Molecular Vision 2005; 11:625-631 <http://www.molvis.org/molvis/v11/a74/> Received 13 May 2005 | Accepted 24 August 2005 | Published 29 August 2005

Download Reprint

SNPs and interaction analyses of myocilin, optineurin, and apolipoprotein E in primary open angle glaucoma patients

Bao Jian Fan, Dan Yi Wang, Dorothy Shu Ping Fan, Pancy Oi Sin Tam, Dennis Shun Chiu Lam, Clement Chee Yung Tham, Ching Yan Lam, Tung Ching Lau, Chi Pui Pang
 
 

Department of Ophthalmology & Visual Sciences, the Chinese University of Hong Kong, Hong Kong, China

Correspondence to: Prof. C. P. Pang, Department of Ophthalmology & Visual Sciences, The Chinese University of Hong Kong, Hong Kong Eye Hospital, 147K Argyle Street, Kowloon, Hong Kong; Phone: +852 27623169; FAX: +852 27159490; email: cppang@cuhk.edu.hk

Abstract

Purpose: To evaluate the association of myocilin (MYOC), optineurin (OPTN), and apolipoprotein E (APOE) genes and their interactions in primary open angle glaucoma (POAG).

Methods: A cohort of 400 unrelated POAG patients (294 high tension glaucoma, HTG, and 106 normal tension glaucoma, NTG) and 281 unrelated control subjects were recruited. All coding exons and splicing junctions in MYOC and OPTN were screened for sequence alterations. Common polymorphisms in APOE were genotyped. Single genes were investigated by univariate and haplotype analysis, and gene-gene interactions by logistic regression and stratified analysis. Multiple comparisons were corrected by the Bonferroni method. Bioinformatics analysis was performed to assess the conservation of mutation sites across species and to predict putative motifs and secondary structures in mutated proteins.

Results: Disease-causing mutations in MYOC and OPTN were identified in 1.75% and 1% of POAG patients, respectively. Most of these mutations were highly conserved across species, many predicted to create new motifs or change protein secondary structures. No individual MYOC polymorphisms significantly contributed to HTG or NTG. A haplotype containing the minor allele of the MYOC IVS2+35A>G increased NTG risk (p=0.0001). Three OPTN polymorphisms, T34T, IVS5+38T>G, and IVS8-53T>C increased NTG risk (p<0.0008), while IVS5+38T>G increased HTG risk (p=0.0006). One haplotype that contains the minor alleles of 3 OPTN polymorphisms, T34T, IVS5+38T>G, and IVS7+24G>A, increased NTG risk (p=0.0002). APOE ε4 carriers had a decreased NTG risk (p=0.007). Possible gene-gene interactions were found between MYOC, OPTN, and APOE.

Conclusions: Disease-causing mutations in MYOC and OPTN accounted for only a small proportion of Chinese POAG patients. Common polymorphisms in MYOC, OPTN, and APOE might interactively contribute to POAG, indicating a polygenic etiology.

Introduction

Glaucoma is a group of diseases causing optic neuropathy and is characterized by optic disc cupping and loss of visual field. It is the second leading cause of blindness worldwide, estimated to affect about 70 million people, with 6.7 million of these being bilaterally blind [1]. Primary open angle glaucoma (POAG; OMIM 137760) is a major primary type of glaucoma.

