Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/ou_press/genome-wide-association-analyses-identify-new-loci-influencing-GRLLffj6br
- Publisher
- Oxford University Press
- Copyright
- © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com
- ISSN
- 0964-6906
- eISSN
- 1460-2083
- DOI
- 10.1093/hmg/ddy111
- Publisher site
- See Article on Publisher Site
Abstract
Abstract Elevated intraocular pressure (IOP) is a significant risk factor for glaucoma, the leading cause of irreversible blindness worldwide. While previous studies have identified numerous genetic variants associated with IOP, these loci only explain a fraction of IOP heritability. Recently established of biobank repositories have resulted in large amounts of data, enabling the identification of the remaining heritability for complex traits. Here, we describe the largest genome-wide association study of IOP to date using participants of European ancestry from the UK Biobank. We identified 671 directly genotyped variants that are significantly associated with IOP (P < 5 × 10−8). In addition to 103 novel loci, the top ranked novel IOP genes are LMX1B, NR1H3, MADD and SEPT9. We replicated these findings in an external population and examined the pleiotropic nature of these loci. These discoveries not only further our understanding of the genetic architecture of IOP, but also shed new light on the biological processes underlying glaucoma. Introduction Intraocular pressure (IOP) is the amount of fluid pressure in the eye which is mainly determined by the balance between aqueous humor production by the ciliary body and drainage through the trabecular meshwork (1). Elevated IOP is a major risk factor for glaucoma, a group of progressive optic neuropathies and the leading cause of irreversible blindness worldwide. Currently, IOP is the only modifiable risk factor for glaucoma and lowering IOP helps to prevent the onset and delay the progression of glaucoma (2–4). Identifying factors that contribute to IOP may aid in uncovering the biological mechanisms regulating this trait, provide new management avenues for IOP, and potentially glaucoma. Epidemiological studies have identified numerous factors associated with IOP, including age, body mass index, systolic blood pressure and type 2 diabetes (5–7). In addition to these systemic factors, studies have provided evidence of a genetic component for IOP determination, with heritability estimates ranging from 29% to 67% (8,9). Several genome-wide association studies (GWASs) have been conducted to identify genetic variants associated with IOP. Initial GWASs identified many genetic loci associated with IOP, including GAS7, TMCO1, GLCCI1-ICA1, ADAMTS18-NUDT7, FOXP1, FAM125B and ARHGEF12 (10–13). Meta-analyses consisting of subjects of European and Asian descent identified additional genomic regions, such as CAV1-CAV2, ABO, FNDC3B, RAPSN, PKHD1 and ADAMTS8 (14–16). Moreover, a recent multi-ethnic meta-analysis identified HIVEP3, ANTXR1, AFAP1, ARID5B, FOXO1 and INCA1 as additional IOP-related loci (17). While these studies identified numerous loci associated with IOP, their findings only explain 2.80–3.66% of the variation in this trait (17), indicating that there are additional variants to discover. The recent establishment of large biobank repositories has resulted in a plethora of both clinical and genomic data. The scope and scale of these biobanks will enable researchers to investigate the association between genetic variants and a multitude of complex traits, increasing the statistical power to detect previously unidentified associations and consequently, to further uncover the remaining heritability for various traits. The UK Biobank is one of the largest biobanks containing both ophthalmologic and genetic data, making it possible to identify novel genetic variants associated with ocular traits, including IOP. Due to the large sample size of the UK Biobank, we believe examining genetic variants associated with IOP will yield novel genomic loci associated with this trait. Therefore, the purpose of this study is to perform a GWAS of IOP using data from the UK Biobank. To the best of our knowledge, the current study represents the largest GWAS of IOP. Findings from this study will elucidate the genetic variants regulating this trait, and subsequently, may further our understanding of the underlying biological mechanisms of glaucoma. Results Study sample The characteristics of the study sample are summarized in Table 1. Overall, a total of 115 486 European white study participants were included in this analysis. The mean (standard deviation) age and IOP are 58.1 (7.9) years and 15.9 (3.5) mmHg, respectively. The proportion of females in this study sample is 53.7%. On an average, male subjects exhibited higher IOP compared with females [16.0 (3.6) mmHg versus 15.8 (3.4) mmHg; P < 0.0001]. Table 1. Descriptive statistics of the study sample Sample size Female, % Age (years), mean (SD) IOP (mmHg), mean (SD) IOP (mmHg) range 115, 486 53.7 58.1 (7.9) 15.9 (3.5) 7.0–39.0 Sample size Female, % Age (years), mean (SD) IOP (mmHg), mean (SD) IOP (mmHg) range 115, 486 53.7 58.1 (7.9) 15.9 (3.5) 7.0–39.0 Table 1. Descriptive statistics of the study sample Sample size Female, % Age (years), mean (SD) IOP (mmHg), mean (SD) IOP (mmHg) range 115, 486 53.7 58.1 (7.9) 15.9 (3.5) 7.0–39.0 Sample size Female, % Age (years), mean (SD) IOP (mmHg), mean (SD) IOP (mmHg) range 115, 486 53.7 58.1 (7.9) 15.9 (3.5) 7.0–39.0 Genome-wide association results For this analysis, the genomic control inflation factor, λ, was 1.16 (Supplementary Material, Fig. S1). Inflation of the genomic control inflation factor may in part be due to the large sample size. The LD score regression intercept, an alternative approach to measure deviation, was 1.004, implying that inflation is primarily due to polygenicity and not confounding. Therefore, our analysis is properly controlled for population structure and cryptic relatedness. Figure 1 presents a Manhattan plot of the genome-wide P values for directly genotyped genetic variants. We identified 671 genome-wide significant genetic variants associated with IOP (P < 5 × 10−8), 66.2% of them are located within functional regions of the genome and 33.8% are located in intergenic regions (Supplementary Material, Table S1). Moreover, these genetic variants map to 149 loci (defined as ±500 kb), including 103 novel loci (Supplementary Material, Table S2). Table 2 presents the summary results for the top genotyped genetic variants for the most significant genes (P < 1 × 10−20). The most significant novel genetic variant is rs10760442 (P = 6.8 × 10−31, GRCh37/hg19 position 129383900), located on chromosome 9q33.3 in an intron of the LMX1B gene. The effect allele G [allele frequency (AF) = 0.63] is associated with an increase in IOP (ß = 0.16). The second most significant novel genetic variant, rs10838681 (P = 9.5 × 10−25, GRCh37/hg19 position 47275064), is located on chromosome 11p11.2 in an intron of NR1H3. The effect allele G (AF = 0.73) is associated with a decrease in IOP (ß = −0.16). The third most significant novel genetic variant is rs326214 (P = 1.0 × 10−23, GRCh37/hg19 position 47298360), located in an exon of MADD on chromosome 11p11.2. The effect allele G (AF = 0.32) is associated with an increase in IOP (ß = 0.15). The fourth most significant genetic variant is rs9038 (P = 1.5 × 10−23, GRCh37/hg19 position 75495397), located on chromosome 17q25.3 in the 3′-UTR of SEPT9. The effect allele T (AF = 