Abstract

Primary open-angle glaucoma (POAG) is a hereditary neurodegenerative disease, characterized by optic nerve changes including increased excavation, notching and optic disc hemorrhages. The excavation can be described by the vertical cup–disc ratio (VCDR). Previously, genome-wide significant evidence for the association of rs10483727 in SIX1–SIX6 locus with VCDR and subsequent POAG was found. Using 1000 genomes-based imputation of four independent population-based cohorts in the Netherlands, we identified a missense variant rs33912345 (His141Asn) in SIX6 associated with VCDR (Pmeta = 7.74 × 10−7, n = 11 473) and POAG (Pmeta = 6.09 × 10−3, n = 292). Exome sequencing analysis revealed another missense variant rs146737847 (Glu129Lys) also in SIX6 associated with VCDR (P = 5.09 × 10−3, n = 1208). These two findings point to SIX6 as the responsible gene for the previously reported association signal. Functional characterization of SIX6 in zebrafish revealed that knockdown of six6b led to a small eye phenotype. Histological analysis showed retinal lamination, implying an apparent normal development of the eye, but an underdeveloped lens, and reduced optic nerve diameter. Expression analysis of morphants at 3 dpf showed a 5.5-fold up-regulation of cdkn2b, a cyclin-dependent kinase inhibitor, involved in cell cycle regulation and previously associated with VCDR and POAG in genome-wide association studies (GWASs). Since both six6b and cdkn2b play a key role in cell proliferation, we assessed the proliferative activity in the eye of morphants and found an alteration in the proliferative pattern of retinal cells. Our findings in humans and zebrafish suggest a functional involvement of six6b in early eye development, and open new insights into the genetic architecture of POAG.

INTRODUCTION

Primary open-angle glaucoma (POAG) is a neurodegenerative disease (1) characterized by loss of retinal ganglion cells, optic nerve degeneration, and as a consequence, visual field loss and eventually blindness. It is recognized as a complex disease in which multiple genetic and environmental factors interact (2–4). Known risk factors include age, race, myopia, high intraocular pressure (IOP), decreased central corneal thickness and positive family history (1). First degree relatives of affected individuals are estimated to have a 10-fold increased risk of POAG compared with the general population (5). Heritability estimates of related quantitative traits as optic disc area (ODA), vertical cup–disc ratio (VCDR), IOP and central corneal thickness are high (52–59%; 48–80%; 35–42% and 68–72%, respectively (6–9)).

Developments in the field of genomics have opened opportunities to uncover the genetic mechanisms involved in POAG. To date, at least 20 genetic loci have been linked to POAG, including three causative genes (MYOC, OPTN, WDR36) (10–12). In addition, genome-wide association studies (GWASs) have allowed the identification of candidate genes, such as CAV1/CAV2 (13,14), which are expressed in the trabecular meshwork as well as in retinal ganglion cells, and TMCO1 which is also expressed in retinal ganglion cells (15).

As in other disorders, GWAS of POAG have targeted on endophenotypes, i.e. heritable quantitative determinants of POAG. This approach has facilitated the identification of genes that were subsequently implicated in POAG. Three of the six loci that have been associated with neurodegeneration of the optic nerve (VCDR) (16) were also associated with POAG (17). The strongest associations were seen for the single-nucleotide polymorphism (SNP) rs1900004, within 10 kb of the ATOH7 gene; the rs10483727 near SIX1-SIX6 genes and the rs1063192, in the 3′UTR region of CDKN2B gene.

Two of these variants point to pathways involved in growth and development of the optic nerve. ATOH7 and SIX1–SIX6 genes are transcription factors involved in eye development (18–22). The role of ATOH7 in eye development and retinal ganglion cell differentiation has been well characterized in several animal models (18–20). SIX1 and SIX6 are homeoprotein members of the SIX/sine oculis family of homeobox transcription factors (22). SIX1 is highly expressed in skeletal muscle (23) and has been involved in myogenesis (24), while SIX6 is highly expressed in the developing eye (25,26). To date, it is not clear which one of the two genes is causally related to optic nerve degeneration and POAG, asking for a more in-depth analysis of the SIX1–SIX6 region.

The protein encoded by CDKN2B gene, the third POAG gene identified by GWAS of VCDR, has emerged as a key protein in the pathogenesis of optic nerve degeneration and POAG in different populations (15–17,27–33). CDKN2B gene is a member of the family of cyclin-dependent kinase (CDK) inhibitors which play a role in cell cycle regulation, influencing the proliferation/differentiation balance (34). In the Six6 null mice it was demonstrated that Six6 repress the transcription of members of the Cdkn1 family, particularly Cdkn1b (35). Therefore, it may be speculated that the SIX1–SIX6 region is potentially related to the CDKN2B regulation, although this has not been described.

In this study, we investigated the SIX1–SIX6 locus in depth in human and zebrafish. First, we fine-mapped the region and performed a conditional, and exome sequencing analysis. As we found two variants that pointed to SIX6 as best candidate gene, we next characterized the effect of six6b knockdown on eye development of zebrafish and evaluated expression levels of several target genes in morphants, including cdkn2b.

RESULTS

Fine mapping of the SIX1–SIX6 region

The results of the analyses in the Rotterdam Study (RS) and the Erasmus Rucphen Family (ERF) study are shown in Table 1. The IOP analysis included 5782 (RS-I), 2116 (RS-II), 2038 (RS-III) and 2589 (ERF) participants. The optic nerve head analyses included 5322 (RS-I), 2054 (RS-II), 1966 (RS-III) and 2131 (ERF) participants with reliable data. The case–control studies consisted of 188 cases and 5548 controls from the RS-I and 104 cases and 2126 controls from the GRIP/ERF study.

Table 1.

Evidence of association of SIX6 with VCDR and POAG in RS and ERF

  RS-I
 
RS-II
 
RS-III
 
ERF
 
Meta-analysis
 
 MA MAF β SE P MAF β SE P MAF β SE P MAF β SE P β SE P-meta 
IOP (∼ age, sex, PC1, PC2)                  
rs33912345 0.41 0.034 0.064 5.8 x 10−01 0.40 −0.039 0.096 6.8 x 10−01 0.39 −0.070 0.093 4.5 x 10−01 0.45 −0.251 0.087 4.1 x 10−03 −0.062 0.041 1.3 x 10−01 
rs10483727 0.41 0.035 0.064 5.8 x 10−01 0.40 −0.037 0.096 7.0 x 10−01 0.39 −0.066 0.093 4.8 x 10−01 0.44 −0.303 0.106 4.3 x 10−03 −0.054 0.042 2.0 x 10−01 
DA (∼ age, sex, PC1, PC2)                  
rs33912345 0.41 −0.020 0.009 3.2 x 10−02 0.40 −0.031 0.015 3.4 x 10−02 0.39 −0.019 0.014 1.6 x 10−01 0.46 −0.003 0.011 8.0 x 10−01 −0.017 0.006 3.5 x 10−03 
rs10483727 0.41 −0.020 0.009 2.7 x 10−02 0.40 −0.033 0.015 2.6 x 10−02 0.39 −0.018 0.014 1.7 x 10−01 0.45 −0.006 0.014 6.5 x 10−01 −0.019 0.006 1.7 x 10−03 
VCDR (∼ age, sex, PC1, PC2)                  
rs33912345 0.41 0.013 0.003 4.2 x 10−07 0.40 0.008 0.004 4.5 x 10−02 0.39 0.002 0.007 7.8 x 10−01 0.46 0.014 0.007 3.7 x 10−02 0.011 0.002 7.7 x 10−07 
rs10483727 0.41 0.013 0.003 3.1 x 10−07 0.40 0.009 0.004 3.1 x 10−02 0.39 0.001 0.007 8.4 x 10−01 0.45 0.017 0.008 3.5 x 10−02 0.011 0.002 5.6 x 10−07 
VCDR (∼ age, sex, PC1, PC2, spherical equivalent)                
rs33912345 0.41 0.013 0.003 2.9 x 10−07 0.40 0.009 0.004 2.6 x 10−02 0.39 0.001 0.007 9.1 x 10−01 0.46 0.014 0.007 3.7 x 10−02 0.011 0.002 4.8 x 10−07 
rs10483727 0.41 0.013 0.003 2.2 x 10−07 0.40 0.010 0.004 1.8 x 10−02 0.39 0.000 0.007 9.7 x 10−01 0.45 0.017 0.008 3.5 x 10−02 0.011 0.002 3.5 x 10−07 
VCDR (∼ age, sex, PC1, PC2, rs33912345)                
rs10483727 0.41 0.000 0.003 9.3 x 10−01 0.40 0.001 0.004 8.7 x 10−01 0.39 0.000 0.007 9.4 x 10−01 0.44 0.001 0.008 9.3 x 10−01 0.000 0.002 8.7 x 10−01 
POAG (∼ age, sex, PC1, PC2)                  
rs33912345 0.41 0.252 0.106 1.8 x 10−02 NA NA NA NA NA NA NA NA 0.46 0.230 0.164 1.6 x 10−01 0.244 0.089 6.1 x 10−03 
rs10483727 0.41 0.246 0.106 2.0 x 10−02 NA NA NA NA NA NA NA NA 0.45 0.379 0.194 5.1 x 10−02 0.277 0.093 2.9 x 10−03 
  RS-I
 
