A Genome-Wide Association Study Identifies a Novel Susceptibility Locus at The DLGAP1 Gene For Resistant Hypertension in The Japanese Population


 Numerous genetic variants associated with hypertension and blood pressure are known, but there is a paucity of evidence from genetic studies of resistant hypertension, especially in the Asian populations. To identify novel genetic loci associated with resistant hypertension in the Japanese population, we conducted a genome-wide association study with 2,705 resistant hypertension cases and 21,296 mild hypertension controls, all from BioBank Japan. We identified one novel susceptibility locus, rs1442386 on chromosome 18p11.3 (DLGAP1), achieving genome-wide significance (odds ratio (95% CI) = 0.85 (0.81–0.90), P = 3.75 × 10−8) and 17 loci showing suggestive association, including rs62525059 of 8q24.3 (CYP11B2). We further detected biological processes associated with resistant hypertension, including chemical synaptic transmission, regulation of transmembrane transport, neuron development and neurological system processes, highlighting the importance of the nervous system. This study provides insights into the etiology of resistant hypertension in the Japanese population.


Introduction
Resistant hypertension is a medical disorder in which patients require at least four antihypertensive drugs of different classes for blood pressure control1,2, and is an increasingly important clinical problem. The exact prevalence of resistant hypertension is unknown. Data from clinical trials, however, suggest that it is not uncommon, affecting about 20 to 30% of patients with hypertension3, 4. In the US National Health and Nutrition Examination Survey (NHANES), the prevalence of uncontrolled hypertension despite use of at least three antihypertensive drugs was reported to increase from 16% of patients treated for hypertension in 1998-2004 to 28% in 2005-20085. Also, Sara dis et al recently estimated that the prevalence of resistant hypertension is 8-12% of adult patients with hypertension, using data from NHANES6.
The etiology of resistant hypertension is unknown, but appears to be multifactorial7, with lifestyle-related, hormonal, and genetic factors having possible important roles. In addition to previously known risk factors, including older age and obesity, medical conditions such as impaired renal function and diabetes mellitus are also associated with resistant hypertension4, 8-10. Elevation of circulating aldosterone level has been identi ed in the majority of patients with resistant hypertension11-13, drawing attention to the importance of aldosterone in the pathogenesis of resistant hypertension14. Also, genetic factors are believed to play a role in the disease etiology15. The speculation that the role of genetic factors may be greater in patients with resistant hypertension than in those with general hypertension would not be inconsistent with excessive blood pressure phenotypes of resistant hypertension. Although previous studies have identi ed numerous genetic variants associated with hypertension and blood pressure16-19, there is a paucity of evidence from genetic studies of resistant hypertension. The available genetic data regarding resistant hypertension are limited and primarily focused on candidate genes20-25. Additionally, pharmacogenomics research on resistant hypertension is in progress15. Recently, a few studies with a comprehensive genetic approach, genome-wide association studies (GWASs), have identi ed some signi cant loci for susceptibility to resistant hypertension in the US population26-28. However, there is still little evidence from genome-wide investigations of resistant hypertension, especially in the Asian population. Therefore, we performed a GWAS to identify novel genetic loci associated with resistant hypertension in the Japanese population.

Data source
The genotyping data and clinical information in this GWAS were obtained from BioBank Japan. The BioBank Japan project, which began in 2003, is a collaborative network of 66 hospitals in all areas of Japan that has collected genomic DNA, serum and clinical information from approximately 270,000 patients diagnosed with any of 51 diseases45,46. All subjects received a detailed explanation, and all signed a written informed consent form. The study protocol conformed to the ethical guidelines of the Declaration of Helsinki and was approved by the Ethics Committees of all participating institutions, including the Institute of Medical Science, the University of Tokyo, the Center for Integrative Medical Sciences, RIKEN, and Nihon University School of Medicine.
Study populationsand phenotype de nitions First, we identi ed 78,463 Japanese patients with hypertension who participated in the BioBank Japan project, using drug prescription data, ful lling the following criteria: 1) Patients who had received at least one prescription of any antihypertensive drug between 2003 and 2012, as listed in Supplementary Table S10. Direct alpha antagonists such as phentolamine and  phenoxybenzamine were excluded from this drug list because they are mainly used for the treatment of pheochromocytoma   and hypertensive emergencies26. Second, we selected study subjects, aged over 40 years, with resistant hypertension as cases and mild hypertension as controls, ful lling the following criteria: 1) The resistant hypertension group was de ned as patients who had received four or more classes of antihypertensive drugs for at least one year, whatever their blood pressure. 2) The control group was de ned as patients with mild hypertension who had received one antihypertensive drug for at least one year. Patients who had received at least one prescription of two or more antihypertensive drugs were excluded from the control group. Consequently, we identi ed a total of 25,450 patients with hypertension (3,978 cases and 21,472 controls) who ful lled the above criteria, and their prescription rates for each class of antihypertensive drugs are listed in Supplementary Table S9. After quality-control and PCA described in detail below, 2,705 cases and 21,296 controls were used to perform a GWAS. The characteristics of the study subjects are summarized in Table 2.

