Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Aug 13.
Published in final edited form as: J Hypertens. 2021 Apr 1;39(4):633–642. doi: 10.1097/HJH.0000000000002701

Associations of biogeographic ancestry with hypertension traits

Jacob M Keaton a,b,c,d,e, Jacklyn N Hellwege b,c,d,e, Ayush Giri c,d,e,f, Eric S Torstenson b,e, Csaba P Kovesdy g, Yan V Sun h, Peter WF Wilson i, Christopher J O’Donnell j, Todd L Edwards b,c,d,e,*, Adriana M Hung e,k,*, Digna R Velez Edwards c,d,e,f,l,*, on behalf of the Million Veteran Program
PMCID: PMC8362794  NIHMSID: NIHMS1724038  PMID: 33534346

Abstract

Objectives:

Ethnic disparities in hypertension prevalence are well documented, though the influence of genetic ancestry is unclear. The aim of this study was to evaluate associations of geographic genetic ancestry with hypertension and underlying blood pressure traits.

Methods:

We tested genetically inferred ancestry proportions from five 1000 Genomes reference populations (GBR, PEL, YRI, CHB, and LWK) for association with four continuous blood pressure (BP) traits (SBP, DBP, PP, MAP) and the dichotomous outcomes hypertension and apparent treatment-resistant hypertension in 220 495 European American, 59 927 African American, and 21 273 Hispanic American individuals from the Million Veteran Program. Ethnicity stratified results were meta-analyzed to report effect estimates per 10% difference for a given ancestry proportion in all samples.

Results:

Percentage GBR was negatively associated with BP (P = 2.13 × 10−19, 7.92 × 10−8, 4.41 × 10−11, and 3.57 × 10−13 for SBP, DBP, PP, and MAP, respectively; coefficient range −0.10 to −0.21 mmHg per 10% increase in ancestry proportion) and was protective against hypertension [P = 2.59 × 10−5, odds ratio (OR) = 0.98] relative to other ancestries. YRI percentage was positively associated with BP (P = 1.63 × 10−23, 1.94 × 10−26, 0.012, and 3.26 × 10−29 for SBP, DBP, PP, and MAP, respectively; coefficient range 0.06–0.32 mmHg per 10% increase in ancestry proportion) and was positively associated with hypertension risk (P = 3.10 × 10−11, OR = 1.04) and apparent treatment-resistant hypertension risk (P = 1.86 × 10−4, OR = 1.04) compared with other ancestries. Percentage PEL was inversely associated with DBP (P = 2.84 × 10−5, beta = −0.11 mmHg per 10% increase in ancestry proportion).

Conclusion:

These results demonstrate that risk for BP traits varies significantly by genetic ancestry. Our findings provide insight into the geographic origin of genetic factors underlying hypertension risk and establish that a portion of BP trait ethnic disparities are because of genetic differences between ancestries.

Keywords: admixture, blood pressure, genetic ancestry, hypertension, resistant hypertension

INTRODUCTION

Hypertension accounts for 45% of heart disease mortality and 51% of stroke mortality [1], whereas complications of hypertension account for 9.4 million deaths annually and cost billions of dollars in lost productivity and healthcare spending worldwide [2,3]. Common measures of hypertension include SBP, DBP, pulse pressure (PP), and mean arterial pressure (MAP). Together, these traits capture the systemic, pulsatile, and steady components that characterize blood pressure.

Although clinical factors, such as age, sex, and ethnicity as well as modifiable risk factors including physical activity, obesity, diet, alcohol consumption, and smoking behavior contribute to risk [4–7], the genetic component to variance in hypertension traits is well documented. Heritability estimates for hypertension traits range from 18 to 56% [8–10]. Genome-wide association studies (GWAS) have identified more than 900 loci for hypertension and related traits [11,12].

Ethnic disparities in hypertension prevalence are well known with prevalence among African Americans (40.5%) much higher compared with European Americans (27.4%) and Hispanic Americans (25.1%) [13]. Research on the genetic factors underlying these differences has been limited. European Americans are predominantly of European ancestry, African Americans are typically an admixture of African ancestry and European ancestry, and Hispanic Americans have a combination of Native American, African, and European ancestry [14]. However, ancestry proportions vary widely across individuals from each of these three ethnicities. We hypothesize that genetically inferred ancestry, not self-reported or administratively determined ethnicity, is a risk factor for hypertension and represents the geographic origin of alleles that contribute to those disparities. Here, we report the results of trans-ethnic association analyses of ancestry proportions based on five 1000 Genome reference populations representing European, Native American, West African, East African, and East Asian ancestries estimated in 301 695 VA Million Veteran Program (MVP) participants with blood pressure traits, hypertension, and apparent treatment-resistant hypertension.

RESEARCH DESIGN AND METHODS

Study participants

MVP participants were recruited as previously described [15]. Briefly, self-reported or administratively determined African American, European American, and Hispanic American MVP participants were ascertained as active Veterans Health Administration (VHA) users from selected VHA healthcare sites who provided informed consent [15]. Blood pressure measurements, hypertension status, and apparent treatment-resistant hypertension status were obtained from electronic health records (EHRs) for study participants over the age of 18 years.

SBP was calculated for each individual as the median across measurements censoring emergency department measures, inpatient measures, measures if pain score was at least 5, and measures occurring at or after International Classification of Diseases ninth edition (ICD-9) diagnostic codes 585 (chronic kidney disease), 405 (secondary hypertension), or 428 (heart failure). We then queried the EHR for the earliest instance of the median SBP from eligible measures and selected the coetaneous DBP measure. For patients on antihypertension drugs prior to the measure, 15 mmHg was added to SBP and 10 mmHg was added to DBP [16,17]. PP was calculated as the difference of SBP and DBP. MAP was calculated as the sum of SBP and twice DBP quantity divided by three.