RS-II
 
RS-III
 
ERF
 
Meta-analysis
 
 MA MAF β SE P MAF β SE P MAF β SE P MAF β SE P β SE P-meta 
IOP (∼ age, sex, PC1, PC2)                  
rs33912345 0.41 0.034 0.064 5.8 x 10−01 0.40 −0.039 0.096 6.8 x 10−01 0.39 −0.070 0.093 4.5 x 10−01 0.45 −0.251 0.087 4.1 x 10−03 −0.062 0.041 1.3 x 10−01 
rs10483727 0.41 0.035 0.064 5.8 x 10−01 0.40 −0.037 0.096 7.0 x 10−01 0.39 −0.066 0.093 4.8 x 10−01 0.44 −0.303 0.106 4.3 x 10−03 −0.054 0.042 2.0 x 10−01 
DA (∼ age, sex, PC1, PC2)                  
rs33912345 0.41 −0.020 0.009 3.2 x 10−02 0.40 −0.031 0.015 3.4 x 10−02 0.39 −0.019 0.014 1.6 x 10−01 0.46 −0.003 0.011 8.0 x 10−01 −0.017 0.006 3.5 x 10−03 
rs10483727 0.41 −0.020 0.009 2.7 x 10−02 0.40 −0.033 0.015 2.6 x 10−02 0.39 −0.018 0.014 1.7 x 10−01 0.45 −0.006 0.014 6.5 x 10−01 −0.019 0.006 1.7 x 10−03 
VCDR (∼ age, sex, PC1, PC2)                  
rs33912345 0.41 0.013 0.003 4.2 x 10−07 0.40 0.008 0.004 4.5 x 10−02 0.39 0.002 0.007 7.8 x 10−01 0.46 0.014 0.007 3.7 x 10−02 0.011 0.002 7.7 x 10−07 
rs10483727 0.41 0.013 0.003 3.1 x 10−07 0.40 0.009 0.004 3.1 x 10−02 0.39 0.001 0.007 8.4 x 10−01 0.45 0.017 0.008 3.5 x 10−02 0.011 0.002 5.6 x 10−07 
VCDR (∼ age, sex, PC1, PC2, spherical equivalent)                
rs33912345 0.41 0.013 0.003 2.9 x 10−07 0.40 0.009 0.004 2.6 x 10−02 0.39 0.001 0.007 9.1 x 10−01 0.46 0.014 0.007 3.7 x 10−02 0.011 0.002 4.8 x 10−07 
rs10483727 0.41 0.013 0.003 2.2 x 10−07 0.40 0.010 0.004 1.8 x 10−02 0.39 0.000 0.007 9.7 x 10−01 0.45 0.017 0.008 3.5 x 10−02 0.011 0.002 3.5 x 10−07 
VCDR (∼ age, sex, PC1, PC2, rs33912345)                
rs10483727 0.41 0.000 0.003 9.3 x 10−01 0.40 0.001 0.004 8.7 x 10−01 0.39 0.000 0.007 9.4 x 10−01 0.44 0.001 0.008 9.3 x 10−01 0.000 0.002 8.7 x 10−01 
POAG (∼ age, sex, PC1, PC2)                  
rs33912345 0.41 0.252 0.106 1.8 x 10−02 NA NA NA NA NA NA NA NA 0.46 0.230 0.164 1.6 x 10−01 0.244 0.089 6.1 x 10−03 
rs10483727 0.41 0.246 0.106 2.0 x 10−02 NA NA NA NA NA NA NA NA 0.45 0.379 0.194 5.1 x 10−02 0.277 0.093 2.9 x 10−03 

Fine mapping of the SIX1–SIX6 region based on imputations of the SNPs using 1000 Genomes Project revealed a missense variant rs33912345 in SIX6, this variant is in complete linkage disequilibrium with the previous reported intergenic variant rs10483727 (the two D’ and r2 = 1). Both the rs10483727 and the missense variant in SIX6 rs33912345 were associated with VCDR (Pmeta = 5.56 × 10−7 and 7.74 × 10−07, respectively), ODA (Pmeta = 1.69 × 10−3 and 3.51 × 10−3, respectively) and POAG (Pmeta = 2.95 × 10−3 and 6.09 × 10−3, respectively) (see Table 1). The significant association between rs10483727 and VCDR disappeared after adjustment for rs33912345 (Pmeta = 8.72 × 10−1), indicating that the SIX6 variant is most likely responsible for the association. This variant determines an amino acid change (His141Asn) in the homeobox DNA-binding domain of SIX6. PolyPhen-2 (36) predicted a benign effect (score = 0) for this evolutionarily conserved variant, based on the Genomic Evolutionary Rate Profiling (GERP) (37) score = 5.38. No significant association was found with variants in SIX1.

Exome sequencing of SIX1–SIX6 region

To determine whether there were other unobserved rare variants in SIX6, we explored the presence of exonic variants associated with VCDR in this region. In exome sequence data from 1208 individuals of the ERF study, we found another missense variant, rs146737847 (MAF = 0.0023), in SIX6 (P = 1.25 × 10−3). This evolutionarily conserved variant (GERP = 5.38) determines an amino acid change (Glu129Lys) in the homeobox DNA-binding domain of SIX6, which PolyPhen-2 predicted as probably damaging (score = 0.971). When conducting a conditional analysis on rs33912345, the rs146737847 variant remained associated with VCDR (P = 4.58 × 10−3), confirming that both variants are independent of each other. Direction of the effect of both variants was the same; rs33912345 (β = 0.011, se = 0.002) and rs146737847 (β = 0.183, se = 0.055). No exonic variants in SIX1 were significantly associated with VCDR

eQTL and expression analyses

No significant cis or trans-eQTLs effects for rs10483727, rs33912345 or rs14673784 were found in the GTEx project database (38). The Ocular Tissue Database (39) was used to compared the expression levels of SIX1 and SIX6 in different human ocular tissues. We found that SIX6 is highly expressed in relevant tissues for POAG, including cornea, optic nerve, trabecular meshwork and retina, compared with SIX1 (PLIER number 27.33 versus 9.65; 22.25 versus 18.32; 25.80 versus 15.43 and 26.08 versus 14.73; respectively).

Exclusion of an eQTL effect of the evaluated SNPs on the gene expression of SIX1 or SIX6, together with the fine-mapping and exome sequencing data, suggests that the variants in SIX6 rather than SIX1 explain the observed association signal reported in previous studies in the region.

Identification and characterization of zebrafish six6b

A zebrafish six6 ortholog was retrieved from the Ensembl database. Two genes were found, six6a and six6b. The protein encoded by six6a together with the protein encoded by six6b was blasted against the human SIX6 protein, showing that six6a and six6b independently have 91% identity to the human SIX6, while the overall identity of both zebrafish six6 to the human protein is 88% (Supplementary Material, Fig. S1). In this study, we evaluated the effect of six6b depletion.

To evaluate the role of six6b at different time points of the development, a quantitative reverse transcription PCR (RT-qPCR) was performed. Taking reference expression levels at 1 day post fertilization, we found that over time expression of six6b gradually increases, reaching a peak at the larvae stage of 3 dpf, then decreasing at 5 dpf (Fig. 3A).

six6b is required for normal eye development in zebrafish

To evaluate the function of six6b during embryonic development, two non-overlapping antisense MOs were designed, the first one targeting the AUG translation initiation site (six6b AUG-MO) and the other targeting the exon1/intron1 splice site (six6b SB-MO). Since embryos injected with either six6b AUG-MO or six6b SB-MO developed similar dose-dependent (2–12 ng) small eye size compared with controls, the six6b SB-MO (6 ng) was selected for further studies as it provided the possibility of quantifying the efficiency of knockdown by RT-qPCR. Embryos injected with the highest concentration of the MOs (12 ng) showed a severe phenotype (twisted trunk, small head and eye, massive heart edema and lethality at ∼4 dpf, data not shown).

Eye size difference between injected and control embryos was quantified at 5 dpf since the phenotype is then well defined and easy to evaluate (Fig. 1A). Seventy-eight percent of the injected embryos showed a small eye phenotype (n = 286). RT-qPCR analysis indicated six6b SB-MO reduced six6b mRNA levels by 70% (Supplementary Material, Fig. S2). To elucidate whether morphants showed small eye phenotype secondary to six6b knockdown as opposed to an overall delay in development, we measured the ratio of the eye size to body length (E:B) (40). The E:B ratio was measured in 30 embryos of each group (Fig. 1B). Morphants showed small eyes compared with controls, suggesting an eye development delay (mean E:B = 0.064 versus 0.092, P = 4.96 × 10−20). Other abnormalities were observed as pericardial edema, diffuse pigmentation that resembles an expanded melanophore phenotype, small head, and in some embryos short jaw. To exclude the presence of small head as a principal cause of a small eye, we measured the ratio of the eye size to head length (E:H) in ∼30 embryos of each group (i.e. wild-type (WT), six6b SB-MO injected and six6b SB-MO + p53-MO co injected). This measure provides the possibility of evaluating the impact of six6b knockdown on eye size independent of the head size reduction. The E:H ratio was significantly smaller in morphants compared with WT embryos (mean E:H = 0.445 versus 0.530, P = 2.59 × 10−10; Supplementary Material, Fig. S3).

Figure 1.

Phenotype characterization of six6b SB-MO injected fish. (A) six6b knockdown results in a small eye phenotype, other abnormalities including pericardial edema, diffused pigmentation small head and short jaw were also observed. The small eye phenotype was not rescued after co-injection with p53-MO indicating a specific effect of six6b SB-MO on eye size. (B) Ratio of eye to body size (E:B) in morphants and p53 co-injected compared with WT embryos, n = 30 for each group. P-values are indicated in the text. (C) Co-injection of wt six6b mRNA and six6b SB-MO partially rescued the small eye phenotype. The mean eye area improves from 1.425.747 to 1.749.203 µm2, P = 0.001.

Figure 1.

Phenotype characterization of six6b SB-MO injected fish. (A) six6b knockdown results in a small eye phenotype, other abnormalities including pericardial edema, diffused pigmentation small head and short jaw were also observed. The small eye phenotype was not rescued after co-injection with p53-MO indicating a specific effect of six6b SB-MO on eye size. (B) Ratio of eye to body size (E:B) in morphants and p53 co-injected compared with WT embryos, n = 30 for each group. P-values are indicated in the text. (C) Co-injection of wt six6b mRNA and six6b SB-MO partially rescued the small eye phenotype. The mean eye area improves from 1.425.747 to 1.749.203 µm2, P = 0.001.