Genotyping and imputation
Genotyping was performed using the Illumina HumanOmniExpressExome BeadChip, or Illumina HumanOmniExpress and HumanExome BeadChip. We aligned the probe sequence in the manifest les of the genotyping array to the GRCh37.3 reference using BLAST to convert genotypes into forward strands. For sample quality control, we excluded samples with (i) sample call rate < 0.98, (ii) closely related samples identi ed using identity by state (IBS) using PLINK47, and (iii) outliers from the East Asian Cluster using PCA for genotype by smartpca48. Then, we applied quality control for genetic variants and excluded SNPs with (i) SNP call rate < 0.99 in both cases and controls, (ii) Hardy-Weinberg equilibrium P ≤ 1 × 10 −6 and MAF < 1%. A Q-Q plot was constructed using observed P values against expected P values and an in ation factor value (λ-value) that was calculated to assess potential population strati cation of the study subjects49.
We pre-phased the genotypes with SHAPEIT50 and imputed dosages with IMPUTE251 using the 1000 Genomes Projects Phase III as a reference52, which was supplied by IMPUTE2 site. For subsequent analysis, we used genotypes with an imputation quality of info ≥ 0.8 and MAF ≥ 5%.

Analysis of GWAS data
For general statistical analysis, we used R statistical environment version 3.4.3 or PLINK1.07. GWAS was used to perform association analysis using imputed allele dosages by snptest53. We set the threshold for genome-wide signi cance at the level of P < 5 × 10 −8 and the threshold for suggestive signi cance at P ≤ 1 × 10 −5 . A Manhattan plot was generated using R software to visualize the results. Regional association plots were generated using LocusZoom54. The online tool HaploReg v4.1 (https://pubs.broadinstitute.org/mammals/haploreg/haploreg.php) was used to explore the genes nearest to the index SNPs, and genes containing a missense mutation in high linkage disequilibrium (LD) (r2 > 0.8) with the GWAS SNPs55. The effects of GWAS SNPs on expression in eQTL studies of different tissues were extracted from the query results of HaploReg.
We de ned an associated locus as a genomic region within ± 1 Mb from the lead SNP. We de ned a locus as novel when it did not include any variants that were previously reported to be signi cantly associated with blood pressure phenotypes, hypertension, or resistant hypertension (P in previous GWAS < 5.0 × 10 -8 ).

Replication of previously reported variants by this GWAS
To verify previously reported variants showing an association (P < 1 × 10 −5 ) with resistant hypertension in a multi-ethnic GWAS dataset26-28,31, we evaluated 26 SNPs in the present Japanese GWAS dataset. Also, we further examined SNPs, previously associated with hypertension or blood pressure phenotypes, in the current GWAS dataset. From the NHGRI European Bioinformatics Institute (NHGRI-EBI) GWAS Catalogue (https://www.ebi.ac.uk/gwas/, accessed July 2020), 172 and 2074 variants showing an association (P < 5 × 10 −8 ) with hypertension and blood pressure phenotypes (including blood pressure, systolic blood pressure, diastolic blood pressure, and mean arterial pressure), respectively were obtained. We then performed a look-up of these previously associated hypertension or blood pressure variants in the current Japanese GWAS dataset.
Gene and pathway-based analysis using VEGAS2 software We performed gene-based association testing using VEGAS2 (version 2) software56. VEGAS2 is an extension of the VErsatile Gene-based Association Study approach which uses 1000 genomes as reference data to estimate linkage disequilibrium between variants within a gene. Based on SNP association P values of GWAS data, the software calculated empirical genebased P values by a simulation procedure. We performed analysis using 1000 Genomes phase 3 East Asian populations (1000G EAS). Gene boundaries were set to ±50 kb of each gene. Up to 10 6 simulations were performed per gene. A total of 24,098 genes were tested, and genes with P < 2.07 × 10 -6 (Bonferroni correction for multiple testing, i.e., 0.05/24,098) were considered to be signi cantly associated with resistant hypertension. Subsequently, we performed pathway analysis using the VEGAS2Pathway approach57. VEGAS2Pathway performs pathway-based association testing and calculates empirical P-values of association for each pathway, while accounting for LD between variants within a gene and between neighboring genes, gene size, and pathway size by using a resampling of gene-based test statistics. The Biosystems gene-pathway annotation le was obtained from the VEGAS2 o cial site (https://vegas2.qimrberghofer.edu.au/biosystems20160324.vegas2pathSYM). The signi cance threshold of the empirical P-value in the pathway analysis was set at 1 × 10 -5 while taking into account the multiple testing of correlated pathways (0.05/5000 independent tests)57.