Hypertension (HTN) cases were defined as participants with a hypertension ICD-9 code in their EHR, on an antihypertensive drug, or having two SBP measures at least 140 mmHg and/or two DBP measures at least 90 mmHg. The definition of hypertension is based on American Heart Association guidelines prior to November 2017. Normotensive controls were defined as participants not on an antihypertensive medication, no hypertension ICD-9 diagnostic code in their EHR, and no SBP measures greater than 140 mmHg or DBP measures greater than 90 mmHg. Apparent treatment-resistant hypertension (rHTN) cases were defined as participants failing to achieve controlled BP on three antihypertensive drugs including a thiazide diuretic or taking four or more medications regardless of achieving control. Non-rHTN controls were hypertensive patients who achieved BP control on one or two medications.

Single nucleotide polymorphism genotyping and quality control

DNA was extracted from whole blood drawn from MVP participants at the Central Biorepository in Boston, Massachusetts, USA and shipped to two external centers for genotyping. Genotyping of the MVP samples was performed using a custom Affymetrix Axiom Biobank Array as previously described [15]. The custom array includes assays for approximately 723 000 genetic variants and is enriched for validated GWAS single nucleotide polymorphisms (SNPs), exome content, and Hispanic and African ancestry markers. Standard quality control procedures and genotype calling of the data in batches using algorithms from Affymetrix Power Tools Suite (v1.18) were performed by the MVP genomics working. Sample exclusion criteria included genotypic duplicates, excess heterozygosity, call rate below 97.5%, and discordance between genetically inferred sex and phenotypic sex. Closely related individuals (halfway between second and third degree relatives or closer) as measured using the KING software program [18] were removed. Variants with low call rate or that diverged from allele frequencies in 1000 Genomes Project [19] reference data were excluded from subsequent analyses.

Estimation of global genetic ancestry proportions

Ancestry proportions were estimated using ADMIXTURE [20] (k = 5) with genotypes from the selected 1000 Genomes Project [19] phase 3, version five populations British in England and Scotland (GBR), Peruvians from Lima, Peru (PEL), Yoruba in Ibadan, Nigeria (YRI), Han Chinese in Beijing, China (CHB), and Luhya in Webuye, Kenya (LWK) as reference [20]. Genotyped variants passing quality control in each MVP population (European American, African American, and Hispanic American) intersecting with reference variants were pruned for linkage disequilibrium and used as input.

Analytic methods

Associations with global genetic ancestry proportions were computed using R [21]. Continuous blood pressure traits (SBP, DBP, PP, and MAP) were modeled using linear regression against each ancestry proportion (GBR, PEL, YRI, CHB, and LWK) for each ethnicity (African American, European American, and Hispanic American). Dichotomous hypertension outcomes (HTN, rHTN) were modeled using logistic regression against each ancestry proportion for each ethnicity. All models were adjusted for age, age2, sex, and BMI. Ancestry-specific effects within ethnicity groups were combined by fixed-effect inverse variance weighting meta-analysis for each hypertension trait using PLINK 1.9 [22]. Effect estimates are reported per 10% increase for a given inferred ancestry proportion in all samples.

Mendelian randomization

The causality of genetic ancestry for hypertension was assessed in an independent study, Vanderbilt’s BioVU resource. The BioVU repository is a collection of stored DNA linked to de-identified EHRs at Vanderbilt University Medical Center, a resource, which currently includes more than 240 000 samples for the investigation of phenotype–genotype associations [23]. Analyses included 75 392 individuals of European, African, and Asian descent genotyped using the Illumina Multi-Ethnic Genotyping Array (MEGA; Illumina, Inc., San Diego, California, USA). Samples with a call rate below 98% or discordant between genetically inferred sex and phenotypic sex were excluded from analyses. SNPs with a call rate below 98% were also excluded. Genotype data were pruned for linkage disequilibrium using a window size of 50 base pairs (bp) shifting by 10 bp at an r2 threshold of 0.1. Genotype data were merged with 1000 Genomes reference data and randomly thinned to include 100 000 variants [24].

Unsupervised ADMIXTURE analysis (k = 3) of 1000 Genomes reference genotype data was performed and African, European, and Asian ancestry proportions were calculated. These ancestry proportions were then projected onto BioVU samples in ADMIXTURE.

Missing genotypes in BioVU were imputed using the Michigan Imputation Server (MIS) (https://imputationserver.sph.umich.edu/) based on the minimac4 algorithm and the 1000 genomes phase 3 version 5 genotype reference set [24,25]. Linear regression of African (AFR) ancestry as the outcome was modeled to estimate genetic effects with covariate adjustments for age, sex, and BMI using SNPTEST v2.5.4-beta [26]. Estimates for the association of 54 549 linkage disequilibrium-pruned variants with AFR ancestry (P < 5 × 10−8) were considered in as instruments in Mendelian randomization. Selected instruments met a minimum imputation quality threshold (info >0.4).

Two-sample Mendelian randomization was performed in R using the MRbase package [27]. Estimates for the association of AFR ancestry variants with SBP in MVP were obtained from GWAS summary statistics published by Giri et al. [28]. Mendelian randomization estimates were transformed to represent change in SBP in mmHg per 10% increase in ancestry proportion. Analyses were performed using the inverse variance weighted method. Sensitivity analyses included Mendelian randomization-Egger regression to address horizontal pleiotropy and to support directionality of the causal relationship.

