Secondary Logo

Journal Logo

Original Study

Systemic Genotype-Phenotype Analysis of MYOC Variants Based on Exome Sequencing and Literature Review

Li, Xueqing MD; Xiao, Xueshan; Li, Shiqiang MD; Sun, Wenmin MD, PhD; Wang, Panfeng MD, PhD; Zhang, Qingjiong MD, PhD

Author Information
Asia-Pacific Journal of Ophthalmology: March-April 2021 - Volume 10 - Issue 2 - p 173-182
doi: 10.1097/APO.0000000000000382

Abstract

Glaucoma is the second leading cause of blindness worldwide and characterized by progressive degeneration of retinal ganglion cells and optic nerve. It is estimated that the increasing number of glaucoma patients will reach to 111.8 million in 2040.1,2 Primary open-angle glaucoma (POAG) and primary angle-closure glaucoma (PACG) are 2 major subtypes of primary glaucoma and are divided based on the status of the anterior chamber angle. POAG is the more common type affecting approximately 3.05% of individuals worldwide and the prevalence rates vary among different ethnic groups.2 An elevation in intraocular pressure (IOP) is the main risk of POAG. Meanwhile, age is another key factor associated with POAG due to its progressive nature. Patients with an onset age of less than 35 years were further classified as juvenile-onset POAG (JOAG),3 an earlier subtype of POAG where a heritable element plays a more important role.

Most POAG subjects are complex heterogenous disorders with genetic variants as multifactorial risk factors.4 A small subset of POAG subjects has been reported to be transmitted as Mendelian traits, mostly through autosomal dominant inheritance. Mutations in 4 genes, including MYOC,5OPTN,6TBK1,7 and WDR36,8 have been reported responsible for autosomal dominant POAG (adPOAG). Of the 4, mutations in MYOC are most frequently reported, accounting for 2%–4% POAG, 8% JOAG, or 22% familial POAG.9–11

Variants in MYOC is common, and most of the variants are unlikely intolerant based on the Genome Aggregation Database (gnomAD; https://gnomad.broadinstitute.org/). Some rare variants in MYOC were observed with a similar frequency among POAG, PACG, and other unrelated population based on our in-house exome sequencing data.12 Glaucoma was not observed in an individual with a complete heterozygous deletion of MYOC,13 nor in homozygous Myoc knock-out mice.14 These lines of evidence suggest that many of the MYOC variants, either missense or predicted loss-of-function (pLoF), may not contribute to POAG or JOAG. This kind of problem has been revealed in a relatively conserved gene, such as CRX (much more conserved than MYOC), where most rare missense variants outside of homeodomain, as well as N-terminal truncation variants involving the homeodomain, are unlikely pathogenic.15 Therefore, a systemic evaluation of MYOC variants and an understanding of the characteristic nature of POAG-causative variants may be valuable, not only in the era of wide-spread clinical gene testing, but also in enriching POAG cases of an unknown cause in searching for new causative genes.

In this study, MYOC variants were collected from in-house exome sequencing data of 7092 individuals with different forms of eye conditions. The variants distribution in these individuals, genotype-phenotype analysis, and a series of bioinformatics analyses were used to classify potential pathogenic variants for POAG. Subsequently, previously reported MYOC mutations were reclassified based on our data, as well as on the Genome Aggregation Database (gnomAD) data. Our results suggest that only part of MYOC rare variants with a specific bioinformatics nature are likely pathogenic to cause adPOAG. Some controversial pathogenic variants as well as variants, as a risk factor for complex POAG, may need further clarification.

METHODS

Subject Ascertainment

Totally 7092 cases, including 455 with POAG and 6637 with various eye conditions, were recruited from Zhongshan Ophthalmic Center. Written informed consent adherent to the tenets of the Declaration of Helsinki was obtained from all participating individuals or their parents before collecting genomic DNA from peripheral venous blood and clinical data. The procedure for extracting genomic DNA has been reported.16 This study was approved by the Institutional Review Board of the Zhongshan Ophthalmic Center. The diagnosis of POAG was made on the grounds of structural and functional evidence of glaucomatous change as previously reported.12,17

Exome Sequencing Analysis

The whole-exome sequencing (WES) and targeted exome sequencing (TES) were performed in DNA samples from 7092 cases. The procedures of exome sequencing were described briefly in our previous study.18

Variants in MYOC were selected from WES or TES data and further analyzed according to the following steps: (1) synonymous variants and variants in untranslated or intronic regions without significant effects on splicing according to the Berkeley Drosophila Genome Project (BDGP; in the public domain, http://www.fruitfly.org/) were excluded; (2) polymorphisms with a minor allele frequency (MAF) equal to or greater than 0.01 either in our in-house controls or in the gnomAD database were compared between the 455 POAG and 6637 controls; (3) rare missense variants were predicted based on bioinformatics online tools including REVEL (https://sites.google.com/site/revelgenomics/),19 CADD (https://cadd.gs.washington.edu/),20 SIFT (http://sift.jcvi.org/www/SIFT_enst_submit.html),21 PolyPhen2 (http://genetics.bwh.harvard.edu/pph2/index.shtml),22 and PROVEAN (http://provean.jcvi.org).23 These 5 in silico tools work on different in silico algorithms to assess the impact of missense variants. Sanger sequencing was used to validate the variants after multiple analyses.24 Rare variants with MAF less than 0.01 were evaluated according to the evidence of pathogenicity assessment from the American College of Medical Genetics and Genomics (ACMG) as well as a series of evidence of benign impact (Table 1).