Data availability
Individual genotyping data and clinical information that support the ndings of this study are publicly available at the National Bioscience Database Center with accession code hum0014 (http://humandbs.biosciencedbc.jp/).

Genome-wide association studyfor resistant hypertension
To clarify the genetic architecture of resistant hypertension, we conducted a GWAS in a Japanese population consisting of 2,705 resistant hypertension cases and 21,296 mild hypertension controls. We evaluated the possibility of population substructure for our sample population by comparison to HapMap samples using principal component analysis (PCA). Although all cases and controls were clustered in the Asian population, a very small portion of the samples was clustered in the Chinese population ( Supplementary Fig. S1a and S1b). We then selected only samples from the major Japanese cluster for further analysis. After whole-genome imputation using the 1000 Genomes Projects as a reference, we examined the association of 6,012,291 SNPs with minor allele frequency (MAF) of more than 5% and an estimated imputation accuracy of greater than 0.8. The quantile-quantile (Q-Q) plot shows the distribution of observed versus expected P values, while the corresponding genomic in ation factor (λ GC ) of 1.057 suggests a low possibility of false-positive associations resulting from population strati cation or cryptic relatedness ( Supplementary Fig. S2). The Manhattan plot, plotting -log 10 (P value) from the GWAS and imputation analysis against the chromosome position, is shown in Fig. 1. Our GWAS identi ed one genetic locus achieving genome-wide signi cance (P < 5× 10 −8 ) and 17 loci showing suggestive association (P < 1 × 10 −5 ) with resistant hypertension in the Japanese population (Table 1). We examined each locus for whether it included any variants that were previously reported to be signi cantly associated with blood pressure phenotypes, hypertension, or resistant hypertension in the 1Mbanking region of each lead variant. We detected one novel locus with signi cant association and three novel loci with suggestive association. The lead variant of the novel signi cant locus was rs1442386 (odds ratio (95% CI) = 0.85 (0.81-0.90), P = 3.75 × 10 −8 ), which is located in the intron region of DLG associated protein 1 (DLGAP1) on chromosome 18p11.3 (Fig. 2). The three novel suggestive loci were PQLC3 (2p25), LOC105369874 (12q14), and MED4 (13q14) locus. The other 14 suggestive loci included variants previously reported to be associated with hypertension or blood pressure phenotypes. For example, variants of the CYP11B2 locus have been frequently validated to be associated with hypertension in multiple populations29,30.

Functional annotation and expression quantitative trait loci (eQTL) analysis
We used HaploReg to perform functional analysis of a total of 18 lead variants showing association with resistant hypertension in the GWAS results. All of these variants were located in non-coding regions (nine intronic and nine intergenic) (Supplementary Table S2). Seven variants were located in gene expression regulatory motifs, such as enhancers, promoters, open chromatins and protein-binding sites in various tissue types. We found that several variants have been identi ed as eQTLs of their nearest genes in various tissue types (Supplementary Table S3). Among them, two variants including rs2075571 of 1q22 and rs9271382 of 6p21 had associations (P < 0.05) with the expression levels of some genes, such as rs2075571 at the THBS3 locus showing an association with THBS3, GBA, RP11-263K19.6, GBAP1 and MUC1 expression, and rs9271382 at the HLA-DQA1 locus showing an association with HLA-DQA1, HLA-DQA2, HLA-DQB1, HLA-DQB1-AS1, HLA-DQB2, HLA-DRB1, HLA-DRB5, and HLA-DRB6 expression in various tissue types. Functional analysis of the rs1442386 variant which reached genome-wide signi cance showed one altered regulatory motif (GCM; glia cells missing) and a signi cant association with DLGAP1 expression in whole blood at P = 0.0018.