POAG is genetically heterogeneous [2]. Seven chromosomal loci have been identified: GLC1A, GLC1B, GLC1C, GLC1D, GLC1E, GLC1F, and GLC1G [3-7]. Putative loci on chromosomes 2p14, 14q11, 14q21-22, 17p13, 17q25, and 19q12-14 have been reported in adult-onset POAG in a genome-wide scan study involving 113 sib-pairs [8]. A genome-wide scan on 146 POAG families of African descent suggested possible linkage to 2q33.3-37.3 and 10p12-13 [9], although a study of 8 Finnish POAG families revealed no association with these sites [10]. A genome-wide scan in 25 families with juvenile-onset POAG identified possible glaucoma loci on 9q22 and 20p12 [11]. The first POAG associated gene was identified as myocilin (MYOC; OMIM 601652) in the GLC1A locus. In Caucasians about 2%-4% of POAG cases are due to MYOC mutations [12], although it can be as high as 36% in juvenile hereditary POAG families [13]. In our previous studies, the prevalence of MYOC mutations is about 1.1%-1.5% in Chinese POAG patients [14,15]. The second gene for POAG was identified as optineurin (OPTN; OMIM 602432) in GLC1E. Mutations in OPTN had been found in 16.7% of 54 families with hereditary and adult-onset POAG and 12% of 124 sporadic patients with POAG, most of them with intraocular pressure (IOP) less than 22 mm Hg [16]. We found OPTN mutations accounting for 1.6% of sporadic Chinese POAG patients [17]. However, later studies reported no glaucoma causing mutations in OPTN among Caucasian POAG patients, one on 801 patients of variable age onset [18] and the other 86 adult-onset patients [19]. No OPTN mutation was found in 313 sporadic Japanese patients, 148 of them with normal tension glaucoma (NTG) [20]. Prevalence of OPTN mutations that cause glaucoma is still to be confirmed.

Apolipoprotein E (APOE; OMIM 107741) had been reported to be a potent modifier gene for POAG [21]. The promoter polymorphism -219T>G was associated with increased optic nerve damage, while -491A>T interacted with -1000C>G to increase IOP in POAG patients. In another study the APOE ε4 allele increased risk of NTG in 70 Tasmania patients [22]. However, two recent studies on Caucasian POAG patients, one involving 155 NTG patients [23] and the other with 137 mixed POAG and NTG patients [24], did not reveal an association between the APOE ε2/ε3/ε4 genotype with glaucoma.

To date, MYOC is the undisputed disease-causing gene for POAG but most studies showed that it accounts for only a minor number of POAG patients. Glaucoma-causing effects of OPTN and APOE have been inconsistently reported. In this study, we evaluated the association of MYOC, OPTN and APOE genes and their possible interactions in POAG in a large Chinese sample.

Methods

Study subjects

Unrelated patients with POAG were recruited from the Eye Clinic of the Prince of Wales Hospital, Hong Kong. POAG was defined as meeting all of the following criteria: exclusion of secondary causes (e.g., trauma, uveitis, or steroid-induced glaucoma), anterior chamber angle open (grade III or IV gonioscopy), characteristic optic disc changes (e.g., vertical cup-disc ratio >0.5, disk hemorrhage, or thin or notched neuroretinal rim), and characteristic visual field changes with reference to Anderson's criteria for minimal abnormality in glaucoma [25]. Visual acuity was determined using Snellen eye chart, IOP by applanation tonometry, and visual field by Humphrey's perimeter with the Glaucoma Hemifield Test. Patients with congenital glaucoma were excluded. Unrelated control subjects were recruited from people who attended the clinic for conditions of senile cataract, floaters, refractive errors, or itchy eyes. They did not have systemic diseases and were excluded from glaucoma using the same criteria of diagnosis as the POAG patients after going through the same procedure of ophthalmic examination. The project was approved by the Ethics Committee for Human Research, the Chinese University of Hong Kong. Informed consent was obtained from all study subjects after explanation of the nature and possible consequences of the study, in accordance with the tenets of the Declaration of Helsinki.

A total of 400 unrelated POAG patients were included in this study, 250 males and 150 females. The highest IOP ranged from 10 to 67 mm Hg (mean±SD: 26.3±9.2 mm Hg): 294 (73.5%) patients were high tension glaucoma (HTG) with highest IOP greater than or equal to 22 mm Hg, and 106 (26.5%) normal tension glaucoma (NTG) having highest IOP <22 mm Hg. Age at diagnosis ranged from 8 to 91 years (mean±SD: 61.0±17.2 years): 80 (20%) were juvenile-onset POAG with age at diagnosis <35 years, and 320 (80%) were adult-onset POAG with age at diagnosis greater than or equal to 35 years. There were 281 unrelated control subjects, 180 males and 101 females, age ranging from 50 to 90 years (mean±SD: 69.8±9.8 years), and IOP<22 mm Hg. All study subjects were Hong Kong Chinese. Genomic DNA was extracted from 200 μl venous whole blood using Qiamp Blood Kit (Qiagen, Hilden, Germany). A subset of these patients had been previously reported: 187 POAG patients for MYOC mutations [15] and 119 POAG patients for OPTN mutations [17].

