https://orcid.org/0000-0002-2836-7053
Figures
Abstract
Objective
Hypertension is more prevalent in African Americans (AA) than other ethnic groups. Genome-wide association studies (GWAS) have identified loci associated with hypertension and other cardio-metabolic traits like type 2 diabetes, coronary artery disease, and body mass index (BMI), however the AA population is underrepresented in these studies. In this study, we examined a large AA cohort for the generalizability of 14 Metabochip array SNPs with previously reported European hypertension associations.
Methods
To evaluate associations, we analyzed genotype data of 14 SNPs for their associations with a diagnosis of hypertension, systolic blood pressure (SBP), and diastolic blood pressure (DBP) in a case-control study of an AA population (N = 9,534). We also performed an age-stratified analysis (>30, 30≥59 and ≥60 years) following the hypertension definition described by the 8th Joint National Committee (JNC). Associations were adjusted for BMI, age, age2, sex, clinical confounders, and genetic ancestry using multivariable regression models to estimate odds ratios (ORs) and beta-coefficients. Analyses stratified by sex were also conducted. Meta-analyses (including both BioVU and COGENT-BP cohorts) were performed using a random-effects model.
Results
We found rs880315 to be associated with systolic hypertension (SBP≥140 mmHg) in the entire cohort (OR = 1.14, p = 0.003) and within women only (OR = 1.16, p = 0.012). Variant rs17080093 associated with lower SBP and DBP (β = -2.99, p = 0.0352 and - β = 1.69, p = 0.0184) among younger individuals, particularly in younger women (β = -3.92, p = 0.0025 and β = -1.87, p = 0.0241 for SBP and DBP respectively). SNP rs1530440 associated with higher SBP and DBP measurements (younger individuals β = 4.1, p = 0.039 and β = 2.5, p = 0.043 for SBP and DBP; (younger women β = 4.5, p = 0.025 and β = 2.9, p = 0.028 for SBP and DBP), and hypertension risk in older women (OR = 1.4, p = 0.050). rs16948048 increases hypertension risk in younger individuals (OR = 1.31, p = 0.011). Among mid-age women rs880315 associated with higher risk of hypertension (OR = 1.20, p = 0.027). rs1361831 associated with DBP (β = -1.96, p = 0.02) among individuals older than 60 years. rs3096277 increases hypertension risk among older individuals (OR = 1.26 p = 0.0015), however, this variant also reduces SBP among younger women (β = -2.63, p = 0.0102).
Conclusion
These findings suggest that European-descent and AA populations share genetic loci that contribute to blood pressure traits and hypertension. However, the OR and beta-coefficient estimates differ, and some are age-dependent. Additional genetic studies of hypertension in AA are warranted to identify new loci associated with hypertension and blood pressure traits in this population.
Citation: Kaur H, Crawford DC, Liang J, Benchek P, COGENT BP Consortium, Zhu X, et al. (2021) Replication of European hypertension associations in a case-control study of 9,534 African Americans. PLoS ONE 16(11): e0259962. https://doi.org/10.1371/journal.pone.0259962
Editor: Heming Wang, Brigham and Women’s Hospital and Harvard Medical School, UNITED STATES
Received: February 9, 2021; Accepted: October 29, 2021; Published: November 18, 2021
Copyright: © 2021 Kaur et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The Database of Genotypes and Phenotypes (dbGaP) accession ID associated with the EAGLE BioVU dataset is PAGE-BioVu phs000559. Details of COGENT study meta-analysis summary statistics are described within the manuscript.
Funding: This study was funded in part by the National Human Genome Research Institute (NHGRI) grant U01HG004798 (to DCC) and by the National Institute of General Medical Sciences (NIGMS) grant - R01GM126249 (to WSB and DCC). This work was funded, in part, by institutional funds from the CWRU Cleveland Institute for Computational Biology and the Clinical and Translational Science Collaborative of Cleveland UL1TR000439 from the National Center for Advancing Translational Sciences (NCATS) component of the National Institutes of Health (NIH). Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NCATS or NIH. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Co-author Dana C. Crawford is a PLOS ONE Editorial Board member. Co-author William S. Bush is a PLOS ONE Statistical Advisory Board member. These memberships do not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials.
Introduction
Persistently elevated blood pressure, or hypertension [1], is one of the major preventable risk factors for heart disease, as well as a leading contributor to mortality globally [2]. It affects approximately 50 million individuals in the US and accounts for 4.5% of the global burden of all diseases. Hypertension commonly co-exists with comorbidities, including diabetes, obesity, chronic kidney disease, coronary heart disease, depression, and HIV that could associate with poorer health outcomes of hypertensive individuals [3]. The National Health and Nutrition Examination Survey (NHANES 2011–2012), conducted by the Centers for Disease Control (CDC), described the highest prevalence of hypertension in the US to be among African-American (AA) adults (42.1%), compared with non-Hispanic whites (28.0%), Hispanic (26.0%), and non-Hispanic Asian (24.7%) adults. Compared to other ethnic groups, AAs have higher mean blood pressure (BP) (both systolic [SBP] and diastolic [DBP]); with earlier age of onset of hypertension [4], and most importantly, a nearly three-fold higher death rate from high blood pressure [5]. The disparity in hypertension risk among AAs is likely due to a complex combination of socioeconomic, environmental (e.g., perceived racial discrimination), and genetic factors which may also influence downstream development of comorbid conditions [6–13]. Much of the differences in hypertension prevalence in AA remain unexplained [14].
The heritability for both SBP and DBP traits is high—30–55% across global populations [15]. Genome-wide association studies (GWAS) and admixture mapping studies in European-descent populations have identified over 200 genetic loci [16–20] that explain only 3.5% of the inter-individual variation in BP traits (SBP and DBP) [21, 22]. Therefore, the small effects of genetic variants could be one of the potential factors leading to differential hypertension prevalence across populations. African-descent populations are the most ancestral, and AAs are an admixed population with strong components of West African descent; thus, for many variants, it remains unknown if loci identified in European-descent individuals have equivalent influence on hypertension in individuals of AA ancestry [23, 24]. Genetic signatures for hypertension may be more variable in AA than in the more heavily studied European-descent populations; hence, additional studies of hypertension genetics within AA populations are needed.