TABLE 1 - Forty-five Rare Variants and Eight Polymorphsims in MYOC Detected in the Present Study
Position NM_000261 NM_000261 G CTR In Silico Prediction GnomAD GnomAD
at chr1 Change Effect AC AC Diagnosis REVEL CADD AF EA_AF Evidence HGMD Ref.
Pathogenic
 171605559 c.1021T > C p.S341P 1 0 POAG 0.821 23.200 D D D / / Ma, Fy DM 34
 171605481 c.1099G > A p.G367R 2 0 POAG 0.787 25.600 D D D / / Ma, Fy DM 31
 171605471 c.1109C > T p.P370L 1 0 JOAG 0.890 24.500 D D D / / Ma, Fy DM 30
Likely pathogenic
 171605820 c.760C > A p.P254T 1 0 JOAG 0.933 24.400 D D D / / Ma Novel /
 171605447 c.1133A > G p.D378G 1 0 JOAG 0.950 26.800 D D D / / Ma DM? 12
 171605282 c.1298G > A p.C433Y 1 0 POAG 0.956 27.800 D D D / / Ma Novel /
 171605124 c.1456C > T p.L486F 1 0 POAG 0.891 25.100 D D D / / Ma DM? 12
VUS-a
 171605823 c.757G > A p.E253K 1 0 POAG 0.310 22.400 T B N / / Mc / /
 171605694 c.886C > T p.R296C 0 2 CTR 0.827 26.100 D D D 0.000016 0.000000 Ma, U3, Fn DM 43
 171605651 c.929G > A p.G310D 0 1 CTR 0.694 25.000 D D D / / Ma, U2 / /
 171605489 c.1091G > T p.G364V 0 1 CTR 0.896 25.500 D D D / / Ma, U2 DM 39
 171605420 c.1160G > A p.G387D 2 1 POAG, CTR 0.958 24.100 D D D 0.000012 0.000163 Ma, U2 DM? 12
 171605340 c.1240G > A p.E414K 1 0 POAG 0.092 11.710 T B N / / Mc, Fn / /
 171605146 c.1434C > G p.D478E 0 1 CTR 0.832 24.500 D D D / / Ma, U2 / /
 171605094 c.1486A > C p.T496P 1 0 POAG 0.704 23.2 T PD N / / Mc, Fn / /
VUS-b
 171621598 c.154A > G p.S52G 1 11 POAG, CTR 0.036 22.000 D B N 0.000004 0.000000 Mc, U3 / /
 171621594 c.158T > C p.V53A 1 14 JOAG, CTR 0.369 25.600 D D N 0.000060 0.000852 Mb, U3, Fn DM 29
 171621508 c.244C > T p.R82C 1 1 JOAG, CTR 0.268 22.800 D PD N 0.000074 0.000050 Mc, U2, Fn DM 5,10
 171621279 c.473G > A p.R158Q 0 1 CTR 0.289 10.460 D B N 0.000004 0.000000 Mc, U2, Fn DM 40
 171607823 c.644T > C p.L215P 2 13 POAG, CTR 0.724 27.600 D D D 0.000071 0.001002 Ma, U3, Fn DM? 40
 171605702 c.878C > A p.T293K 0 1 CTR 0.541 13.100 T B N 0.000012 0.000000 Mc, U2 DM 5,10
 171605682 c.898G > A p.E300K 0 2 CTR 0.734 23.200 D PD D 0.000036 0.000502 Ma, U2, Fn R 41
 171605642 c.938C > T p.S313F 2 11 POAG, CTR 0.554 23.200 T D N 0.000060 0.000802 Mc, U3, Fn DM 28
 171605556 c.1024A > G p.R342G 1 1 POAG, CTR 0.590 18.850 D B D 0.000012 0.000163 Mc, U2, Fn / /
 171605168 c.1412A > G p.Y471C 0 7 CTR 0.848 25.100 D D D 0.000028 0.000381 Ma, U3, Fn DM 41
Likely benign
 171621595 c.157G > A p.V53M 2 2 CTR 0.137 25.900 D D N / / Mb, U3, Fn / /
 171607808 c.659C > T p.T220I 1 1 CTR 0.556 25.100 D D N 0.000024 0.000000 Mb, U2 / /
 171607737 c.730G > A p.G244R 1 1 CTR 0.578 25.900 D PD D 0.000008 0.000000 Mb, U2 / /
 171605606 c.974C > T p.T325M 1 1 CTR 0.777 24.400 D D D 0.000028 0.000054 Mb, U2 / /
 171605117 c.1463C > T p.A488V 1 1 CTR 0.715 25.400 D D N 0.000020 0.000272 Ma, U1, U2 / /
 171605088 c.1492G > C p.D498H 1 1 CTR 0.656 25.900 D D N / / Mb, U2 / /
Benign
 171621705 c.47C > T p.P16L 0 2 CTR 0.062 18.800 T B N 0.000008 0.000109 Mc, U3 / /
 171621402 c.350T > C p.L117P 0 1 CTR 0.063 14.320 D B N / / Mc, U2 / /
 171621370 c.382C > T p.R128W 0 2 CTR 0.207 22.300 D PD N 0.000004 0.000000 Mc, U2 / /
 171621349 c.403A > T p.T135S 0 1 CTR 0.080 0.809 T B N 0.000004 0.000054 Mc, U2 / /
 171621279 c.473G > T p.R158L 0 1 CTR 0.289 10.460 D B N 0.000004 0.000000 Mc, U2 / /
 171621180 c.572C > T p.T191I 0 1 CTR 0.116 7.685 T B N 0.000004 0.000000 Mc, U2 / /
 171607856 c.611C > T p.T204M 0 7 CTR 0.242 0.176 T B N 0.000103 0.000702 Mc, U3 / /
 171605805 c.775A > T p.T259S 0 1 CTR 0.387 22.100 D B N / / Mc, U2 / /
 171605320 c.1260C > G p.N420K 0 2 CTR 0.323 20.700 T PD D 0.000032 0.000000 Mc, U1, U2 / /
 171605235 c.1345G > A p.V449I 0 1 CTR 0.181 0.389 T B N 0.000056 0.000000 Mc, U2 / /
 171621481 c.271C > T p.R91∗† 1 1 JOAG, CTR / 35.000 / / / 0.000032 0.000109 / DM 12
 171621313 c.439C > T p.R147 0 1 CTR / / / / / 0.000124 0.000050 / / /
 171621286 c.466A > T p.K156 0 1 CTR / / / / / 0.000004 0.000054 / / /
 171607788 c.679delA p.I227Ffs2 0 1 CTR / / / / / 0.000008 0.000000 / / /
Polymorphisms (MAF ≥ 0.01 in gnomAD database or in-house controls)
 171621718 c.34G > C p.G12R 5 111 POAG, CTR 0.330 4.770 T B N 0.000653 0.008922 Mc, U3, Fn DM 29
 171621695 c.57G > T p.Q19H 2 55 POAG, CTR 0.256 17.690 T B N 0.000285 0.003559 Mc, U3, Fn DM 28
 171621616 c.136C > T p.R46 5 121 POAG, CTR / 36.000 / / / 0.000675 0.009224 / DM 5,10
 171621525 c.227G > A p.R76K 65 895 POAG, CTR 0.203 15.430 D B N 0.142531 0.081696 Mc, U3 DP 5,10
 171621205 c.547G > A p.G183S 1 16 POAG, CTR 0.126 16.970 T B N 0.000048 0.000652 Mc, U3, Fn / /
 171607843 c.624C > G p.D208E 4 51 POAG, CTR 0.126 9.525 T B N 0.000004 0.000000 Mc, U3, Fn DM? 27
 171605526 c.1054G > A p.E352K 0 1 CTR 0.514 22.900 D D N 0.001192 0.000000 Mc, U2 DM 25
 171605522 c.1058C > T p.T353I 11 169 POAG, CTR 0.316 16.450 T PD N 0.000675 0.009122 Mc, U3 DM 26
/, not available. variant reported previously in our cohort. Diagnose abbreviation: AC, allele count; AF, allele frequency; CADD, Combined Annotation Dependent Depletion; CTR, control; DM, damaging mutations; G, glaucoma; HGMD, Human Gene Mutation Database; JOAG, juvenile open-angle glaucoma; MAF, minor allele frequency; POAG, primary open-angle glaucoma; REVEL, Rare Exome Variant Ensemble Learner; VUS, variants with uncertain significance. SIFT; Polyphen2; PROVEAN.B, benign; D, damaging in SIFT and probably damaging in Polyphen2; DP, disease-associated polymorphism; N, neutral; PD, possible damaging; R, retired; T, tolerant.Variants with a score of equal or above 95th percentile of REVEL or CADD were defined as damaging.Percentiles score of REVEL: 75% = 0.33, 90% = 0.541, 95% = 0.664; percentiles score of CADD: 75% = 16.24, 90% = 22.8, 95% = 24.7.Missense variants predicted to be damaging by at least 4 (Ma), 3 (Mb), and 2 or fewer (Mc) in silico tools.Variants also detected in unaffected family members (U1), or in 1 (U2) or 2 (U3) relative controls without POAG.Variants with family evidence (Fy, segregated or de novo) or without family evidence (Fn, no segregation information).