We performed two control experiments; first, a specific p53-targeting MO (4 ng) was co-injected with the six6b SB-MO (6 ng) to exclude whether the phenotype observed was due to off-target effects mediated by induction of p53 expression. The mean ratio E:B in p53-MO co-injected embryos was 0.061, showing a small eye phenotype comparable with six6b knockdown (P = 2.41 × 10−31). Excluding a small eye size as a consequence of p53-mediated apoptosis pathway. Second, we conducted rescue experiments using WT six6b mRNA. Since no difference in the body length was observed between WT and morphants at 5 dpf (mean = 3.952 mm versus 3.908 mm, respectively; P = 1.95 × 10−1), assessment of rescue experiments was focused on eye dimensions only. Co-injection of six6b SB-MO with wt-six6b mRNA resulted in a partial, but significant, rescue of small eye phenotype (mean eye area improved from 1.425.747to 1.749.203 µm2, P = 1 × 10−3; Fig. 1C and Supplementary Material, Fig. S4).

Loss of six6b results in immature and underdeveloped eyes

In order to carry out detailed assessment of the six6b SB-MO small eye phenotype, histological analysis was performed on embryos at 3 and 5 dpf (Fig. 2A–G). Histology revealed that retinal lamination in morphants occurred. Therefore, the ganglion cell layer (GCL), inner nuclear layer (INL), and outer nuclear layer (ONL) were visualized by H&E staining. Although the optic nerve was seen leaving the retina, its thickness was decreased compared with WT embryos. A semi-quantitative analysis on serial sections of six microns showed that the optic nerve thickness on WT embryos (n = 4) was around 21 µm, while in morphants (n = 4) it was ∼12 µm. In addition, the lens of 3 dpf morphants was immature and cellularized compared with controls of the same age, supporting that eyes of morphants were delayed in development. Eyes of morphants at 5 dpf showed a conserved general architecture, leaving small size as the principal difference between control embryos and morphants at this age.

Figure 2.

six6b knockdown results in small and underdeveloped eyes. (AD) Representative images of histological sections of WT (A and B) and morphants (C and D) at 3 dpf. Though retinal lamination occurred in morphants, note the cellularized and underdeveloped lens. Panels (A) and (C) were hematoxylin–eosin (H&E) stained. (EH) Histological sections of WT (E and F) and morphants (G and H) at 5 dpf. Note that at this age small eye size was the principal difference between the groups, the lens of morphants resembles the WT lens. Panels (B, D, F and H) show immunohistochemical localization of six6/six3 in both WT and morphants. The optic nerve is marked by arrowheads; L = Lens.

Figure 2.

six6b knockdown results in small and underdeveloped eyes. (AD) Representative images of histological sections of WT (A and B) and morphants (C and D) at 3 dpf. Though retinal lamination occurred in morphants, note the cellularized and underdeveloped lens. Panels (A) and (C) were hematoxylin–eosin (H&E) stained. (EH) Histological sections of WT (E and F) and morphants (G and H) at 5 dpf. Note that at this age small eye size was the principal difference between the groups, the lens of morphants resembles the WT lens. Panels (B, D, F and H) show immunohistochemical localization of six6/six3 in both WT and morphants. The optic nerve is marked by arrowheads; L = Lens.

Immunohistochemical localization of six6 was studied using a rabbit anti-SIX6 polyclonal antibody on sections at 3 and 5 dpf (Fig. 2B–H). Since the rabbit anti-SIX6 can recognize the protein products of the closely related six3a and six3b as well, the anti-SIX6 immunoreactivity was only used to investigate the localization of six6/3 proteins in the zebrafish retina, not for the efficiency of six6b knockdown. Both WT and morphants were analyzed. Consistent with previous studies in chick embryos, mouse, and human fetal and adult eye (21,22,41–43), we found six6/3 protein localized in the nuclei of the GCL, the nuclei of potential bipolar, amacrine and horizontal cells located in the INL, the nuclei of the photoreceptors and the optic nerve (Fig. 2B–F). In general, labeling in WT embryos at 3 dpf was stronger than at 5 dpf. Morphants at 3 dpf presented less intense labeling, as predicted from expression data.

Expression patterns of six3a and six3b in six6b morphants

The SIX gene family of homeoproteins is required in both Drosophila and vertebrates for eye development. Similar to SIX6, SIX3 has also been involved in eye development (26,44–46), both genes share high homology and are expressed during early stages (21,25). To determine whether zebrafish six3 can compensate six6b depletion, RT-qPCR of morphants at 3 dpf was performed. Two orthologs of the human SIX3 were retrieved from Ensembl, six3a (ENSDARG00000058008) and six3b (ENSDARG00000054879). RT-qPCR of morphants compared with WT showed an expected down-regulation of six6b. Surprisingly, we found down-regulation of six3a in morphants as well, while expression levels of six3b were not significantly deregulated (Fig. 3B).

Figure 3.

six6, six3a/six3b and cdkn2b mRNA expression change. (A) WT relative expression of six6b, six3a and six3b over time from 1 to 5 dpf. Relative expression was calculated by setting the 1 dpf expression level at 1. Samples expression were normalized to the control gene β-actin (B) six6b knockdown in response to six6b SB-MO at 3 dpf. In addition, morphants showed downregulation of six3a but no six3b (C) Overexpression of cdkn2b in response to six6b depletion. All sample expression from experiments B and C were normalized to the control gene sdha. Relative expression was calculated by setting the WT expression level at 1. Values represent means ± s.e.m. *P < 0.05; **P < 0.01.

Figure 3.

six6, six3a/six3b and cdkn2b mRNA expression change. (A) WT relative expression of six6b, six3a and six3b over time from 1 to 5 dpf. Relative expression was calculated by setting the 1 dpf expression level at 1. Samples expression were normalized to the control gene β-actin (B) six6b knockdown in response to six6b SB-MO at 3 dpf. In addition, morphants showed downregulation of six3a but no six3b (C) Overexpression of cdkn2b in response to six6b depletion. All sample expression from experiments B and C were normalized to the control gene sdha. Relative expression was calculated by setting the WT expression level at 1. Values represent means ± s.e.m. *P < 0.05; **P < 0.01.

Knockdown of six6b up-regulate expression of the cell cycle inhibitor cdkn2b in zebrafish

Since Six6/Dach interacts with some members of the family of cyclin-dependent kinases (CDK) inhibitors (35), we decided to evaluate whether knockdown of six6b in zebrafish alters the expression of cdkn2b, another member of the CDK inhibitors previously associated in our GWAS studies with VCDR (16) and POAG (17). RT-qPCR was performed at 3 dpf on injected and WT embryos, showing a 5.5-fold overexpression of cdkn2b in morphants compared with WT (P = 1.56 × 10−02), while cdkn2c and cdkn2d (taken as a control) remain unaffected (Fig. 3C). Expression levels of dacha and dachb at 3 dpf showed no differences between morphants and controls (data not shown).

Effect of six6b knockdown on cell proliferation in the developing eye of zebrafish

To evaluate whether up-regulation of cdkn2b in morphants leads to premature cell cycle exit, we examined the expression of the proliferating cell nuclear antigen (PCNA) in eye by immunostaining. WT embryos and morphants at 2, 3, 4 and 5 dpf were analyzed.

In WT fish at 2 dpf, almost all retinal progenitor cells showed a strong PCNA immunoreactivity (Fig. 4A). As expected, in WT embryos at 3, 4 and 5 dpf, cells concentrated in the ciliary marginal zone were PCNA-positive (Fig. 4B–D). The ciliary marginal zone is a proliferative zone described in eye of cold-blooded vertebrates (47). Thus, this finding is consistent with normal retinal differentiation processes in zebrafish (48,49). In contrast, morphants at 3 and 4 dpf showed numerous PCNA-positive cells in the photoreceptor and inner nuclear layers (Fig. 4F and G), suggesting that proliferation in morphants was not restricted to the ciliary marginal zone; on the contrary, retinas of morphants at 3 and 4 dpf showed a PCNA immunoreactivity pattern that resembled 2 dpf WT retinas. At 5 dpf morphants showed PCNA immunoreactivity mostly restricted to the ciliary marginal zone comparable with WT embryos (Fig. 4H). We also investigated whether increased apoptosis was the cause of small eye phenotype. Caspase-3 staining did not show an increased apoptosis rate in morphants compared with WT embryos (Supplementary Material, Fig. S5).

Figure 4.

six6b is involved in proliferation and differentiation of retinal cells in zebrafish. (AD) Representative images of PCNA immunoreactivity in WT embryos at 2, 3, 4 and 5 dpf; (A) In WT embryos at 2 dpf almost all cells were PCNA-positive, the lens was cellularized and there was not retinal lamination, while at 3, 4 and 5 dpf (B–D) proliferation is restricted to the ciliary marginal zone, marked by arrowheads. Retinal lamination is clear and the lens is mature and not cellularized. In contrast, morphants at 3 and 4 dpf present proliferating cells located outside the ciliary marginal zone (F and G). Note that at 5 dpf the morphants resembles the WT retina, although some cells located in the central retinal showed a positive PCNA signal (H).

Figure 4.

six6b is involved in proliferation and differentiation of retinal cells in zebrafish. (AD) Representative images of PCNA immunoreactivity in WT embryos at 2, 3, 4 and 5 dpf; (A) In WT embryos at 2 dpf almost all cells were PCNA-positive, the lens was cellularized and there was not retinal lamination, while at 3, 4 and 5 dpf (B–D) proliferation is restricted to the ciliary marginal zone, marked by arrowheads. Retinal lamination is clear and the lens is mature and not cellularized. In contrast, morphants at 3 and 4 dpf present proliferating cells located outside the ciliary marginal zone (F and G). Note that at 5 dpf the morphants resembles the WT retina, although some cells located in the central retinal showed a positive PCNA signal (H).

DISCUSSION

In this study, we investigated the SIX1–SIX6 region in more detail to understand its role in optic nerve degeneration. Fine mapping and conditional analysis revealed a missense variant, rs33912345, in the SIX6 gene. Further analysis of the SIX6 exonic region revealed another rare missense variant, rs146737847. Both variants are located in the homeobox DNA-binding domain of SIX6 and are highly evolutionarily conserved. No common or rare variants in SIX1 were significantly associated to VCDR or POAG.