RESULTS

Global genetic ancestry proportions in MVP participants corresponding to five 1000 Genomes reference populations (GBR, PEL, YRI, CHB, and LWK) were tested for association with SBP, DBP, PP, MAP, HTN, and rHTN. Collectively, these analyses included 301 695 individuals from three ethnicities (African American, European American, and Hispanic American). Characteristics of study participants by ethnicity are presented in Table 1. Most participants were men (92%). Participants had an average age of 59 ± 13 years (mean ± SD) and an average BMI of 30 ± 5.8 kg/m2. On average, 69% of study participants were diagnosed with hypertension with proportions higher among African Americans (72%) and European Americans (71%) compared with Hispanic Americans (45%). The increased proportion of participants on antihypertensive medication among African Americans (74%) and Hispanic Americans (93%) compared with European Americans (69%) may be because of treatment bias caused by clinician awareness of ethnic disparity in hypertension risk. Apparent treatment-resistant hypertension was diagnosed in 6% of study participants with proportions higher in African Americans (8%) compared with European Americans (5%) and Hispanic Americans (4%). Among European Americanss, the largest ancestry proportion was GBR with an average proportion of 0.96 ± 0.08, compared with YRI in African Americans (0.72 ± 0.15) and GBR in Hispanic Americans (0.52 ± 0.21).

TABLE 1.

Demographics of the study participants

Characteristic European Americans (N = 220 495) African Americans (N = 59 927) Hispanic Americans (N = 21,273)
Male (%) 204 840 (92.9%) 52 196 (87.1%) 19 380 (91.1%)
Age (years) mean (SD) 58.9 (12.6) 60.6 (11.4) 52.7 (14.5)
BMI (kg/m2) mean (SD) 30.2 (5.8) 30.1 (6.0) 30.6 (5.6)
Type 2 diabetes (%) 50 714 (23.0%) 16 720 (27.9%) 5786 (27.2%)
Hypertensive (%) 156 641 (71.0%) 43 098 (71.9%) 9477 (44.5%)
On antihypertensive medication (%) 107 873 (68.9%) 31 748 (73.7%) 8831 (93.2%)
Drug-resistant hypertensive (%) 11 762 (5.3%) 5071 (8.5%) 901 (4.2%)
SBP (mmHg) mean (SD) 138 (15) 140 (17) 135 (16)
DBP (mmHg) mean (SD) 82 (11) 85 (13) 82 (11)
PP (mmHg) mean (SD) 57 (13) 55 (12) 53 (12)
MAP (mmHg) mean (SD) 100 (11) 103 (12) 99 (12)
Genetic ancestry proportions – mean (SD)
 Peruvians from Lima (PEL) 0.01 (0.05) 0.01 (0.02) 0.35 (0.23)
 British in UK (GBR) 0.96 (0.08) 0.18 (0.14) 0.52 (0.21)
 Han Chinese in Beijing (CHB) 0.01 (0.03) 0.01 (0.02) 0.03 (0.07)
 Yoruban in Nigeria (YRI) 0.01 (0.04) 0.72 (0.15) 0.07 (0.13)
 Luhya in Kenya (LWK) 0.01 (0.01) 0.08 (0.06) 0.03 (0.04)
Open in a new tab

MAP, mean arterial pressure; PP, pulse pressure; SD, standard deviation.

Among European American participants, we observed an association of GBR ancestry with lower SBP (P = 2.58 × 10−3, beta = −0.11 mmHg per 10% increase in ancestry proportion) and PP (P = 1.73 × 10−3, beta = −0.09 mmHg per 10% increase in ancestry proportion) (Figs. 1–6, Supplementary Tables 1 and 4, http://links.lww.com/HJH/B500). Among African Americans, GBR ancestry was also associated with lower SBP (P = 7.08 × 10−21, beta = −0.41 mmHg per 10% increase in ancestry proportion), as well as lower DBP (P = 7.36 × 10−26, beta = −0.35 mmHg per 10% increase in ancestry proportion), and MAP (P = 8.80 × 10−28, beta = −0.52 mmHg per 10% increase in ancestry proportion) and was protective for HTN (P = 2.01 × 10−10, OR = 0.95 per 10% increase in ancestry proportion) and rHTN (P = 1.90 × 10−3, OR = 0.96 per 10% increase in ancestry proportion) (Figs. 1–6, Supplementary Tables 1 and 4, http://links.lww.com/HJH/B500). In contrast, YRI ancestry was associated with higher SBP (P = 3.13 × 10−19, beta = 0.38 mmHg per 10% increase in ancestry proportion), DBP (P = 7.08 × 10−24, beta = 0.33 mmHg per 10% increase in ancestry proportion), and MAP (P = 5.95 × 10−26, beta = 0.35 mmHg per 10% increase in ancestry proportion) and risk for HTN (P = 3.84 × 10−10, OR = 1.05 per 10% increase in ancestry proportion) and rHTN (P = 5.80 × 10−3, OR = 1.04 per 10% increase in ancestry proportion) among African American participants (Figs. 1–6, Supplementary Tables 2 and 5, http://links.lww.com/HJH/B500).

FIGURE 1.

FIGURE 1

Open in a new tab

Ancestry associations with SBP. Forest plots show mean and 95% confidence interval of ancestry associations from meta-analysis and ethnicity-specific analyses.

FIGURE 6.

FIGURE 6

Open in a new tab

Ancestry associations with drug-resistant hypertension. Forest plots show mean and 95% confidence interval of ancestry associations from meta-analysis and ethnicity-specific analyses.