In the early guidelines of the United States Joint National Committee (JNC) on hypertension, more emphasis was placed on DBP than on SBP as a predictor of cardiovascular events. The Framingham studies, which first identified hypertension as a significant health concern, have contributed significantly to the "Seventh Report of the JNC on prevention, detection, evaluation, and treatment of high blood pressure" (JNC-7) in 2003 [25]. This report redefined clinical criteria for hypertension diagnosis to include SBP≥140 mmHg and/or DBP≥90 mmHg. Age is also a well-recognized predictor of elevated blood pressure [26–28]. Published studies also showed that the effect of genetics on blood pressure varies by age [29, 30] and demonstrated that age is a critical player in hypertension development. In 2014, panel members of JNC-8 recommended for the first time age-stratified thresholds for hypertension diagnosis and initiation of treatment, mainly to reduce potential adverse effects in the elderly—SBP/DBP ≥140/90 mmHg (SBP≥140 mmHg and/or DBP≥90 mmHg) for adults aged less than 60 years, and SBP/DBP≥150/90 mmHg (SBP≥150 mmHg and/or DBP≥90 mmHg) for individuals aged 60 years and older [31, 32]. Furthermore, differences in hypertension and associated cardiovascular disease by sex are well documented [33], however sex-specific differences due to genetic architecture in the AA population warrant further investigation.
As the vast majority of GWAS of hypertension have been performed in European-descent populations, relatively little is known about whether risk estimates for these genetic polymorphisms generalize to the AA population. Furthermore, no genetic study has examined variant associations using the revised hypertension definition from JNC-8 guidelines. In this study, we selected 14 single-nucleotide polymorphisms (SNPs) available on the Illumina Metabochip array with reported European BP associations. We attempted to replicate these associations in the AA population for SBP, DBP, and following JNC-7 and JNC-8 diagnostic criteria for hypertension. We successfully replicated associations of three variant associations with hypertension, SBP, and DBP, indicating their impact on blood pressure measurements in AAs as well.
Material and methods
Study population and hypertension phenotypes
The study population was drawn from EAGLE-BioVU, the Epidemiologic Architecture for Genes Linked to Environment study accessing BioVU, the Vanderbilt University Medical Center’s (VUMC) biorepository linked to de-identified electronic health records (EHRs) as previously described [34, 35]. Race/ethnicity within BioVU is defined by administrative assignment, which shows strong agreement with genetic ancestry for the AA population [36, 37]. Two outcomes of the study were SBP and DBP, which were obtained at the first clinic visit (baseline). Due to limited information about medication use in this study, the first recorded SBP and DBP were considered as pre-treatment measures (the best available surrogate for pre-treatment measures). The third outcome of the study was a hypertension diagnosis based on SBP and DBP measurements; for the purposes of this GWAS, all individuals were defined as hypertensive (SBP≥140 mmHg) or non-hypertensive, regardless of comorbidities. All clinical and demographic data were obtained from de-identified individuals’ EHRs. DNA was isolated from discarded blood drawn from these de-identified patients as part of routine clinical care. Detailed clinical and demographic information was extracted from patient EHRs for research purposes. Genetic ancestry of participants was further evaluated by principal components (PCs) of ancestry-informative markers, and via local ancestry inference using RFMix, as previously described [38].
The Continental Origins and Genetic Epidemiology Network-Blood Presure (COGENT-BP) Study, consisting of 29,378 individuals across 19 studies was included to perform meta-analysis. In brief, all individuals were >20 years old. For individuals reporting the use of antihypertensive medications, SBP and DBP were adjusted by the addition of 15 mmHg and 10 mmHg for SBP and DBP, respectively [21, 39]. Significant outliers (individuals with SBP or DBP >4 standard deviations (SDs) from the mean) were excluded. Each COGENT-BP sub-study received institutional-review-board approval of its consent procedures, examination and surveillance components, data security measures, and DNA collection with approved use for genetic research. All participants in each COGENT-BP sub-study provided written, informed consent [21].
Genotyping
Genotyping was performed using the custom Illumina Metabochip array, which includes approximately 200,000 SNPs selected to fine-map loci genome-wide that were previously associated with metabolic traits. Genotyping was performed according to the manufacturer’s protocol (IlluminaTM) [40, 41] at the Vanderbilt University Center for Human Genetics Research (CHGR) DNA Resources Core. Details of the genotype calling process are described elsewhere [40]. In brief, genotypes were called using both the GenCall 2.0 algorithm (as part of Illumina Genome Studio) and the GenoSNP algorithm [42], and discordance between the two methods was used as a quality check (QC) filter. QC statistics were calculated using GenomeStudio and PLINK [43], including call rate, Hardy-Weinberg equilibrium deviations, and duplicate genotype inconsistencies. PCs were generated to estimate genetic ancestry based on HapMap/1,000 Genomes reference populations. We performed model-based clustering, using the mclust R package to differentiate five ancestry groups: European descent, African descent, East Asian descent, South Asian descent, and Hispanic descent [37], and restricted all analyses to individuals within the African-descent cluster. Genotyping for the COGENT-BP study was performed using a variety of Affymetrix and Illumina arrays, all imputed to the Combined HapMap phase II+III reference as described in [21].
SNP selection and annotation
To assess hypertension and blood pressure-related trait associations which were previously identified in European-descent populations and available within our Metabochip variants, we first tabulated already-published, most significant associations within a given genomic region [19], and among them, 14 were present on the Metabochip array (S1 Table in S1 File). Among these 14 SNPs, rs1327235 was previously linked with SBP and DBP in AA (S2 Table in S1 File) [23]. All 14 SNPs have minor allele frequencies greater than 0.1 in our AA population (Table 1) and were evaluated using the HaploReg database to select the strongly linked SNPs (r2 ≥ 0.8) with the index SNP in African-descent populations from the 1,000 Genomes Project [44]. None of these 14 SNPs were in linkage disequilibrium (LD) with each other in the AA study population.