In line with previous findings, the rs33912345 has been described in patients with severe eye malformations, including anophthalmia, microphthalmia and coloboma (22,50,51). It is in complete linkage disequilibrium with the reported SNP rs10483727, which is known to be associated with VCDR and POAG (16,17,30). The rs146737847 variant has been reported in POAG cases of United States at the American Society of Human Genetics meeting 2012 (http://www.ashg.org/2012meeting/abstracts/fulltext/f120122353.htm).

SIX6 in adult human tissues is highly expressed in the choroid, ciliary body, sclera, optic nerve head and retina, particularly GCL, INL and ONL (43). Mouse, chick and Xenopus studies have demonstrated a role of Six6 during eye development (22,26,35,52), whereas Six1 is highly expressed in skeletal muscle (23) and plays a role in myogenesis (24). Both our findings and the known biological function support the notion that SIX6, rather than SIX1, may play a role in VCDR and POAG.

To evaluate the function of the SIX6 gene on eye development and POAG. We used zebrafish as animal model due to its numerous advantages in comparison to mammalian models (53). We assessed the effect of six6b knockdown in zebrafish. Deficiency of six6b in zebrafish led to a small eye phenotype, associated with small head and pericardial edema. Although morphants showed a small head, we found that the eye of morphants remains smaller independently of the head size. Other abnormalities, including diffused pigmentation were also observed. A clear connection between the visual system, particularly retinal-hypothalamic projections, and the control of pigment cells has been established (54–56). In this study, six6b depleted embryos showed diffused pigmentation, a feature present in mutants who lack the ability to sense ambient light (57). This may indicate an abnormal eye function in six6b morphants. Furthermore, histological sections of six6b morphants at 3 dpf showed a rudimentary lens, which indicates a developmental delay of the eye. Also, the optic nerve had a reduced thickness when compared with age-matched controls, which is relevant in the context of the neurodegenerative disease POAG. This finding is consistent with previous analysis of Six6 null mice (35), which showed retinal hypoplasia, with often absence of optic nerve and chiasm. All these results support our hypothesis that the main effect of six6b knockdown is in the ocular system.

In flies, sine oculis (so) and Optix play a central role in eye development (25,58). In vertebrates, the Six3 and Six6 are the homologs of Optix. Both the Six3 and Six6 show high homology and are expressed during early stages of eye development (25,26). Six3 has a wide role in forebrain development, whereas Six6 has a more specific role in eye development (35,45,46,52,59–61). We evaluated the effect of six6b knockdown on expression levels of six3a and six3b. No significant changes in expression levels of six3b were found in morphants compared with WT embryos. Interestingly, we found down-regulation of six3a in six6b morphants. However, 14 mismatches between the six6b SB-MO and the six3a gene confirmed that our MO did not target six3a. Down-regulation of six3a in six6b morphants might suggest a complex feedback mechanism between six3a/six3b and six6b in zebrafish. Though both Six3 and Six6 might function through similar (62) and different (63) mechanisms, our results show that in zebrafish six6b is needed for normal eye development. Neither six3a nor six3b can compensate for six6b deficiency, resulting in a small eye phenotype.

Studies in medaka fish indicate that Six3 and the replication–initiation inhibitor gemini act antagonistically to regulate the balance between proliferation and differentiation during eye development (62). Six6 also interacts with gemini (62), supporting a potential role of Six6 in cell proliferation and differentiation. Additionally, in the Six6 null mice it was demonstrated that Six6 interacts with Dach1, a transcriptional co-repressor, to suppress Cdkn1b transcription in vivo (35). Cdkn1b is a member of the CDK inhibitor family and controls the cell cycle progression, arresting cell proliferation. CDKN2B, a gene known to be associated with VCDR and POAG (16,17,27,28,30,64), is also a member of the CDK inhibitor family. CDKN2B controls G1 progression of the cell cycle and is highly induced by TGF-β (65), which in turn is one of the pathways involved in POAG pathogenesis (17). We demonstrated a 5.5-fold overexpression of cdkn2b in six6b morphants compared with WT embryos. This finding is in accordance with a rat model of glaucoma (15), in which overexpression of Cdkn2b was described after induction of high IOP. Up-regulation of cdkn2b arrests cell cycle progression causing an abnormal proliferation and small eye phenotype. Based on our results, we conclude that Six6 repress the expression not only of Cdkn1b, as previously reported (35), but also cdkn2b. This reveals a new interaction between the two known pathways associated with POAG (developmental and TGF-β signaling pathways) (Fig. 5A; www.ingenuity.com).

Figure 5.

six6b and cdkn2b interaction. (A) Ingenuity diagram of the biological interaction of the known genes identified through the GWAS of optic disc parameters (VCDR, ODA and IOP); the bold dashed line shows the new interaction found in this study using a zebrafish model. The Ingenuity knowledge base is a repository of biological interactions and functional annotations of in vivo and in vitro experiments (www.ingenuity.com). The diagram was generated using the function ‘Path Explorer’. (B) six6b represses cdkn2b transcription in association with the co-represor dach. six6b and dach form a complex that interacts with the regulatory region of cdkn2b. This interaction blocks cdkn2b transcription and thus controls the proliferative state of retinal progenitor cells. VCDR, vertical cup–disc ratio; ODA, optic disc area; IOP, intraocular pressure.

Figure 5.

six6b and cdkn2b interaction. (A) Ingenuity diagram of the biological interaction of the known genes identified through the GWAS of optic disc parameters (VCDR, ODA and IOP); the bold dashed line shows the new interaction found in this study using a zebrafish model. The Ingenuity knowledge base is a repository of biological interactions and functional annotations of in vivo and in vitro experiments (www.ingenuity.com). The diagram was generated using the function ‘Path Explorer’. (B) six6b represses cdkn2b transcription in association with the co-represor dach. six6b and dach form a complex that interacts with the regulatory region of cdkn2b. This interaction blocks cdkn2b transcription and thus controls the proliferative state of retinal progenitor cells. VCDR, vertical cup–disc ratio; ODA, optic disc area; IOP, intraocular pressure.

Next, we evaluated the proliferative activity of retinal cells through PCNA assessment. PCNA allows detection of cells in either G1 or S phase of the cell cycle. We found that knockdown of six6b changes the pattern of positive PCNA labeling. Similar results were found in a previous study, in which a collagen gene (drCol15a1b) was knockdown in zebrafish (66). It is well established that retinal progenitor cells start a differentiation process when they exit the cell cycle (66,67). Analysis of drCol15a1b morphants indicated that cells arrested in a proliferative state were undifferentiated (66). Our results demonstrate an abnormal proliferation in the presence of an efficient knockdown of six6b with a recovery at 5 dpf. This could be explained by a reduced morpholino activity, as a consequence of its dilution due to cell growth and division. Based on our findings, we hypothesized that decreased expression of six6b may lead to fewer differentiated cells as depicted by our proposed model in Figure 5B. This needs to be assessed in future studies by evaluating the presence of retinal-differentiated cells in the eyes of six6b morphants. Caspase-3 assessment did not show an increased number of apoptotic cells in the eye of six6b-depleted embryos.

In conclusion, by combining fine-mapping and exome sequence analysis we found that SIX6, rather than SIX1, is responsible for the association signal previously reported in the SIX1/SIX6 region. We found two variants located in the DNA-binding homeodomain of SIX6, which were significantly associated with VCDR. The rs33912345 variant was also associated with POAG in our study. Knockdown of six6b in zebrafish led to a small eye phenotype, as a result of an abnormal proliferative pattern in retinal cells. This alteration may also be associated with a disturbance in the differentiation process of retinal progenitor cells. Additionally, we were able to find an in vivo interaction between proteins identified by the GWAS (SIX6 and CDKN2B) for POAG. Further biological research is needed to verify the exact mechanism behind this interaction. Our findings from genetic association and functional studies demonstrate that SIX6 is a susceptibility gene for POAG and is involved in early eye development. This study provides new insight into the complex genetic architecture of POAG.

MATERIALS AND METHODS

Human and fine-mapping studies

Study populations

The RS is a prospective cohort study in the district Ommoord of Rotterdam. The rationale and study design are described elsewhere (68). The original cohort (RS-I) consisted of 7983 residents aged ≥55 years. In 2000, the cohort was extended with 3011 residents aged ≥55 years (RS-II). In 2006, the cohort was further expanded with 3932 residents aged 45–54 years (RS-III). Details about examinations in each one of the cohorts are described elsewhere (16).

The ERF Study is a family-based study in a genetically isolated population in the southwest of the Netherlands. It includes over 3000 participants with age varying between 18 and 86 years old. The cross-sectional examinations took place between 2002 and 2005. The rationale and study design have been described elsewhere (69,70).

An independent series of 104 glaucoma cases from an isolated population (the Genetic Research in an Isolated Population [GRIP] study) was used for a case–control analysis. These patients were recruited in three local hospitals in the geographical region of the ERF study. They did not participate in the ERF study. Participants from the ERF study were used as the control population.

Ophthalmic examinations

The ophthalmic assessment in the RS and ERF study included a medical history, autorefraction, keratometry, IOP measurements, visual field testing, fundus photography and optic nerve head imaging. Details about IOP and optic disc measurements are described elsewhere (16,71). In RS-I, glaucoma diagnosis was based on glaucomatous visual field loss (GVFL). The visual field of each eye was screened using a 52-point supra-threshold test that covered the central visual field with a radius of 24°, and that tested the same locations as used in the Glaucoma Hemifield Test. Participants with a reproducible visual field loss in a second supra-threshold test, were evaluated with a Goldmann kinetic perimetry or full-threshold HFA test by a skilled perimetrist on both eyes. We incorporated prevalent glaucoma cases, (72) as well as incident cases (73) based on GVFL during at least one of the examination rounds. VCDR per se was not an inclusion criteria of the diagnosis. In GRIP/ERF, the diagnosis was based on the optic disc appearance, visual field testing and angle assessment.

The study protocols were approved by the Medical Ethics Committee of the Erasmus University, and all participants have given a written informed consent in accordance with the Declaration of Helsinki.