Sequence analysis of MYOC

The coding regions of MYOC, exons 1-3, including splicing junctions, were screened for sequence alterations using polymerase chain reaction (PCR), high-throughput conformation sensitive gel electrophoresis (HTCSGE), and a cost saving sequencing protocol [15] on an ABI 377XL automated DNA sequencer (Perkin-Elmer Applied Biosystems, Foster City, CA). Sequence data were aligned using Sequence Navigator analysis software (version 1.0.1; Perkin-Elmer Applied Biosystems) and compared to the published MYOC gene sequence (GenBank AB006688).

Sequence analysis of OPTN

All coding exons, 4-16, of OPTN and their splicing junctions were screened for sequence alterations by PCR, HTCSGE, and direct DNA sequencing [17]. Sequence data were compared to the published OPTN gene sequence (GenBank AF420371).

Polymorphisms of APOE

The promoter polymorphisms -491A>T, -427T>C, and -219T>G, and the ε2/ε3/ε4 genotype on exon 4 were investigated by PCR followed by restriction endonuclease assays and polyacrylamide gel electrophoresis [26,27].

Crosschecking of sequence studies

More than one quarter of the PCR products in each of the sequence studies on MYOC, OPTN, and APOE, with no regard of the respective HTCSGE or restriction endonuclease assay results, were selected at random for direct DNA sequencing. Complete matching of results was obtained.

Statistical analysis

SAS (release 8.2; SAS Institute, Cary, NC) was used for statistical analyses. Hardy-Weinberg equilibrium for each polymorphism was tested by χ² test. A χ² test or Fisher's exact test was used to compare allele or genotype frequencies of HTG or NTG patients with control subjects. Expectation-maximization (EM)-based haplotype frequency estimations and permutation-based hypothesis testing procedures were performed in HTG, NTG and control subjects using the GENECOUNTING program (version 1.3) [28].

Logistic regression analysis was used to search for gene-gene interactions. Disease status was set as dependent variable (HTG or NTG=1, control=0), and gene polymorphisms as independent variables (homozygote=2, heterozygote=1, wildtype=0). Logistic regression models were built using various gene polymorphisms along with their interaction items. A stepwise regression approach was used to optimize the analysis. To reduce false-positive interactions, stratified analysis was used to verify the gene-gene interactions identified from logistic regression analysis. We stratified the study subjects according to the genotype of one gene and performed analysis of the other gene in different strata defined by the genotype of the former gene. The Breslow-Day test was used to test the homogeneity of the odds ratios (ORs) in different strata. The confidence intervals (CIs) for ORs were calculated by the Mantel-Haenszel method. An α level of 0.05 was chosen; the Bonferroni method was used to adjust for multiple comparisons.

Bioinformatics analysis

Firstly, to assess whether the normal residues at disease-causing mutation sites were conserved across species, the following protein entries in Swiss-Prot or TrEMBL were selected: MYOC_Human (Q99972), MYOC_Monkey (Q863A3), MYOC_Bovine (Q9XTA3), MYOC_Mouse (O70624), MYOC_Rat (Q9R1J4), MYOC_Rabbit (Q866N2), MYOC_Pig (AAN59763), MYOC_Cat (AAS68633), OPTN_Human (Q96CV9), OPTN_Macmu (Q861Q8), OPTN_Macfa (Q95KA2), OPTN_Ponpy (Q5R923), OPTN_Pig (Q7YS99), OPTN_Rat (Q8R5M4), OPTN_Mouse (Q8K3K8), OPTN_Chick (Q90Z16) and OPTN_Zebrafish (Q5RI56). The sequences were aligned by the multiple alignment tool ClustalW [29]. Secondly, to predict putative motifs in the mutated human protein sequences, the ScanProsite tool was used including patterns with the high probability of occurrence [30]. Finally, the PSIPRED program was applied to predict the secondary structure of the mutated human proteins [31].