0.61) is associated with an increase in IOP (ß = 0.14). Table 2. Lead genotyped SNPs for top-ranking genes associated with IOP UK Biobank Replication SNP Chr Pos Locus Alleles AF1 ß SE P ß SE P rs4656461 1 165687205 LOC440700-TMCO1 G/A 0.12 0.29 0.02 9.8 × 10−44 0.25 0.04 1.4 × 10−10 rs7555523 1 165718979 TMCO1 C/A 0.12 0.31 0.02 1.3 × 10−43 0.26 0.04 2.4 × 10−11 rs1700874 1 219182858 MIR548F3 A/G 0.93 0.28 0.03 2.7 × 10−26 0.18 0.05 5.5 × 10−4 rs4380442 3 171930381 FNDC3B G/A 0.10 −0.26 0.02 1.1 × 10−28 −0.16 0.04 1.4 × 10−4 rs6816389 4 7864457 AFAP1 T/C 0.56 0.17 0.01 1.6 × 10−34 0.10 0.03 2.7 × 10−4 rs1363919 5 52625853 LOC257396-FST T/G 0.52 0.13 0.01 6.5 × 10−24 0.08 0.03 1.5 × 10−3 rs7739648 6 1540483 FOXF2-FOXCUT A/G 0.36 −0.14 0.01 3.6 × 10−21 −0.10 0.03 5.3 × 10−4 rs6946058 7 115839253 TFEC-TES A/C 0.73 0.15 0.02 4.0 × 10−22 0.08 0.03 7.6 × 10−3 rs8940 7 116146074 CAV2 C/G 0.82 −0.23 0.02 3.6 × 10−38 −0.20 0.03 1.5 × 10−9 rs4236601 7 116162729 CAV2-CAV1 G/A 0.73 −0.22 0.02 1.0 × 10−42 −0.19 0.03 2.0 × 10−11 rs13270051 8 108288752 ANGPT1 C/T 0.87 0.23 0.02 9.4 × 10−31 0.09 0.04 1.6 × 10−2 rs1570204 9 4216751 GLIS3 T/C 0.65 −0.17 0.01 7.1 × 10−31 −0.11 0.03 8.2 × 10−5 rs10760442 9 129383900 LMX1B G/A 0.63 0.16 0.01 6.8 × 10−31 0.11 0.03 1.0 × 10−4 rs12778514 10 63832279 ARID5B G/A 0.59 0.14 0.01 1.0 × 10−25 0.09 0.03 1.2 × 10−3 rs1222926 10 95039397 CYP26A1-MYOF C/A 0.36 0.15 0.01 8.5 × 10−27 0.08 0.03 2.7 × 10−3 rs7940065 11 770007 PDDC1 C/T 0.51 0.14 0.01 3.1 × 10−22 0.00 0.03 9.6 × 10−1 rs4963156 11 780827 LOC171391 C/T 0.54 0.14 0.01 3.6 × 10−23 0.00 0.03 9.0 × 10−1 rs10902223 11 817786 RPLP2-PNPLA2 C/T 0.44 0.14 0.01 7.1 × 10−21 0.02 0.03 4.1 × 10−1 rs10838681 11 47275064 NR1H3 G/A 0.73 −0.16 0.02 9.5 × 10−25 −0.11 0.03 1.2 × 10−4 rs326214 11 47298360 MADD G/A 0.32 0.15 0.01 1.0 × 10−23 0.10 0.03 2.3 × 10−4 rs1052373 11 47354787 MYBPC3 C/T 0.68 −0.15 0.02 1.6 × 10−22 −0.10 0.03 1.7 × 10−4 rs3740689 11 47380593 SPI1 G/A 0.41 0.15 0.01 2.6 × 10−26 0.08 0.03 2.1 × 10−3 rs11824864 11 47411736 SPI1-MIR4487 A/G 0.86 −0.20 0.02 8.5 × 10−23 −0.16 0.04 7.4 × 10−6 rs111228939 11 47456067 PSMC3-RAPSN T/C 0.87 −0.21 0.02 6.7 × 10−24 −0.21 0.05 3.0 × 10−6 rs4752843 11 47531884 CELF1 T/G 0.86 −0.19 0.02 3.6 × 10−22 −0.18 0.04 1.6 × 10−6 rs4147730 11 47605427 NDUFS3 G/A 0.86 −0.20 0.02 3.6 × 10−23 −0.18 0.04 1.0 × 10−6 rs4752805 11 48018355 PTPRJ A/G 0.75 −0.15 0.02 1.3 × 10−22 −0.14 0.03 7.1 × 10−6 rs2305013 11 120340060 ARHGEF12 A/T 0.95 −0.32 0.03 4.9 × 10−25 −0.28 0.07 1.8 × 10−5 rs72755233 15 100692953 ADAMTS17 G/A 0.89 −0.31 0.02 4.1 × 10−45 −0.19 0.06 2.3 × 10−3 rs76953588 16 77611387 ADAMTS18-NUDT7 T/C 0.92 −0.26 0.03 9.8 × 10−26 −0.19 0.05 2.1 × 10−4 rs9938149 16 88331640 LOC101928880-ZNF469 C/A 0.37 −0.14 0.01 4.2 × 10−24 −0.07 0.03 7.9 × 10−3 rs12150284 17 10031090 GAS7 C/T 0.63 0.26 0.01 4.6 × 10−77 0.20 0.03 1.7 × 10−12 rs9038 17 75495397 SEPT9 T/C 0.61 0.14 0.01 1.5 × 10−23 0.07 0.03 2.1 × 10−2 UK Biobank Replication SNP Chr Pos Locus Alleles AF1 ß SE P ß SE P rs4656461 1 165687205 LOC440700-TMCO1 G/A 0.12 0.29 0.02 9.8 × 10−44 0.25 0.04 1.4 × 10−10 rs7555523 1 165718979 TMCO1 C/A 0.12 0.31 0.02 1.3 × 10−43 0.26 0.04 2.4 × 10−11 rs1700874 1 219182858 MIR548F3 A/G 0.93 0.28 0.03 2.7 × 10−26 0.18 0.05 5.5 × 10−4 rs4380442 3 171930381 FNDC3B G/A 0.10 −0.26 0.02 1.1 × 10−28 −0.16 0.04 1.4 × 10−4 rs6816389 4 7864457 AFAP1 T/C 0.56 0.17 0.01 1.6 × 10−34 0.10 0.03 2.7 × 10−4 rs1363919 5 52625853 LOC257396-FST T/G 0.52 0.13 0.01 6.5 × 10−24 0.08 0.03 1.5 × 10−3 rs7739648 6 1540483 FOXF2-FOXCUT A/G 0.36 −0.14 0.01 3.6 × 10−21 −0.10 0.03 5.3 × 10−4 rs6946058 7 115839253 TFEC-TES A/C 0.73 0.15 0.02 4.0 × 10−22 0.08 0.03 7.6 × 10−3 rs8940 7 116146074 CAV2 C/G 0.82 −0.23 0.02 3.6 × 10−38 −0.20 0.03 1.5 × 10−9 rs4236601 7 116162729 CAV2-CAV1 G/A 0.73 −0.22 0.02 1.0 × 10−42 −0.19 0.03 2.0 × 10−11 rs13270051 8 108288752 ANGPT1 C/T 0.87 0.23 0.02 9.4 × 10−31 0.09 0.04 1.6 × 10−2 rs1570204 9 4216751 GLIS3 T/C 0.65 −0.17 0.01 7.1 × 10−31 −0.11 0.03 8.2 × 10−5 rs10760442 9 129383900 LMX1B G/A 0.63 0.16 0.01 6.8 × 10−31 0.11 0.03 1.0 × 10−4 rs12778514 10 63832279 ARID5B G/A 0.59 0.14 0.01 1.0 × 10−25 0.09 0.03 1.2 × 10−3 rs1222926 10 95039397 CYP26A1-MYOF C/A 0.36 0.15 0.01 8.5 × 10−27 0.08 0.03 2.7 × 10−3 rs7940065 11 770007 PDDC1 C/T 0.51 0.14 0.01 3.1 × 10−22 0.00 0.03 9.6 × 10−1 rs4963156 11 780827 LOC171391 C/T 0.54 0.14 0.01 3.6 × 10−23 0.00 0.03 9.0 × 10−1 rs10902223 11 817786 RPLP2-PNPLA2 C/T 0.44 0.14 0.01 7.1 × 10−21 0.02 0.03 4.1 × 10−1 rs10838681 11 47275064 NR1H3 G/A 0.73 −0.16 0.02 9.5 × 10−25 −0.11 0.03 1.2 × 10−4 rs326214 11 47298360 MADD G/A 0.32 0.15 0.01 1.0 × 10−23 0.10 0.03 2.3 × 10−4 rs1052373 11 47354787 MYBPC3 C/T 0.68 −0.15 0.02 1.6 × 10−22 −0.10 0.03 1.7 × 10−4 rs3740689 11 47380593 SPI1 G/A 0.41 0.15 0.01 2.6 × 10−26 0.08 0.03 2.1 × 10−3 rs11824864 11 47411736 SPI1-MIR4487 A/G 0.86 −0.20 0.02 8.5 × 10−23 −0.16 0.04 7.4 × 10−6 rs111228939 11 47456067 PSMC3-RAPSN T/C 0.87 −0.21 0.02 6.7 × 10−24 −0.21 0.05 3.0 × 10−6 rs4752843 11 47531884 CELF1 T/G 0.86 −0.19 0.02 3.6 × 10−22 −0.18 0.04 1.6 × 10−6 rs4147730 11 47605427 NDUFS3 G/A 0.86 −0.20 0.02 3.6 × 10−23 −0.18 0.04 1.0 × 10−6 rs4752805 11 48018355 PTPRJ A/G 0.75 −0.15 0.02 1.3 × 10−22 −0.14 0.03 7.1 × 10−6 rs2305013 11 120340060 ARHGEF12 A/T 0.95 −0.32 0.03 4.9 × 10−25 −0.28 0.07 1.8 × 10−5 rs72755233 15 100692953 ADAMTS17 G/A 0.89 −0.31 0.02 4.1 × 10−45 −0.19 0.06 2.3 × 10−3 rs76953588 16 77611387 ADAMTS18-NUDT7 T/C 0.92 −0.26 0.03 9.8 × 10−26 −0.19 0.05 2.1 × 10−4 rs9938149 16 88331640 LOC101928880-ZNF469 C/A 0.37 −0.14 0.01 4.2 × 10−24 −0.07 0.03 7.9 × 10−3 rs12150284 17 10031090 GAS7 C/T 0.63 0.26 0.01 4.6 × 10−77 0.20 0.03 1.7 × 10−12 rs9038 17 75495397 SEPT9 T/C 0.61 0.14 0.01 1.5 × 10−23 0.07 0.03 2.1 × 10−2 SNPs with a P < 1.0 × 10−20 are presented. Gene name is in boldface if the genetic variant is located within the gene. Genomic positions are according to GRCh37/hg19. Chr, chromosome; Pos, position; AF1, allele 1 frequency; SE, standard error. Table 2. Lead genotyped SNPs for top-ranking genes associated with IOP UK Biobank Replication SNP Chr Pos Locus Alleles AF1 ß SE P ß SE P rs4656461 1 165687205 LOC440700-TMCO1 G/A 0.12 0.29 0.02 9.8 × 10−44 0.25 0.04 1.4 × 10−10 rs7555523 1 165718979 TMCO1 C/A 0.12 0.31 0.02 1.3 × 10−43 0.26 0.04 2.4 × 10−11 rs1700874 1 219182858 MIR548F3 A/G 0.93 0.28 0.03 2.7 × 10−26 0.18 0.05 5.5 × 10−4 rs4380442 3 171930381 FNDC3B G/A 0.10 −0.26 0.02 1.1 × 10−28 −0.16 0.04 1.4 × 10−4 rs6816389 4 7864457 AFAP1 T/C 0.56 0.17 0.01 1.6 × 10−34 0.10 0.03 2.7 × 10−4 rs1363919 5 52625853 LOC257396-FST T/G 0.52 0.13 0.01 6.5 × 10−24 0.08 0.03 1.5 × 10−3 rs7739648 6 1540483 FOXF2-FOXCUT A/G 0.36 −0.14 0.01 3.6 × 10−21 −0.10 0.03 5.3 × 10−4 rs6946058 7 115839253 TFEC-TES A/C 0.73 0.15 0.02 4.0 × 10−22 0.08 0.03 7.6 × 10−3 rs8940 7 116146074 CAV2 C/G 0.82 −0.23 0.02 3.6 × 10−38 −0.20 0.03 1.5 × 10−9 rs4236601 7 116162729 CAV2-CAV1 G/A 0.73 −0.22 0.02 1.0 × 10−42 −0.19 0.03 2.0 × 10−11 rs13270051 8 108288752 ANGPT1 C/T 0.87 0.23 0.02 9.4 × 10−31 0.09 0.04 1.6 × 10−2 rs1570204 9 4216751 GLIS3 T/C 0.65 −0.17 0.01 7.1 × 10−31 −0.11 0.03 8.2 × 10−5 rs10760442 9 129383900 LMX1B G/A 0.63 0.16 0.01 6.8 × 10−31 0.11 0.03 1.0 × 10−4 rs12778514 10 63832279 ARID5B G/A 0.59 0.14 0.01 1.0 × 10−25 0.09 0.03 1.2 × 10−3 rs1222926 10 95039397 CYP26A1-MYOF C/A 0.36 0.15 0.01 8.5 × 10−27 0.08 0.03 2.7 × 10−3 rs7940065 11 770007 PDDC1 C/T 0.51 0.14 0.01 3.1 × 10−22 0.00 0.03 9.6 × 10−1 rs4963156 11 780827 LOC171391 C/T 0.54 0.14 