86;r2�0.73); when genotype at rs591874 was included in the
Table 1. Ambulatory BPs in Members of 248 ParticipatingFamilies
Variable N LQ Median UQSibling
Correlation
Age, y 1425* 35.7 50.9 60.9
Clinic systolicBP, mm Hg
1182† 121.3 134.0 153.0 0.362
Clinic diastolicBP, mm Hg
1182† 73.7 82.0 92.0 0.264
Daytime systolicBP, mm Hg
1182† 123.3 134.3 149.7 0.265
Daytime diastolicBP, mm Hg
1182† 73.3 80.7 92.0 0.214
Night systolicBP, mm Hg
932‡ 104.6 114.0 128.0 0.319
Night diastolicBP, mm Hg
932‡ 60.6 67.0 75.0 0.329
LQ indicates lower quartile; UQ, upper quartile.*52.4% were female and 36.1% were classified as hypertensive.†Comprising 958 quantitative and 224 nonquantitative recordings.‡Comprising 770 quantitative and 162 nonquantitative recordings.
Table 2. Association Between rs591874 Genotypes and BPs
Data are means�SEs (n) of log-transformed covariate-adjusted residuals for each BP phenotype. A includes untreated participants only for all 6 of the principalBP phenotypes. B shows the data, in treated and untreated participants, regarding the associations that were significant after adjustment for multiple comparisonsin the untreated participants.
Palomino-Doza et al Hypertension and P2X Genes 983
by guest on June 4, 2018http://hyper.ahajournals.org/
model, the effect of rs656612 was no longer significant.Some nominally significant associations (P�0.05) were pres-ent for other SNPs (Table S6); however, after correcting formultiple tests using QVALUE, only the associations betweenrs591874 and rs656612 with diastolic BP remained signifi-cant, whereas other associations became nonsignificant at the5% false discovery rate level. As expected given the loss ofapproximately half of the genotype/phenotype informationentailed in performing the TDT analyses, these did notprovide additional strong evidence of association(0.04�P�0.1).
In supplementary analyses including participants on andoff treatment (using identical statistical methodology), theassociation between the marker rs591874 with day and nightdiastolic BP became, if anything, slightly more significant(P�0.0028 and 0.00098, respectively, among 1126 and 894genotyped and phenotyped individuals; Table 2). This wasalso the case for the association between night diastolic BPand the marker rs656612 (P�0.0016; N�888).
DiscussionWe have found association between common variation in thegenomic segment including the P2X7 and P2X4 genes andBP. Each minor allele of the markers rs591874 and rs656612increases diastolic BP by �0.2 SD in a codominant fashion.These associations remain significant after correction formultiple testing. The effects are small, genotypes accountingfor only 1% of the trait’s variability. The SNPs rs591874 andrs656612 are both located in intron 1 of the P2X7 gene, andthe degree of linkage disequilibrium is high between them(r2�0.75). When both SNPs were included in statisticalmodels, the association appeared principally to be attributableto the rs591874 genotype. Despite the suggestion fromprevious linkage studies that the region of chromosome 22harboring the P2X6 receptor may be involved in susceptibil-ity to hypertension, we found no significant associations withSNPs in P2X6 and BP.
The association that we observed between rs591874 andBP was strongest for nighttime diastolic BP, although it wasnominally significant for both systolic and diastolic BPs,whether measured by the clinic readings, which initiated the24-hour ambulatory monitoring, by the daytime, or by thenighttime ambulatory readings. Our data, in keeping withsome previous studies, show that the heritability of nighttimediastolic ambulatory BP is significantly higher than the otherreadings, possibly because of the lesser effect of randomenvironmental variables, such as physical activity and mentalstressors during the night.27 Our result is, therefore, mostlikely to reflect the increased power of the study to detectgenetic association with the more heritable nighttime diastol-ic values, despite the fact that the numbers of individuals withquantitative nighttime monitoring values available were sig-nificantly lower than the numbers with daytime monitoringsor clinic readings. It seems physiologically less likely that ourresult arises from specific genetic effects on diastolic (asopposed to systolic) and nighttime (as opposed to daytime orclinic) BP, although this cannot be absolutely ruled out.Meta-analysis has shown recently that nighttime BP is abetter predictor of adverse outcome than daytime or clinic
BP, suggesting that genetic associations with nighttime pres-sures might be particularly important.28
Because P2X7 and P2X4 lie adjacent to each other onchromosome 12, it is conceivable that the association that wehave observed between rs591874 and BP arises through aregulatory effect on either or both of these 2 genes. Withrespect to P2X7, among the SNPs we typed was rs3751143,which has been shown by others to result in an �50%decrease in receptor function in the heterozygous state and acomplete loss of receptor function in the homozygous state.5
We observed no association between BP and this SNP, so itseems unlikely that loss of P2X7 function is the mechanismaccounting for our results. We performed in silico analyses toaddress the possibilities that the associated SNPs, which areintronic, could mediate differences in splicing or modifytranscription factor binding sites. Cross-species sequencealignment29 shows quite considerable sequence conservationin intron 1 of P2X7, including the rs591874 SNP, suggestinga potential regulatory role. RESCUE-ESE30 and AutomatedSplice Site Analysis programs31 show that neither of theSNPs are in consensus splice sequences or generate crypticsplice sites, but the minor allele of the rs591874 SNPabolishes a binding site for the splicing protein SC35. Geneticvariants altering SC35 binding sites can have large effects ongene expression.32 Additional studies to quantify the effect ofthe associated SNPs and of additional SNPs (particularly inthe 24-kb regions between the 2 genes, which was notcovered with tagSNPs in this study) on the expression of bothP2X7 and P2X4 may allow the detection of regulatory effectsmediating BP differences.