Consistent with observations in European Americans and African Americans, GBR ancestry was associated with lower SBP (P = 2.56 × 10−4, beta = −0.17 mmHg per 10% increase in ancestry proportion) and PP (P = 3.75 × 10−11, beta = −0.23 mmHg per 10% increase in ancestry proportion), and YRI ancestry was associated with risk for rHTN (P = 0.03, OR = 1.10 per 10% increase in ancestry proportion) among Hispanic American participants (Figs. 1–6, Supplementary Tables 3 and 6, http://links.lww.com/HJH/B500).

Meta-analysis of the three ethnicity strata (European American, African American, and Hispanic American) further supported the inference that AFR ancestry negatively influences hypertension traits and GBR ancestry is protective. GBR ancestry was associated with lower SBP (P = 2.13 × 10−19, beta = −0.21 mmHg per 10% increase in ancestry proportion), DBP (P = 7.92 × 10−8, beta = −0.10 mmHg per 10% increase in ancestry proportion), PP (P = 4.41 × 10−11, beta = −0.12 mmHg per 10% increase in ancestry proportion), and MAP (P = 3.57 × 10−13, beta = −0.13 mmHg per 10% increase in ancestry proportion) and was protective for HTN (P = 2.59 × 10−5, OR = 0.98 per 10% increase in ancestry proportion) (Figs. 1–6, Tables 2 and 3). YRI ancestry was associated with lower SBP (P = 1.63 × 10−23, beta = 0.32 mmHg per 10% increase in ancestry proportion), DBP (P = 1.94 × 10−26, beta = 0.26 mmHg per 10% increase in ancestry proportion), PP (P = 0.01, beta = 0.06 mmHg per 10% increase in ancestry proportion), and MAP (P = 3.26 × 10−29, beta = 0.28 mmHg per 10% increase in ancestry proportion) and risk for HTN (P = 3.10 × 10−11, OR = 1.04 per 10% increase in ancestry proportion) and for rHTN (P = 1.86 × 10−4, OR = 1.04 per 10% increase in ancestry proportion) (Figs. 1–6, Tables 2 and 3).

TABLE 2.

Meta-analyzed ancestry associations with blood pressure traits

Trait SBP DBP PP MAP
N 301 695 301 315 301 301 301 369
British in UK (GBR) Beta (SE) −0.21 (0.02) −0.10 (0.02) −0.12 (0.02) −0.13 (0.02)
P 2.13E-19 7.92E-08 4.41E-11 3.57E-13
P HET <1.00E-04 <1.00E-04 9.00E-04 <1.00E-04
Peruvians from Lima (PEL) Beta (SE) 0.09 (0.03) −0.11 (0.03) 0.21 (0.03) −0.04 (0.03)
P 6.68E-03 2.84E-05 3.69E-15 1.40E-01
P HET 1.40E-03 5.34E-01 4.00E-04 8.14E-02
Yoruban in Nigeria (YRI) Beta (SE) 0.32 (0.03) 0.26 (0.02) 0.06 (0.02) 0.28 (0.02)
P 1.63E-23 1.94E-26 1.21E-02 3.26E-29
P HET <1.00E-04 1.60E-03 <1.00E-04 <1.00E-04
Han Chinese in Beijing (CHB) Beta (SE) 0.05 (0.08) 0.13 (0.06) −0.07 (0.06) 0.10 (0.06)
P 4.77E-01 2.13E-02 2.39E-01 8.40E-02
P HET 6.00E-04 1.93E-01 4.80E-03 1.44E-02
Luhya in Kenya (LWK) Beta (SE) −0.13 (0.09) −0.05 (0.07) −0.07 (0.07) −0.10 (0.07)
P 1.46E-01 4.35E-01 2.72E-01 1.38E-01
P HET <1.00E-04 1.00E-04 <1.00E-04 <1.00E-04
Open in a new tab

PHET, P value for heterogeneity.

TABLE 3.

Meta-analyzed ancestry associations with hypertension and apparent treatment-resistant hypertension

Trait HTN rHTN
N cases 20 9216 17 734
N controls 77 793 57 501
GBR OR (95% Cl) 0.98 (0.97–0.99) 0.98 (0.97–1.00)
P 2.59E-05 5.54E-02
P HET <1.00E-04 5.04E-02
PEL OR (95% Cl) 1.00 (0.99–1.02) 0.98 (0.95–1.00)
P 6.19E-01 9.53E-02
P HET 2.33E-02 9.87E-01
YRI OR (95% Cl) 1.04 (1.03–1.06) 1.04 (1.02–1.07)
P 3.10E-11 1.86E-04
P HET 9.43E-02 4.59E-01
CHB OR (95% Cl) 0.97 (0.94–1.00) 1.02 (0.95–1.09)
P 5.61E-02 6.41E-01
P HET 5.38E-01 7.51E-01
LWK OR (95% Cl) 0.97 (0.94–1.00) 1.02 (0.96–1.08)
P 8.25E-02 5.02E-01
P HET <1.00E-04 2.06E-02
Open in a new tab

CI, confidence interval; OR, odds ratio; PHET, P value for heterogeneity.

Mendelian randomization analyses were performed to examine the causal relationship between African biogeographic ancestry and hypertension. Inverse variance-weighted Mendelian randomization showed a significant increase in SBP (P = 7.71 × 10−8, beta = 1.10 × 10−3 per 10% increase in AFR ancestry). The Egger intercept did not indicate the presence of horizontal pleiotropy (P = 0.32, intercept 4.52 × 10−4).