Results

Univariate analysis of individual polymorphisms in MYOC

A total of 22 sequence polymorphisms of MYOC were identified (Table 1). All sequence polymorphisms followed Hardy-Weinberg equilibrium. Q19H, R91X, C245Y, E300K, S313F, and Y471C were disease-causing mutations, accounting for 1.75% (7/400) of POAG patients. Except for Q19H, which was identified in two patients, all the others were found only in one patient. All disease-causing mutations were found only in HTG patients. There was no significant association of all these polymorphisms with HTG or NTG (p>0.0023, the Bonferroni corrected significance level; Table 1).

Univariate analysis of individual polymorphisms in OPTN

A total of 21 polymorphisms of OPTN were identified (Table 2). All sequence polymorphisms followed the Hardy-Weinberg equilibrium. I88T, E103D, and H486R were disease-causing mutations, accounting for 1% (4/400) of POAG patients. I88T was identified in one HTG patient and one NTG patient, respectively, the other two mutations were each found in only one NTG patient. The frequency of allele G carriers for IVS5+38T>G was higher in HTG (OR=2.3, 95% CI: 1.4, 3.8; p=0.0006) than in controls. When compared with controls, higher frequencies were found in NTG for T34T (OR=2.3, 95% CI: 1.4, 3.7; p=0.0008), IVS5+38T>G (OR=5.0, 95% CI: 2.9, 8.9; p<0.0001) and IVS8-53T>C (OR=4.0, 95% CI: 1.8, 8.7; p=0.0002). There was no significant association of other polymorphisms with HTG or NTG (p>0.0024, the Bonferroni corrected significance level; Table 2).

Univariate analysis of individual polymorphisms in APOE

All four informative polymorphisms of APOE, -491A>T, -427T>C, -219T>G, and ε2/ε3/ε4 genotype investigated in this study followed the Hardy-Weinberg equilibrium (Table 3). Compared to control subjects, when analyzed with non-ε4-carriers (genotypes 2/3+3/3), the frequency of ε4-carriers (genotypes 2/4+3/4+4/4) was significantly lower in NTG (OR=0.4, 95% CI: 0.2, 0.8; p=0.007). The distributions of -219T>G, -427T>C, and -491A>T were not statistically different between patients with HTG or NTG and control subjects (p>0.0125, the Bonferroni corrected significance level, Table 3).

Bioinformatics analysis of disease-causing mutations

In the MYOC gene, 6 disease-causing mutations were identified. The normal residues at these mutation sites were highly conserved across species including human, monkey, bovine, rabbit, mouse, rat, pig, and cat, with an exception that the mutation Q19H was not conserved in mouse. Q19H is located within the signal peptide (residues 1-32). The nonsense mutation R91X truncates the MYOC protein into a polypeptide with only 58 amino acids, creating a new casein kinase II phosphorylation site (residues 88-91: TkaX). The missense mutations C245Y, E300K, S313F, and Y471C are located within the olfactomedin-like domain (residues 245-504). C245Y creates a new tyrosine sulfation site (residues 238-252: sgegdtgYgelvwvg) and was predicted to change the MYOC secondary structure from coiled-coil to β-strand at residue 247. The E300K mutation was predicted to change the protein secondary structure from β-strand to coiled-coil at residues 288-289 and 327. Similarly, S313F would change the secondary structure from β-strand to coiled-coil at residues 314 and 319.

In OPTN, 3 disease-causing mutations were identified. The normal residues at the mutation sites of E103D and H486R were highly conserved across species including human, monkey, mouse, rat, orangutan, pig, chick, and zebrafish. I88T was conserved only across human, rat, and orangutan. Mutations I88T and E103D are located within the region that interacts with RAB8 (residues 58-209) while H486R was within the region that interacts with Huntingtin (residues 411-577). Moreover, the OPTN secondary structure is predicted to change from α-helix to coiled-coil at residues 101-102 due to the E103D mutation.