Gene-based association analysis
We used VEGAS2 to obtain gene-based P values for phenotypic association from SNP-based P values, using the 1000 Genomes Projects EAS phase 3 reference set. The genes of which the gene-based P values exceeded a Bonferroni-corrected threshold of P < 2.07 × 10 -6 are given in Supplementary Table S4. Gene-based tests identi ed 21 genes associated with resistant hypertension, including GBX1, AGAP3, ASB10, ABCF2 and TMUB1 on chromosome 7q36, ESRP1 and LOC100288748 on 8q22, CYP11B2, CYP11B, GML, LY6D, LYNX1_1, LYNX1_2 and LOC100133669 on 8q24, DLGAP1, DLGAP1-AS3, DLGAP1-AS4 and MIR6718 on 18p11, ERG on 21q22, and GAB4 and CECR7 on 22q11. These genes were located not only at one locus with signi cant association, but also at ve loci with suggestive association, in the current GWAS. For four lead variants, rs253447 of 5q31, rs77163128 of 7p12, rs200741614 of 12q14, and rs11619475 of 13q14, no genes were identi ed by VEGAS2, because these variants were intergenic and were located over 50kb outside the neighboring genes, resulting in them being outside the subject for gene-based association analysis.

Pathway-based association analysis
To further investigate the biological processes involved in resistant hypertension, we performed pathway-based association analysis using the VEGAS2Pathway approach. Figure 3 shows 35 Gene Ontology (GO) terms of biological process (BP) and cellular component (CC) that reached a genome-wide, pathway-based signi cant P value of less than 1 × 10 −5 (Supplementary Table S5). Among them, we observed three prominent sets of GO terms that were highly associated with resistant hypertension. The most numerous set consisted of synapse (GO:0045202) and excitatory synapse (GO:0060076), especially involving postsynaptic compartments (CC term) for chemical synaptic transmission (GO:0007268) (BP term). Subsequently, a set of plasma membrane region (GO:0098590), postsynaptic membrane (GO:0045211), and membrane region (GO:0098589) for regulation of transmembrane transport (GO:0034762), and a set of neuron part (GO:0097458) and neuron projection (GO:0043005) for neuron development (GO:0048666) and neurological system process (GO:0050877) were also highly associated. These results suggest important pathways of the nervous system that may be involved in resistant hypertension.

Evaluation of previously reported variants
To verify previously reported loci showing an association with resistant hypertension, we performed analysis in the current Japanese GWAS dataset (Supplementary Table S6). These 26 SNPs have been previously evaluated in a multi-ethnic GWAS dataset including Caucasian, Hispanic, and African American subjects26-28,31. These SNPs, however, did not show a signi cant association with resistant hypertension in the Japanese population. We further examined whether variants previously associated with blood pressure phenotypes or hypertension showed an association with resistant hypertension in the current GWAS dataset. A total of 2074 and 172 SNPs were selected as variants previously associated with blood pressure and hypertension from the NHGRI-EBI GWAS catalog, respectively (listed in Supplementary Tables S7 and S8). Among these blood pressure-associated variants, eight at three loci showed suggestive associations with resistant hypertension (P < 1 × 10 -5 ) (Supplementary Table S7). The most signi cant association was rs62525059 of 8q24 at the CYP11B2 locus (P = 2.58 × 10 −7 ). The next suggestive associated variants were rs4072037 of 1q22 near MUC1/GBAP1 (P = 5.30 × 10 −6 ) and rs3774427 of 3p21 near CACNA1D (P = 6.51 × 10 −6 ). Also, two loci, including rs62525059 (CYP11B2) and rs3774427 (CACNA1D), showed a suggestive association with resistant hypertension in variants previously associated with hypertension (Supplementary Table  S8), the same as those previously associated with blood pressure. These results suggest the possibility that CYP11B2 (the aldosterone synthase gene) and CACNA1D (a member of the voltage-gated calcium channel gene family) may be involved in the development not only of hypertension, but also of resistant hypertension. However, the current GWAS data that were used to assess the association with resistant hypertension did not successfully replicate a large number of previous GWAS ndings. Most of the previously reported variants associated with blood pressure were established from studies of quantitative traits of blood pressure phenotypes. In addition, previous GWASs of hypertension frequently adopted non-hypertensive subjects or general populations as the control. On the contrary, the present study evaluated a binary outcome using mild hypertensive controls, which may have led to a reduction in statistical power. These differences in our GWAS data may have resulted in the discrepancy in genetic correlations from previous ndings in quantitative outcomes or studies using non-hypertensive controls.