Statistical analyses
Allele and genotype frequencies of all the variants were calculated using PLINK software [45]. SNPs were tested for HWE using the χ2-test at 0.001 level of significance. Genetic Principal Components (gPCs) were computed using EIGENSOFT, as previously described [40]. To negate the effect of other clinical phenotypes on hypertension associations, we estimated clinical Principal Components (cPCs) as described in Zhang et al [46] by first generating a binary array of International Classification of Diseases, 9th edition, Clinical Modification (ICD-9-CM)-based codes where individuals with three or more instances of the code on separate visit dates are positive (1), and all others are negative (0). Using these binary phenotype arrays, we then computed cPCs using the prcomp function in R and selected the top 10. Multivariable logistic and linear regression modeling was used to evaluate association of SNPs with hypertension, SBP, or DBP (all three dependent variables), using BMI, age, age2, sex, cPC1-cPC10, and gPC1-gPC10 as model covariates. The age-stratified analysis was also conducted based on JNC-8 definitions (SBP/DBP ≥140/90 mmHg for adults aged less than 60 years, and SBP/DBP≥150/90 mmHg for individuals aged 60 years and older). Sex-differences in such phenotypes are of growing interest and even understudied. Therefore, sex-stratified and-adjusted studies were also explored. In all models, the odds ratio (OR) and beta-coefficient (β) and their 95% confidence interval (CI) were estimated. Meta-analyses among BioVU and COGENT-BP cohorts were performed using the random-effects model in the metafor R package [21].
Ethical statement
The VUMC protocol for EAGLE-BioVu is considered non-human subjects research (The Code of Federal Regulations, 45 CFR 46.102 (f)) [36, 47]. The COGENT-BP study (for which we accessed meta-analysis results) was approved by the Institutional Review Board (IRB) # 04-95-72 and study-related committees (see Appendix I in S1 File). All participants of the COGENT-BP study provided informed consent for DNA research and data are publicly available in dbGaP. Also, we have followed recommendations from the STrengthening the Reporting of Genetic Association Studies Statement in this study (see Appendix II in S1 File) [48].
Results
Among 9,534 AA individuals from EAGLE-BioVU with available EHR data, 65.3% were female. Mean age, SBP, and DBP were 46.1 years, 129.8 mmHg, and 78.3 mmHg, respectively (Table 2). Following the JNC-8 guidelines, breakdowns by age strata show the expected trend of increasing SBP, DBP, and hypertension prevalence with age.
We identified 14 SNPs previously associated with hypertension in European-ancestry populations that were also genotyped on the Metabochip and available for analysis in our dataset. We observed similar allele frequencies between our study population and the African (AFR) samples from the 1,000 Genomes Project, but notable differences in frequencies of these SNPs in other populations (American, Asian and European) does exist (Table 1). We replicated one known European hypertension association to genetic variants in our AA cohort and observed two borderline associations. We defined criteria for hypertension according to two guidelines, 1.) JNC-7 –SBP ≥140 mmHg and, 2.) JNC-8 –a) younger age group (<30 years)–SBP≥140 mmHg, DBP≥90 mmHg, b) mid-age group (≥30 to <60 years)–DBP≥90 mmHg, c) older age group (≥60 years)–SBP≥150 mmHg and/or DBP≥90 mmHg. The G allele of rs880315 higher risk of hypertension (JNC-7) (OR [95% CI] = 1.14 [1.05, 1.25], p = 0.003,) and A allele of rs3096277 showed borderline association with hypertension (JNC-7) (OR [CI 95%] = 1.07 [1.00, 1.16], p = 0.06,) across the entire AA cohort (Table 3A). Furthermore, the G allele of SNP rs1361831, which previously associated with SBP and DBP in European populations, showed similarly low but non-significant association p-values for SBP and DBP (β (95% CI) = -1.19 (-2.50, 0.13); p = 0.08 and -0.78 (-1.58, 0.02); p = 0.06, respectively). All models were adjusted for BMI, age, age2, and sex, and for genetic ancestry using genetic principal components (gPCs) and clinical comorbidities using clinical principal components (cPCs).
We next investigated the role of SNPs according to the age stratifications outlined by the JNC-8 definition (2014). We found no association with hypertension in individuals under age 30 (N = 2,123) (Tables 1 and 3B). However, in this strata SNP rs17080093 (A allele) was associated with both SBP and DBP (β = -2.99 (-5.48, -0.51), and β = -1.69 (-3.26, -0.12) respectively, both p<0.05). Among mid-aged individuals ((≥30 to <60 years) (N = 5,351), rs880315 associated with higher risk of hypertension (OR (95% CI) = 1.13 (1.01, 1.26), p = 0.03) (Tables 1 and 3C). Among older individuals ≥ 60 years (N = 2,060), rs1361831 significantly associated with DBP (β (95% CI) = -1.96 (-3.65, -0.27), p = 0.02), and rs3096277 was significantly associated with hypertension (OR (95%CI) = 1.26 (1.09, 1.45), p = 0.001) (Tables 1 and 3D) after adjustment for BMI, clinical PCs, genetic PCs and sex. We also evaluated the impact of local ancestry on each SNP effect and found that the observed associations did not change after these adjustments (S3 Table in S1 File). This suggests that the effects of associated alleles were not due solely to the inheritance of the alleles from a European background via admixture.
In sex-stratified analysis, we found G allele of SNP rs880315 was associated with increased risk of hypertension in women (OR = 1.16, 95% CI = 1.03–1.31, p<0.01). Though the direction of the effect was same in men, it was not statistically significant (OR = 1.11, 95%CI = 0.96–1.28, p = 0.15) (Table 4A). Moreover, we did not observed any differences in SBP and DBP among men and women. Age-stratified analysis by sex revealed that allele A of rs17080093 significantly associated with decrease in SBP and DBP among younger women (β = -3.92, 95%CI = -6.46, -1.38 and β = -1.87, 95%CI = -3.50, -0.25 respectively, both p<0.05) and weakly associated with increase in SBP and DBP among mid-aged men (though not statistically significant β = 2.90, 95%CI = -0.24, 6.03; p = 0.07 and β = 1.91, 95%CI = -0.04, 3.87, p = 0.06). Among individuals age 30–60 years rs880315 was associated with increased risk of hypertension (OR = 1.20, 95% CI = 1.04, 1.38, p = 0.01) (Table 4B and 4C). rs16948048 found to be significantly differ among hypertensive and non-hypertensive group of younger individuals (OR = 1.31, 95% CI = 0.04, 1.77, p = 0.011). Younger individuals, particularly women with A allele of SNP rs1530440 were likely to have higher both SBP and DBP measurements (younger individuals β = 4.1, and β = 2.5, for SBP and DBP, respectively and younger women β = 4.5, and β = 2.9, for SBP and DBP, respectively, all p<0.05) (Table 4B). However, the same SNP was associated with higher risk of hypertension development in older women (OR = 1.4, p = 0.050) (Table 4D). Allele A of rs3096277 found to be associated with a decrease in SBP among younger women (β = -2.63, 95% CI = -4.63, -0.62, p = 0.01) however this allele increased SBP among younger men (β = 5.05, 95% CI = -0.80, 10.91, p = 0.09) (Table 4B and 4C). Allele G of rs1361831 associated with decreased DBP in older women (β = -2.42, 95% CI = -4.56, -0.27, p = 0.03) (Table 4D). Sex-stratified models were adjusted for age, age2, BMI, clinical PCs, and genetic PCs, and age-sex stratified models adjusted BMI, clinical PCs, and genetic PCs.