Haplotype analysis of common polymorphisms in MYOC

Six common polymorphisms with a prevalence of more than 1% in the study subjects were included in the haplotype analysis of MYOC. They were -83G>A, R46X, R76K, IVS2+35A>G, T353I, and 1515+73G>C. Global test showed that the difference of overall haplotype frequency profiles was significant between NTG and controls (p=0.03). The frequency of one haplotytpe that contains the minor allele of IVS2+35A>G and the major alleles in the other 5 loci was significantly higher in NTG than in controls (OR=2.0; p=0.0001, Bonferroni corrected significance level 0.003).

Haplotype analysis of common polymorphisms in OPTN

Twelve common polymorphisms with a prevalence of more than 1% in the study subjects were included in the haplotype analysis of OPTN. They were T34T, T49T, M98K, IVS5+38T>G, IVS6-5T>C, IVS6-10G>A, IVS7+24G>A, IVS8+20G>A, IVS8-53T>C, IVS15+10G>A, IVS15-48C>A, and R545Q. When compared to controls, global tests showed a significant difference of overall haplotype frequency profiles in NTG (p<0.0001). The frequency of one haplotype that contains the minor alleles of 3 OPTN polymorphisms (T34T, IVS5+38T>G, and IVS7+24G>A) and the major alleles in the other 9 loci was higher in NTG than in controls (OR=8.0; p=0.0002, Bonferroni corrected significance level 0.0007).

Haplotype analysis of common polymorphisms in APOE

All four polymorphisms investigated in this study were included in the haplotype analysis of APOE. Global test showed that the difference of overall haplotype frequency profiles was marginally significant between NTG and controls (p=0.04). However, all the permutation-based p values for individual haplotypes were higher than the Bonferroni corrected significance level of 0.003.

Analysis of gene-gene interactions

Logistic regression analysis identified two pairs of significant gene-gene interactions for HTG and three pairs for NTG. Among them, one pair of interaction was found between MYOC and OPTN while the others were between APOE and MYOC or OPTN. Stratified analysis further verified all these significant gene-gene interactions (Table 4).

Discussion

As far as we know, this is the first report on a systemic evaluation of the association and interactions of MYOC, OPTN, and APOE in POAG based on a large sample. According to Alward's criteria [32], coding sequence changes were assumed to be disease-causing mutations if they are (1) expected to alter the amino acid sequence of the corresponding protein, and (2) more commonly observed in patients with POAG than in controls. To meet the latter criterion, a polymorphism needs to be completely absent from the control population or significantly more common in the POAG population. Based on such requirements, disease-causing mutations in MYOC and OPTN were identified in 1.75% and 1%, respectively, of our unrelated POAG patients. Therefore, disease-causing mutations in MYOC and OPTN accounted for only a small proportion of Chinese POAG patients. Bioinformatics analysis showed that most of these disease-causing mutations were highly conserved across species, indicating these residues to be structurally or functionally essential and important. Many are predicted to create new motifs or change protein secondary structures. Further experimental investigation is required to define the actual changes. By using transfected COS-7 cells we recently found that the C245Y mutant MYOC protein formed homomultimeric complexes that migrated at molecular weights larger than their wildtype counterparts and these mutant complexes remained sequestered intracellularly. This cysteine to tyrosine mutation may cause the protein to fail to fold or oligomerize correctly [33]. Further functional studies of the other mutations reported in this study would help decipher the mechanisms by which these mutants cause glaucoma.

Our univariate analysis showed that when compared with controls, carriers with T34T or IVS8-53T>C in OPTN were more susceptible to NTG (Table 2). When compared with controls, carriers with IVS5+38T>G in OPTN conferred susceptibility to both NTG and HTG (Table 2). The APOE ε4 carriers had a decreased NTG risk (Table 3). Haplotype analysis revealed that one haplotype that contains the minor allele of the MYOC polymorphism, IVS2+35A>G, increased NTG risk. Another haplotype that contains the minor alleles of 3 OPTN polymorphisms, T34T, IVS5+38T>G, and IVS7+24G>A, significantly increased NTG risk by 8 fold. These findings indicate that besides disease-causing mutations, noncoding or silent sequence changes in MYOC and OPTN may also modify the susceptibility of POAG.