0.01 3.6 × 10−23 0.00 0.03 9.0 × 10−1 rs10902223 11 817786 RPLP2-PNPLA2 C/T 0.44 0.14 0.01 7.1 × 10−21 0.02 0.03 4.1 × 10−1 rs10838681 11 47275064 NR1H3 G/A 0.73 −0.16 0.02 9.5 × 10−25 −0.11 0.03 1.2 × 10−4 rs326214 11 47298360 MADD G/A 0.32 0.15 0.01 1.0 × 10−23 0.10 0.03 2.3 × 10−4 rs1052373 11 47354787 MYBPC3 C/T 0.68 −0.15 0.02 1.6 × 10−22 −0.10 0.03 1.7 × 10−4 rs3740689 11 47380593 SPI1 G/A 0.41 0.15 0.01 2.6 × 10−26 0.08 0.03 2.1 × 10−3 rs11824864 11 47411736 SPI1-MIR4487 A/G 0.86 −0.20 0.02 8.5 × 10−23 −0.16 0.04 7.4 × 10−6 rs111228939 11 47456067 PSMC3-RAPSN T/C 0.87 −0.21 0.02 6.7 × 10−24 −0.21 0.05 3.0 × 10−6 rs4752843 11 47531884 CELF1 T/G 0.86 −0.19 0.02 3.6 × 10−22 −0.18 0.04 1.6 × 10−6 rs4147730 11 47605427 NDUFS3 G/A 0.86 −0.20 0.02 3.6 × 10−23 −0.18 0.04 1.0 × 10−6 rs4752805 11 48018355 PTPRJ A/G 0.75 −0.15 0.02 1.3 × 10−22 −0.14 0.03 7.1 × 10−6 rs2305013 11 120340060 ARHGEF12 A/T 0.95 −0.32 0.03 4.9 × 10−25 −0.28 0.07 1.8 × 10−5 rs72755233 15 100692953 ADAMTS17 G/A 0.89 −0.31 0.02 4.1 × 10−45 −0.19 0.06 2.3 × 10−3 rs76953588 16 77611387 ADAMTS18-NUDT7 T/C 0.92 −0.26 0.03 9.8 × 10−26 −0.19 0.05 2.1 × 10−4 rs9938149 16 88331640 LOC101928880-ZNF469 C/A 0.37 −0.14 0.01 4.2 × 10−24 −0.07 0.03 7.9 × 10−3 rs12150284 17 10031090 GAS7 C/T 0.63 0.26 0.01 4.6 × 10−77 0.20 0.03 1.7 × 10−12 rs9038 17 75495397 SEPT9 T/C 0.61 0.14 0.01 1.5 × 10−23 0.07 0.03 2.1 × 10−2 UK Biobank Replication SNP Chr Pos Locus Alleles AF1 ß SE P ß SE P rs4656461 1 165687205 LOC440700-TMCO1 G/A 0.12 0.29 0.02 9.8 × 10−44 0.25 0.04 1.4 × 10−10 rs7555523 1 165718979 TMCO1 C/A 0.12 0.31 0.02 1.3 × 10−43 0.26 0.04 2.4 × 10−11 rs1700874 1 219182858 MIR548F3 A/G 0.93 0.28 0.03 2.7 × 10−26 0.18 0.05 5.5 × 10−4 rs4380442 3 171930381 FNDC3B G/A 0.10 −0.26 0.02 1.1 × 10−28 −0.16 0.04 1.4 × 10−4 rs6816389 4 7864457 AFAP1 T/C 0.56 0.17 0.01 1.6 × 10−34 0.10 0.03 2.7 × 10−4 rs1363919 5 52625853 LOC257396-FST T/G 0.52 0.13 0.01 6.5 × 10−24 0.08 0.03 1.5 × 10−3 rs7739648 6 1540483 FOXF2-FOXCUT A/G 0.36 −0.14 0.01 3.6 × 10−21 −0.10 0.03 5.3 × 10−4 rs6946058 7 115839253 TFEC-TES A/C 0.73 0.15 0.02 4.0 × 10−22 0.08 0.03 7.6 × 10−3 rs8940 7 116146074 CAV2 C/G 0.82 −0.23 0.02 3.6 × 10−38 −0.20 0.03 1.5 × 10−9 rs4236601 7 116162729 CAV2-CAV1 G/A 0.73 −0.22 0.02 1.0 × 10−42 −0.19 0.03 2.0 × 10−11 rs13270051 8 108288752 ANGPT1 C/T 0.87 0.23 0.02 9.4 × 10−31 0.09 0.04 1.6 × 10−2 rs1570204 9 4216751 GLIS3 T/C 0.65 −0.17 0.01 7.1 × 10−31 −0.11 0.03 8.2 × 10−5 rs10760442 9 129383900 LMX1B G/A 0.63 0.16 0.01 6.8 × 10−31 0.11 0.03 1.0 × 10−4 rs12778514 10 63832279 ARID5B G/A 0.59 0.14 0.01 1.0 × 10−25 0.09 0.03 1.2 × 10−3 rs1222926 10 95039397 CYP26A1-MYOF C/A 0.36 0.15 0.01 8.5 × 10−27 0.08 0.03 2.7 × 10−3 rs7940065 11 770007 PDDC1 C/T 0.51 0.14 0.01 3.1 × 10−22 0.00 0.03 9.6 × 10−1 rs4963156 11 780827 LOC171391 C/T 0.54 0.14 0.01 3.6 × 10−23 0.00 0.03 9.0 × 10−1 rs10902223 11 817786 RPLP2-PNPLA2 C/T 0.44 0.14 0.01 7.1 × 10−21 0.02 0.03 4.1 × 10−1 rs10838681 11 47275064 NR1H3 G/A 0.73 −0.16 0.02 9.5 × 10−25 −0.11 0.03 1.2 × 10−4 rs326214 11 47298360 MADD G/A 0.32 0.15 0.01 1.0 × 10−23 0.10 0.03 2.3 × 10−4 rs1052373 11 47354787 MYBPC3 C/T 0.68 −0.15 0.02 1.6 × 10−22 −0.10 0.03 1.7 × 10−4 rs3740689 11 47380593 SPI1 G/A 0.41 0.15 0.01 2.6 × 10−26 0.08 0.03 2.1 × 10−3 rs11824864 11 47411736 SPI1-MIR4487 A/G 0.86 −0.20 0.02 8.5 × 10−23 −0.16 0.04 7.4 × 10−6 rs111228939 11 47456067 PSMC3-RAPSN T/C 0.87 −0.21 0.02 6.7 × 10−24 −0.21 0.05 3.0 × 10−6 rs4752843 11 47531884 CELF1 T/G 0.86 −0.19 0.02 3.6 × 10−22 −0.18 0.04 1.6 × 10−6 rs4147730 11 47605427 NDUFS3 G/A 0.86 −0.20 0.02 3.6 × 10−23 −0.18 0.04 1.0 × 10−6 rs4752805 11 48018355 PTPRJ A/G 0.75 −0.15 0.02 1.3 × 10−22 −0.14 0.03 7.1 × 10−6 rs2305013 11 120340060 ARHGEF12 A/T 0.95 −0.32 0.03 4.9 × 10−25 −0.28 0.07 1.8 × 10−5 rs72755233 15 100692953 ADAMTS17 G/A 0.89 −0.31 0.02 4.1 × 10−45 −0.19 0.06 2.3 × 10−3 rs76953588 16 77611387 ADAMTS18-NUDT7 T/C 0.92 −0.26 0.03 9.8 × 10−26 −0.19 0.05 2.1 × 10−4 rs9938149 16 88331640 LOC101928880-ZNF469 C/A 0.37 −0.14 0.01 4.2 × 10−24 −0.07 0.03 7.9 × 10−3 rs12150284 17 10031090 GAS7 C/T 0.63 0.26 0.01 4.6 × 10−77 0.20 0.03 1.7 × 10−12 rs9038 17 75495397 SEPT9 T/C 0.61 0.14 0.01 1.5 × 10−23 0.07 0.03 2.1 × 10−2 SNPs with a P < 1.0 × 10−20 are presented. Gene name is in boldface if the genetic variant is located within the gene. Genomic positions are according to GRCh37/hg19. Chr, chromosome; Pos, position; AF1, allele 1 frequency; SE, standard error. Figure 1. View largeDownload slide Manhattan plot displaying the –log10(P values) for the association between IOP and the genotyped genetic variants. The solid and dotted horizontal lines represent genome-wide significant associations (P = 5 × 10−8) and suggestive associations (P = 1.0 × 10−6), respectively. Genetic variants are plotted by genomic position. Figure 1. View largeDownload slide Manhattan plot displaying the –log10(P values) for the association between IOP and the genotyped genetic variants. The solid and dotted horizontal lines represent genome-wide significant associations (P = 5 × 10−8) and suggestive associations (P = 1.0 × 10−6), respectively. Genetic variants are plotted by genomic position. Results from imputed genetic variants Using imputed data, we identified 16 854 genetic variants significantly associated with IOP (P < 5 × 10−8), representing 191 loci (Supplementary Material, Table S3). Among the identified regions, 145 are novel loci. Regional association plots were generated for the top four novel genes identified during the analysis of genotyped genetic variants (Fig. 2). After imputation, several imputed genetic variants were more significant than the genotyped variants in all of the identified regions. Figure 2. View largeDownload slide Regional association plots for the top four novel genes associated with IOP. For (A) LMX1B, (B) NR1H3, (C) MADD and (D) SEPT9, the most significant genotyped SNP is plotted in purple. Squares and circles represent genotyped and imputed genetic variants, respectively. Genes are shown below the SNPs with arrows indicating the strand orientation for each gene. The color-coding in each plot represents the level of linkage disequilibrium with the lead SNP in each plot. Figure 2. View largeDownload slide Regional association plots for the top four novel genes associated with IOP. For (A) LMX1B, (B) NR1H3, (C) MADD and (D) SEPT9, the most significant genotyped SNP is plotted in purple. Squares and circles represent genotyped and imputed genetic variants, respectively. Genes are shown below the SNPs with arrows indicating the strand orientation for each gene. The color-coding in each plot represents the level of linkage disequilibrium with the lead SNP in each plot. Conditional analysis We performed conditional analyses on the top four novel genes associated with IOP to determine whether additional genetic variants contribute to the IOP association. After conditioning on the most significant genetic variant in SEPT9 and MADD respectively, all neighboring SNP associations reduced toward the null, suggesting the identified genetic variants are the lead marker of the IOP associations. For the NR1H3 conditional analysis, we identified an additional independent variant, rs60515486, associated with IOP, which suggests additional evidence of association in this region. For the LMX1B conditional analysis, we identified two additional independent variants, rs11795066 and chr9: 129386031, associated with IOP. After conditioning on the lead and secondary genetic variants, the