Discussion
To investigate novel susceptibility loci for resistant hypertension, we performed a GWAS in the Japanese population consisting of 2,705 resistant hypertension cases and 21,296 controls. We identi ed a novel locus at chromosome 18p11.31 (DLGAP1) associated with resistant hypertension that reached genome-wide signi cance. Also, we identi ed 17 loci with suggestive association, 14 of which were previously reported to be associated with hypertension or blood pressure (e.g. the CYP11B2 locus).
The lead SNP (rs1442386) of the most signi cant association locus achieving genome-wide signi cance in this study was in the intron region of the DLGAP1 gene, which is exclusively expressed in brain and encodes disks large-associated protein 132. This variant is putatively GCM motif-altering, and is signi cantly associated with DLGAP1 expression in whole blood cells (Supplementary Table S2). The DLGAP1 protein, which localizes at postsynaptic density and interacts with postsynaptic density 95 (PSD95) protein, is involved in signaling at neuronal postsynaptic densities and maintaining normal brain function and development32. Although the ndings of pathway analyses remain putative, our analyses showed that chemical synaptic transmission and regulation of transmembrane transport, including synaptic and trans-synaptic signaling pathways in the synapse and postsynaptic compartments, were signi cantly associated with resistant hypertension. Additionally, we observed signi cant associations of neuron development and neurological system process pathways with resistant hypertension. Our GWAS ndings, combined with pathway analyses, provide an insight that the DLGAP1 protein at the postsynaptic membrane in the central nervous system may contribute to driving resistant hypertension, and highlight the importance of the nervous system in the etiology of resistant hypertension. Genetic variations of DLGAP1 are known to be associated with several psychiatric disorders, such as obsessive-compulsive disorder, schizophrenia, and major depressive disorder33-35. A worldwide epidemiological study has shown that resistant hypertension is associated with mental stress and anxiety36. These ndings suggest that psychological stress may play a possible role in the pathophysiology of resistant hypertension. Further studies are needed to examine the associations between resistant hypertension and anxiety disorders and the possible roles of each in the development of the other.
The lead SNP (rs62525059) of the next suggestive association locus was located 9kb downstream of CYP11B2 (the aldosterone synthase gene), which is the rate limiting step of aldosterone synthesis in humans37. This variant was reported to be associated with blood pressure and hypertension in East Asian populations, including Japanese38,39. In addition, our replication study revealed that some variants, which were previously associated with both hypertension and blood pressure at the CYP11B2 locus, showed a suggestive association with resistant hypertension in our GWAS. Furthermore, gene-based association analysis in this study revealed that the CYP11B2 gene was signi cantly associated with resistant hypertension, supporting the GWAS results. Previous genetic studies of resistant hypertension revealed the association of variants related to the aldosterone and aldosterone pathways, including the beta and gamma subunits of the epithelial sodium channel (ENaC)22,40, angiotensinogen (ATG)41, and CYP4A1142. A recent study in Brazil showed that plasma aldosterone level was signi cantly associated with the -344 C/T CYP11B2 polymorphism in 62 patients with resistant hypertension25. Regarding clinical features of resistant hypertension, excessive aldosterone is implicated in the pathophysiology of resistant hypertension11. Supporting this, elevation of circulating aldosterone level has been identi ed in the majority of patients with resistant hypertension11-13. Although previous GWASs for resistant hypertension in multi-ethnic populations did not detect any variant with a signi cant or suggestive association near the CYP11B2 region26-28,31, our ndings in the Japanese population, in combination with previous genetic and clinical ndings, suggest that the aldosterone synthase gene (CYP11B2) may be a potential causal gene for resistant hypertension, and support the important role of aldosterone and its pathways in the pathophysiology of resistant hypertension, the same as for hypertension.