We simultaneously analyzed the common polymorphisms in MYOC, OPTN, and APOE using a logistic regression approach to hunt for possible gene-gene interactions. Several interactions among MYOC, OPTN, and APOE were identified. Stratified analysis further verified these interactions. It has been reported that the APOE polymorphism -491A>T interacted with the polymorphism -1000C>G in MYOC, and APOE was a potential modifier for POAG [21]. In the present study, we did not investigate -1000C>G, which we recently showed to have similar frequencies in 212 POAG patients and 221 controls [34]. However, we identified another two interactions between MYOC and APOE, -83G>A with ε2/ε3/ε4 and IVS2+35A>G with -219T>G. We found two interactions between OPTN and APOE (Table 4), further supporting the modifier effects of APOE on POAG. In addition, we found a potential interaction between MYOC and OPTN, T353I with IVS15+10G>A. There was some evidence to suggest that MYOC-T353I, together with other genetic and environmental risk factors including family history, hypertension, and cigarette smoking, might increase POAG risk [35]. In the present study, the polymorphism MYOC-T353I individually remained no significant contribution to POAG risk but there was a significant interaction with OPTN-IVS15+10G>A. Our findings demonstrated that common polymorphisms in MYOC, OPTN, and APOE might interactively contribute to POAG, indicating a polygenic etiology.

Because rare sequence polymorphisms are not informative for both haplotype analysis and logistic regression analysis, in the present study we only included in these two kinds of multivariable analyses common polymorphisms which prevalence is not less than 1% in our study subjects. Meanwhile, we did not match cases and controls in age since only elderly individuals could reasonably serve as normal controls for POAG. However, we found similar results when we analyzed our data by considering age as an independent variable in the logistic regression model (data not shown). It indicated that age was not a confounding factor in our study. In the present study we defined POAG as having both characteristic optic disc changes and visual field damages. It was usually acceptable to diagnose POAG as having a milder cup-disc ratio threshold (>0.5) but with a definite and reliable visual field defect [36]. In our study all patients with a cup-disc ratio between 0.5 and 0.7 had a glaucomatous visual field defect. We have considered using cup-disc ratio >0.7 as a criterion for POAG. But our subgroup analysis involving only patients with cup-disc ratio >0.7 showed similar results with patients with cup-disc ratio >0.5 (data not shown). In addition, noncoding polymorphisms both in MYOC and OPTN affected susceptibility to glaucoma. Whether these MYOC polymorphisms resulted in altered expression levels of MYOC, or the OPTN polymorphisms caused transcription complications, that lead to production of proteins with altered function or stability, remain to be investigated. Further functional studies using either transgenic cell lines or animal models are needed to confirm the gene and gene-gene interactive effects of the sequence polymorphisms.

Acknowledgements

This work was supported in part by a block grant from the University Grants Committee, Hong Kong, and by a direct grant, 2040997, from the Medical Panel, the Chinese University of Hong Kong.

References

1. Quigley HA. Number of people with glaucoma worldwide. Br J Ophthalmol 1996; 80:389-93.

2. Sarfarazi M. Recent advances in molecular genetics of glaucomas. Hum Mol Genet 1997; 6:1667-77.

3. Raymond V. Molecular genetics of the glaucomas: mapping of the first five "GLC" loci. Am J Hum Genet 1997; 60:272-7.

4. Trifan OC, Traboulsi EI, Stoilova D, Alozie I, Nguyen R, Raja S, Sarfarazi M. A third locus (GLC1D) for adult-onset primary open-angle glaucoma maps to the 8q23 region. Am J Ophthalmol 1998; 126:17-28.

5. Sarfarazi M, Child A, Stoilova D, Brice G, Desai T, Trifan OC, Poinoosawmy D, Crick RP. Localization of the fourth locus (GLC1E) for adult-onset primary open-angle glaucoma to the 10p15-p14 region. Am J Hum Genet 1998; 62:641-52.

6. Wirtz MK, Samples JR, Rust K, Lie J, Nordling L, Schilling K, Acott TS, Kramer PL. GLC1F, a new primary open-angle glaucoma locus, maps to 7q35-q36. Arch Ophthalmol 1999; 117:237-41.