Genotyping and statistical analyses of fine-mapping studies

Illumina Infinium II HumanHap550 chip v3.0 array was used for DNA genotyping in RS-I, RS-II and RS-III. In the ERF study, genotyping was performed using Illumina 318 K, Illumina 370 K, Illumina 610 K and Affymetrix 250 K genotyping platforms. Imputation for both studies was done by using 1000 Genomes Phase 1 (v3) as the reference.

We used the mean IOP, VCDR or ODA of both eyes. In cases of missing or unreliable data on one eye, the other eye was taken. Participants with a history of glaucoma laser or surgery were excluded for the IOP analysis. IOP measurements of participants with IOP lowering medication were imputed by adding 25% to the value of the measurement. We used 1000 Genomes Phase 1(v3) to perform fine mapping and modeling of SIX1/SIX6 locus in relation to these continuous outcomes on an assumed additive model for the effect of the risk allele. Additionally, we examined through logistic regression the association between the SIX1/SIX6 locus and primary open-angle glaucoma. All analyses were adjusted for age, sex, the first two principal components (RS) or family structure (ERF). Extra adjustments were done with spherical equivalent, as a measure of myopia, to evaluate whether the missense variant rs33912345 was independently associated with VCDR. The spherical equivalent was calculated for each eye using the standard formula: spherical equivalent = sphere + 1/2 cylinder. Additionally, independence between rs10483727 and rs33912345 was tested, adjusting for the last one. Then, an inverse variance weighted meta-analysis was performed.

Exome sequencing

Exomes of 1309 individuals from the ERF population were sequenced ‘in-house’ at the Center for Biomics of the Cell Biology department of the Erasmus MC, The Netherlands, using the Aligent version V4 capture kit on an Illumina Hiseq2000 sequencer and the TrueSeq Version3 protocol. The sequences reads were aligned to the human genome build 19 (hg19) using Burrows-Wheeler Aligner (74) and NARWHAL pipeline (75). For each sample, at least 4 Gigabases of sequence was aligned to the genome with an average fold-coverage of 74.23x per base. Subsequently, the aligned reads were processed further using the Indel Realigner, Mark Duplicates and Table Recalibration tools from the Genome Analysis Toolkit (GATK) (76) and Picard (http: picard.sourceforge.net) to remove systematic biases and to recalibrate the PHRED quality scores in the alignments. Genetic variants were called using the Unified Genotyper Tool from GATK. Of the 1309 individuals, 1208 had data on VCDR. As the ERF study included related individuals, association analysis was performed in SOLAR using procedure ‘polygenic –screen’, and adjusting for age and sex. In total, two variants in SIX1 and seven variants in SIX6 with a call rate of >0.99 were found. Only one of the nine variants at SIX6, rs146737847, reached significance levels. In addition, conditional analyses were performed with the rs33912345, previously found in the fine-mapping analysis.

eQTL and expression analyses

We used the Genotype-Tissue expression (GTex) project database (www.broadinstitute.org/gtex/) to examine whether rs10483727, rs33912345 or rs146737847 had a cis- or trans-eQTL effect. Although the GTex database does not include eye tissues, it contains more than 10 different brain regions and other tissues. For trans-eQTL analysis, a P-value of 5 × 10−13 was considered significant (38).

Additionally, we compared expression levels of SIX1 and SIX6 using the Ocular Tissue Database (https://genome.uiowa.edu/otdb/). The gene expression is indicated as Affymetrix Probe Logarithmic Intensity Error (PLIER) number. The PLIER numbers were calculated by GC-background correction, PLIER normalization, log transformation and z-score calculation (39). Larger PLIER numbers represent high expression levels of a particular gene in particular tissue.

Statistical analysis of zebrafish experiments

Statistical analysis was performed using SPSS statistics 20. Analysis of E:B and E:H ratios was performed using the Student's t-test. Each group (six6b SB-MO injected, p53 MO injected and six6b SB-MO + p53 MO co-injected) was compared individually with the WT group (n = 30, in each group). All RT-qPCR experiments were undertaken in triplicate. Results were considered statistically significant when P < 0.05, and are denoted in the figures with an asterisk.

Zebrafish maintenance

The zebrafish (Danio rerio) strain used for this work was the Tupfel long fin (TL). Adults were maintained at 28°C on a 14 hour-light/10 h dark cycle. Embryos were collected from natural mating and raised in system water containing methylene blue at 28°C. Developmental stages were determined according to Kimmel (77). All procedures and conditions were in accordance with the Dutch animal welfare legislation. The use of zebrafish for this study was approved by the Institutional Review Board for experimental animals of the Erasmus MC, Rotterdam.

Genetic analysis of the zebrafish SIX6 ortholog

The zebrafish six6 ortholog was retrieved from Ensembl using the comparative genomics tool (78). Two genes were found, six6a (ENSDARG00000025187) and six6b (ENSDARG00000031316). In this study, we reported the effect of six6b knockdown. We sequenced WT zebrafish six6b to corroborate the reference sequence and select the regions for morpholino oligonucleotide (MO) design.

Morpholino and mRNA microinjections

Morpholino antisense oligonucleotides were obtained from Gene-Tools (Philomath, OR, USA). Two morpholinos were designed against six6b. One morpholino was designed to target the translation initiation site: AUG-MO = 5′-AAATTGGCAACTGAAACATGAAGGC-3′; and the second morpholino targeted exon 1 donor site: SB-MO = 5′-TGTAAATCTGGAAAACGCACCTGTT-3′. All MO sequences were aligned with the Danio rerio genome using NCBI and UCSC Blast to confirm specificity to the six6b genomic region and ratify that neither the AUG-MO nor the SB-MO recognized six6a, six3a or six3b. A previously described MO designed to target the translation site of p53 (p53-MO = 5′-GCGCCATTGCTTTGCAAGAATTG-3′ was used for control experiments (79). Morpholinos were reconstituted in distilled water and further diluted in Danieau solution containing 0.1% Phenol Red for microinjections into embryos.

Indicated dosages of MOs were injected into the yolk of one to two cell stage using a pneumatic picopump (World Precision Instruments, Berlin, Germany). Optimum dose was considered when maximal knockdown efficiency, low mortality rate and small eye phenotype without severe malformations were observed on injected embryos. Six nanograms of either SB or AUG-MO were selected as optimal concentration. In separate experiments, 4 ng of p53-MO was co-injected with 6 ng of six6b-specific MO (SB or AUG), in order to detect off-target effects due to activation of p53 expression.

For MO rescue experiments, WT Danio rerio six6b cDNA was ligated into the pCR2.1-TOPO vector (Invitrogen), and subcloned at the site of EcoRI in pCS2+ vector. The fidelity of six6b-pCS2 was verified by direct sequencing. Using as template NsiI linearized six6b-pCS2, WT six6b mRNA was synthesized using a mMESSAGE mMACHINE SP6 kit (Ambion Inc). Double injection of mRNA and MO was carried out in two steps, first 3 ng six6b SB-MO were injected into the yolk, then a second injection with 9 pg six6b mRNA into the cell was performed, all embryos were injected in one-cell stage. Eye size was evaluated at 3 dpf.

RNA isolation and RT-qPCR

Pooled embryos were snap frozen in liquid nitrogen and stored at −80°C. Total RNA from ∼60 WT embryos at 1,2,3,4 and 5 dpf was isolated for time series analyses. For evaluation of six6b knockdown, total RNA was isolated from ∼60 morphants and WT embryos at 3 dpf; in all extractions an RNA-Bee (Tri-Test, Inc) protocol was used. Synthesis of cDNA was performed using Superscript III reverse transcriptase (Invitrogen, CA, USA). To measure mRNA levels, qRT-PCR on cDNA samples was carried out using an SYBR® select Master Mix for CFX (Applied Biosystems, Inc., USA). All samples were analyzed on the Bio-Rad CFX96 qPCR detection system.

Primers used for qRT-PCR were designed using Primer3Plus tool (80). Primers for the reference genes sdha and b-actin were designed using Primer Express software (version 2.0.0). Oligonucleotide sequences are shown in detail in Supplementary Material, Table S1.

Eye measurements

WT and morphant embryos at 5 dpf were euthanized and fixed in 4% paraformaldehyde at 4°C overnight. Photographs were made at a fixed magnification of the dorsal view of the whole embryo with a camera (Olympus DP72) attached to a dissecting microscope (Olympus SZX 16). The cellSense imaging software (Olympus) was used to measure eye, head and body length for each embryo. Eye length was defined as the longest dimension of the elliptical eye, head length was determined from the most anterior part of the head to the otic vesicles, as previously described (81) and body length was measured from the tip of the head to the end of the trunk. The ratio of the eye size to body length (E:B) was measured to determine the relative size of the eye (40). The ratio of the eye size to head length (E:H) was measured to evaluate the effect of six6b depletion on eye size, independently of the head length.

For the assessment of the rescue experiments, three eye measurements from the dorsal view of embryos at 3 pdf were made. First, the eye length (defined previously) and called anterior–posterior length (AP), second the superior–inferior (SI) length defined as the longest dimension along the superior–inferior axis and third the total eye area. All measurements were traced using the Count & Measure function of the cellSense imaging software.

Histological analysis

Histological analysis was performed using standard protocols. Briefly, embryos were fixed in 4% paraformaldehyde at 4°C overnight, embedded in paraffin using standard procedures and cut in 6 µm sections. Subsequently, histological hematoxylin–eosin staining of the sections was carried out using a standard protocol.