Statistical Analysis

The IBM SPSS Statistics Version 25.0 software was used for statistical analysis. The allele frequencies of each polymorphism were compared between POAG and controls using the χ2 test or Fisher Exact Test. Significant p values should be less than 0.003 (0.05/16) after Bonferroni correction. The gnomAD database was also used to calculate the risk of POAG from these polymorphisms. The χ2 test or Fisher Exact Test was also used in analyzing whether the different grades of variants were enriched in POAG compared with controls. With regard to the diagnostic age and maximal IOP among different groups from the literature review, Analysis of Variance was used to illustrate these characteristics.

Overview of Disease-Causing Variants in MYOC

Literatures with variants in MYOC reported previously were searched through PubMed (https://www.ncbi.nlm.nih.gov/pubmed/), Google Scholar (https://scholar.google.com/), and Web of Science (https://apps.webofknowledge.com/) from 1997 to February 2020 using terms as follows: “MYOC and mutation”, “TIGR and mutation”, and “myocilin”. Only literatures published in English were selected for further analysis. All disease-causing variants (DCVs) reported in the literatures were summarized and named based on NM_000261 and HGVS nomenclature (http://varnomen.hgvs.org). Thereafter, these variants were reclassified into the following 3 groups based on our data and gnomAD data: 1) pathogenic and likely pathogenic, 2) variants with uncertain significance (VUS), and 3) likely benign and benign. The distribution of these variants in POAG, in-house controls, and in the gnomAD database was compared to weigh the contributions of different variants to POAG. In addition, the ACMG guideline, familial or segregation information, and overall online prediction tools were used to evaluate systemically the pathogenicity of the variants.

RESULTS

Heterozygous Variants Identified in MYOC and Phenotypes of Patients With Pathogenic Variants

In total, 53 variants were identified in the 7092 probands with various eye conditions, including 48 missenses and 5 truncations (4 nonsenses and 1 frameshift; Table 1, Fig. 1). Eight of the 53, including 7 missense variants and 1 nonsense, were considered as polymorphisms with a MAF greater than 0.01, either in the in-house controls or in the gnomAD database. Subjects with these polymorphisms were equally distributed into different forms of eye conditions without significant enrichment in POAG patients (P > 0.05; Table S1, http://links.lww.com/APJO/A73).5,10,25–29

FIGURE 1
FIGURE 1:
The distribution of MYOC variants detected in our cohort. The horizontal axis is the proband count and the vertical axis represents the scaled down positions of the 504 amino acids of MYOC. The right lines represent variants detected in our cohort that are calculated as pathogenic or likely pathogenic ones. The lines on the left of the exon diagram are these likely benign and benign variants. These variants from 6 groups were distinguished by various colors. The numbers in parentheses represent mutant alleles count detected.

The remaining 45 were all heterozygous and were considered rare variants, including 41 missenses and 4 truncations. These 45 rare variants (as listed in Table 1) were more commonly seen in patients with POAG than in non-POAG controls (P = 3.31E-14; Table S2, http://links.lww.com/APJO/A74) based on our in-house data, suggesting the important role of MYOC variants in POAG. Then individual variants were further evaluated and classified as different subgroups based on the 5 in silico tools, associated phenotypes, frequency in gnomAD, and previously published evidence with POAG (Table 1). Of the 41 rare missense variants, 22 variants that were predicted as deleterious by at least 4 in silico tools were significantly more enriched in the POAG group than in the in-house control group (P = 2.9E-05, Table S3, http://links.lww.com/APJO/A75), whereas variants that were predicted as deleterious by no more than 3 in silico tools showed no significant difference between the 2 groups after Bonferroni correction (P = 0.08). This suggests that variants predicted to be damaging by at least 4 in silico tools were potentially pathogenic.

Seven of the 45 rare variants were classified as pathogenic or likely pathogenic variants (Fig. S1, http://links.lww.com/APJO/A78), including c.760C > A (p.P254T), c.1021T > C (p.S341P), c.1099G > A (p.G367R), c.1109C > T (p.P370L), c.1133A > G (p.D378G), c.1298G > A (p.C433Y), and c.1456C > T (p.L486F).12,30–32 Of the 7, 5 were known while 2 variants c.760C > A (p.P254T) and c.1298G > A (p.C433Y) were novel and detected in a patient with POAG and a patient with JOAG, respectively. For the 2 novel variants, different residues at the same codons were reported as DCVs, such as p.C433R, p.P254R, and p.P254L.33–40 All of the 7 were absent in the gnomAD database, predicted to be damaging by at least 4 in silico tools, and located in the olfactomedin (OLF)-domain. Four of them, p.S341P, p.P370L, p.D378G, and p.L486F, have been described in our previous study.12 Based on ACMG guideline,38 these 7 pathogenic and likely pathogenic variants were also grouped as pathogenic/likely pathogenic, accounting for 1.76% (8/455) of patients with open-angle glaucoma in our cohort (Table 1).

For the remaining 34 missense variants, 18 were temporarily classified into the VUS group (Table 1). In addtion, VUS were further classified into 2 subgroups, VUS-a and VUS-b, in line with the evidence from in silico prediction, clinical phenotype, frequency in gnomAD database, and classification in the Human Gene Mutation Database (HGMD)5,10,12,28,29,39–44 (Table 1). Variants predicted as damaging via at least 4 in silico tools and identified in POAG as well as controls were classified as VUS-a. However, variants distributed without distinction in POAG and controls or only detected in controls from in-house data were grouped into VUS-b, although most were reported as disease-causing mutation (DM) in the HGMD. Interestingly, all the variants in the VUS-a group were also found in the OLF- domain, whereas those in the VUS-b group were scattered around 3 exons of MYOC. Variants in the VUS-a group were significantly enriched in POAG (P = 3.59E-04) while the frequency of variants in the VUS-b group was similar to that in controls (P = 0.08), suggesting that variants in the VUS-a group are potentially pathogenic, although a few were also detected in controls. Apart from the 18 variants, the other 16 missense variants were classified as likely benign or benign due to low prediction scores, absence in POAG, and being frequently observed in the general population (Table 1).

Concerning the 4 rare truncated variants detected, they were not enriched in POAG (P = 0.28), and there was no more family information to support their pathogenicity except that one p.R91 carrier was reported in previous studies.12,41 With the increasing number of participants with WES or TES, another individual diagnosed as having high myopia also carried p.R91. The other 3 truncated variants were all identified in individuals without glaucoma (Table 1).