Given these results, we next sought additional replication of associations via random effects meta-analysis among independent participants of BioVU and the COGENT-BP study. The results of the meta-analysis combining data across samples with pooled estimates of BioVU and COGENT-BP summarized in Table 5. rs880315-T, and rs17080093-T, were significantly associated with SBP, DBP, and hypertension (all p-values < 0.05 for SBP, DBP, and hypertension, respectively). rs1361831-T associated with reduced risk of hypertension development and rs2272007-T significantly increases DBP and hypertension risk.
Discussion
While the genetic risk for hypertension has been studied extensively in multiple European-descent populations, much less is known about the impact of these hypertension SNPs in AA populations. Furthermore, previously reported associations with hypertension phenotypes used older definitions (SBP≥140 mmHg). In this study, we evaluated European-associated SNPs for hypertension and BP within an AA population using multiple definitions of hypertension. We replicated four SNPs associations (which reached nominal significance), including three intronic SNPs in rs880315 (CASZ1), rs3096277 (CDH13), and rs17080093 (PLEKHG1) gene and one intergenic SNP rs1361831 close to the RSPO3 gene. The sex-stratified analysis showed a differential effect of genetic variants on blood pressure traits in AA.
Considering new hypertension diagnostic criteria established by the 8th JNC report, we performed age-stratified analyses. We found variant rs3096277 (CDH13—Cadherin 13) associated with hypertension (SBP≥150 mmHg and/or DBP≥90 mmHg) in individuals aged 60 years and above and SBP in younger women as well. However, this variant strongly correlated with an increased risk of SBP elevation in younger men. Associations between CDH13 SNPs with SBP, DBP, and hypertension are consistent with prior studies, though the specific variants appear to be ancestry-dependent [49]. In a study of Mexican participants, rs3096277 was not associated with BP traits; but another polymorphism from this region rs11646213, which was not present on Metabochip array, was associated with the risk of hypertension development [50]. The pairwise LD analysis using 1,000 Genomes Project genotype data for African Ancestry in Southwest US (ASW) (most representative of the present cohort) revealed that rs11646213 was not in LD (r2 = 0.05) with rs3096277. Moreover, a recent trans-ethnic association study showed another SNP, rs7500448, which was not present on Metabochip array, from the CDH13 gene (not in LD with SNP rs3096277 in ASW (r2 = 0.01) using 1,000 Genomes Project data), associated with SBP and DBP in the Million Veteran Program (MVP) and the UK Biobank (UKB) [51]. The CDH13 gene encodes for a calcium-dependent cell-cell adhesion glycoprotein and is predominantly expressed in the nervous and cardiovascular systems in particular spinal cord and aorta, the carotid, iliac and renal arteries, and the heart [52] and is known to play an essential role in regulating angiogenesis and blood vessel remodeling [53]. This gene has also been associated with different cancer types, such as lung, breast, prostate cancers, and tumor angiogenesis [54–56]. All three SNPs in CDH13, rs3096277, rs7500448, and rs11646213, were associated with different BP traits in different populations and not in LD with each other, suggesting that these variants cannot substitute for one another and could potentially influence the CDH13 gene function independently.
SNP rs880315 in CASZ1 (Zinc Finger Protein Castor Homolog 1) nominally associated with hypertension (DBP≥90 mmHg) in the mid-tier age group (≥30 to <60 years), but not in younger (<30 years) or older individuals (≥60 years). This variant has also been associated with an increased risk of hypertension in women despite similar frequencies in men. Previously, this variant was associated with SBP in women of European ancestry (Women’s Health Study; WHS) [57], and to hypertension, SBP, and DBP in the Japanese population [58]. This gene encodes castor homolog-1, a zinc finger transcription factor that is known to be involved in cell-cycle signaling and apoptosis regulation. This gene is expressed in many tissues, including cardiac myocytes, and the nearby genomic region (chr1:10,630,927–10,790,973) has also been implicated in neuroectodermal tumors [59, 60].
Variant rs17080093 located in the intron of PLEKHG1 (pleckstrin-homology-domain-containing, family G [with RhoGef domain] member 1) was associated with reduced risk of SBP and DBP elevation in the younger age group (<30 years), particularly in younger women. Previously, this gene has been linked to childhood obesity in the Hispanic population [61]. This PLEKHG1 gene plays an essential role in vascular endothelial cell reorientation by targeting RhoA, Rac1 and/or Cdc42, which are involved in cyclic‐stretch‐induced perpendicular reorientation, thus makes it a potential candidate for studies on blood pressure traits [62]. Another SNP in the same gene, rs17080102 (not genotyped by the Metabochip) associated with SBP and DBP in another cohort of AA, European, and East Asian ancestry [23]. Furthermore, this locus showed a significant association with SBP and DBP in a trans-ethnic meta-analysis [17, 20, 23]. Recently, this gene has been associated with maternal preeclampsia, another hypertension-related phenotype [63]. SNP rs17080102 was found to be in moderate LD (r2 = 0.45) with rs17080093 among ASW individuals from the 1,000 Genomes Project. HaploReg reports that the intronic SNP rs17080102 from the PLEKHG1 gene is present in and significantly alters various regulatory motifs (S4 Table in S1 File). Thus, we speculate that the association of rs17080093 to hypertension in our study is tagging the effect of rs17080102. The rs1530440 is intronic SNP in C10orf107, an open reading frame of unknown function was associated with both SBP and DBP in younger women in the studied cohort. Cheh et al reported association of this SNP with DBP in the European population. The ARID5B (AT rich interactive domain 5B (MRF1 like)) gene is located in the close proximity of C10orf107 and its higher expression was reported in cardiovascular tissue and involved in smooth muscle cell differentiation [64]. rs16948048, located on 5’ upstream to ZNF652, showed significant difference among young hypertensive and non-hypertensive individuals. This SNPs has previously been linked to DBP in European population [64] and essential hypertension among Chinese individuals [65] suggesting that tis gene might be involved in the blood pressure traits.