DISCUSSION

Risk factors for hypertension include age, sex, BMI, diet, physical activity, alcohol consumption, smoking behavior, socioeconomic status, and, most importantly for this study, ethnicity and genetic factors [4–7]. The contributions of genetic ancestry to ethnic disparity in hypertension risk are not well understood. We performed association analyses of genetic ancestry corresponding to five representative 1000 Genome populations with four blood pressure traits, hypertension, and apparent treatment-resistant hypertension. Subsequent meta-analyses reveal compelling evidence of West African ancestry as a risk factor for hypertension, and European ancestry as protective against the development of hypertension. A significant BP-lowering association with a consistent direction of effect across our three study populations (European American, African American, Hispanic American) was observed for GBR ancestry with SBP, PP, and MAP and for PEL ancestry with DBP. GBR ancestry was also protective for HTN with a consistent direction of effect across study populations. A significant BP-raising or hypertension risk association with a consistent direction of effect across populations was observed for YRI ancestry with DBP, MAP, HTN, and rHTN.

Increased hypertension risk in African Americans is well documented [29,30]. Historically, individuals from ancestral African populations with traditional lifestyles exhibit lower mean BP with little increase because of increasing age and are at low risk of hypertension [31]. However, several studies have shown variation in hypertension risk among African populations with lower risk within rural populations to higher risk within urban centers [31–35]. These findings combined with the results of our study suggests that the effect of genetic variation on hypertension risk in African ancestry individuals may be modified by environmental factors. Additionally, these results suggest that ancestry proportions influence blood pressure traits regardless of the cultural context in which they may exist, as we see consistent effects in ethnically and culturally distinct groups. This finding also supports a model where adaptive processes in distinct geographic ranges result in geographic differences in allele frequencies for blood pressure genetic variants, resulting in ethnic disparities in modern populations that now share a geographic domain and some lifestyle and cultural exposures.

Our finding of association of YRI ancestry with increased rHTN risk may explain some differences in drug response. Previous studies have shown increased efficacy of hypertension treatments in individuals with specific genotypes at variants with large differences in minor allele frequency (MAF) across ethnic groups. For example, two recent studies in individuals of European descent have shown that a polymorphisms in the aldosterone synthase gene CYP11B2 (rs1799998, 1000 Genomes Project Phase 3 European-ancestry MAF = 0.486, African-ancestry MAF = 0.189) predict response to the hypertension medications candesartan and irbesartan [24,36,37]. In individuals of African descent, a polymorphism of the epithelial sodium channel gene SCNN1B (rs1799979, 1000 Genomes Project Phase 3 European-ancestry MAF = 0.000, African-ancestry MAF = 0.024), predicts response to the hypertension medication amiloride [38].

This study has limitations. Although biogeographic ancestry was computed using genetic markers, it is not clear that the observed effect was not because of environmental factors correlated with genetic ancestry. Any study of genetic ancestry must carefully consider environment, socioeconomic status, and culture to effectively distinguish genetic and nongenetic effects [39]. Environmental factors influencing blood pressure, which may have culturally specific effects (e.g. antihypertensive medication type, specific mental health outcomes, baseline stress levels, sleep deprivation, socioeconomic factors, etc.) were not considered in our analyses because of lack of availability. Further genomic study at the local ancestry level and SNP level, as well as modeling of rich environmental data, is needed to further characterize the exact genetic factors, which may underly our observed effects.

We also observed that maximum mean ancestry proportion was highest among European American participants (GBR = 0.96) compared with African American participants (YRI = 0.72) and Hispanic American participants (GBR = 0.52) because of varying historical admixture within each group. These differences, combined with differences in sample size, lead to an inequality in power between populations to detect associations with BP outcomes. To overcome these limitations, we have meta-analyzed our results and focused on associations with a consistent direction of effect across sample populations. Further development of the phenotyping algorithm is necessary to better capture these factors. MVP is constituted of older and predominantly male individuals, which may limit extrapolation of our results to other populations. This limitation highlights the need for additional studies in other populations to further characterize these findings. Cohort studies sampling the overall population may provide a valuable estimate of hypertension prevalence disparities explained by genetic factors underlying biogeographic ancestry.

Despite these limitations, mendelian randomization analyses suggest that biogeographic ancestry is causal for increased blood pressure. We observed a significant and causal relationship of AFR ancestry with SBP. The mendelian randomization approach is subject to confounding by a pleiotropic mechanism. mendelian randomization–Egger regression sensitivity analysis indicated an intercept near zero, thus suggesting an absence of horizontal pleiotropy.

Although previous studies have identified associations of African ancestry with hypertension risk [40], this study inimitably provides evidence of association of genetically-inferred ancestry estimated from multiple geographic parental populations with hypertension risk and underlying blood pressure traits. These results establish that geographic genetic ancestry is a risk factor for hypertension and underlying BP traits and provide insight into the origin of genetic factors fundamental to hypertension and apparent treatment-resistant hypertension risk. Thus, these findings support hypotheses that some proportion of hypertension and blood pressure trait ethnic disparities are because of genetic differences between geographic groups.

Supplementary Material

Supplementary Tables

FIGURE 2.

FIGURE 2

Open in a new tab

Ancestry associations with DBP. Forest plots show mean and 95% confidence interval of ancestry associations from meta-analysis and ethnicity-specific analyses.

FIGURE 3.

FIGURE 3

Open in a new tab

Ancestry associations with pulse pressure. Forest plots show mean and 95% confidence interval of ancestry associations from meta-analysis and ethnicity-specific analyses.

FIGURE 4.

FIGURE 4

Open in a new tab

Ancestry associations with mean arterial pressure. Forest plots show mean and 95% confidence interval of ancestry associations from meta-analysis and ethnicity-specific analyses.

FIGURE 5.

FIGURE 5

Open in a new tab

Ancestry associations with hypertension. Forest plots show mean and 95% confidence interval of ancestry associations from meta-analysis and ethnicity-specific analyses.