Overall, the 53 variants detected in 7092 individuals were present in 1592 probands and scattered in all 3 exons of MYOC. The 45 rare variants could be divided into 6 subgroups, and distributed differently between POAG and controls (P = 3.46E-09, Fig. 2, A). All 7 pathogenic or likely pathogenic variants and 10 variants of the VUS-a group were exclusively located in the OLF-domain of exon 3 (Fig. 1).

FIGURE 2
FIGURE 2:
Phenotype information in the present and previous studies. A, The distribution of different types of variants in POAG and CTR from our cohort. The vertical axis is the ratio of pedigree from different groups while the number are labeled in every module. POAG, primary open-angle glaucoma; CTR, control subjects with various eye conditions. The statistics is calculated by χ2 test. B, The proportion of 121 disease-causing variants (DCVs) of MYOC detected in patients with various diagnoses (upper) and the proportion of 64 pathogenic/likely pathogenic variants after re-evaluating. C, D, The maximal intraocular pressure (IOP) or IOP when first diagnosed and diagnostic ages recorded in previous studies and the comparison among them based on 3 groups possessing different pathogenicity evidence to POAG. B, benign variants; LB, likely benign variants; LP, likely pathogenic variants; P, pathogenic variants. ∗∗∗∗, the P value is less than 0.0001. ∗∗∗, the P value is less than 0.001. ∗∗, the P value is less than 0.01. ns, no significant difference. ACG, angle-closure glaucoma; NTG, normal-tension glaucoma; PCG, primary congenital glaucoma; VUS, variants with uncertain significance.

The phenotypes of the 7 probands and their available family members with pathogenic/likely pathogenic variants were summarized in Table 2, including 3 male individuals and 4 female individuals. The average age at diagnosis was 32.67 ± 11.02 years and ranged from 18 to 49 years. The mean highest IOP or IOP when first visiting our clinic was 35.13 ± 9.32 mm Hg.

TABLE 2 - Clinical Data of Four New Families With Pathogenic or Likely Pathogenic Variants in Our Cohort
IOP (mm Hg) BCVA VCDR Visual Field
Family ID Clinic Diagnosis Variants Effect Gender Age at Exam OD OS OD OS OD OS OD OS
Pathogenic variants
 10445 POAG OU c.[1099G > A];[=] p.[G367R];[=] M 46 28.7 39.3 0.7 NLP 0.9 1.0 TV CD
 20239 JOAG OU c.[1099G > A];[=] p.[G367R];[=] F 18 26.0 26.0 0.6 0.8 0.4 0.4 Normal Normal
Likely pathogenic variants
 11357 JOAG OU c.[760C > A];[=] p.[P254T];[=] M 19 38.7 41.1 0.9 0.7 0.5 0.9 TV IDD
 12776 JOAG OU c.[1298G > A];[=] p.[C433Y];[=] F 32 18.0 16.0 1.0 1.0 0.6 0.4 Normal IAD
F, female; M, male.BCVA, best-corrected visual acuity; CD, circular defect; IAD, inferior arcuate defect; IDD, interior difuse defect; OD, right eye; OS, left eye; TV, tubular vision; VCDR, vertical cup-to-disc ratio.
Intraocular pressure (IOP) after medication.

Overview of the Prevalence of the DCVs in MYOC From Multiple Population

Thus far, 134 variants reported in the HGMD were classified into 3 groups based on genotype-phenotype and bioinformatics analysis, including 121 MYOC DCVs (Table S4, http://links.lww.com/APJO/A76, and S5, http://links.lww.com/APJO/A77). Among the 121 DCVs, 85.95% (104/121) were missense variants, 2.48% (3/121) were inframe variants, and 11.58% (14/121) were truncated variants, including 6 nonsense variants and 7 frameshift variants, and 1 gross insertion. After re-evaluating the 121 DCVs, 52.89% of variants (64/121) were pathogenic/likely pathogenic variants, 32.23% (39/121) were VUS, and 14.88% (18/121) were likely benign/benign variants.

Truncation Variants Reported in the HGMD

The distribution of 13 truncated DCVs in the literature and c.136C > T(p.R46), which is reported as a polymorphism, were displayed on the left of the exon column (Fig. 3). Truncation variants in MYOC were detected in 184/14,401 patients with POAG, a number significantly higher than the 49/10,379 detected in the controls (P = 9.97E-11). One of the nonsense variants, c.1102C > T (p.Q368), was the most frequently reported and had supportive evidence not only in familial studies but also populational studies. The population studies displayed that the p.Q368 variant was identified in 139/9209 cases and significantly higher than in the control group (P = 8.53E-17). Nevertheless, the age-related penetrance of p.Q368 was 0% at 30 years old and 82% at 65 years.45

FIGURE 3
FIGURE 3:
The comparison of distribution of truncation MYOC variants between literature review and the gnomAD database. The general map of truncated variants from the gnomAD database (on the right of the exon diagram) and literature reviews. (on the right of the exon diagram) Variants with high frequency are labeled and the number behind the variant means the probands count reported. The number of probands based on literature reviews was divided into 2 family and population study according to the research method.