7. Monemi S, Spaeth G, DaSilva A, Popinchalk S, Ilitchev E, Liebmann J, Ritch R, Heon E, Crick RP, Child A, Sarfarazi M. Identification of a novel adult-onset primary open-angle glaucoma (POAG) gene on 5q22.1. Hum Mol Genet 2005; 14:725-33.

8. Wiggs JL, Allingham RR, Hossain A, Kern J, Auguste J, DelBono EA, Broomer B, Graham FL, Hauser M, Pericak-Vance M, Haines JL. Genome-wide scan for adult onset primary open angle glaucoma. Hum Mol Genet 2000; 9:1109-17.

9. Nemesure B, Jiao X, He Q, Leske MC, Wu SY, Hennis A, Mendell N, Redman J, Garchon HJ, Agarwala R, Schaffer AA, Hejtmancik F, Barbados Family Study Group. A genome-wide scan for primary open-angle glaucoma (POAG): the Barbados Family Study of Open-Angle Glaucoma. Hum Genet 2003; 112:600-9.

10. Lemmela S, Ylisaukko-oja T, Forsman E, Jarvela I. Exclusion of 14 candidate loci for primary open angle glaucoma in Finnish families. Mol Vis 2004; 10:260-4 <http://www.molvis.org/molvis/v10/a33/>.

11. Wiggs JL, Lynch S, Ynagi G, Maselli M, Auguste J, Del Bono EA, Olson LM, Haines JL. A genomewide scan identifies novel early-onset primary open-angle glaucoma loci on 9q22 and 20p12. Am J Hum Genet 2004; 74:1314-20.

12. Fingert JH, Heon E, Liebmann JM, Yamamoto T, Craig JE, Rait J, Kawase K, Hoh ST, Buys YM, Dickinson J, Hockey RR, Williams-Lyn D, Trope G, Kitazawa Y, Ritch R, Mackey DA, Alward WL, Sheffield VC, Stone EM. Analysis of myocilin mutations in 1703 glaucoma patients from five different populations. Hum Mol Genet 1999; 8:899-905.

13. Shimizu S, Lichter PR, Johnson AT, Zhou Z, Higashi M, Gottfredsdottir M, Othman M, Moroi SE, Rozsa FW, Schertzer RM, Clarke MS, Schwartz AL, Downs CA, Vollrath D, Richards JE. Age-dependent prevalence of mutations at the GLC1A locus in primary open-angle glaucoma. Am J Ophthalmol 2000; 130:165-77.

14. Lam DS, Leung YF, Chua JK, Baum L, Fan DS, Choy KW, Pang CP. Truncations in the TIGR gene in individuals with and without primary open-angle glaucoma. Invest Ophthalmol Vis Sci 2000; 41:1386-91.

15. Pang CP, Leung YF, Fan B, Baum L, Tong WC, Lee WS, Chua JK, Fan DS, Liu Y, Lam DS. TIGR/MYOC gene sequence alterations in individuals with and without primary open-angle glaucoma. Invest Ophthalmol Vis Sci 2002; 43:3231-5.

16. Rezaie T, Child A, Hitchings R, Brice G, Miller L, Coca-Prados M, Heon E, Krupin T, Ritch R, Kreutzer D, Crick RP, Sarfarazi M. Adult-onset primary open-angle glaucoma caused by mutations in optineurin. Science 2002; 295:1077-9.

17. Leung YF, Fan BJ, Lam DS, Lee WS, Tam PO, Chua JK, Tham CC, Lai JS, Fan DS, Pang CP. Different optineurin mutation pattern in primary open-angle glaucoma. Invest Ophthalmol Vis Sci 2003; 44:3880-4.

18. Alward WL, Kwon YH, Kawase K, Craig JE, Hayreh SS, Johnson AT, Khanna CL, Yamamoto T, Mackey DA, Roos BR, Affatigato LM, Sheffield VC, Stone EM. Evaluation of optineurin sequence variations in 1,048 patients with open-angle glaucoma. Am J Ophthalmol 2003; 136:904-10.

19. Wiggs JL, Auguste J, Allingham RR, Flor JD, Pericak-Vance MA, Rogers K, LaRocque KR, Graham FL, Broomer B, Del Bono E, Haines JL, Hauser M. Lack of association of mutations in optineurin with disease in patients with adult-onset primary open-angle glaucoma. Arch Ophthalmol 2003; 121:1181-3.