To our knowledge, this is the first study to systematicallyexamine polymorphic variation in these P2X receptor genesand BP. The relatively large sample size in the present studymakes a false-positive result because of random chance lesslikely. Our focus on quantitative data decreased the potentialeffect of confounding factors (notably, treatment for hyper-tension, although inclusion of treated individuals strength-ened our result), and the use of ambulatory monitor readingsincreased our power to detect small genetic effects. Detailedphenotyping enabled us to control for covariates, and ouranalyses incorporated an adjustment for multiple compari-sons. However, this study has certain limitations. As with anynovel genetic association, these findings will require replica-tion in other large studies; the effect that we have describedis small, so as with certain other genetic associations (eg,PPARG Pro12Ala and type 2 diabetes), the optimal estimateof the effect may be apparent only after several additionalstudies have been performed. The effect that we show(a change of 2 to 3 mm Hg per allele) is not of sufficient sizeto be of use in genetic screening for BP risk; future fine-mapping studies will be needed to conclusively identify theSNP or set of SNPs at the P2X7/P2X4 locus that has maximaleffect on BP. However, even the most strongly associatedSNP in the region, when it is identified, may not have aneffect that is of substantial size. Our study does not indicatewhether the genetic effect on BP arises from upregulation ordownregulation of P2X signaling; future functional studieswill be required to determine this with security.
984 Hypertension November 2008
by guest on June 4, 2018http://hyper.ahajournals.org/
PerspectivesWe have found that common genetic variation in the regionof the P2X7 and P2X4 genes has a small but significant effecton BP in a white population. The associations seem to followcodominant models and account for �1% of BP variation.Genotypes were associated with differences in diastolic BP of�0.2 SD (�2 to 3 mm�Hg) per allele. This is the first clinicalstudy to substantiate previous investigations in laboratoryanimals and in cell lines that suggest an important role of P2Xsignaling in BP homeostasis.
Sources of FundingThe family collection was funded by the Wellcome Trust. Thegenotyping was funded by the British Heart Foundation. H.W. andB.K. hold British Heart Foundation Chairs.
DisclosuresNone.
References1. Lifton R. Molecular genetics of human blood pressure variation.
Science. 1996;272:676–680.2. Wellcome Trust Case Control Consortium. Genome-wide association
study of 14,000 cases of seven common diseases and 3,000 sharedcontrols. Nature. 2007;447:645–646.
3. Coats AJS. Benefits of ambulatory blood pressure monitoring in thedesign of antihypertensive drug trials. Blood Press Monit. 1996;1:157–160.
4. Conway J, Johnston J, Coats A, Somers V, Sleight P. The use of ambu-latory blood pressure monitoring to improve the accuracy and reduce thenumbers of subjects in clinical trials of antihypertensive agents.J Hypertens. 1988;6:111–116.
5. Gu BJ, Sluyter R, Skarratt KK, Shemon AN, Dao-Ung L-P, Fuller SJ,Barden JA, Clarke AL, Petrou S, Wiley JS. An Arg307 to Gln poly-morphism within the ATP-binding site causes loss of function of thehuman P2X7 receptor. J Biol Chem. 2004;279:31287–31295.
6. Ray FR, Huang W, Slater M, Barden JA. Purinergic receptor distributionin endothelial cells in blood vessels: a basis for selection of coronaryartery grafts. Atherosclerosis. 2002;162:55–61.
7. Yamamoto K, Sokabe T, Matsumoto T, Yoshimura K, Shibata M, OhuraN, Fukuda T, Sato T, Sekine K, Kato S, Isshiki M, Fujita T, KobayashiM, Kawamura K, Masuda H, Kamiya A, Ando J. Impaired flow-dependent control of vascular tone and remodeling in P2X4-deficientmice. Nat Med. 2006;12:16–17.