Missense Variants Detected in Previous Studies With Complicated Evidence

The missense variants contain 104 DCVs and 11 polymorphisms recorded in the HGMD, most of which are clustered in the OLF-domain in MYOC (Fig. 4). Missense variants were reclassified as 61 in the pathogenic/likely pathogenic, 37 in the VUS, and 5 in the likely benign/benign categories according to the comprehensive evaluation of in silico tools, population frequency, and whether classified as DM in the HGMD and phenotypic analysis (Table S4, http://links.lww.com/APJO/A76). Similar to our in-house data, 59 of 61 pathogenic/likely pathogenic missense variants were located in the OLF-domain, whereas 57 missense variants were predicted as damaging via at least 4 in silico tools, and 54 missense variants absented frequencies in the gnomAD (Figs. 1 and 5). What's more, the incomplete penetrance was also reported in c.1130C > T (p.T377 M) commonly.

FIGURE 4
FIGURE 4:
The distribution of MYOC DCVs from literature reviews and whole MYOC variants in the coding region from the gnomAD database after excluding variants with low quality. The numbers in parentheses represents allele count detected.
FIGURE 5
FIGURE 5:
The distribution of DCVs in MYOC based on previous studies. The vertical axis and horizontal represents the same meaning as Figure 1.

Characterized Phenotypes of MYOC Variants in the HGMD

From the literature review, 182 pedigrees were recorded based on the familial study and 395 probands were detected on population studies. Among the 577 index cases and their family members with DCVs, 47.97% (757/1578) of individuals were affected with POAG, 26.55% (419/1578) of individuals were diagnosed with JOAG, 1.52% (24/1578) normal-tension glaucoma (NTG), 6.40% (101/1578) were considered to have ocular hypertension or glaucoma suspect status, 1.33% (21/1578) participants had PCG, and 0.19% (3/1578) affected PACG, pigmentary glaucoma, or other unclassified glaucoma. However, another 15.02% (237/1582) of participants were under the DCV carriers category owing to showing no glaucomatous clinical signs when they underwent a series of ophthalmologic examinations (Fig. 2B).

As one of the early identified genes that caused POAG,39MYOC caused up to no more than 4% of POAG subjects in various populations.10,11 From the previous studies, DCVs in MYOC were detected in 2.45% (78/3190) of POAG and ocular hypertension subjects from US or Canadian, and the rates were 3.79% (55/1490) in Australian, 2.87% (90/3140) in European, 2.32% (91/3925) in East Asians, and 1.97% (33/1671) in Africans, respectively (Table S5, http://links.lww.com/APJO/A77).

Clinical Phenotype Compiled by Different Groups

Individuals with pathogenic/likely pathogenic variants, VUS, and likely benign/benign variants had an average maximal IOP or IOP when first diagnosed of 33.87 ± 9.89 mm Hg, 31.09 ± 9.54, and 29.73 ± 8.23, respectively, based on previous studies. There was a significant difference among the 3 groups in terms of maximal IOP or IOP when first diagnosed (P = 2.00E-04). Meanwhile, the mean age at diagnosis of the 3 groups were 35.47 ± 15.94, 37.97 ± 21.02 and 44.97 ± 19.76 years. The average diagnostic age was also significantly different among the 3 groups. (P = 6.80E-03) (Fig. 2C and D) However, the significant difference was caused by diagnosis age between the pathogenic/likely pathogenic group and likely benign/benign group. Interestingly, the phenotypes of pathogenic/likely pathogenic variants reported were concentrated in the POAG and JOAG groups (Fig. 2B).

DISCUSSION

In the present study, 8 polymorphisms with MAF ≥ 0.01 were distributed indistinctively in POAG and controls from our in-house analysis. In addition, all 45 rare variants were significantly higher in POAG than in controls (P = 3.31E-14) and in PACG (P = 2.22E-04). Interestingly, all 7 pathogenic or likely pathogenic variants and 8 variants of the VUS-a group were enriched in the OLF-domain of exon 3. The conserved OLF-domain was an important domain composed of extracellular proteins and involved in such disorders as glaucoma, cancers, inflammatory bowel disorder, and more.33,46 Pathogenic and likely pathogenic variants differ in whether there is a family history. For variants in the VUS-a group, their contribution to glaucoma is uncertain, as they were identified in the control group and the controls are too young to show glaucomatous signs and symptoms.

Taking the truncation variants into consideration, the evidence thus far has suggested that they scattered in the coding region of MYOC and the more frequently reported, the higher the frequency in the gnomAD database (Figs. 4 and 5). With a high frequency in the population, p.Q368 might be considered as a hot-pot with a high frequency, though incomplete penetrance was observed. Functional research revealed that the nonsense variant affected intracellular sequestration and myocilin secretion, which might be a mechanism of glaucoma caused by mutant MYOC. It is highly necessary to lucubrate the genetic factor in the POAG patient with p.Q368. However, there was no difference between truncation variants within the OLF-domain and those based on the gnomAD database. It suggests that the correlation between positions of truncation variants and their pathogenicity need further clarification.