Our GWAS data did not successfully replicate PTPRD ndings in previous studies on resistant hypertension27,28. This discrepancy may derive in part from differences in race, selection of cases and controls, sample size of each study, and prescription rate for each class of antihypertensive drugs. In previous studies, discovery GWAS was performed using 1529 samples of genotype data of the INternational VErapamil-SR Trandolapril Study (INVEST)-GENEtic Substudy (INVEST-GENES), which collected DNA samples from INVEST study participants43. INVEST was a clinical trial evaluating adverse cardiovascular outcomes in multiple ethnic hypertensive patients with documented coronary artery disease who were randomly assigned to an atenolol-based β-blocker strategy or verapamil-SR-based calcium channel blocker strategy. The prescription rate for β-blockers in previous GWASs was approximately 50 percent in the non-resistant hypertension group from INVEST28. On the other hand, the prescription rate for β-blockers in this study was approximately 5 percent in the control group (Supplementary Table S9). It seems reasonable to speculate that the rate of patients with resistance to β-blockers in previous GWASs was higher than that in our GWAS. As the PTPRD gene is associated with blood pressure response to atenolol27, this difference in the prescription rate of β-blockers may be the reason that previous GWASs discovered an association with resistant hypertension in the PTPRD region, and that we could not nd a susceptibility locus near the PTPRD gene.
Our study has some potential limitations. First, there is a possibility that diagnostic misclassi cation, which may impact on estimation of genetic correlations of GWAS44, may have occurred in our study. Because data on the diagnosis of hypertension were not available in this study, we de ned patients with resistant hypertension based on antihypertensive drug prescription data. Also, data on blood pressure were limited to the rst entry, resulting in incomplete evaluation of blood pressure control after the initiation of antihypertensive medications. Therefore, we adopted a more stringent diagnosis, de ned as patients who had received four or more classes of drugs whatever their blood pressure control. However, this de nition has a concern of potential systematic exclusion of uncontrolled patients who had received three classes of antihypertensive drugs. Second, regarding the control group, we identi ed patients with mild hypertension, de ned as patients who had received one antihypertensive drug for at least one year. This de nition may cause misclassi cation bias in that control samples may have included patients with moderate or severe hypertension, even if a small number of patients, leading to underestimation of genetic correlations. Also, we might have missed some positive results. These concerns call for further studies, such as replication studies using larger samples with diagnosis by physicians, to con rm the validity of our ndings. Third, the ethnicity of all our study subjects was Japanese, and cases were ethnically matched with the controls, limiting the ability to generalize the results. Further studies are needed to con rm the association of the locus and resistant hypertension in other cohorts with different races, because the relevance of our ndings to other ethnic groups remains to be demonstrated.
We identi ed a novel locus at the DLGAP1 gene with susceptibility to resistant hypertension with genome-wide signi cant levels, and 17 loci, including CYP11B2, as having suggestive association with resistant hypertension in the Japanese population. Pathway analysis revealed that chemical synaptic transmission, regulation of transmembrane transport, neuron development and neurological system processes are signi cantly associated with resistant hypertension. The DLGAP1 protein is known to be involved in signaling at neuronal postsynaptic densities and in maintaining normal brain function. Our novel ndings suggest a possible role of DLGAP1, which may contribute to susceptibility to resistant hypertension, possibly via the central nervous system, leading to a new target for drug discovery in the future. In addition, our data suggest that the aldosterone synthase gene (CYP11B2), which is known to be associated with hypertension, may be involved in the development not only of hypertension, but also of resistant hypertension, and underline the importance of this locus in resistant hypertension. SNP, single-nucleotide polymorphism (rsID of lead SNP); Chr; chromosome, Position, physical position of human genome version of GRCh38; RA; risk allele, NRA; non-risk allele, OR; odds ratio, L95; lower 95% con dence limit, U95; upper 95% con dence limit, MAF; minor allele frequency.
Odds ratios (OR) and con dence intervals (CI) were calculated using the non-risk allele as a reference. *: indicates a novel locus for resistant hypertension. The nearest gene is shown as the locus label, but should not be interpreted as the best candidate. A list of all the genes in the 500-kb anking region of the lead SNP is presented in Table S1. Table 2. Baseline characteristics of study population.