Immunohistochemistry

Briefly, deparaffinnized sections were pretreated for antigen retrieval by microwave heating in 0.1 m sodium citrate buffer (ph 6) for time series of 9-3-3 min. Expression pattern of six6 in larvae at 3 dpf was evaluated using a rabbit polyclonal antibody against six6 (1:400, Sigma). Proliferating cells were labeled using a mouse monoclonal antibody to PCNA (1:32000, Sigma). All antibody incubations were carried out overnight at 4°C followed by incubation with a secondary antibody (BrightVision Poly-HRP-anti-rabbit IgG or poly-HRP-Anti-mouse, respectively) and indirect immunoperoxidase labeling and hematoxylin counter stain was performed.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

FUNDING

This study was supported by Stichting Lijf en Leven. The ERF study was supported by grants from the Netherlands Organization for Scientific Research (NWO) and a joint grant from NWO and the Russian Foundation for Basic research (Pionier, 047.016.009, 047.017.043), Erasmus Medical Center, and the Centre for Medical Systems Biology (CMSB; National Genomics Initiative). Exome sequencing analysis in ERF was supported by the ZonMw grant (91111025). The generation and management of GWAS genotype data for the RS is supported by the Netherlands Organization for Scientific Research NWO Investments (nr. 175.010.2005.011, 911-03-012). This study is funded by the Research Institute for Diseases in the Elderly (014-93-015; RIDE2); the Netherlands Genomics Initiative Netherlands Organization for Scientific Research (NWO) (project nr. 050-060-810); CHANCES (nr 242244). The RS is funded by Erasmus Medical Center and Erasmus University, Rotterdam, Netherlands Organization for the Health Research and Development (ZonMw), the Research Institute for Diseases in the Elderly (RIDE), the Ministry of Education, Culture and Science, the Ministry for Health, Welfare and Sports, the European Commission (DG XII), and the Municipality of Rotterdam; M.D. Fonds; Stichting Oogfonds Nederland; Henkes Stichting; Stichting Nederlands Oogheelkundig Onderzoek; Swart van Essen; Bevordering van Volkskracht; Blindenhulp; Landelijke Stichting voor Blinden en Slechtzienden; Rotterdamse Vereniging voor Blindenbelangen; OOG; Algemene Nederlandse Vereniging ter Voorkoming van Blindheid; the Rotterdam Eye Hospital Research Foundation; Medical Workshop; Heidelberg Engineering; Topcon Europe BV.

ACKNOWLEDGEMENTS

For the RS, we thank Ada Hooghart, Corina Brussee, Riet Bernaerts-Biskop and Patricia van Hilten for the ophthalmic data collection; Pascal Arp, Mila Jhamai, Marijn Verkerk, for their help in creating the GWAS database; Carolina Medina-Gomez and Fernando Rivadeneira for their support in creation of 1000 genomes imputed data. The authors are grateful to the study participants, the staff from the RS and the participating general practitioners and pharmacists. For the ERF study, we are grateful to all participants and their relatives, to general practitioners and neurologists for their contributions, to P. Veraart for her help in genealogy, to Jeannette Vergeer for the supervision of the laboratory work, to P. Snijders for his help in data collection and to Elisa van Leeuwen, Maarten Kooijman and Lennart Karssen for their support in creation of 1000 genomes imputed data.

Conflict of Interest statement. None declared.

REFERENCES

1
Kwon
Y.H.
Fingert
J.H.
Kuehn
M.H.
Alward
W.L.
Primary open-angle glaucoma
N. Engl. J. Med.
 , 
2009
, vol. 
360
 (pg. 
1113
-
1124
)
2
Ramdas
W.D.
Amin
N.
van Koolwijk
L.M.
Janssens
A.C.
Demirkan
A.
de Jong
P.T.
Aulchenko
Y.S.
Wolfs
R.C.
Hofman
A.
Rivadeneira
F.
, et al.  . 
Genetic architecture of open angle glaucoma and related determinants
J. Med. Genet.
 , 
2011
, vol. 
48
 (pg. 
190
-
196
)
3
Ramdas
W.D.
Wolfs
R.C.
Hofman
A.
de Jong
P.T.
Vingerling
J.R.
Jansonius
N.M.
Lifestyle and risk of developing open-angle glaucoma: the Rotterdam study
Arch. Ophthalmol.
 , 
2011
, vol. 
129
 (pg. 
767
-
772
)
4
Ramdas
W.D.
Wolfs
R.C.
Hofman
A.
de Jong
P.T.
Vingerling
J.R.
Jansonius
N.M.
Ocular perfusion pressure and the incidence of glaucoma: real effect or artifact? The Rotterdam Study
Invest. Ophthalmol. Vis. Sci.
 , 
2011
, vol. 
52
 (pg. 
6875
-
6881
)
5
Wolfs
R.C.
Klaver
C.C.
Ramrattan
R.S.
van Duijn
C.M.
Hofman
A.
de Jong
P.T.
Genetic risk of primary open-angle glaucoma. Population-based familial aggregation study
Arch. Ophthalmol.
 , 
1998
, vol. 
116
 (pg. 
1640
-
1645
)
6
Schwartz
J.T.
Reuling
F.H.
Feinleib
M.
Size of the physiologic cup of the optic nerve head. hereditary and environmental factors
Arch. Ophthalmol.
 , 
1975
, vol. 
93
 (pg. 
776
-
778
)
7
Freeman
E.E.
Roy-Gagnon
M.H.
Descovich
D.
Masse
H.
Lesk
M.R.
The heritability of glaucoma-related traits corneal hysteresis, central corneal thickness, intraocular pressure, and choroidal blood flow pulsatility
PLoS One
 , 
2013
, vol. 
8
 pg. 
e55573
 
8
Klein
B.E.
Klein
R.
Lee
K.E.
Heritability of risk factors for primary open-angle glaucoma: the Beaver Dam Eye Study
Invest. Ophthalmol. Vis. Sci.
 , 
2004
, vol. 
45
 (pg. 
59
-
62
)
9
van Koolwijk
L.M.
Despriet
D.D.
van Duijn
C.M.
Pardo Cortes
L.M.
Vingerling
J.R.
Aulchenko
Y.S.
Oostra
B.A.
Klaver
C.C.
Lemij
H.G.
Genetic contributions to glaucoma: heritability of intraocular pressure, retinal nerve fiber layer thickness, and optic disc morphology
Invest. Ophthalmol. Vis. Sci.
 , 
2007
, vol. 
48
 (pg. 
3669
-
3676
)
10
Stone
E.M.
Fingert
J.H.
Alward
W.L.
Nguyen
T.D.
Polansky
J.R.
Sunden
S.L.
Nishimura
D.
Clark
A.F.
Nystuen
A.
Nichols
B.E.
, et al.  . 
Identification of a gene that causes primary open angle glaucoma
Science
 , 
1997
, vol. 
275
 (pg. 
668
-
670
)
11
Rezaie
T.
Child
A.
Hitchings
R.
Brice
G.
Miller
L.
Coca-Prados
M.
Heon
E.
Krupin
T.
Ritch
R.
Kreutzer
D.
, et al.  . 
Adult-onset primary open-angle glaucoma caused by mutations in optineurin
Science
 , 
2002
, vol. 
295
 (pg. 
1077
-
1079
)
12
Monemi
S.
Spaeth
G.
DaSilva
A.
Popinchalk
S.
Ilitchev
E.
Liebmann
J.
Ritch
R.
Heon
E.
Crick
R.P.
Child
A.
, et al.  . 
Identification of a novel adult-onset primary open-angle glaucoma (POAG) gene on 5q22.1
Hum. Mol. Genet.
 , 
2005
, vol. 
14
 (pg. 
725
-
733
)
13
Wiggs
J.L.
Kang
J.H.
Yaspan
B.L.
Mirel
D.B.
Laurie
C.
Crenshaw
A.
Brodeur
W.
Gogarten
S.
Olson
L.M.
Abdrabou
W.
, et al.  . 
Common variants near CAV1 and CAV2 are associated with primary open-angle glaucoma in Caucasians from the USA
Hum. Mol. Genet.
 , 
2011
, vol. 
20
 (pg. 
4707
-
4713
)
14
Thorleifsson
G.
Walters
G.B.
Hewitt
A.W.
Masson
G.
Helgason
A.
DeWan
A.
Sigurdsson
A.
Jonasdottir
A.
Gudjonsson
S.A.
Magnusson
K.P.
, et al.  . 
Common variants near CAV1 and CAV2 are associated with primary open-angle glaucoma
Nat. Genet.
 , 
2010
, vol. 
42
 (pg. 
906
-
909
)
15
Burdon
K.P.
Macgregor
S.
Hewitt
A.W.
Sharma
S.
Chidlow
G.
Mills
R.A.
Danoy
P.
Casson
R.
Viswanathan
A.C.
Liu
J.Z.
, et al.  . 
Genome-wide association study identifies susceptibility loci for open angle glaucoma at TMCO1 and CDKN2B-AS1
Nat. Genet.
 , 
2011
, vol. 
43
 (pg. 
574
-
578
)
16
Ramdas
W.D.
van Koolwijk
L.M.
Ikram
M.K.
Jansonius
N.M.
de Jong
P.T.
Bergen
A.A.
Isaacs
A.
Amin
N.
Aulchenko
Y.S.
Wolfs
R.C.
, et al.  . 
A genome-wide association study of optic disc parameters
PLoS Genet.
 , 
2010
, vol. 
6
 pg. 
e1000978
 