associations for the neighboring SNPs reduced toward the null (Supplementary Material, Fig. S2). Replication of findings To replicate our significant GWAS findings, we compared our results with the summary statistics provided by Springelkamp et al. (16). We used the simpleM method (18–20) to identify the effective number of independent tests and adjusted for multiple testing. Out of 671 genotype significant variants, 651 overlapped with the Springelkamp et al. (16) data, 632 of which were in same effect direction [409 (65%) with nominal significance P < 0.05]. Using the simpleM method, we estimated the effective number of tests and calculated the adjusted Bonferroni correction threshold as P < 1.0 × 10−4. A total of 89 SNPs were replicated in the Springelkamp data at the P < 1.0 × 10−4 significance level. They mapped to 69 unique genes and 55 of them are novel. Out of all 16 854 imputed significant variants, 15 062 were present in Springelkamp et al. (16) data, 14 587 of which exhibited the same direction of effect [9196 (63%) with nominal significance P < 0.05]. A total of 2464 were significant at the P < 1.0 × 10−4 significance level. A total of 160 unique genes, 131 of which are novel, were replicated in the Springelkamp data. Analysis of previously reported IOP loci We evaluated 95 previously reported IOP genetic variants using this study population. Through simpleM (18–20), we identified 76 independent tests yielding a corrected P of 6.6 × 10−4 (0.05/76). Of the previously reported genetic variants, 71 variants were significant and 64 (67.4%) exhibited consistent direction of effects in the current study (Supplementary Material, Table S4). To replicate genomic regions associated with IOP, we examined genetic variants surrounding the previously reported variants. Using this approach and threshold, we replicated 74 regions. IOP heritability estimates We estimated the heritability of IOP among the study participants using BOLT-REML (21). The estimated heritability of IOP using all genotyped genetic variants was 40.4%. When restricted to only significant genetic variants (n = 671), these variants explained 7.2% of the variance. Pleiotropy among significant loci To assess the pleiotropic nature of significant IOP loci with a replicable significant threshold of 5 × 10−8, we examined these loci in the GWAS Catalog. Of the 671 variants (found in 149 unique loci from significant genotyped variants) examined, 68 SNPs had exact matches in the GWAS Catalog (31 January 2018 version). Figure 3 presents the network of related traits and diseases for these 68 SNPs. Numerous neurological disorders related to eye diseases were directly matched to the included SNPs, such as primary open-angle, primary angle closure and high pressure glaucoma, as well as age-related macular degeneration. Moreover, ocular parameters, including central corneal thickness, axial length, optic cup area and iris characteristics, were also mapped. Additionally, SNPs matched to digestive and immune disorders, cancer, and cardiovascular and hematological measurements, including blood pressure, body mass index and type 2 diabetes (Supplementary Material, Table S5). Figure 3. View largeDownload slide A network of the pleiotropy effects of our genome-wide significant SNPs in the GWAS Catalog. A network of traits and diseases for 68 SNPs that matched directly to other phenotypes through the GWAS Catalog. Yellow, orange and red ovals denote SNPs, individual traits and categories, respectively. Figure 3. View largeDownload slide A network of the pleiotropy effects of our genome-wide significant SNPs in the GWAS Catalog. A network of traits and diseases for 68 SNPs that matched directly to other phenotypes through the GWAS Catalog. Yellow, orange and red ovals denote SNPs, individual traits and categories, respectively. Enrichment analysis Table 3 presents the top five GLAD4U diseases and Reactome pathways from WebGestalt using genes derived directly from genotyped genetic variants. The most significant GLAD4U disease associated with IOP is glaucoma (P = 9.59 × 10−5), open-angle glaucoma (P = 9.59 × 10−5), ocular hypertension (P = 3.56 × 10−4), von Willebrand disease (P = 2.67 × 10−3) and ectopia lentis (P = 6.07 × 10−3). Unsurprisingly, the top three diseases relate to glaucoma and have corresponding risk factors. Moreover, the von Willebrand factor is present in the endothelium of Schlemm’s canal and was upregulated in primary open-angle glaucoma cases (22). Mutations in several genes have also been associated with both glaucoma and ectopic lentis, suggesting similar biological mechanisms underlying these two conditions (23). The most significant Reactome pathway associated with IOP is the olfactory signaling pathway (P = 3.91 × 10−3), followed by defective B3GALTL causes Peters-plus syndrome (P = 5.83 × 10−3), O-glycosylation of TSR domain-containing proteins (P = 5.83 × 10−3), ABC transporters in lipid homeostasis (P = 2.93 × 10−2) and extracellular matrix organization (P = 2.93 × 10−2). Olfactory functioning is associated with several neurological diseases, including glaucoma, suggesting that glaucoma may be associated with multisensory manifestations (24). Peters-plus syndrome, an autosomal recessive disease characterized by ocular abnormalities, short stature and developmental delays, is associated with numerous eye conditions, including glaucoma (25). The thrombospondin type 1 repeat superfamily plays a role in neuronal development and ocular homeostasis (26,27). Several ABC transporters have been associated with ocular traits, including IOP (16,28). The extracellular matrix turnover in the trabecular meshwork has been suggested to regulate IOP (29). En masse, results from the enrichment analysis identified biologically relevant diseases and pathways related to IOP. Table 3. Top enrichment results associated with IOP Term Observed/total genes P-value Adjusted P-value Disease Glaucoma 12/133 7.64 × 10−8 9.59 × 10−5 Glaucoma, open angle 10/86 8.61 × 10−8 9.59 × 10−5 Ocular hypertension 10/103 4.80 × 10−7 3.56 × 10−4 von Willebrand disease 7/56 4.80 × 10−6 2.67 × 10−3 Ectopia lentis 4/13 1.36 × 10−5 6.07 × 10−3 Reactome Olfactory signaling pathway 19/387 3.00 × 10−6 3.91 × 10−3 Defective B3GALTL causes Peters-plus syndrome 6/35 9.59 × 10−6 5.83 × 10−3 O-glycosylation of TSR domain-containing proteins 6/37 1.34 × 10−5 5.83 × 10−3 ABC transports in lipid homeostasis 4/18 1.10 × 10−4 2.93 × 10−2 Extracellular matrix organization 14/300 1.12 × 10−4 2.93 × 10−2 Term Observed/total genes P-value Adjusted P-value Disease Glaucoma 12/133 7.64 × 10−8 9.59 × 10−5 Glaucoma, open angle 10/86 8.61 × 10−8 9.59 × 10−5 Ocular hypertension 10/103 4.80 × 10−7 3.56 × 10−4 von Willebrand disease 7/56 4.80 × 10−6 2.67 × 10−3 Ectopia lentis 4/13 1.36 × 10−5 6.07 × 10−3 Reactome Olfactory signaling pathway 19/387 3.00 × 10−6 3.91 × 10−3 Defective B3GALTL causes Peters-plus syndrome 6/35 9.59 × 10−6 5.83 × 10−3 O-glycosylation of TSR domain-containing proteins 6/37 1.34 × 10−5 5.83 × 10−3 ABC transports in lipid homeostasis 4/18 1.10 × 10−4 2.93 × 10−2 Extracellular matrix organization 14/300 1.12 × 10−4 2.93 × 10−2 Table 3. Top enrichment results associated with IOP Term Observed/total genes P-value Adjusted P-value Disease Glaucoma 12/133 7.64 × 10−8 9.59 × 10−5 Glaucoma, open angle 10/86 8.61 × 10−8 9.59 × 10−5 Ocular hypertension 10/103 4.80 × 10−7 3.56 × 10−4 von Willebrand disease 7/56 4.80 × 10−6 2.67 × 10−3 Ectopia lentis 4/13 1.36 × 10−5 6.07 × 10−3 Reactome