ACKNOWLEDGEMENTS

Million Veteran Program (MVP):

MVP Executive Committee

  1. Co-Chair: J. Michael Gaziano, MD, MPH

  2. Co-Chair: Rachel Ramoni, DMD, ScD

  3. Jim Breeling, MD (ex-officio)

  4. Kyong-Mi Chang, MD

  5. Grant Huang, PhD

  6. Sumitra Muralidhar, PhD

  7. Christopher J. O’Donnell, MD, MPH

  8. Philip S. Tsao, PhD

MVP Program Office

  1. Sumitra Muralidhar, PhD

  2. Jennifer Moser, PhD

MVP Recruitment/Enrollment

  1. Recruitment/Enrollment Director/Deputy Director, Boston - Stacey B. Whitbourne, PhD; Jessica V. Brewer, MPH

  2. MVP Coordinating Centers
    1. Clinical Epidemiology Research Center (CERC), West Haven - John Concato, MD, MPH
    2. Cooperative Studies Program Clinical Research Pharmacy Coordinating Center, Albuquerque - Stuart Warren, J.D., Pharm D; Dean P. Argyres, MS
    3. Genomics Coordinating Center, Palo Alto - Philip S. Tsao, PhD
    4. Massachusetts Veterans Epidemiology Research Information Center (MAVERIC), Boston - J. Michael Gaziano, MD, MPH
    5. MVP Information Center, Canandaigua - Brady Stephens, MS

  3. Core Biorepository, Boston - Mary T. Brophy MD, MPH; Donald E. Humphries, PhD

  4. MVP Informatics, Boston - Nhan Do, MD; Shahpoor Shayan

  5. Data Operations/Analytics, Boston - Xuan-Mai T. Nguyen, PhD

MVP Science

  1. Genomics - Christopher J. O’Donnell, MD, MPH; Saiju Pyarajan PhD; Philip S. Tsao, PhD

  2. Phenomics - Kelly Cho, MPH, PhD

  3. Data and Computational Sciences - Saiju Pyarajan, PhD

  4. Statistical Genetics - Elizabeth Hauser, PhD; Yan Sun, PhD; Hongyu Zhao, PhD

MVP Local Site Investigators

- Atlanta VA Medical Center (Peter Wilson)