17
Ramdas
W.D.
van Koolwijk
L.M.
Lemij
H.G.
Pasutto
F.
Cree
A.J.
Thorleifsson
G.
Janssen
S.F.
Jacoline
T.B.
Amin
N.
Rivadeneira
F.
, et al.  . 
Common genetic variants associated with open-angle glaucoma
Hum. Mol. Genet.
 , 
2011
, vol. 
20
 (pg. 
2464
-
2471
)
18
Brown
N.L.
Patel
S.
Brzezinski
J.
Glaser
T.
Math5 is required for retinal ganglion cell and optic nerve formation
Development
 , 
2001
, vol. 
128
 (pg. 
2497
-
2508
)
19
Mao
C.A.
Cho
J.H.
Wang
J.
Gao
Z.
Pan
P.
Tsai
W.W.
Frishman
L.J.
Klein
W.H.
Reprogramming amacrine and photoreceptor progenitors into retinal ganglion cells by replacing Neurod1 with Atoh7
Development
 , 
2013
, vol. 
140
 (pg. 
541
-
551
)
20
Prasov
L.
Nagy
M.
Rudolph
D.D.
Glaser
T.
Math5 (Atoh7) gene dosage limits retinal ganglion cell genesis
Neuroreport
 , 
2012
, vol. 
23
 (pg. 
631
-
634
)
21
Jean
D.
Bernier
G.
Gruss
P.
Six6 (Optx2) is a novel murine Six3-related homeobox gene that demarcates the presumptive pituitary/hypothalamic axis and the ventral optic stalk
Mech. Dev.
 , 
1999
, vol. 
84
 (pg. 
31
-
40
)
22
Gallardo
M.E.
Lopez-Rios
J.
Fernaud-Espinosa
I.
Granadino
B.
Sanz
R.
Ramos
C.
Ayuso
C.
Seller
M.J.
Brunner
H.G.
Bovolenta
P.
, et al.  . 
Genomic cloning and characterization of the human homeobox gene SIX6 reveals a cluster of SIX genes in chromosome 14 and associates SIX6 hemizygosity with bilateral anophthalmia and pituitary anomalies
Genomics
 , 
1999
, vol. 
61
 (pg. 
82
-
91
)
23
Boucher
C.A.
Carey
N.
Edwards
Y.H.
Siciliano
M.J.
Johnson
K.J.
Cloning of the human SIX1 gene and its assignment to chromosome 14
Genomics
 , 
1996
, vol. 
33
 (pg. 
140
-
142
)
24
Ridgeway
A.G.
Skerjanc
I.S.
Pax3 is essential for skeletal myogenesis and the expression of Six1 and Eya2
J. Biol. Chem.
 , 
2001
, vol. 
276
 (pg. 
19033
-
19039
)
25
Serikaku
M.A.
O'Tousa
J.E.
Sine oculis is a homeobox gene required for Drosophila visual system development
Genetics
 , 
1994
, vol. 
138
 (pg. 
1137
-
1150
)
26
Oliver
G.
Mailhos
A.
Wehr
R.
Copeland
N.G.
Jenkins
N.A.
Gruss
P.
Six3, a murine homologue of the sine oculis gene, demarcates the most anterior border of the developing neural plate and is expressed during eye development
Development
 , 
1995
, vol. 
121
 (pg. 
4045
-
4055
)
27
Fan
B.J.
Wang
D.Y.
Pasquale
L.R.
Haines
J.L.
Wiggs
J.L.
Genetic variants associated with optic nerve vertical cup-to-disc ratio are risk factors for primary open angle glaucoma in a US Caucasian population
Invest. Ophthalmol. Vis. Sci.
 , 
2011
, vol. 
52
 (pg. 
1788
-
1792
)
28
Cao
D.
Jiao
X.
Liu
X.
Hennis
A.
Leske
M.C.
Nemesure
B.
Hejtmancik
J.F.
CDKN2B polymorphism is associated with primary open-angle glaucoma (POAG) in the Afro-Caribbean population of Barbados, West Indies
PLoS One
 , 
2012
, vol. 
7
 pg. 
e39278
 
29
Pasquale
L.R.
Loomis
S.J.
Kang
J.H.
Yaspan
B.L.
Abdrabou
W.
Budenz
D.L.
Chen
T.C.
Delbono
E.
Friedman
D.S.
Gaasterland
D.
, et al.  . 
CDKN2B-AS1 genotype-glaucoma feature correlations in primary open-angle glaucoma patients from the United States
Am. J. Ophthalmol.
 , 
2013
, vol. 
155
 (pg. 
342
-
353 e345
)
30
Osman
W.
Low
S.K.
Takahashi
A.
Kubo
M.
Nakamura
Y.
A genome-wide association study in the Japanese population confirms 9p21 and 14q23 as susceptibility loci for primary open angle glaucoma
Hum. Mol. Genet.
 , 
2012
, vol. 
21
 (pg. 
2836
-
2842
)
31
Burdon
K.P.
Crawford
A.
Casson
R.J.
Hewitt
A.W.
Landers
J.
Danoy
P.
Mackey
D.A.
Mitchell
P.
Healey
P.R.
Craig
J.E.
Glaucoma risk alleles at CDKN2B-AS1 are associated with lower intraocular pressure, normal-tension glaucoma, and advanced glaucoma
Ophthalmology
 , 
2012
, vol. 
119
 (pg. 
1539
-
1545
)
32
Nakano
M.
Ikeda
Y.
Tokuda
Y.
Fuwa
M.
Omi
N.
Ueno
M.
Imai
K.
Adachi
H.
Kageyama
M.
Mori
K.
, et al.  . 
Common variants in CDKN2B-AS1 associated with optic-nerve vulnerability of glaucoma identified by genome-wide association studies in Japanese
PLoS One
 , 
2012
, vol. 
7
 pg. 
e33389
 
33
Wiggs
J.L.
Yaspan
B.L.
Hauser
M.A.
Kang
J.H.
Allingham
R.R.
Olson
L.M.
Abdrabou
W.
Fan
B.J.
Wang
D.Y.
Brodeur
W.
, et al.  . 
Common variants at 9p21 and 8q22 are associated with increased susceptibility to optic nerve degeneration in glaucoma
PLoS Genet.
 , 
2012
, vol. 
8
 pg. 
e1002654
 
34
Krimpenfort
P.
Ijpenberg
A.
Song
J.Y.
van der Valk
M.
Nawijn
M.
Zevenhoven
J.
Berns
A.
p15Ink4b is a critical tumour suppressor in the absence of p16Ink4a
Nature
 , 
2007
, vol. 
448
 (pg. 
943
-
946
)
35
Li
X.
Perissi
V.
Liu
F.
Rose
D.W.
Rosenfeld
M.G.
Tissue-specific regulation of retinal and pituitary precursor cell proliferation
Science
 , 
2002
, vol. 
297
 (pg. 
1180
-
1183
)
36
Adzhubei
I.A.
Schmidt
S.
Peshkin
L.
Ramensky
V.E.
Gerasimova
A.
Bork
P.
Kondrashov
A.S.
Sunyaev
S.R.
A method and server for predicting damaging missense mutations
Nat. Methods
 , 
2010
, vol. 
7
 (pg. 
248
-
249
)
37
Cooper
G.M.
Goode
D.L.
Ng
S.B.
Sidow
A.
Bamshad
M.J.
Shendure
J.
Nickerson
D.A.
Single-nucleotide evolutionary constraint scores highlight disease-causing mutations
Nat. Methods
 , 
2010
, vol. 
7
 (pg. 
250
-
251
)
38
Consortium
G.T.
The Genotype-Tissue Expression (GTEx) project
Nat. Genet.
 , 
2013
, vol. 
45
 (pg. 
580
-
585
)
39
Wagner
A.H.
Anand
V.N.
Wang
W.H.
Chatterton
J.E.
Sun
D.
Shepard
A.R.
Jacobson
N.
Pang
I.H.
Deluca
A.P.
Casavant
T.L.
, et al.  . 
Exon-level expression profiling of ocular tissues
Exp. Eye Res.
 , 
2013
, vol. 
111
 (pg. 
105
-
111
)
40
Veth
K.N.
Willer
J.R.
Collery
R.F.
Gray
M.P.
Willer
G.B.
Wagner
D.S.
Mullins
M.C.
Udvadia
A.J.
Smith
R.S.
John
S.W.
, et al.  . 
Mutations in zebrafish lrp2 result in adult-onset ocular pathogenesis that models myopia and other risk factors for glaucoma
PLoS Genet.
 , 
2011
, vol. 
7
 pg. 
e1001310
 
41
Toy
J.
Yang
J.M.
Leppert
G.S.
Sundin
O.H.
The optx2 homeobox gene is expressed in early precursors of the eye and activates retina-specific genes
Proc. Natl. Acad. Sci. U S A.
 , 
1998
, vol. 
95
 (pg. 
10643
-
10648
)
42
Lopez-Rios
J.
Gallardo
M.E.
Rodriguez de Cordoba
S.
Bovolenta
P.
Six9 (Optx2), a new member of the six gene family of transcription factors, is expressed at early stages of vertebrate ocular and pituitary development
Mech. Dev.
 , 
1999
, vol. 
83
 (pg. 
155
-
159
)
43
Aijaz
S.
Allen
J.
Tregidgo
R.
van Heyningen
V.
Hanson
I.
Clark
B.J.
Expression analysis of SIX3 and SIX6 in human tissues reveals differences in expression and a novel correlation between the expression of SIX3 and the genes encoding isocitrate dehyhrogenase and cadherin 18
Genomics
 , 
2005
, vol. 
86
 (pg. 
86
-
99
)
44
Zhou
X.
Hollemann
T.
Pieler
T.
Gruss
P.
Cloning and expression of xSix3, the Xenopus homologue of murine Six3
Mech. Dev.
 , 
2000
, vol. 
91
 (pg. 
327
-
330
)
45
Lagutin
O.
Zhu
C.C.
Furuta
Y.
Rowitch
D.H.
McMahon
A.P.
Oliver
G.
Six3 promotes the formation of ectopic optic vesicle-like structures in mouse embryos
Dev. Dyn.
 , 
2001
, vol. 
221
 (pg. 
342
-
349
)
46
Carl
M.
Loosli
F.
Wittbrodt
J.
Six3 inactivation reveals its essential role for the formation and patterning of the vertebrate eye
Development
 , 
2002
, vol. 
129
 (pg. 
4057
-
4063
)
47
Amato
M.A.
Arnault
E.
Perron
M.
Retinal stem cells in vertebrates: parallels and divergences
Int. J. Dev. Biol.
 , 
2004
, vol. 
48
 (pg. 
993
-
1001
)
48
Raymond
P.A.
Barthel
L.K.
Bernardos
R.L.
Perkowski
J.J.
Molecular characterization of retinal stem cells and their niches in adult zebrafish
BMC Dev. Biol.
 , 
2006
, vol. 
6
 pg. 
36
 