8. Hu B, Mei Q-B, Yao X-J, Smith E, Barry WH, Liang BT. A novelcontractile phenotype with cardiac transgenic expression of the humanP2X4 receptor. FASEB J. 2001;15:2739–2741.
9. Wildman SS, Marks J, Churchill LJ, Peppiatt CM, Chraibi A, Shirley DG,Horisberger J-D, King BF, Unwin RJ. Regulatory interdependence ofcloned epithelial Na� channels and P2X receptors. J Am Soc Nephrol.2005;16:2586–2597.
10. Zhang Y, Sanchez D, Gorelik J, Klenerman D, Lab M, Edwards C,Korchev Y. Basolateral P2X4-like receptors regulate the extracellularATP-stimulated epithelial Na� channel activity in renal epithelia. Am JPhysiol Renal Physiol. 2007;292:F1734–F1740.
11. North RA. Molecular physiology of P2X receptors. Physiol Rev. 2002;82:1013–1067.
12. Perola M, Kainulainen K, Pajukanta P, Terwilliger JD, Hiekkalinna T,Ellonen P, Kaprio J, Koskenvuo M, Kontula K, Peltonen L. Genome-widescan of predisposing loci for increased diastolic blood pressure in Finnishsiblings. J Hypertens. 2000;18:1579–1585.
13. 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. 2000;102:1956–1963.
14. De Lange M, Spector TD, Andrew T. Genome-wide scan for bloodpressure suggests linkage to chromosome 11, and replication of loci on16, 17, and 22. Hypertension. 2004;44:872–877.
15. Keavney B, Mayosi B, Gaukrodger N, Imrie H, Baker M, Fraser R,Ingram M, Watkins H, Farrall M, Davies E, Connell J. Genetic variationat the locus encompassing 11-beta hydroxylase and aldosterone synthaseaccounts for heritability in cortisol precursor (11-deoxycortisol) urinarymetabolite excretion. J Clin Endocrinol Metab. 2005;90:1072–1077.
16. Keavney B, Bird R, Caiazza A, Casadei B, Conway J. Measurement ofblood pressure using the auscultatory and oscillometric methods in thesame cuff deflation: validation and field trial of the A&D TM2421monitor. J Hum Hypertens. 2000;14:573–579.
17. The International Hapmap Consortium. The international hapmap project.Nature. 2003;426:789–796.
18. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visual-ization of LD and haplotype maps. Bioinformatics. 2005;21:263–265.
19. Birney E, Andrews TD, Bevan P, Caccamo M, Chen Y, Clarke L, CoatesG, Cuff J, Curwen V, Cutts T, Down T, Eyras E, Fernandez-Suarez XM,Gane P, Gibbins B, Gilbert J, Hammond M, Hotz HR, Iyer V, Jekosch K,Kahari A, Kasprzyk A, Keefe D, Keenan S, Lehvaslaiho H, McVicker G,Melsopp C, Meidl P, Mongin E, Pettett R, Potter S, Proctor G, Rae M,Searle S, Slater G, Smedley D, Smith J, Spooner W, Stabenau A, StalkerJ, Storey R, Ureta-Vidal A, Woodwark KC, Cameron G, Durbin R, CoxA, Hubbard T, Clamp M. An overview of Ensembl. Genome Research.2004;14:925–928.
20. Wigginton JE, Abecasis GR. PEDSTATS. descriptive statistics, graphicsand quality assessment for gene mapping data. Bioinformatics. 2005;21:3445–3447.
22. Abecasis GR, Cardon LR, Cookson WOC. A general test of associationfor quantitative traits in nuclear families. Am J Hum Genet. 2000;66:279–292.
23. Benjamini Y, Hochberg Y. Multiple hypotheses testing with weights.Scand J Stat. 1997;24:407–418.
24. Storey J. The positive false discovery rate: a Bayesian interpretation andthe q-value. Ann Stat. 2003;31:2013–2035.
25. Knuiman M, Divitini ML, Welborn T, Bartholomew HC. Familial cor-relations, cohabitation effects, and heritability for cardiovascular riskfactors. Ann Epidemiol. 1996;6:188–194.
26. Lee K, Klein BE, Klein R. Familial agregation of components of themultiple metabolic syndrome in the Framingham Heart and OffspringCohorts: Genetic Analysis Workshop Problem I. BMC Genetics. 2003;4(suppl-1):S94–S98.
27. Boggia J, Li Y, Thijs L, Hansen TW, Kikuya M, Bjorklund-Bodegard K,Richart T, Ohkubo T, Kuznetsova T, Torp-Pedersen C, Lind L, Ibsen H,Imai Y, Wang J, Sandoya E, O’Brien E, Staessen JA. Prognostic accuracyof day versus night ambulatory blood pressure: a cohort study. Lancet.2007;370:1219–1229.