  1. Bay Pines VA Healthcare System (Rachel McArdle)

  2. Birmingham VA Medical Center (Louis Dellitalia)

  3. Cincinnati VA Medical Center (John Harley)

  4. Clement J. Zablocki VA Medical Center (Jeffrey Whittle)

  5. Durham VA Medical Center (Jean Beckham)

  6. Edith Nourse Rogers Memorial Veterans Hospital (John Wells)

  7. Edward Hines, Jr. VA Medical Center (Salvador Gutierrez)

  8. Fayetteville VA Medical Center (Gretchen Gibson)

  9. VA Healthcare Upstate New York (Laurence Kaminsky)

  10. New Mexico VA Healthcare System (Gerardo Villareal)

  11. VA Boston Healthcare System (Scott Kinlay)

  12. VA Western New York Healthcare System (Junzhe Xu)

  13. Ralph H. Johnson VA Medical Center (Mark Hamner)

  14. Wm. Jennings Bryan Dorn VA Medical Center (Kathlyn Sue Haddock)

  15. VA North Texas Healthcare System (Sujata Bhushan)

  16. Hampton VA Medical Center (Pran Iruvanti)

  17. Hunter Holmes McGuire VA Medical Center (Michael Godschalk)

  18. Iowa City VA Healthcare System (Zuhair Ballas)

  19. Jack C. Montgomery VA Medical Center (Malcolm Buford)

  20. James A. Haley Veterans’ Hospital (Stephen Mastorides)

  21. Louisville VA Medical Center (Jon Klein)

  22. Manchester VA Medical Center (Nora Ratcliffe)

  23. Miami VA Healthcare System (Hermes Florez)

  24. Michael E. DeBakey VA Medical Center (Alan Swann)

  25. Minneapolis VA Healthcare System (Maureen Murdoch)

  26. N. FL/S. GA Veterans Health System (Peruvemba Sriram)

  27. Northport VA Medical Center (Shing Yeh)

  28. Overton Brooks VA Medical Center (Ronald Wash-burn)

  29. Philadelphia VA Medical Center (Darshana Jhala)

  30. Phoenix VA Healthcare System (Samuel Aguayo)

  31. Portland VA Medical Center (David Cohen)

  32. Providence VA Medical Center (Satish Sharma)

  33. Richard Roudebush VA Medical Center (John Call-aghan)

  34. Salem VA Medical Center (Kris Ann Oursler)

  35. San Francisco VA Healthcare System (Mary Whooley)

  36. South Texas Veterans Healthcare System (Sunil Ahuja)

  37. Southeast Louisiana Veterans Healthcare System (Amparo Gutierrez)

  38. Southern Arizona VA Healthcare System (Ronald Schifman)

  39. Sioux Falls VA Healthcare System (Jennifer Greco)

  40. St. Louis VA Healthcare System (Michael Rauchman)

  41. Syracuse VA Medical Center (Richard Servatius)

  42. VA Eastern Kansas Healthcare System (Mary Oehlert)

  43. VA Greater Los Angeles Healthcare System (Agnes Wallbom)

  44. VA Loma Linda Healthcare System (Ronald Fernando)

  45. VA Long Beach Healthcare System (Timothy Morgan)

  46. VA Maine Healthcare System (Todd Stapley)

  47. VA New York Harbor Healthcare System (Scott Sherman)

  48. VA Pacific Islands Healthcare System (Gwenevere Anderson)

  49. VA Palo Alto Healthcare System (Philip Tsao)

  50. VA Pittsburgh Healthcare System (Elif Sonel)

  51. VA Puget Sound Healthcare System (Edward Boyko)

  52. VA Salt Lake City Healthcare System (Laurence Meyer)

  53. VA San Diego Healthcare System (Samir Gupta)

  54. VA Southern Nevada Healthcare System (Joseph Fayad)

  55. VA Tennessee Valley Healthcare System (Adriana Hung)

  56. Washington DC VA Medical Center (Jack Lichy)

  57. W.G. (Bill) Hefner VA Medical Center (Robin Hurley)

  58. White River Junction VA Medical Center (Brooks Robey)

  59. William S. Middleton Memorial Veterans Hospital (Robert Striker)

The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the U.S. Department of Health and Human Services. This publication does not represent the views of the Department of Veterans Affairs or the United States Government.

L.E. and D.R.V.E. were supported by NIH/NHLBI grant HL121429. This research is based on data from the Million Veteran Program, Office of Research and Development, Veterans Health Administration, and was supported by funding from the Department of Veterans Affairs Office of Research and Development, Million Veteran Program Grant I01BX003360 to A.M.H. J.M.K. is supported by the Vanderbilt Genomic Medicine Training Program, funded by T32HG008341 (PI: JC Denny). J.N.H. is supported by the Vanderbilt Molecular and Genetic Epidemiology of Cancer (MAGEC) training program, funded by T32C7A160056 (PI: X-O Shu). The work was in part supported by the Building Interdisciplinary Research Careers in Women’s Healthcareer development program’s 2K12HD043483-17 (PI: KE Hartmann) to A.G. Y.V.S. and P.W.F.W. were funded by the Veterans Affairs Merit Award I01-01BX003340 (PI: PWF Wilson). Y.V.S. was supported by NIH grant NR013520. This work was supported using resources and facilities of the VA Informatics and Computing Infrastructure (VINCI), VA HSR RES 13-457. This publication does not represent the views of the Department of Veterans Affairs or the United States Government.

Abbreviations:

BP

blood pressure

CHB

Han Chinese in Beijing, China

EHR

electronic health record

GBR

British in England and Scotland

GWAS

genome-wide association study

HTN

hypertension

ICD-9

International Classification of Diseases ninth edition

LWK

Luhya in Webuye, Kenya

MAP

mean arterial pressure

MVP

Million Veteran Program

PEL

Peruvians from Lima, Peru

PP

pulse pressure

rHTN

apparent treatment-resistant hypertension

VHA

Veterans Health Administration

YRI

Yoruba in Ibadan, Nigeria

Footnotes

Conflicts of interest

There are no conflicts of interest.