Olfactory signaling pathway 19/387 3.00 × 10−6 3.91 × 10−3 Defective B3GALTL causes Peters-plus syndrome 6/35 9.59 × 10−6 5.83 × 10−3 O-glycosylation of TSR domain-containing proteins 6/37 1.34 × 10−5 5.83 × 10−3 ABC transports in lipid homeostasis 4/18 1.10 × 10−4 2.93 × 10−2 Extracellular matrix organization 14/300 1.12 × 10−4 2.93 × 10−2 Term Observed/total genes P-value Adjusted P-value Disease Glaucoma 12/133 7.64 × 10−8 9.59 × 10−5 Glaucoma, open angle 10/86 8.61 × 10−8 9.59 × 10−5 Ocular hypertension 10/103 4.80 × 10−7 3.56 × 10−4 von Willebrand disease 7/56 4.80 × 10−6 2.67 × 10−3 Ectopia lentis 4/13 1.36 × 10−5 6.07 × 10−3 Reactome Olfactory signaling pathway 19/387 3.00 × 10−6 3.91 × 10−3 Defective B3GALTL causes Peters-plus syndrome 6/35 9.59 × 10−6 5.83 × 10−3 O-glycosylation of TSR domain-containing proteins 6/37 1.34 × 10−5 5.83 × 10−3 ABC transports in lipid homeostasis 4/18 1.10 × 10−4 2.93 × 10−2 Extracellular matrix organization 14/300 1.12 × 10−4 2.93 × 10−2 Discussion In this study, we conducted the largest GWAS of IOP to date using data from the UK Biobank. Out of the 671 directly genotyped variants associated with IOP we discovered that mapped to 149 loci, 103 loci were novel. This discovery more than doubled the number of known IOP loci. The most significant of these novel genes are LMX1B, NR1H3, MADD and SEPT9. We also replicated 74 previously identified regions associated with IOP. Moreover, results from our enrichment analysis identified several biologically relevant diseases and pathways associated with IOP. Findings from this study demonstrate the polygenic and pleiotrophic nature of IOP loci. Large biobanks have provided an invaluable resource for uncovering the missing heritability of traits and furthering our understanding of the genetic architecture of complex traits. Early GWASs of IOP consisted of several thousand study participants and identified numerous common variants with large effect sizes. These studies, however, were powered to primarily identify the ‘low hanging fruits.’ As such, several strategies hypothesized to identify the ‘mid-hanging fruits’ to uncover the remaining heritability. One strategy included increasing the sample size to increase the statistical power to identify novel variants. By using a sample size nearly twice as large as the most recent IOP GWAS, all genotyped variants were able to account for 40.4% of IOP heritability, while genome-wide significant variants explained 7.2%. Our study illustrates the ability to harvest ‘mid-hanging fruits’ for complex traits by increasing the sample size of a study. The most significant novel genes associated with IOP were LMX1B and NR1H3. LMX1B is a member of the LIM-homeodomain family of transcription factors and plays an essential role in the normal development of serotonergic neurons and the anterior segment of the eye (30,31). Mutations in LMX1B are associated with nail-patella syndrome, of which ocular hypertension and glaucoma are frequently seen in individuals with this syndrome (32,33). Moreover, LMX1B haplotypes were shown to influence glaucoma susceptibility, indicating that alterations in LMX1B may result in glaucomatous damage (34). Transcripts of LMX1B were also detected in numerous ocular tissues, including the ciliary body and the trabecular meshwork, tissues related to the production and drainage of the aqueous humor, respectively (31). NR1H3, also known as the liver x receptor alpha, belongs to the nuclear receptor superfamily that regulates lipid homeostasis and inflammation and may be associated with neuronal degeneration (35). Moreover, activation of the liver x receptors mitigate ocular inflammation and decrease the expression of proinflammatory genes (36). Together, these novel genes represent biologically relevant genes for IOP and may further our understanding of the underlying mechanisms regulating this trait. The third and fourth most significant novel genes were MADD and SEPT9. MADD, from the MAP kinase activating death domain, encodes a protein that interacts with tumor necrosis factor alpha receptor 1 and aids in apoptosis (37). Protein isoforms of MADD were upregulated in glaucomatous retinas (38). Similarly, tumor necrosis factor alpha receptor 1 and the optic nerve head were also upregulated in glaucomatous retinas (38–40). SEPT9 is a member of the septin family, a group of proteins implicated in the formation of the cytoskeleton, cell division and cancer, which interacts with the SEPT2 and SEPT7 proteins. SEPT2, SEPT7 and SEPT9 form a complex in the primary cilium of retinal pigmented epithelial cells (41). Moreover, both the SEPT2 and SEPT7 proteins were detected in the optic nerve head of glaucoma cases and controls (42). These findings suggest that MADD and SEPT9 may relate to pathways associated with neurodegeneration and photoreception. Understanding the pleiotropy of human traits is increasingly important, given the movement toward precision medicine (43). Apropos to the current study, numerous genes associated with IOP also relate to other ocular parameters and systemic factors. Blood pressure is directly associated with IOP and individuals with hypertension have a higher risk of glaucoma (44,45). Such findings suggest that blood pressure treatment may influence IOP and subsequently, an individual’s risk for glaucoma. Similarly, body mass index is positively associated with IOP and interventions to modify body mass index may lead to reductions in IOP (6). Moreover, individuals with diabetes exhibit higher IOP compared with those without diabetes and diabetics are more susceptible to glaucoma (5,46). As such, diabetes interventions may also impact IOP and glaucoma risk. Approved drugs meant for treating a certain disease may possibly be used for IOP management through repositioning based on pleiotropy information. Evidently, elucidating the pleiotropic nature of human traits will uncover the effect of a genetic variant on multiple biological pathways of different traits. This may lead to novel insights for disease prevention and treatment. Despite the aforementioned findings, this study is not without its limitations. Our study did not focus on rare variants with MAF < 0.5%, which may be more appropriate for sequencing data. With the advent of low cost sequencing, however, the ability to discover rare variants will uncover additional heritability. Also, we do not have IOP medication information. Although this is a prospective cohort study and the prevalence of glaucoma is low in our data, the effect of this omission is likely to be insignificant and small. Additionally, the study sample for this analysis consisted of European participants and our findings may not be fully generalizable to other ethnic groups. The nonreplication of some genome-wide significant loci may be due to the limited sample size and different study design of the replication set available. Studies in other ethnic populations are needed to evaluate the transferability of these findings and may lead to the identification of additional novel genetic variants. In conclusion, in the largest GWAS of IOP, we identified 103 novel loci associated with IOP using data from the UK Biobank, substantially increasing the number of loci associated with this trait. In addition, we were able to replicate numerous associations with previously reported IOP genomic regions. We also identified pleiotropic genes that relate to factors associated with IOP, as well as to biologically relevant diseases and pathways through