It is suggested that MYOC DCVs result in misfolded protein and can further be degraded in the endoplasmic reticulum. Dimers or multimers may be the important form of myocilin's function, and the gain of function is recognized as the mechanism of DCVs.47,48 Glaucoma caused by mutant MYOC is considered an autosomal dominant inheritance disease, and a missense variant occurring in the OLF-domain led to a misfolding dimer or multimer. However, the misfolding protein interfered with these normal one while the truncation variants, which led to a truncated protein, degraded intracellularly and had less impact on the wild type dimer or multimer.49 The mechanism also needs more evidence to clarify. Moreover, the myocilin also expresses in myelin sheath in peripheral nerves, which provide a new insight that variants in myocilin may reduce the tolerance of optic nerve to pressure.50

From the variants reported and the present study, it is clear that the OLF-domain plays an important role in the function of myocilin, and variants identified in the region suggest they are pathogenic and could lead to glaucoma. The pathogenicity of missense variants could be more accurate when combining multiple lines of computational data on different principles. Heterozygous truncation variants have a mild influence on the occurrence of phenotypes because truncated protein degrades intracellularly. Therefore, the wild type amino acid could also form a dimer or multimer with proper function. This may cause subjects with truncation variant in MYOC to show a normal phenotype. Thus, truncation variants need more genetic evidence to clarify their potential pathogenicity.

Acknowledgments

The authors are grateful to the patients and their family members for their participation.