20. Tang S, Toda Y, Kashiwagi K, Mabuchi F, Iijima H, Tsukahara S, Yamagata Z. The association between Japanese primary open-angle glaucoma and normal tension glaucoma patients and the optineurin gene. Hum Genet 2003; 113:276-9.

21. Copin B, Brezin AP, Valtot F, Dascotte JC, Bechetoille A, Garchon HJ. Apolipoprotein E-promoter single-nucleotide polymorphisms affect the phenotype of primary open-angle glaucoma and demonstrate interaction with the myocilin gene. Am J Hum Genet 2002; 70:1575-81.

22. Vickers JC, Craig JE, Stankovich J, McCormack GH, West AK, Dickinson JL, McCartney PJ, Coote MA, Healey DL, Mackey DA. The apolipoprotein epsilon4 gene is associated with elevated risk of normal tension glaucoma. Mol Vis 2002; 8:389-93 <http://www.molvis.org/molvis/v8/a46/>.

23. Lake S, Liverani E, Desai M, Casson R, James B, Clark A, Salmon JF. Normal tension glaucoma is not associated with the common apolipoprotein E gene polymorphisms. Br J Ophthalmol 2004; 88:491-3.

24. Ressiniotis T, Griffiths PG, Birch M, Keers S, Chinnery PF. The role of apolipoprotein E gene polymorphisms in primary open-angle glaucoma. Arch Ophthalmol 2004; 122:258-61.

25. Anderson DR. Automated static perimetry. St. Louis: Mosby-Year Book; 1992: p 12.

26. Mak YT, Chiu H, Woo J, Kay R, Chan YS, Hui E, Sze KH, Lum C, Kwok T, Pang CP. Apolipoprotein E genotype and Alzheimer's disease in Hong Kong elderly Chinese. Neurology 1996; 46:146-9.

27. Rebeck GW, Cheung BS, Growdon WB, Deng A, Akuthota P, Locascio J, Greenberg SM, Hyman BT. Lack of independent associations of apolipoprotein E promoter and intron 1 polymorphisms with Alzheimer's disease. Neurosci Lett 1999; 272:155-8.

28. Zhao JH, Sham PC. Generic number systems and haplotype analysis. Comput Methods Programs Biomed 2003; 70:1-9.

29. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 1994; 22:4673-80.

30. Bairoch A. PROSITE: a dictionary of sites and patterns in proteins. Nucleic Acids Res 1991; 19 Suppl:2241-5.

31. McGuffin LJ, Bryson K, Jones DT. The PSIPRED protein structure prediction server. Bioinformatics 2000; 16:404-5.

32. Alward WL, Kwon YH, Khanna CL, Johnson AT, Hayreh SS, Zimmerman MB, Narkiewicz J, Andorf JL, Moore PA, Fingert JH, Sheffield VC, Stone EM. Variations in the myocilin gene in patients with open-angle glaucoma. Arch Ophthalmol 2002; 120:1189-97.

33. Fan BJ, Leung DYL, Wang DY, Gobeil S, Raymond V, Tam POS, Lam DSC, Pang CP. A novel myocilin mutation in a Chinese family with juvenile-onset open angle glaucoma. Arch Ophthalmol. In press 2005.

34. Fan BJ, Leung YF, Pang CP, Fan DS, Wang DY, Tong WC, Tam PO, Chua JK, Lau TC, Lam DS. Polymorphisms in the myocilin promoter unrelated to the risk and severity of primary open-angle glaucoma. J Glaucoma 2004; 13:377-84.

35. Fan BJ, Leung YF, Wang N, Lam SC, Liu Y, Tam OS, Pang CP. Genetic and environmental risk factors for primary open-angle glaucoma. Chin Med J (Engl) 2004; 117:706-10.

36. Buhrmann RR, Quigley HA, Barron Y, West SK, Oliva MS, Mmbaga BB. Prevalence of glaucoma in a rural East African population. Invest Ophthalmol Vis Sci 2000; 41:40-8.