49
Kassen
S.C.
Thummel
R.
Burket
C.T.
Campochiaro
L.A.
Harding
M.J.
Hyde
D.R.
The Tg(ccnb1:EGFP) transgenic zebrafish line labels proliferating cells during retinal development and regeneration
Mol. Vis.
 , 
2008
, vol. 
14
 (pg. 
951
-
963
)
50
Gallardo
M.E.
Rodriguez De Cordoba
S.
Schneider
A.S.
Dwyer
M.A.
Ayuso
C.
Bovolenta
P.
Analysis of the developmental SIX6 homeobox gene in patients with anophthalmia/microphthalmia
Am. J. Med. Genet. A
 , 
2004
, vol. 
129A
 (pg. 
92
-
94
)
51
Aijaz
S.
Clark
B.J.
Williamson
K.
van Heyningen
V.
Morrison
D.
Fitzpatrick
D.
Collin
R.
Ragge
N.
Christoforou
A.
Brown
A.
, et al.  . 
Absence of SIX6 mutations in microphthalmia, anophthalmia, and coloboma
Invest. Ophthalmol. Vis. Sci.
 , 
2004
, vol. 
45
 (pg. 
3871
-
3876
)
52
Zuber
M.E.
Perron
M.
Philpott
A.
Bang
A.
Harris
W.A.
Giant eyes in Xenopus laevis by overexpression of XOptx2
Cell
 , 
1999
, vol. 
98
 (pg. 
341
-
352
)
53
Lieschke
G.J.
Currie
P.D.
Animal models of human disease: zebrafish swim into view
Nat. Rev. Genet.
 , 
2007
, vol. 
8
 (pg. 
353
-
367
)
54
Guo
S.
Linking genes to brain, behavior and neurological diseases: what can we learn from zebrafish?
Genes Brain Behav.
 , 
2004
, vol. 
3
 (pg. 
63
-
74
)
55
Malicki
J.
Neuhauss
S.C.
Schier
A.F.
Solnica-Krezel
L.
Stemple
D.L.
Stainier
D.Y.
Abdelilah
S.
Zwartkruis
F.
Rangini
Z.
Driever
W.
Mutations affecting development of the zebrafish retina
Development
 , 
1996
, vol. 
123
 (pg. 
263
-
273
)
56
Neuhauss
S.C.
Behavioral genetic approaches to visual system development and function in zebrafish
J. Neurobiol.
 , 
2003
, vol. 
54
 (pg. 
148
-
160
)
57
Haffter
P.
Granato
M.
Brand
M.
Mullins
M.C.
Hammerschmidt
M.
Kane
D.A.
Odenthal
J.
van Eeden
F.J.
Jiang
Y.J.
Heisenberg
C.P.
, et al.  . 
The identification of genes with unique and essential functions in the development of the zebrafish, Danio rerio
Development
 , 
1996
, vol. 
123
 (pg. 
1
-
36
)
58
Cheyette
B.N.
Green
P.J.
Martin
K.
Garren
H.
Hartenstein
V.
Zipursky
S.L.
The Drosophila sine oculis locus encodes a homeodomain-containing protein required for the development of the entire visual system
Neuron
 , 
1994
, vol. 
12
 (pg. 
977
-
996
)
59
Kobayashi
M.
Toyama
R.
Takeda
H.
Dawid
I.B.
Kawakami
K.
Overexpression of the forebrain-specific homeobox gene six3 induces rostral forebrain enlargement in zebrafish
Development
 , 
1998
, vol. 
125
 (pg. 
2973
-
2982
)
60
Loosli
F.
Winkler
S.
Wittbrodt
J.
Six3 overexpression initiates the formation of ectopic retina
Genes Dev.
 , 
1999
, vol. 
13
 (pg. 
649
-
654
)
61
Bernier
G.
Panitz
F.
Zhou
X.
Hollemann
T.
Gruss
P.
Pieler
T.
Expanded retina territory by midbrain transformation upon overexpression of Six6 (Optx2) in Xenopus embryos
Mech. Dev.
 , 
2000
, vol. 
93
 (pg. 
59
-
69
)
62
Del Bene
F.
Tessmar-Raible
K.
Wittbrodt
J.
Direct interaction of geminin and Six3 in eye development
Nature
 , 
2004
, vol. 
427
 (pg. 
745
-
749
)
63
Lopez-Rios
J.
Tessmar
K.
Loosli
F.
Wittbrodt
J.
Bovolenta
P.
Six3 and Six6 activity is modulated by members of the groucho family
Development
 , 
2003
, vol. 
130
 (pg. 
185
-
195
)
64
Dimasi
D.P.
Burdon
K.P.
Hewitt
A.W.
Fitzgerald
J.
Wang
J.J.
Healey
P.R.
Mitchell
P.
Mackey
D.A.
Craig
J.E.
Genetic investigation into the endophenotypic status of central corneal thickness and optic disc parameters in relation to open-angle glaucoma
Am. J. Ophthalmol.
 , 
2012
, vol. 
154
 (pg. 
833
-
842 e832
)
65
Hannon
G.J.
Beach
D.
p15INK4B is a potential effector of TGF-beta-induced cell cycle arrest
Nature
 , 
1994
, vol. 
371
 (pg. 
257
-
261
)
66
Gonzalez-Nunez
V.
Nocco
V.
Budd
A.
Characterization of drCol 15a1b: a novel component of the stem cell niche in the zebrafish retina
Stem Cells
 , 
2010
, vol. 
28
 (pg. 
1399
-
1411
)
67
Agathocleous
M.
Harris
W.A.
From progenitors to differentiated cells in the vertebrate retina
Annu. Rev. Cell Dev. Biol.
 , 
2009
, vol. 
25
 (pg. 
45
-
69
)
68
Hofman
A.
van Duijn
C.M.
Franco
O.H.
Ikram
M.A.
Janssen
H.L.
Klaver
C.C.
Kuipers
E.J.
Nijsten
T.E.
Stricker
B.H.
Tiemeier
H.
, et al.  . 
The Rotterdam Study: 2012 objectives and design update
Eur. J. Epidemiol.
 , 
2011
, vol. 
26
 (pg. 
657
-
686
)
69
Aulchenko
Y.S.
Heutink
P.
Mackay
I.
Bertoli-Avella
A.M.
Pullen
J.
Vaessen
N.
Rademaker
T.A.
Sandkuijl
L.A.
Cardon
L.
Oostra
B.
, et al.  . 
Linkage disequilibrium in young genetically isolated Dutch population
Eur. J. Hum. Genet.
 , 
2004
, vol. 
12
 (pg. 
527
-
534
)
70
Pardo
L.M.
MacKay
I.
Oostra
B.
van Duijn
C.M.
Aulchenko
Y.S.
The effect of genetic drift in a young genetically isolated population
Ann. Hum. Genet.
 , 
2005
, vol. 
69
 (pg. 
288
-
295
)
71
Rolando
M.
Iester
M.
Campagna
P.
Borgia
L.
Traverso
C.
Calabria
G.
Measurement variability in digital analysis of optic discs
Doc. Ophthalmol.
 , 
1994
, vol. 
85
 (pg. 
211
-
222
)
72
Dielemans
I.
Vingerling
J.R.
Wolfs
R.C.
Hofman
A.
Grobbee
D.E.
de Jong
P.T.
The prevalence of primary open-angle glaucoma in a population-based study in The Netherlands. The Rotterdam Study
Ophthalmology
 , 
1994
, vol. 
101
 (pg. 
1851
-
1855
)
73
Czudowska
M.A.
Ramdas
W.D.
Wolfs
R.C.
Hofman
A.
De Jong
P.T.
Vingerling
J.R.
Jansonius
N.M.
Incidence of glaucomatous visual field loss: a ten-year follow-up from the Rotterdam Study
Ophthalmology
 , 
2010
, vol. 
117
 (pg. 
1705
-
1712
)
74
Li
H.
Durbin
R.
Fast and accurate short read alignment with Burrows–Wheeler transform
Bioinformatics
 , 
2009
, vol. 
25
 (pg. 
1754
-
1760
)
75
Brouwer
R.W.
van den Hout
M.C.
Grosveld
F.G.
van Ijcken
W.F.
NARWHAL, a primary analysis pipeline for NGS data
Bioinformatics
 , 
2012
, vol. 
28
 (pg. 
284
-
285
)
76
McKenna
A.
Hanna
M.
Banks
E.
Sivachenko
A.
Cibulskis
K.
Kernytsky
A.
Garimella
K.
Altshuler
D.
Gabriel
S.
Daly
M.
, et al.  . 
The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data
Genome Res.
 , 
2010
, vol. 
20
 (pg. 
1297
-
1303
)
77
Kimmel
C.B.
Ballard
W.W.
Kimmel
S.R.
Ullmann
B.
Schilling
T.F.
Stages of embryonic development of the zebrafish
Dev. Dyn.
 , 
1995
, vol. 
203
 (pg. 
253
-
310
)
78
Vilella
A.J.
Severin
J.
Ureta-Vidal
A.
Heng
L.
Durbin
R.
Birney
E.
EnsemblCompara GeneTrees: complete, duplication-aware phylogenetic trees in vertebrates
Genome Res.
 , 
2009
, vol. 
19
 (pg. 
327
-
335
)
79
Robu
M.E.
Larson
J.D.
Nasevicius
A.
Beiraghi
S.
Brenner
C.
Farber
S.A.
Ekker
S.C.
p53 activation by knockdown technologies
PLoS Genet.
 , 
2007
, vol. 
3
 pg. 
e78
 
80
Untergasser
A.
Nijveen
H.
Rao
X.
Bisseling
T.
Geurts
R.
Leunissen
J.A.
Primer3Plus, an enhanced web interface to Primer3
Nucleic Acid Res.
 , 
2007
, vol. 
35
 (pg. 
W71
-
W74
)
81
van de Water
S.
van de Wetering
M.
Joore
J.
Esseling
J.
Bink
R.
Clevers
H.
Zivkovic
D.
Ectopic Wnt signal determines the eyeless phenotype of zebrafish masterblind mutant
Development
 , 
2001
, vol. 
128
 (pg. 
3877
-
3888
)

Author notes

These two authors contributed equally as senior authors.

Supplementary data