REFERENCES

1. Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006; 90:262–267.
2. Tham YC, Li X, Wong TY, et al. Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis. Ophthalmology 2014; 121:2081–2090.
3. Goldwyn R, Waltman SR, Becker B. Primary open-angle glaucoma in adolescents and young adults. Arch Ophthalmol 1970; 84:579–582.
4. Wiggs JL, Pasquale LR. Genetics of glaucoma. Hum Mol Genet 2017; 26:R21–R27.
5. Alward WL, Fingert JH, Coote MA, et al. Clinical features associated with mutations in the chromosome 1 open-angle glaucoma gene (GLC1A). N Engl J Med 1998; 338:1022–1027.
6. Rezaie T, Child A, Hitchings R, et al. Adult-onset primary open-angle glaucoma caused by mutations in optineurin. Science 2002; 295:1077–1079.
7. Fingert JH, Robin AL, Stone JL, et al. Copy number variations on chromosome 12q14 in patients with normal tension glaucoma. Hum Mol Genet 2011; 20:2482–2494.
8. Monemi S, Spaeth G, DaSilva A, et al. Identification of a novel adult-onset primary open-angle glaucoma (POAG) gene on 5q22.1. Hum Mol Genet 2005; 14:725–733.
9. Wiggs JL, Allingham RR, Vollrath D, et al. Prevalence of mutations in TIGR/Myocilin in patients with adult and juvenile primary open-angle glaucoma. Am J Hum Genet 1998; 63:1549–1552.
10. Fingert JH, Heon E, Liebmann JM, et al. Analysis of myocilin mutations in 1703 glaucoma patients from five different populations. Hum Mol Genet 1999; 8:899–905.
11. Faucher M, Anctil JL, Rodrigue MA, et al. Founder TIGR/myocilin mutations for glaucoma in the Quebec population. Hum Mol Genet 2002; 11:2077–2090.
12. Huang X, Li M, Guo X, et al. Mutation analysis of seven known glaucoma-associated genes in Chinese patients with glaucoma. Invest Ophthalmol Vis Sci 2014; 55:3594–3602.
13. Wiggs JL, Vollrath D. Molecular and clinical evaluation of a patient hemizygous for TIGR/MYOC. Arch Ophthalmol 2001; 119:1674–1678.
14. Kim BS, Savinova OV, Reedy MV, et al. Targeted disruption of the myocilin gene (Myoc) suggests that human glaucoma-causing mutations are gain of function. Mol Cell Biol 2001; 21:7707–7713.
15. Yi Z, Xiao X, Li S, et al. Pathogenicity discrimination and genetic test reference for CRX variants based on genotype-phenotype analysis. Exp Eye Res 2019; 189:107846.
16. Wang Q, Wang P, Li S, et al. Mitochondrial DNA haplogroup distribution in Chaoshanese with and without myopia. Mol Vis 2010; 16:303–309.
17. Foster PJ, Buhrmann R, Quigley HA, et al. The definition and classification of glaucoma in prevalence surveys. Br J Ophthalmol 2002; 86:238–242.
18. Sun W, Huang L, Xu Y, et al. Exome sequencing on 298 probands with early-onset high myopia: approximately one-fourth show potential pathogenic mutations in RetNet Genes. Invest Ophthalmol Vis Sci 2015; 56:8365–8372.
19. Ioannidis NM, Rothstein JH, Pejaver V, et al. REVEL: an ensemble method for predicting the pathogenicity of rare missense variants. Am J Hum Genet 2016; 99:877–885.
20. Rentzsch P, Witten D, Cooper GM, et al. CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Res 2019; 47:D886–D894.
21. Ng PC, Henikoff S. SIFT: Predicting amino acid changes that affect protein function. Nucleic Acids Res 2003; 31:3812–3814.
22. Adzhubei I, Jordan DM, Sunyaev SR. Predicting functional effect of human missense mutations using PolyPhen-2. Curr Protoc Hum Genet 2013.
23. Choi Y, Sims GE, Murphy S, et al. Predicting the functional effect of amino acid substitutions and indels. PLoS One 2012; 7:e46688.
24. Li J, Jiang D, Xiao X, et al. Evaluation of 12 myopia-associated genes in Chinese patients with high myopia. Invest Ophthalmol Vis Sci 2015; 56:722–729.
25. Allingham RR, Wiggs JL, De La Paz MA, et al. Gln368STOP myocilin mutation in families with late-onset primary open-angle glaucoma. Invest Ophthalmol Vis Sci 1998; 39:2288–2295.
26. Yoon SJ, Kim HS, Moon JI, et al. Mutations of the TIGR/MYOC gene in primary open-angle glaucoma in Korea. Am J Hum Genet 1999; 64:1775–1778.
27. Lam DS, Leung YF, Chua JK, et al. Truncations in the TIGR gene in individuals with and without primary open-angle glaucoma. Invest Ophthalmol Vis Sci 2000; 41:1386–1391.