Variant rs1361831 (intergenic region, chr6:126,859,944) located near to RSPO3-chr6:127,118,671–127,199,481; (R-spondin family member 3), gene significantly associated with DBP in the older age (≥60) group of the studied cohort. A large-scale genome-wide association study revealed a relationship between a genomic region close to this gene and renal-function–related traits, including blood urea nitrogen in East Asian populations [66]. This locus showed an association with the waist-to-hip ratio in the GIANT study, which included individuals of European and European-American descent [67]. SNP rs13209747 (not genotyped by the Metabochip) is located near this RSPO3 associated with SBP and DBP in the COGENT-BP study in AA and East Asian individuals and trans-ethnic meta-analysis [23]. rs1361831 and rs13209747 were found to be in strong LD (r2 = 0.80 in ASW, using 1,000 Genome Project data), suggesting that these were introduced recently into the population and may substitute for each other. A recently published trans-ethnic study showed an association of novel missense and rare markers in this region with SBP, DBP, and pulse pressure (the difference between SBP and DBP after addition of the constant values) in male veterans of European, African, Hispanic, Asian, and Native American descent [51]. Our results showed an opposite direction of effect from the COGENT-BP replication results and from this trans-ethnic meta-analysis, possibly due to differences in the male/female ratio between the studies. In the trans-ethnic study, the sample was heavily skewed towards males with African ancestry; however, our study was approximately 65% female. These results may indicate effect modification of this SNP by sex and age. Moreover, additional adjustments with local ancestry suggested African ancestry also contributes to observed allelic effects, and associations were not due to the inheritance of European genetic allele only. Our meta-analysis results indirectly linked CASZ1, PLEKHG1, C20orf187, RSPO3, ULK4, and LRRC10B genes with altered blood pressure levels among AA population that merits in-vitro validation of these findings. Also, these genetic variants would be helpful in stratifying “at-risk” individuals predispose to higher risk of elevated SBP, DBP and hypertension development based on their genetic architecture.
Differences in allele frequency of variants among different populations could lead to clinical heterogeneity. Additionally, age-dependent exposure to various non-genetic risk factors for hypertension could be linked to the differential prevalence of blood pressure/hypertension [68]. Our reported associations have less extreme p-values than associations made in more extensive European studies, and these differences are likely attributed to reduced statistical power in our study. AA is an admixed population of African and European ancestry. Consequently, variants associated with hypertension or other BP traits (SBP and DBP) in European-ancestry cannot necessarily be generalized to other populations. However, replication of associated variants would help us to identify essential and significant trans-population genetic variants. Here, we replicated European hypertension variant associations in AA. Our findings indicate remarkable differences in allele frequencies and p-values, which further strengthen the concept of population-specific hypertension pathogenesis components. Therefore, the investigation of genetically diverse populations is necessary to identify causal variants in affected populations. Our present cohort of approximately 10,000 AA individuals has a reasonable power to detect common variants; however, to evaluate the possible effect of other hypertension-associated variants found in European-descent studies, larger samples are needed. Another reason for the observed differences could be adjustments for potential confounders such as age, sex, BMI, use of hypertensive drugs, and ethnicity/race that affects study outcomes. In the EAGLE-BioVU dataset, we considered presumed pre-treatment SBP and DBP measurements for hypertension diagnosis; therefore, we used SBP and DBP measurements without adjustment for medication effects. However, we made adjustments for clinical PCs. On the other hand, the COGENT-BP study, consistent with other published studies in European ancestry, adjusted measured SBP and DBP for hypertensive medication usage (+15 and +10 mmHg, respectively) [21].
Strengths
The present study was conceived as a replication of SNPs previously associated with hypertension from European-descent studies in an AA cohort to assess their generalizability. For the present study, hypertension outcome has different diagnosis criteria, according to the revised JNC-8 guidelines; however, observed associations with hypertension did not change with different hypertension diagnosis criteria. We also performed an age-stratified analysis, which was not performed in other studies investigating the association of genetic markers with hypertension. Furthermore, unlike other genetic epidemiological studies, we have included clinical PCs to adjust for confounding comorbidities, which is not typical in other studies. These findings suggest that previously associated variants with blood pressure traits in European populations may be generalized to African American decent individuals. Age-stratified associations further confirm that age plays a significant role in hypertension risk. Therefore, identifying individuals, from various age groups, with a greater risk for hypertension development may help target preventative measures.
The genetic admixture may confound genetic association studies, and thus, we have included genetic PCs to control population stratification or hidden population substructures. Furthermore, we used the local ancestry association approach to check the robustness of our associations. After adjusting for local ancestry for AA, our SNP associations remain unchanged. Hence, the association of SNPs in the present cohort shows that the risk alleles’ effect is similar to the background of African ancestry and not likely due to European admixture.
Limitations
Our study has a few limitations that we need to acknowledge. First, SNPs selected were limited to those available on the Metabochip array. Also, as smoking status was not well documented in the EHR, it could not be reliably included in the model. We did not adjust for medication use, and instead used the first SBP and DBP measurements to define hypertensive and non-hypertensive individuals; we considered the first measurements to be the best approximation for pre-therapy values. While this may introduce a bias, we would expect this to lower statistical power rather than increasing the false-positive rate. As this study aimed to replicate hypertension-associated SNPs found previously in a new (AA) population, the p-values were not corrected for multiple testing. Finally, the meta-analysis was based on the unadjusted estimates, and stratified analysis on the basis of various confounding variables such as age and sex could not be conducted.
Conclusion
Future studies to investigate age- and sex-stratified associations of genetic variants with hypertension diagnosis, SBP, and DBP measurements, including pre and post-treatment of antihypertensive medications, are warranted as other AA populations may have different genetic substructure and not be truly comparable. The study of multiple blood pressure readings over considerable time, including medication compliance and genetic variants, would be informative for the early diagnosis of individuals at hypertension risk in the understudied AA population. Furthermore, genetic loci stratified investigations aimed at determining the role of the respective genetic variance in hypertension could get precise treatment recommendations as part of personalized medicine that may be developed in the near future.