28. Fagard RH, Celis H, Thijs L, Staessen JA, Clement DL, De Buyzere ML,De Bacquer DA. Daytime and nighttime blood pressure as predictors ofdeath and cause-specific cardiovascular events in hypertension. Hyper-tension. 2008;51:55–61.
29. Madden TL, Tatusov RL, Zhang J. Applications of network BLASTserver. Meth Enzymol. 1996;266:131–141.
30. Cartegni L, Wang J, Zhu Z, Zhang MQ, Krainer AR. ESEfinder: a webresource to identify exonic splicing enhancers. Nucleic Acids Res. 2003;31:3568–3571.
32. Gabut M, Mine M, Marsac C, Brivet M, Tazi J, Soret J. The SR proteinSC35 is responsible for aberrant splicing of the E1? Pyruvate dehydro-genase mRNA in a case of mental retardation with lactic acidosis. MolCell Biol. 2005;25:3286–3294.
Palomino-Doza et al Hypertension and P2X Genes 985
by guest on June 4, 2018http://hyper.ahajournals.org/
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Hypertension doi: 10.1161/HYPERTENSIONAHA.108.1132822008;52:980-985; originally published online October 13, 2008;Hypertension.
http://hyper.ahajournals.org/content/52/5/980World Wide Web at:
The online version of this article, along with updated information and services, is located on the
is online at: Hypertension Information about subscribing to Subscriptions:
http://www.lww.com/reprints Information about reprints can be found online at: Reprints:
document. Permissions and Rights Question and Answer this process is available in the
click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,
can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialHypertensionin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:
by guest on June 4, 2018http://hyper.ahajournals.org/
AMBULATORY BLOOD PRESSURE IS ASSOCIATED WITH POLYMORPHIC
VARIATION IN P2X RECEPTOR GENES
Addison Julian Palomino Doza, Thahira J. Rahman, Peter J. Avery, Bongani M. Mayosi, Martin
Farrall, Hugh Watkins, Christopher R.W. Edwards, and Bernard Keavney
Contents: Supplementary Methods
Supplementary Tables I – VI
Supplementary Figure I
Supplementary Methods: I. Ambulatory blood pressure monitoring protocol Three readings were taken by the A&D TM-2421 monitor with the patient in a relaxed seated
position at the start of the monitoring period. Simultaneous auscultation was carried out by a trained
observer, to confirm satisfactory (within 5mmHg) agreement between the monitor and auscultatory
values; if this criterion was not met, the cuff was repositioned until satisfactory agreement was
obtained. The three readings which had satisfactory agreement between the monitor and the
observer in the final cuff position are referred to as “clinic readings”. The monitor was programmed
to record blood pressure every half-hour during the daytime and every hour during the night, and a
recording was considered of satisfactory technical quality if at least 20 daytime ambulatory data
points were available for analysis. Patients also recorded the time they went to bed and rose in the
morning to enable individualised calculation of the “daytime” and “night-time” periods. Mean values
for systolic and diastolic blood pressures for the clinic, daytime and night-time periods were
analysed for association with genotypes.
II. Calculation of q-values
Web reference for QVALUE software: http://genomics.princeton.edu/storeylab/qvalue/
The false discovery rate estimates the proportion of false positives among the “discoveries”, that is,
the significant p-values obtained, when multiple tests are being carried out. It differs from the false
positive rate, which estimates the proportion of true null hypotheses that would be incorrectly called
as positives when a particular threshold is applied. The FDR can therefore be expressed as the
number of false positives divided by the number of significant tests. The q-value is the FDR
analogue of the p-value. The q-value of an individual hypothesis test is the minimum FDR value at
which that test can be deemed to be significant: hence, a q-value of 0.05 for a particular test would
indicate that in the context of the multiple tests performed, there is a one in twenty chance that that
test is a false positive. The QVALUE software takes as input a vector of p-values for a series of
hypothesis tests (here, the p-values corresponding to the genetic association tests performed), and
utilises these to calculate the corresponding q-values for each test. The parameter π0, which
represents the proportion of true null hypotheses in the study, can be calculated using a bootstrap
method which is implemented in the QVALUE software.
Supplementary Table S1. List of polymorphisms genotyped by SEQUENOM.
SNP Gene / Location (bp)
Forward primer (5’→3’) Reverse primer (5’→3’) Annealing temp. (oC) / Conc. of MgCl2
SNP ID, function and genomic position of the SNP as in the ENSEMBL database, PCR primer sequences, annealing temperature and concentration of MgCl2 used for PCR, size of PCR amplicon and restriction enzymes used for the RFLP are shown
Supplementary Table S3: Genotype distribution at typed markers
Gene SNP Genotypes No Hom wild type % Heterozygous % Hom non wild type %P2X7 rs591874 1338 727 54.335 494 36.921 117 8.744