the enrichment analysis. This study identified multiple novel genes associated with IOP, furthering our understanding of the biological mechanisms regulating this trait and potentially, the pathogenesis of glaucoma. Materials and Methods Ethics statement The UK Biobank received approval from the North West Multi-centre Research Ethics Committee and all study participants provided informed consent. Study sample This study was performed using de-identified data from the UK Biobank, as described elsewhere previously (47,48). Briefly, the UK Biobank is a population-based prospective study of 500 000 study participants living in the United Kingdom. Participants between 40 and 69 years of age who were registered with the National Health Service and living less than 25 miles from an assessment center were recruited. Baseline data, collected between 2006 and 2010, consisted of questionnaires regarding lifestyle and medical history, physical measurements, and collection of blood, saliva and urine samples. Additionally, 117 649 study participants at six centers partook in the eye and vision component of the study, where numerous ocular measurements were obtained, including IOP. For the current analysis, we included 115 486 participants who self-reported as white. IOP measurements Initial IOP measurements for each eye were obtained with the Ocular Response Analyzer (Reichert Corp., Philadelphia, PA) between 2009 and 2010, with a repeat assessment between 2012 and 2013. The average Goldmann-corrected IOP of both eyes was used for downstream analysis. If only one IOP measurement was obtained, this measurement was used as a surrogate for the final value. If IOP measurements were available from the repeat assessment, the average of the IOP measurements from the first and second visit was used as the final IOP measurement (Supplementary Material, Fig. S3). The use of a long-term average for quantitative traits has been shown to increase statistical power (49). Study participants who received eye surgery within 4 weeks prior to the ocular assessment or those with possible eye infections did not receive IOP measurements. Furthermore, we excluded study participants with extreme IOP mean values, i.e. IOP < 7 mmHg and IOP > 40 mmHg. Genotyping, imputation and quality control The latest genetic data released on March 2018 was used for this analysis. The genotyping, imputation and quality control protocols are described elsewhere (50). Briefly, this data release includes 488 377 UK Biobank study participants genotyped on either the UK Biobank Axiom Array (825 927 markers; n = 438 427) or the UK BiLEVE Axiom Array (807 411 markers; n = 49 950). Quality control was performed on the genetic markers and individual samples. Individual markers were tested for batch, plate, sex and array effects, as well as departures from the Hardy–Weinberg equilibrium (HWE) and discordance across control replicates and were either excluded or set to missing. Moreover, a small set of duplicate samples were removed. After the above quality control parameters, 805 426 markers from both arrays for 488 377 participants were available in the release. In the current study, autosomal genotyped genetic variants with a minor allele frequency (MAF) ≥ 0.5% and HWE ≥ 1 × 10−12 were retained for downstream analysis, yielding 681 497 variants. We further used imputed genotypes to integrate additional genetic variants not directly genotyped on either array. Phasing of the overlapping genetic variants from the two arrays was performed using SHAPEIT3 (51) with the 1000 Genomes Project (phase 3) dataset. Imputation was conducted using IMPUTE4, a modified version of IMPUTE2, and the 1000 Genomes Project (phase 3), UK10K and Haplotype Reference Consortium reference panels, resulting in 92 693 895 autosomal SNPs, short indels and large structural variants. We applied further quality control parameters to exclude low quality variants (info score < 0.3) and rare variants (MAF < 0.5%), resulting in approximately 11.9 million variants for further analysis. Statistical analysis We performed genetic association analyses using linear mixed models implemented by BOLT-LMM (v2.3) (52), accounting for population structure and cryptic relatedness. Assuming an additive genetic model, we adjusted for age, sex, genotyping array and the first 10 principal components of genetic ancestry (as provided in the data release). Genetic variants with P < 5 × 10−8 were declared genome-wide significant. Independent loci were identified using the PLINK (53) clumping procedure (–clump-r2 0.000001–clump-kb 500). We performed LD score regression using LDSC (v1.0.0) (54). To identify the effective number of independent tests to correct for multiple testing during replication, we used the program simpleM (18–20). SNPs with effects in the same direction as the discovery and P < 0.05 were declared nominally suggestive. Conditional analyses were performed by including the lead genetic variant into the regression model for identified genomic loci. We estimated the heritability of IOP using BOLT-REML (21) and GCTA (55). Enrichment analysis To identify diseases and biological pathways associated with IOP, we performed gene-based association tests and enrichment analysis among unrelated study participants using SKAT-O (56) and WebGestalt (57), respectively. We mapped all genotyped SNPs to genes according to their hg19 assembly genomic position. To capture regulatory elements, we extended gene boundaries to include SNPs ± 50 kb of a gene. For running SKAT-O (56), we adjusted for age, gender and the first 10 principal components of genetic ancestry. Enrichment of GLAD4U, a tool that prioritizes gene lists related to diseases from PubMed, diseases and Reactome pathways associated with IOP were performed using WebGestalt (57). Biological terms were declared significant and reported if P < 0.05 after Benjamin–Hochberg multiple testing correction. Supplementary Material Supplementary Material is available at HMG online. Web Resources The URLs for downloaded data and programs: ANNOVAR, http://annovar.openbioinformatics.org/ BOLT-LMM, https://data.broadinstitute.org/alkesgroup/BOLT-LMM/ GCTA, http://cnsgenomics.com/software/gcta/ GWAS Catalog, https://www.ebi.ac.uk/gwas/ PLINK, https://www.cog-genomics.org/plink2 simpleM, http://simplem.sourceforge.net WebGestalt, http://www.webgestalt.org/LocusZoom, http://locuszoom.org/ Acknowledgements We thank the study participants from the UK Biobank and the staff who aided in data collection and processing. Conflict of Interest statement: None declared. Funding This work was supported by National Institutes of Health (NIH; Bethesda, MD) grants R01EY022651, R01EY027315 and P30EY001792 (departmental core grant). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. References 1 Weinreb R.N. , Leung C.K. , Crowston J.G. , Medeiros F.A. , Friedman D.S. , Wiggs J.L. , Martin K.R. ( 2016 ) Primary open-angle glaucoma . Nat. Rev. Dis. Primers , 2 , 16067 . Google Scholar CrossRef Search ADS PubMed 2 Collaborative Normal-Tension Glaucoma Study Group . ( 1998 ) Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures. Collaborative Normal-Tension Glaucoma Study Group . Am. J. Ophthalmol ., 126 , 487 – 497 . CrossRef Search ADS PubMed 3 Kass M.A. , Heuer D.K. , Higginbotham E.J. , Johnson C.A. , Keltner J.L. , Miller J.P. , Parrish R.K. 2nd , Wilson M.R. , Gordon M.O. ( 2002 ) The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma . Arch. Ophthalmol ., 120 , 701 – 713 . discussion 829–730. Google Scholar CrossRef Search ADS PubMed 4 Heijl A. , Leske M.C. , Bengtsson B. , Hyman L. , Bengtsson B. , Hussein M. and Early Manifest Glaucoma Trial Group . ( 2002 ) Reduction of intraocular pressure and glaucoma progression: results from the Early Manifest Glaucoma Trial . Arch. Ophthalmol ., 120 , 1268 – 1279 . Google Scholar CrossRef Search ADS PubMed 5 Memarzadeh F. , Ying-Lai M. , Azen S.P. , Varma R. and Los Angeles Latino Eye Study Group . ( 2008 ) Associations with intraocular pressure in Latinos: the Los Angeles Latino Eye Study . Am. J. Ophthalmol ., 146 , 69 – 76 . Google Scholar CrossRef Search ADS PubMed 6 Cohen E. , Kramer M. , Shochat T. , Goldberg E. , Garty M. , Krause I. ( 2016 ) Relationship between body mass index and intraocular pressure in men and women: a population-based study . J. Glaucoma , 25 , e509 – e513 . Google Scholar CrossRef Search ADS PubMed 7 Kawase K. , Tomidokoro A. , Araie M. , Iwase A. , Yamamoto T. , Tajimi Study Group and Japan Glaucoma Society . ( 2008 ) Ocular and systemic factors related to intraocular pressure in Japanese adults: the Tajimi study . Br. J. Ophthalmol ., 92 , 1175 – 1179 . Google Scholar CrossRef Search ADS PubMed 8 Chang T.C. , Congdon N.G. , Wojciechowski R. , Munoz B. , Gilbert D. , Chen P. , Friedman D.S. , West S.K. ( 2005 ) Determinants and heritability of intraocular pressure and cup-to-disc ratio in a defined older population . Ophthalmology , 112 , 1186 – 1191 . Google Scholar CrossRef Search ADS PubMed 9 Zheng Y. , Xiang F. , Huang W. , Huang G. , Yin Q. , He M. ( 2009 ) Distribution and heritability of intraocular pressure in Chinese children: the Guangzhou twin eye study . Invest. Ophthalmol. Vis. Sci ., 50 , 2040 – 2043 . Google Scholar CrossRef Search ADS PubMed 10 van Koolwijk L.M. , Ramdas W.D. , Ikram M.K. , Jansonius N.M. , Pasutto F. , Hysi P.G. , Macgregor S. , Janssen S.F. , Hewitt A.W. , Viswanathan A.C. et al. ( 2012 ) Common genetic determinants of intraocular pressure and primary open-angle glaucoma . PLoS Genet ., 8 , e1002611 . Google Scholar CrossRef Search ADS PubMed 11 Strange A. , Bellenguez C. , Sim X. , Luben R. , Hysi P.G. , Ramdas W.D. , van Koolwijk L.M.E. , Freeman C. , Pirinen M. , Su Z. et al. ( 2013 ) Genome-wide association study of intraocular pressure identifies the GLCCI1/ICA1 region as a glaucoma susceptibility locus . Hum. Mol. Genet ., 22 , 4653 – 4660 . Google Scholar CrossRef Search ADS PubMed 12 Nag A. , Venturini C. , Small K.S. International Glaucoma Genetics Consortium Young T.L. , Viswanathan A.C. , Mackey D.A. , Hysi P.G. , Hammond C. ( 2014 ) A genome-wide association study of intra-ocular pressure suggests a novel association in the gene FAM125B in the TwinsUK cohort . Hum. Mol. Genet ., 23 , 3343 – 3348 . Google Scholar CrossRef Search ADS PubMed 13 Springelkamp H. , Iglesias A.I. , Cuellar-Partida G. , Amin N. , Burdon K.P. , van Leeuwen E.M. , Gharahkhani P. , Mishra A. , van der Lee S.J. , Hewitt A.W. et al. ( 2015 ) ARHGEF12 influences the risk of glaucoma by increasing intraocular pressure . Hum. Mol. Genet ., 24 , 2689 – 2699 . Google Scholar CrossRef Search ADS PubMed 14 Hysi P.G. , Cheng C.Y. , Springelkamp H. , Macgregor S. , Bailey J.N.C. , Wojciechowski R. , Vitart V. , Nag A. , Hewitt A.W. , Hohn R. et al. ( 2014 ) Genome-wide analysis of multi-ancestry cohorts identifies new loci influencing intraocular pressure and susceptibility to glaucoma . Nat. Genet ., 46 , 1126 – 1130 . Google Scholar CrossRef Search ADS PubMed 15 Ozel A.B. , Moroi S.E. , Reed D.M. , Nika M. , Schmidt C.M. , Akbari S. , Scott K. , Rozsa F. , Pawar H. , Musch D.C. et al. ( 2014 ) Genome-wide association study and meta-analysis of intraocular pressure . Hum. Genet ., 133 , 41 – 57 . Google Scholar CrossRef Search ADS PubMed 16 Springelkamp H. , Iglesias A.I. , Mishra A. , Hohn R. , Wojciechowski R. , Khawaja A.P. , Nag A. , Wang Y.X. , Wang J.J. , Cuellar-Partida G. et al. ( 2017 ) New insights into the genetics of primary open-angle glaucoma based on meta-analyses of intraocular pressure and optic disc characteristics . Hum. Mol. Genet ., 26 , 438 – 453 . Google Scholar PubMed 17 Choquet H. , Thai K.K. , Yin J. , Hoffmann T.J. , Kvale M.N. , Banda Y. , Schaefer C. , Risch N. , Nair K.S. , Melles R. et al. ( 2017 ) A large multi-ethnic genome-wide association study identifies novel genetic loci for intraocular pressure . Nat. Commun ., 8 , 2108 . Google Scholar CrossRef Search ADS PubMed 18 Gao X. , Starmer J. , Martin E.R. ( 2008 ) A multiple testing correction method for genetic association studies using correlated single nucleotide polymorphisms . Genet. Epidemiol ., 32 , 361 – 369 . Google Scholar CrossRef Search ADS PubMed 19 Gao X. , Becker L.C. , Becker D.M. , Starmer J.D. , Province M.A. ( 2010 ) Avoiding the high Bonferroni penalty in genome-wide association studies . Genet. Epidemiol ., 34 , 100 – 105 . Google Scholar PubMed 20 Gao X. ( 2011 ) Multiple testing corrections for imputed SNPs . Genet. Epidemiol ., 35 , 154 – 158 . Google Scholar CrossRef Search ADS PubMed 21 Loh P.-R. , Bhatia G. , Gusev A. , Finucane H.K. , Bulik-Sullivan B.K. , Pollack S.J. , de Candia T.R. , Lee S.H. , Wray N.R. , Kendler K.S. et al. ( 2015 ) Contrasting genetic architectures of schizophrenia and other complex diseases using fast variance-components analysis . Nat. Genet ., 47 , 1385 – 1392 . Google Scholar CrossRef Search ADS PubMed 22 Rokicki W. , Zaba M. , Mrukwa-Kominek E. , Wygledowska-Promienska D. , Brzozowa M. , Reichman-Warmusz E. , Wojnicz R. ( 2014 ) von Willebrand factor in iris vasculature of glaucoma patients . Med. Sci. Monit ., 20 , 1051 – 1055 . Google Scholar CrossRef Search ADS PubMed 23 Morales J. , Al-Sharif L. , Khalil D.S. , Shinwari J.M. , Bavi P. , Al-Mahrouqi R.A. , Al-Rajhi A. , Alkuraya F.S. , Meyer B.F. , Al Tassan N. ( 2009 ) Homozygous mutations in ADAMTS10 and ADAMTS17 cause lenticular myopia, ectopia lentis, glaucoma, spherophakia, and short stature . Am. J. Hum. Genet ., 85 , 558 – 568 . Google Scholar CrossRef Search ADS PubMed 24 O'Hare F. , Rance G. , McKendrick A.M. , Crowston J.G. ( 2012 ) Is primary open-angle glaucoma part of a generalized sensory neurodegeneration? A review of the evidence . Clin. Exp. Ophthalmol ., 40 , 895 – 905 . Google Scholar CrossRef Search ADS PubMed 25 Maillette de Buy Wenniger-Prick L.J. , Hennekam R.C. ( 2002 ) The Peters’ plus syndrome: a review . Ann. Genet ., 45 , 97 – 103 . Google Scholar CrossRef Search ADS PubMed 26 Adams J.C. , Tucker R.P. ( 2000 ) The thrombospondin type 1 repeat (TSR) superfamily: diverse proteins with related roles in neuronal development . Dev. Dyn ., 218 , 280 – 299 . Google Scholar CrossRef Search ADS PubMed 27 Hiscott P. , Paraoan L. , Choudhary A. , Ordonez J.L. , Al-Khaier A. , Armstrong D.J. ( 2006 ) Thrombospondin 1, thrombospondin 2 and the eye . Prog. Retin. Eye Res ., 25 , 1 – 18 . Google Scholar CrossRef Search ADS PubMed 28 Tarling E.J. , de Aguiar Vallim T.Q. , Edwards P.A. ( 2013 ) Role of ABC transporters in lipid transport and human disease . Trends Endocrinol. Metab ., 24 , 342 – 350 . Google Scholar CrossRef Search ADS PubMed 29 Acott T.S. , Kelley M.J. ( 2008 ) Extracellular matrix in the trabecular meshwork . Exp. Eye Res ., 86 , 543 – 561 . Google Scholar CrossRef Search ADS PubMed 30 Ding Y.Q. , Marklund U. , Yuan W. , Yin J. , Wegman L. , Ericson J. , Deneris E. , Johnson R.L. , Chen Z.F. ( 2003 ) Lmx1b is essential for the development of serotonergic neurons . Nat. Neurosci ., 6 , 933 – 938 . Google Scholar CrossRef Search ADS PubMed 31 Pressman C.L. , Chen H. , Johnson R.L. ( 2000 ) LMX1B, a LIM homeodomain class transcription factor, is necessary for normal development of multiple tissues in the anterior segment of the murine eye . Genesis , 26 , 15 – 25 . Google Scholar CrossRef Search ADS PubMed 32 Vollrath D. , Jaramillo-Babb V.L. , Clough M.V. , McIntosh I. , Scott K.M. , Lichter P.R. , Richards J.E. ( 1998 ) Loss-of-function mutations in the LIM-homeodomain gene, LMX1B, in nail-patella syndrome . Hum. Mol. Genet ., 7 , 1091 – 1098 . Google Scholar CrossRef Search ADS PubMed 33 Mimiwati Z. , Mackey D.A. , Craig J.E. , Mackinnon J.R. , Rait J.L. , Liebelt J.E. , Ayala-Lugo R. , Vollrath D. , Richards J.E. ( 2006 ) Nail-patella syndrome and its association with glaucoma: a review of eight families . Br. J. Ophthalmol ., 90 , 1505 – 1509 . Google Scholar CrossRef Search ADS PubMed 34 Park S. , Jamshidi Y. , Vaideanu D. , Bitner-Glindzicz M. , Fraser S. , Sowden J.C. ( 2009 ) Genetic risk for primary open-angle glaucoma determined by LMX1B haplotypes . Invest. Ophthalmol. Vis. Sci ., 50 , 1522 – 1530 . Google Scholar CrossRef Search ADS PubMed 35 Wang L. , Schuster G.U. , Hultenby K. , Zhang Q. , Andersson S. , Gustafsson J.A. ( 2002 ) Liver X receptors in the central nervous system: from lipid homeostasis to neuronal degeneration . Proc. Natl. Acad. Sci. U. S. A ., 99 , 13878 – 13883 . Google Scholar CrossRef Search ADS PubMed 36 Yang H. , Zheng S. , Qiu Y. , Yang Y. , Wang C. , Yang P. , Li Q. , Lei B. ( 2014 ) Activation of liver X receptor alleviates ocular inflammation in experimental autoimmune uveitis . Invest. Ophthalmol. Vis. Sci ., 55 , 2795 – 2804 . Google Scholar CrossRef Search ADS PubMed 37 Schievella A.R. , Chen J.H. , Graham J.R. , Lin L.L. ( 1997 ) MADD, a novel death domain protein that interacts with the type 1 tumor necrosis factor receptor and activates mitogen-activated protein kinase . J. Biol. Chem ., 272 , 12069 – 12075 . Google Scholar CrossRef Search ADS PubMed 38 Yang X. , Luo C. , Cai J. , Powell D.W. , Yu D. , Kuehn M.H. , Tezel G. ( 2011 ) Neurodegenerative and inflammatory pathway components linked to TNF-alpha/TNFR1 signaling in the glaucomatous human retina . Invest. Ophthalmol. Vis. Sci ., 52 , 8442 – 8454 . Google Scholar CrossRef Search ADS PubMed 39 Yan X. , Tezel G. , Wax M.B. , Edward D.P. ( 2000 ) Matrix metalloproteinases and tumor necrosis factor alpha in glaucomatous optic nerve head . Arch. Ophthalmol ., 118 , 666 – 673 . Google Scholar CrossRef Search ADS PubMed 40 Tezel G. , Li L.Y. , Patil R.V. , Wax M.B. ( 2001 ) TNF-alpha and TNF-alpha receptor-1 in the retina of normal and glaucomatous eyes . Invest. Ophthalmol. Vis. Sci ., 42 , 1787 – 1794 . Google Scholar PubMed 41 Ghossoub R. , Hu Q. , Failler M. , Rouyez M.-C. , Spitzbarth B. , Mostowy S. , Wolfrum U. , Saunier S. , Cossart P. , James Nelson W. et al. ( 2013 ) Septins 2, 7 and 9 and MAP4 colocalize along the axoneme in the primary cilium and control ciliary length . J. Cell Sci ., 126 , 2583 – 2594 . Google Scholar CrossRef Search ADS PubMed 42 Bhattacharya S.K. , Crabb J.S. , Bonilha V.L. , Gu X. , Takahara H. , Crabb J.W. ( 2006 ) Proteomics implicates peptidyl arginine deiminase 2 and optic nerve citrullination in glaucoma pathogenesis . Invest. Ophthalmol. Vis. Sci ., 47 , 2508 – 2514 . Google Scholar CrossRef Search ADS PubMed 43 Gratten J. , Visscher P.M. ( 2016 ) Genetic pleiotropy in complex traits and diseases: implications for genomic medicine . Genome Med ., 8 , 78 . Google Scholar CrossRef Search ADS PubMed 44 Klein B.E. , Klein R. , Knudtson M.D. ( 2005 ) Intraocular pressure and systemic blood pressure: longitudinal perspective: the Beaver Dam Eye Study . Br. J. Ophthalmol ., 89 , 284 – 287 . Google Scholar CrossRef Search ADS PubMed 45 Bae H.W. , Lee N. , Lee H.S. , Hong S. , Seong G.J. , Kim C.Y. ( 2014 ) Systemic hypertension as a risk factor for open-angle glaucoma: a meta-analysis of population-based studies . PLoS One , 9 , e108226 . Google Scholar CrossRef Search ADS PubMed 46 Chopra V. , Varma R. , Francis B.A. , Wu J. , Torres M. , Azen S.P. and Los Angeles Latino Eye Study Group . ( 2008 ) Type 2 diabetes mellitus and the risk of open-angle glaucoma the Los Angeles Latino Eye Study . Ophthalmology , 115 , 227 – 232 . e221. Google Scholar CrossRef Search ADS PubMed 47 Allen N.E. , Sudlow C. , Peakman T. , Collins R. and UK Biobank . ( 2014 ) UK Biobank data: come and get it . Sci. Transl. Med ., 6 , 224ed4 . Google Scholar CrossRef Search ADS PubMed 48 Sudlow C. , Gallacher J. , Allen N. , Beral V. , Burton P. , Danesh J. , Downey P. , Elliott P. , Green J. , Landray M. et al. ( 2015 ) UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age . PLoS Med ., 12 , e1001779 . Google Scholar CrossRef Search ADS PubMed 49 Ganesh S.K. , Chasman D.I. , Larson M.G. , Guo X. , Verwoert G. , Bis J.C. , Gu X. , Smith A.V. , Yang M.L. , Zhang Y. et al. ( 2014 ) Effects of long-term averaging of quantitative blood pressure traits on the detection of genetic associations . Am. J. Hum. Genet ., 95 , 49 – 65 . Google Scholar CrossRef Search ADS PubMed 50 Bycroft C. , Freeman C. , Petkova D. , Band G. , Elliott L.T. , Sharp K. , Motyer A. , Vukcevic D. , Delaneau O. , O'Connell J. et al. ( 2017 ) Genome-wide genetic data on ∼500,000 UK Biobank participants. bioRxiv , doi: https://doi.org/10.1101/166298. 51 O'Connell J. , Sharp K. , Shrine N. , Wain L. , Hall I. , Tobin M. , Zagury J.F. , Delaneau O. , Marchini J. ( 2016 ) Haplotype estimation for biobank-scale data sets . Nat. Genet ., 48 , 817 – 820 . Google Scholar CrossRef Search ADS PubMed 52 Loh P.R. , Tucker G. , Bulik-Sullivan B.K. , Vilhjalmsson B.J. , Finucane H.K. , Salem R.M. , Chasman D.I. , Ridker P.M. , Neale B.M. , Berger B. et al. ( 2015 ) Efficient Bayesian mixed-model analysis increases association power in large cohorts . Nat. Genet ., 47 , 284 – 290 . Google Scholar CrossRef Search ADS PubMed 53 Chang C.C. , Chow C.C. , Tellier L.C. , Vattikuti S. , Purcell S.M. , Lee J.J. ( 2015 ) Second-generation PLINK: rising to the challenge of larger and richer datasets . Gigascience , 4 , 7 . Google Scholar CrossRef Search ADS PubMed 54 Bulik-Sullivan B.K. , Loh P.R. , Finucane H.K. , Ripke S. , Yang J. , Schizophrenia Working Group of the Psychiatric Genomics C. , Patterson N. , Daly M.J. , Price A.L. , Neale B.M. ( 2015 ) LD score regression distinguishes confounding from polygenicity in genome-wide association studies . Nat. Genet ., 47 , 291 – 295 . Google Scholar CrossRef Search ADS PubMed 55 Yang J. , Lee S.H. , Goddard M.E. , Visscher P.M. ( 2011 ) GCTA: a tool for genome-wide complex trait analysis . Am. J. Hum. Genet ., 88 , 76 – 82 . Google Scholar CrossRef Search ADS PubMed 56 Lee S. , Emond M.J. , Bamshad M.J. , Barnes K.C. , Rieder M.J. , Nickerson D.A. N.G.E.S.P.-E.L.P. Team Christiani D.C. , Wurfel M.M. , Lin X. ( 2012 ) Optimal unified approach for rare-variant association testing with application to small-sample case-control whole-exome sequencing studies . Am. J. Hum. Genet ., 91 , 224 – 237 . Google Scholar CrossRef Search ADS PubMed 57 Wang J. , Vasaikar S. , Shi Z. , Greer M. , Zhang B. ( 2017 ) WebGestalt 2017: a more comprehensive, powerful, flexible and interactive gene set enrichment analysis toolkit . Nucleic Acids Res ., 45 , W130 – W137 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Journal
Human Molecular Genetics – Oxford University Press
Published: Mar 28, 2018