Acknowledgments
We thank the COGENT-BP consortium for allowing us access to the summary statistics. COGENT-BP investigators are listed in the Appendix III in S1 File.
References
- 1. Mucci N, Giorgi G, De Pasquale Ceratti S, Fiz-Perez J, Mucci F, Arcangeli G. Anxiety, Stress-Related Factors, and Blood Pressure in Young Adults. Front Psychol. 2016;7:1682. Epub 2016/11/15. pmid:27840615; PMCID: PMC5083786.
- 2. Franklin SS, Wong ND. Hypertension and cardiovascular disease: contributions of the framingham heart study. Glob Heart. 2013;8(1):49–57. Epub 2013/03/01. pmid:25690263.
- 3. Mohamed SF, Uthman OA, Caleyachetty R, Chumo I, Mutua MK, Asiki G, et al. Uncontrolled hypertension among patients with comorbidities in sub-Saharan Africa: protocol for a systematic review and meta-analysis. Syst Rev. 2020;9(1):16. Epub 2020/01/18. pmid:31948464; PMCID: PMC6964205.
- 4. Lackland DT. Racial differences in hypertension: implications for high blood pressure management. Am J Med Sci. 2014;348(2):135–8. Epub 2014/07/02. pmid:24983758; PMCID: PMC4108512.
- 5. Rosamond W, Flegal K, Furie K, Go A, Greenlund K, Haase N, et al. Heart disease and stroke statistics—2008 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2008;117(4):e25–146. Epub 2007/12/19. pmid:18086926.
- 6. Tarver-Carr ME, Powe NR, Eberhardt MS, LaVeist TA, Kington RS, Coresh J, et al. Excess risk of chronic kidney disease among African-American versus white subjects in the United States: a population-based study of potential explanatory factors. J Am Soc Nephrol. 2002;13(9):2363–70. Epub 2002/08/23. pmid:12191981.
- 7. Norris K, Nissenson AR. Race, gender, and socioeconomic disparities in CKD in the United States. J Am Soc Nephrol. 2008;19(7):1261–70. Epub 2008/06/06. pmid:18525000.
- 8. Freedman BI, Murea M. Target organ damage in African American hypertension: role of APOL1. Curr Hypertens Rep. 2012;14(1):21–8. Epub 2011/11/10. pmid:22068337; PMCID: PMC3253170.
- 9. Martins D, Norris K. Hypertension treatment in African Americans: physiology is less important than sociology. Cleve Clin J Med. 2004;71(9):735–43. Epub 2004/10/14. pmid:15478705.
- 10. Suthanthiran M, Li B, Song JO, Ding R, Sharma VK, Schwartz JE, et al. Transforming growth factor-beta 1 hyperexpression in African-American hypertensives: A novel mediator of hypertension and/or target organ damage. Proc Natl Acad Sci U S A. 2000;97(7):3479–84. Epub 2000/03/22. pmid:10725360; PMCID: PMC16265.
- 11. O’Toole JF, Sedor JR. Kidney disease: new technologies translate mechanisms to cure. J Clin Invest. 2014;124(6):2294–8. Epub 2014/06/04. pmid:24892702; PMCID: PMC4089455.
- 12. Ferdinand KC, Townsend RR. Hypertension in the US Black population: risk factors, complications, and potential impact of central aortic pressure on effective treatment. Cardiovasc Drugs Ther. 2012;26(2):157–65. Epub 2012/01/17. pmid:22246101.
- 13. Dolezsar CM, McGrath JJ, Herzig AJM, Miller SB. Perceived racial discrimination and hypertension: a comprehensive systematic review. Health Psychol. 2014;33(1):20–34. Epub 2014/01/15. pmid:24417692; PMCID: PMC5756074.
- 14. Fuchs FD. Why do black Americans have higher prevalence of hypertension?: an enigma still unsolved. Hypertension. 2011;57(3):379–80. Epub 2011/02/09. pmid:21300666.
- 15. Padmanabhan S, Caulfield M, Dominiczak AF. Genetic and molecular aspects of hypertension. Circ Res. 2015;116(6):937–59. Epub 2015/03/15. pmid:25767282.
- 16. Simino J, Shi G, Weder A, Boerwinkle E, Hunt SC, Rao DC. Body mass index modulates blood pressure heritability: the Family Blood Pressure Program. Am J Hypertens. 2014;27(4):610–9. Epub 2013/09/14. pmid:24029162; PMCID: PMC3958601.
- 17. Johnson T, Gaunt TR, Newhouse SJ, Padmanabhan S, Tomaszewski M, Kumari M, et al. Blood pressure loci identified with a gene-centric array. Am J Hum Genet. 2011;89(6):688–700. Epub 2011/11/22. pmid:22100073; PMCID: PMC3234370.
- 18. Franklin SS, Thijs L, Hansen TW, O’Brien E, Staessen JA. White-coat hypertension: new insights from recent studies. Hypertension. 2013;62(6):982–7. Epub 2013/09/18. pmid:24041952.
- 19. Ehret GB, Ferreira T, Chasman DI, Jackson AU, Schmidt EM, Johnson T, et al. The genetics of blood pressure regulation and its target organs from association studies in 342,415 individuals. Nat Genet. 2016;48(10):1171–84. Epub 2016/09/13. pmid:27618452; PMCID: PMC5042863.
- 20. International Consortium for Blood Pressure Genome-Wide Association S, Ehret GB, Munroe PB, Rice KM, Bochud M, Johnson AD, et al. Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk. Nature. 2011;478(7367):103–9. Epub 2011/09/13. pmid:21909115; PMCID: PMC3340926.
- 21. Liang J, Le TH, Edwards DRV, Tayo BO, Gaulton KJ, Smith JA, et al. Single-trait and multi-trait genome-wide association analyses identify novel loci for blood pressure in African-ancestry populations. PLoS Genet. 2017;13(5):e1006728. Epub 2017/05/13. pmid:28498854; PMCID: PMC5446189.
- 22. 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(7):e0133031. Epub 2015/07/15. pmid:26162070; PMCID: PMC4498745.