28. Fan BJ, Wang DY, Fan DS, et al. SNPs and interaction analyses of myocilin, optineurin, and apolipoprotein E in primary open angle glaucoma patients. Mol Vis 2005; 11:625–631.
29. Jia LY, Tam PO, Chiang SW, et al. Multiple gene polymorphisms analysis revealed a different profile of genetic polymorphisms of primary open-angle glaucoma in northern Chinese. Mol Vis 2009; 15:89–98.
30. Adam MF, Belmouden A, Binisti P, et al. Recurrent mutations in a single exon encoding the evolutionarily conserved olfactomedin-homology domain of TIGR in familial open-angle glaucoma. Hum Mol Genet 1997; 6:2091–2097.
31. Suzuki Y, Shirato S, Taniguchi F, et al. Mutations in the TIGR gene in familial primary open-angle glaucoma in Japan. Am J Hum Genet 1997; 61:1202–1204.
32. Kee C, Ahn BH. TIGR gene in primary open-angle glaucoma and steroid-induced glaucoma. Korean J Ophthalmol 1997; 11:75–78.
33. Vasconcellos JP, Melo MB, Costa VP, et al. Novel mutation in the MYOC gene in primary open glaucoma patients. J Med Genet 2000; 37:301–303.
34. Gobeil S, Letartre L, Raymond V. Functional analysis of the glaucoma-causing TIGR/myocilin protein: integrity of amino-terminal coiled-coil regions and olfactomedin homology domain is essential for extracellular adhesion and secretion. Exp Eye Res 2006; 82:1017–1029.
35. Donegan RK, Hill SE, Freeman DM, et al. Structural basis for misfolding in myocilin-associated glaucoma. Hum Mol Genet 2015; 24:2111–2124.
36. Yang Y, Shi Y, Huang X, et al. Identification of a novel MYOC mutation in a Chinese family with primary open-angle glaucoma. Gene 2015; 571:188–193.
37. Souzeau E, Burdon KP, Ridge B, et al. A novel de novo Myocilin variant in a patient with sporadic juvenile open angle glaucoma. BMC Med Genet 2016; 17:30.
38. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med 2015; 17:405–424.
39. Stone EM, Fingert JH, Alward WL, et al. Identification of a gene that causes primary open angle glaucoma. Science 1997; 275:668–670.
40. Kubota R, Mashima Y, Ohtake Y, et al. Novel mutations in the myocilin gene in Japanese glaucoma patients. Hum Mutat 2000; 16:270.
41. Pang CP, Leung YF, Fan B, et al. TIGR/MYOC gene sequence alterations in individuals with and without primary open-angle glaucoma. Invest Ophthalmol Vis Sci 2002; 43:3231–3235.
42. Chen Y, Jiang D, Yu L, et al. CYP1B1 and MYOC mutations in 116 Chinese patients with primary congenital glaucoma. Arch Ophthalmol 2008; 126:1443–1447.
43. Ennis S, Gibson J, Griffiths H, et al. Prevalence of myocilin gene mutations in a novel UK cohort of POAG patients. Eye (Lond) 2010; 24:328–333.
44. Liu W, Liu Y, Challa P, et al. Low prevalence of myocilin mutations in an African American population with primary open-angle glaucoma. Mol Vis 2012; 18:2241–2246.
45. Craig JE, Baird PN, Healey DL, et al. Evidence for genetic heterogeneity within eight glaucoma families, with the GLC1A Gln368STOP mutation being an important phenotypic modifier. Ophthalmology 2001; 108:1607–1620.
46. Yokoe H, Anholt RR. Molecular cloning of olfactomedin, an extracellular matrix protein specific to olfactory neuroepithelium. Proc Natl Acad Sci U S A 1993; 90:4655–4659.
47. Caballero M, Rowlette LL, Borras T. Altered secretion of a TIGR/MYOC mutant lacking the olfactomedin domain. Biochim Biophys Acta 2000; 1502:447–460.
48. Fautsch MP, Johnson DH. Characterization of myocilin-myocilin interactions. Invest Ophthalmol Vis Sci 2001; 42:2324–2331.
49. Tamm ER. Myocilin and glaucoma: facts and ideas. Prog Retin Eye Res 2002; 21:395–428.
50. Ohlmann A, Goldwich A, Flugel-Koch C, et al. Secreted glycoprotein myocilin is a component of the myelin sheath in peripheral nerves. Glia 2003; 43:128–140.
Keywords:

exome sequencing; myocilin gene; olfactomedin domain; primary open-angle glaucoma; variant

Supplemental Digital Content

Copyright © 2021 Asia-Pacific Academy of Ophthalmology. Published by Wolters Kluwer Health, Inc. on behalf of the Asia-Pacific Academy of Ophthalmology.