REFERENCES

  • 1.Causes of Death 2008. World Health Organization. Available at: http://www.who.int/healthinfo/global_burden_disease/cod_2008_sources_methods.pdf. [Accessed 5 October 2020]
  • 2.Lim SS, Vos T, Flaxman AD, Danaei G, Shibuya K, Adair-Rohani H, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet Lond Engl 2012; 380:2224–2260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Organisation mondiale de la santé. Global status report on noncom-municable diseases 2010. Geneva: World Health Organization; 2011. [Google Scholar]
  • 4.Rossier BC, Bochud M, Devuyst O. The hypertension pandemic: an evolutionary perspective. Physiol Bethesda Md 2017; 32:112–125. [DOI] [PubMed] [Google Scholar]
  • 5.Slama M, Susic D, Frohlich ED. Prevention of hypertension. Curr Opin Cardiol 2002; 17:531–536. [DOI] [PubMed] [Google Scholar]
  • 6.Pipe A Tobacco addiction and hypertension. J Hum Hypertens 1996; 10 (Suppl 2):S13–S16. [PubMed] [Google Scholar]
  • 7.Grotto I, Huerta M, Sharabi Y. Hypertension and socioeconomic status. Curr Opin Cardiol 2008; 23:335–339. [DOI] [PubMed] [Google Scholar]
  • 8.Levy D, DeStefano AL, Larson MG, O’Donnell CJ, Lifton RP, Gavras H, et al. Evidence for a gene influencing blood pressure on chromosome 17. Genome scan linkage results for longitudinal blood pressure phenotypes in subjects from the Framingham Heart Study. Hypertension 2000; 36:477–483. [DOI] [PubMed] [Google Scholar]
  • 9.Salfati E, Morrison AC, Boerwinkle E, Chakravarti A. Direct estimates of the genomic contributions to blood pressure heritability within a population-based cohort (ARIC). PloS One 2015; 10:e0133031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Alwan H, Ehret G, Ponte B, Pruijm M, Ackermann D, Guessous I, et al. Heritability of ambulatory and office blood pressure in the Swiss population. J Hypertens 2015; 33:2061–2067. [DOI] [PubMed] [Google Scholar]
  • 11.MacArthur J, Bowler E, Cerezo M, Gil L, Hall P, Hastings E, et al. The new NHGRI-EBI Catalog of published genome-wide association studies (GWAS Catalog). Nucleic Acids Res 2017; 45:D896–D901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Evangelou E, Warren HR, Mosen-Ansorena D, Mifsud B, Pazoki R, Gao H, et al. Genetic analysis of over 1 million people identifies 535 new loci associated with blood pressure traits. Nat Genet 2018; 50:1412–1425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Centers for Disease Control and Prevention (CDC). Racial/ethnic disparities in prevalence, treatment, and control of hypertension–United States, 1999–2002. MMWR Morb Mortal Wkly Rep 2005; 54:7–9. [PubMed] [Google Scholar]
  • 14.Bryc K, Durand EY, Macpherson JM, Reich D, Mountain JL. The Genetic Ancestry of African Americans, Latinos, and European Americans across the United States. Am J Hum Genet 2015; 96:37–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Gaziano JM, Concato J, Brophy M, Fiore L, Pyarajan S, Breeling J, et al. Million Veteran Program: a mega-biobank to study genetic influences on health and disease. J Clin Epidemiol 2016; 70:214–223. [DOI] [PubMed] [Google Scholar]
  • 16.Levy D, Ehret GB, Rice K, Verwoert GC, Launer LJ, Dehghan A, et al. Genome-wide association study of blood pressure and hypertension. Nat Genet 2009; 41:677–687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Taylor JY, Schwander K, Kardia SLR, Arnett D, Liang J, Hunt SC, et al. A Genome-wide study of blood pressure in African Americans accounting for gene-smoking interaction. Sci Rep 2016; 6:18812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Manichaikul A, Mychaleckyj JC, Rich SS, Daly K, Sale M, Chen W-M. Robust relationship inference in genome-wide association studies. Bioinforma Oxf Engl 2010; 26:2867–2873. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.The 1000 Genomes Project Consortium: Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO. A global reference for human genetic variation. Nature 2015; 526:68–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Alexander DH, Novembre J, Lange K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res 2009; 19:1655–1664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2015. ; Available at: https://www.R-project.org/. [Accessed 5 October 2020]. [Google Scholar]
  • 22.Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets. GigaScience 2015; 4:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Roden DM, Pulley JM, Basford MA, Bernard GR, Clayton EW, Balser JR, Masys DR. Development of a large-scale de-identified DNA biobank to enable personalized medicine. Clin Pharmacol Ther 2008; 84:362–369. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.1000 Genomes Project Consortium: Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, et al. A global reference for human genetic variation. Nature 2015; 526:68–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Das S, Forer L, Schönherr S, Sidore C, Locke AE, Kwong A, et al. Next-generation genotype imputation service and methods. Nat Genet 2016; 48:1284–1287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Marchini J, Howie B, Myers S, McVean G, Donnelly P. A new multipoint method for genome-wide association studies by imputation of genotypes. Nat Genet 2007; 39:906–913. [DOI] [PubMed] [Google Scholar]
  • 27.Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. Elife 2018; 7:e34408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Giri A, Hellwege JN, Keaton JM, Park J, Qiu C, Warren HR, et al. Transethnic association study of blood pressure determinants in over 750,000 individuals. Nat Genet 2019; 51:51–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Hall WD, Ferrario CM, Moore MA, Hall JE, Flack JM, Cooper W, et al. Hypertension-related morbidity and mortality in the southeastern United States. Am J Med Sci 1997; 313:195–209. [DOI] [PubMed] [Google Scholar]
  • 30.Comstock GW. An epidemiologic study of blood pressure levels in a biracial community in the southern United States. 1956.Am J Epidemiol 1995; 141:584–628. [DOI] [PubMed] [Google Scholar]
  • 31.Kaufman J, Barkey N. Hypertension in Africa: an overview of prevalence rates and causal risk factors. Ethn Dis 1993; 3 (Suppl):S83–S101. [PubMed] [Google Scholar]
  • 32.Lisk DR, Williams DE, Slattery J. Blood pressure and hypertension in rural and urban Sierra Leoneans. Ethn Dis 1999; 9:254–263. [PubMed] [Google Scholar]
  • 33.Kaufman JS, Owoaje EE, Rotimi CN, Cooper RS. Blood pressure change in Africa: case study from Nigeria. Hum Biol 1999; 71:641–657. [PubMed] [Google Scholar]
  • 34.van der Sande MA, Milligan PJ, Nyan OA, Rowley JT, Banya WA, Ceesay SM, et al. Blood pressure patterns and cardiovascular risk factors in rural and urban gambian communities. J Hum Hypertens 2000; 14:489–496. [DOI] [PubMed] [Google Scholar]
  • 35.Njelekela M, Sato T, Nara Y, Miki T, Kuga S, Noguchi T, et al. Nutritional variation and cardiovascular risk factors in Tanzania–rural-urban difference. S Afr Med J 2003; 93:295–299. [PubMed] [Google Scholar]
  • 36.Kurland L, Melhus H, Karlsson J, Kahan T, Malmqvist K, Ohman P, et al. Aldosterone synthase (CYP11B2) −344 C/T polymorphism is related to antihypertensive response: result from the Swedish Irbesartan Left Ventricular Hypertrophy Investigation versus Atenolol (SILVHIA) trial. Am J Hypertens 2002; 15:389–393. [DOI] [PubMed] [Google Scholar]
  • 37.Ortlepp JR, Hanrath P, Mevissen V, Kiel G, Borggrefe M, Hoffmann R. Variants of the CYP11B2 gene predict response to therapy with candesartan. Eur J Pharmacol 2002; 445:151–152. [DOI] [PubMed] [Google Scholar]
  • 38.Baker EH, Duggal A, Dong Y, Ireson NJ, Wood M, Markandu ND, et al. Amiloride, a specific drug for hypertension in black people with T594 M variant? Hypertens Dallas Tex 1979 2002; 40:13–17. [DOI] [PubMed] [Google Scholar]
  • 39.Mersha TB, Abebe T. Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities. Hum Genomics 2015; 9:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Zhu X, Luke A, Cooper RS, Quertermous T, Hanis C, Mosley T, et al. Admixture mapping for hypertension loci with genome-scan markers. Nat Genet 2005; 37:177–181. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Tables
NIHMS1724038-supplement-Supplementary_Tables.xlsx (16.3KB, xlsx)

RESOURCES