- 23. Franceschini N, Fox E, Zhang Z, Edwards TL, Nalls MA, Sung YJ, et al. Genome-wide association analysis of blood-pressure traits in African-ancestry individuals reveals common associated genes in African and non-African populations. Am J Hum Genet. 2013;93(3):545–54. Epub 2013/08/27. pmid:23972371; PMCID: PMC3769920.
- 24. Adeyemo A, Gerry N, Chen G, Herbert A, Doumatey A, Huang H, et al. A genome-wide association study of hypertension and blood pressure in African Americans. PLoS Genet. 2009;5(7):e1000564. Epub 2009/07/18. pmid:19609347; PMCID: PMC2702100.
- 25. Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G, et al. 2007 Guidelines for the Management of Arterial Hypertension: The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens. 2007;25(6):1105–87. Epub 2007/06/15. pmid:17563527.
- 26. Pinto E. Blood pressure and ageing. Postgrad Med J. 2007;83(976):109–14. Epub 2007/02/20. pmid:17308214; PMCID: PMC2805932.
- 27. Kshirsagar AV, Chiu YL, Bomback AS, August PA, Viera AJ, Colindres RE, et al. A hypertension risk score for middle-aged and older adults. J Clin Hypertens (Greenwich). 2010;12(10):800–8. Epub 2010/10/30. pmid:21029343; PMCID: PMC3683833.
- 28. Buford TW. Hypertension and aging. Ageing Res Rev. 2016;26:96–111. Epub 2016/02/03. pmid:26835847; PMCID: PMC4768730.
- 29. Shi G, Gu CC, Kraja AT, Arnett DK, Myers RH, Pankow JS, et al. Genetic effect on blood pressure is modulated by age: the Hypertension Genetic Epidemiology Network Study. Hypertension. 2009;53(1):35–41. Epub 2008/11/26. pmid:19029486; PMCID: PMC2633773.
- 30. Kannel WB. Hypertension: reflections on risks and prognostication. Med Clin North Am. 2009;93(3):541–58, Table of Contents. Epub 2009/05/12. pmid:19427490; PMCID: PMC3719976.
- 31. Shrout T, Rudy DW, Piascik MT. Hypertension update, JNC8 and beyond. Curr Opin Pharmacol. 2017;33:41–6. Epub 2017/05/20. pmid:28525818.
- 32. Zhang PY. Review of new hypertension guidelines. Eur Rev Med Pharmacol Sci. 2015;19(2):312–5. Epub 2015/02/17. pmid:25683948.
- 33. Colafella KMM, Denton KM. Sex-specific differences in hypertension and associated cardiovascular disease. Nat Rev Nephrol. 2018;14(3):185–201. Epub 2018/01/31. pmid:29380817.
- 34. Crawford DC, Goodloe R, Farber-Eger E, Boston J, Pendergrass SA, Haines JL, et al. Leveraging Epidemiologic and Clinical Collections for Genomic Studies of Complex Traits. Hum Hered. 2015;79(3–4):137–46. Epub 2015/07/24. pmid:26201699; PMCID: PMC4528966.
- 35. Roden DM, Pulley JM, Basford MA, Bernard GR, Clayton EW, Balser JR, et al. Development of a large-scale de-identified DNA biobank to enable personalized medicine. Clin Pharmacol Ther. 2008;84(3):362–9. Epub 2008/05/27. pmid:18500243; PMCID: PMC3763939.
- 36. Dumitrescu L, Ritchie MD, Denny JC, El Rouby NM, McDonough CW, Bradford Y, et al. Genome-wide study of resistant hypertension identified from electronic health records. PLoS One. 2017;12(2):e0171745. Epub 2017/02/22. pmid:28222112; PMCID: PMC5319785.
- 37. Hall JB, Dumitrescu L, Dilks HH, Crawford DC, Bush WS. Accuracy of administratively-assigned ancestry for diverse populations in an electronic medical record-linked biobank. PLoS One. 2014;9(6):e99161. Epub 2014/06/05. pmid:24896101; PMCID: PMC4045967.
- 38. Fish AE, Crawford DC, Capra JA, Bush WS. Local ancestry transitions modify snp-trait associations. Pac Symp Biocomput. 2018;23:424–35. Epub 2017/12/09. pmid:29218902; PMCID: PMC5728664.
- 39. 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(6):677–87. Epub 2009/05/12. pmid:19430479; PMCID: PMC2998712.
- 40. Buyske S, Wu Y, Carty CL, Cheng I, Assimes TL, Dumitrescu L, et al. Evaluation of the metabochip genotyping array in African Americans and implications for fine mapping of GWAS-identified loci: the PAGE study. PLoS One. 2012;7(4):e35651. Epub 2012/04/28. pmid:22539988; PMCID: PMC3335090.
- 41. Voight BF, Kang HM, Ding J, Palmer CD, Sidore C, Chines PS, et al. The metabochip, a custom genotyping array for genetic studies of metabolic, cardiovascular, and anthropometric traits. PLoS Genet. 2012;8(8):e1002793. Epub 2012/08/10. pmid:22876189; PMCID: PMC3410907.
- 42. Giannoulatou E, Yau C, Colella S, Ragoussis J, Holmes CC. GenoSNP: a variational Bayes within-sample SNP genotyping algorithm that does not require a reference population. Bioinformatics. 2008;24(19):2209–14. Epub 2008/07/26. pmid:18653518.
- 43. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–75. Epub 2007/08/19. pmid:17701901; PMCID: PMC1950838.
- 44. Ward LD, Kellis M. HaploReg: a resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of genetically linked variants. Nucleic Acids Res. 2012;40(Database issue):D930-4. Epub 2011/11/09. pmid:22064851; PMCID: PMC3245002.
- 45. 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. Epub 2015/02/28. pmid:25722852; PMCID: PMC4342193.
- 46. Zhang Z, Adela Castelló A. Principal components analysis in clinical studies. Ann Transl Med. 2017 Sep; 5(17): 351. pmid:28936445; PMCID: PMC5599285
- 47. Pulley J, Clayton E, Bernard GR, Roden DM, Masys DR. Principles of human subjects protections applied in an opt-out, de-identified biobank. Clin Transl Sci. 2010;3(1):42–8. Epub 2010/05/07. pmid:20443953; PMCID: PMC3075971.
- 48. Little J, Higgins JP, Ioannidis JP, Moher D, Gagnon F, von Elm E, et al. STrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statement. PLoS Med. 2009;6(2):e22. Epub 2009/02/06. pmid:19192942; PMCID: PMC2634792.
- 49. Levy D, Larson MG, Benjamin EJ, Newton-Cheh C, Wang TJ, Hwang SJ, et al. Framingham Heart Study 100K Project: genome-wide associations for blood pressure and arterial stiffness. BMC Med Genet. 2007;8 Suppl 1:S3. Epub 2007/10/16. pmid:17903302; PMCID: PMC1995621.
- 50. Vargas-Alarcon G, Martinez-Rodriguez N, Velazquez-Cruz R, Perez-Mendez O, Posadas-Sanchez R, Posadas-Romero C, et al. The T>A (rs11646213) gene polymorphism of cadherin-13 (CDH13) gene is associated with decreased risk of developing hypertension in Mexican population. Immunobiology. 2017;222(10):973–8. Epub 2016/09/30. pmid:27682011.
- 51. Giri A, Hellwege JN, Keaton JM, Park J, Qiu C, Warren HR, et al. Trans-ethnic association study of blood pressure determinants in over 750,000 individuals. Nat Genet. 2019;51(1):51–62. Epub 2018/12/24. pmid:30578418; PMCID: PMC6365102.
- 52. Ivanov D, Philippova M, Antropova J, Gubaeva F, Iljinskaya O, Tararak E, et al. Expression of cell adhesion molecule T-cadherin in the human vasculature. Histochem Cell Biol. 2001;115(3):231–42. Epub 2001/05/01. pmid:11326751.
- 53. Rubina K, Kalinina N, Potekhina A, Efimenko A, Semina E, Poliakov A, et al. T-cadherin suppresses angiogenesis in vivo by inhibiting migration of endothelial cells. Angiogenesis. 2007;10(3):183–95. Epub 2007/05/09. pmid:17486418.
- 54. Sato M, Mori Y, Sakurada A, Fujimura S, Horii A. The H-cadherin (CDH13) gene is inactivated in human lung cancer. Hum Genet. 1998;103(1):96–101. Epub 1998/09/16. pmid:9737784
- 55. Thomas G, Jacobs KB, Yeager M, Kraft P, Wacholder S, Orr N, et al. Multiple loci identified in a genome-wide association study of prostate cancer. Nat Genet. 2008;40(3):310–5. Epub 2008/02/12. pmid:18264096.
- 56. Hebbard LW, Garlatti M, Young LJ, Cardiff RD, Oshima RG, Ranscht B. T-cadherin supports angiogenesis and adiponectin association with the vasculature in a mouse mammary tumor model. Cancer Res. 2008;68(5):1407–16. Epub 2008/03/05. pmid:18316604; PMCID: PMC2676344.
- 57. Ho JE, Levy D, Rose L, Johnson AD, Ridker PM, Chasman DI. Discovery and replication of novel blood pressure genetic loci in the Women’s Genome Health Study. J Hypertens. 2011;29(1):62–9. Epub 2010/11/04. pmid:21045733; PMCID: PMC3005130.
- 58. Takeuchi F, Isono M, Katsuya T, Yamamoto K, Yokota M, Sugiyama T, et al. Blood pressure and hypertension are associated with 7 loci in the Japanese population. Circulation. 2010;121(21):2302–9. Epub 2010/05/19. pmid:20479155.
- 59. Yuan ZR, Wang R, Solomon J, Luo X, Sun H, Zhang L, et al. Identification and characterization of survival-related gene, a novel cell survival gene controlling apoptosis and tumorigenesis. Cancer Res. 2005;65(23):10716–24. Epub 2005/12/03. pmid:16322216.
- 60. Liu Z, Yang X, Tan F, Cullion K, Thiele CJ. Molecular cloning and characterization of human Castor, a novel human gene upregulated during cell differentiation. Biochem Biophys Res Commun. 2006;344(3):834–44. Epub 2006/04/25. pmid:16631614.
- 61. Comuzzie AG, Cole SA, Laston SL, Voruganti VS, Haack K, Gibbs RA, et al. Novel genetic loci identified for the pathophysiology of childhood obesity in the Hispanic population. PLoS One. 2012;7(12):e51954. Epub 2012/12/20. pmid:23251661; PMCID: PMC3522587.
- 62. Abiko H, Fujiwara S, Ohashi K, Hiatari R, Mashiko T, Sakamoto N, et al. Rho guanine nucleotide exchange factors involved in cyclic-stretch-induced reorientation of vascular endothelial cells. J Cell Sci. 2015;128(9):1683–95. Epub 2015/03/22. pmid:25795300.
- 63. Gray KJ, Kovacheva VP, Mirzakhani H, Bjonnes AC, Almoguera B, DeWan AT, et al. Gene-Centric Analysis of Preeclampsia Identifies Maternal Association at PLEKHG1. Hypertension. 2018;72(2):408–16. Epub 2018/07/04. pmid:29967039; PMCID: PMC6043396.
- 64. Newton-Cheh C, Johnson T, Gateva V, Tobin MD, Bochud M, Coin L, et al. Nat Genet. 2009 Jun; 41(6): 666–676. pmid:19430483; PMCID: PMC2891673
- 65. Niu W, Zhang Y, Ji K, Gu M, Gao P, Zhu D. Clin Chim Acta. 2010 Oct 9;411(19–20):1491–5. Epub 2010 Jun 10. pmid:20542020.
- 66. Kim YJ, Go MJ, Hu C, Hong CB, Kim YK, Lee JY, et al. Large-scale genome-wide association studies in East Asians identify new genetic loci influencing metabolic traits. Nat Genet. 2011;43(10):990–5. Epub 2011/09/13. pmid:21909109.
- 67. Heid IM, Jackson AU, Randall JC, Winkler TW, Qi L, Steinthorsdottir V, et al. Meta-analysis identifies 13 new loci associated with waist-hip ratio and reveals sexual dimorphism in the genetic basis of fat distribution. Nat Genet. 2010;42(11):949–60. Epub 2010/10/12. pmid:20935629; PMCID: PMC3000924.
- 68. Niiranen TJ, Henglin M, Claggett B, Muggeo VMR, McCabe E, Jain M, et al. Trajectories of Blood Pressure Elevation Preceding Hypertension Onset: An Analysis of the Framingham Heart Study Original Cohort. JAMA Cardiol. 2018;3(5):427–31. Epub 2018/ pmid:29562081