Association between SRBD1 and ELOVL5 Gene Polymorphisms and Primary Open-Angle Glaucoma

Abstract

Purpose.: The S1 RNA binding domain 1 (SRBD1) and elongation of long-chain fatty acids family member 5 (ELOVL5) have been reported to be susceptibility genes for early-onset normal-tension glaucoma (NTG). The present study we conducted to assess whether these genes were associated with primary open-angle glaucoma (POAG), including late-onset NTG and high-tension glaucoma (HTG).

Methods.: Three hundred seventy Japanese patients with POAG, including 158 NTG and 212 HTG patients and 191 control subjects were analyzed for SRBD1 (rs3213787) and ELOVL5 (rs735860) gene polymorphisms.

Results.: The A allele frequencies of rs3213787 were significantly higher in NTG (98.4%, P = 0.0003) and HTG (97.6%, P = 0.0013) patients than in the control subjects (92.7%). The A allele frequency was significantly higher (P = 0.014), even though the NTG patients were limited to those diagnosed at ages older than 60 years. The POAG patients with the CC or CT risk genotypes of rs735860 were significantly older (P = 0.032, analysis of variance, P = 0.043 and P = 0.015, respectively) than were the POAG patients with the TT genotype, and the frequency of a family history of glaucoma in POAG patients with the CC risk genotype was significantly higher (P = 0.015) than that in POAG patients with the TT genotype.

Conclusions.: SRBD1 gene polymorphism is associated with the development of HTG as well as NTG, including late-onset NTG. Typical POAG associated with ELOVL5 gene polymorphism may have a late rather than an early onset.

Glaucoma includes a group of eye disorders characterized by visual field defects, apoptosis of the retinal ganglion cells, and progressive degeneration of the optic nerve. Primary open-angle glaucoma (POAG) represents the most prevalent form of glaucoma and is clinically classified as high-tension glaucoma (HTG), in which elevated intraocular pressure (IOP) is a major feature, and normal-tension glaucoma (NTG), in which IOP is consistently within the statistically normal population range. It is recognized that POAG is a multifactorial disorder involving the role of multiple genes. ¹ To date, mutations in three genes (myocilin, ² optineurin, ³ and WD repeat domain 36 ⁴ ) and several susceptibility gene polymorphisms, such as optic atrophy 1 ⁵ and estrogen receptor beta, ⁶ have been reported to contribute to the development of POAG. However, these mutations and polymorphisms do not account for all POAG cases, and other unknown genetic factors are considered to contribute to the development of POAG. In the Japanese population, the prevalence of NTG is higher than in other ethnic populations (92% of Japanese patients with POAG have NTG), ⁷ and it is important to elucidate the genetic factors that are responsible for the development of NTG. 

To identify susceptibility genes for early-onset NTG, we previously performed a genome-wide association study with more than 500,000 single-nucleotide polymorphisms (SNPs; GeneChip Human Mapping 500K Array Set; Affymetrix Inc., Santa Clara, CA), and found that S1 RNA binding domain 1 (SRBD1) and elongation of long-chain fatty acids family member 5 (ELOVL5) were new susceptibility genes for early-onset NTG. ⁸ As the next step, the present study was conducted to investigate whether the SRBD1 and ELOVL5 gene polymorphisms were associated with POAG, including late-onset NTG and HTG. Furthermore, we also investigated in detail the association between these gene polymorphisms and the phenotypic features of patients with POAG. 

Materials and Methods

Subjects

Japanese patients with POAG were recruited from the ophthalmology practices at the University of Yamanashi Hospital, Enzan Municipal Hospital, Uenohara City Hospital, and Oizumi Clinic in Yamanashi or Nagano Prefectures, Japan. A diagnosis of POAG was made when open angles were detected on a gonioscopic examination and the typical glaucomatous cupping of the optic disc (thinning of the optic disc rim and/or enlargement of the optic disc cupping) with compatible visual field defects (nasal step and/or partial arcuate visual field defect) was detected by automated static perimetry (Humphrey Visual Field Analyzer 30-2: HFA30-2; Humphrey Instruments-Carl Zeiss Meditec, Dubin, CA). In addition, patients with HTG had evidence of at least one previous measurement of IOP that was more than 21 mm Hg with a Goldmann applanation tonometer. Patients with NTG showed an IOP of less than 21 mm Hg each time they were tested. Patients were excluded if they had a history of eye surgery, including laser treatment, before the diagnosis of POAG. Control subjects recruited from the participating institutions were Japanese who were older than 40 years, had an IOP below 21 mm Hg, exhibited no glaucomatous cupping of the optic disc (no thinning of disc rim and cup-to-disc ratio less than 0.4), and had no family history of glaucoma. All subjects received comprehensive ophthalmic examinations, including both slit-lamp biomicroscopy and funduscopy, and were also interviewed to determine whether they had any relatives with glaucoma. The subjects who reported having such a relative were then recorded as having a family history of glaucoma. The study protocol was approved by the Ethics Committee of University of Yamanashi, and informed consent was obtained from all study participants. The study was conducted in accordance with the Declaration of Helsinki. 

Genomic DNA Genotyping

Peripheral blood was collected, and genomic DNA was purified (Flexi Gene DNA Kit; Qiagen, Valencia, CA). To assess the association between the SRBD1 and ELOVL5 gene polymorphisms and POAG, we selected for genotyping the rs3213787 SNP, located in intron 17 of the SRBD1 gene on chromosome 2, region p21, and the rs735860 SNP, located in the 3′-UTR of the ELOVL5 gene on chromosome 6, region p21.1-p12.1. The SNPs were selected and genotyped with an allele-specific primer real-time PCR method, because the most statistically significant association was observed for these SNPs, which had 2.80- and 1.69-fold increased risks of early-onset NTG for the A and C alleles respectively, as previously described. ⁸ The following primers were used for amplification, and the predicted amplicon lengths for rs3213787 and rs735860 were 105 and 212 base pairs, respectively: for rs3213787: A allele–specific forward primer, AATGTATAAACCCATAGACGTTCCCTA; G allele–specific forward primer, AATGTATAAACCCATAGACGTTCCCTG; and common reverse primer, TCACAGAATCTTGAGTTTAACTGGC and for rs735860: C allele–specific forward primer, CCTTGGTCCTGCTCCGTCC; T allele–specific forward primer, CCTTGGTCCTGCTCCGTCT; and common reverse primer, GACCCAGAGTGCCAGAACA. 

Allelic discrimination for these SNPs was achieved using PCR amplification of specific alleles. Briefly, the first nucleotide difference (A or G) between sense primers used to discriminate between the major and minor alleles for rs3213787 is located on the 3′ end. The second primer base change (T to C) located 3 bases from the 3′ end generates an internal primer–template mismatch, and this prevents the amplification of the nonmatching primer. These changes were made to prevent the generation of possible spurious products that could otherwise occur by the annealing and extension of a G allele–specific primer to the first-round product of an A allele–specific primer. PCR amplifications with A and G allele–specific primers were separately performed to determine the genotypes for this SNP. To perform accurate genotyping, we used the amplification products accurately genotyped by direct sequencing as positive controls for this method. The genotyping for rs735860 was performed in the same manner. 

Statistical Analysis

A χ² analysis of Hardy-Weinberg equilibrium was performed for patients and control subjects. Genotype and allele frequency differences between NTG or HTG patients and control subjects were estimated using the χ² test and Fisher's exact test, respectively. In addition, a logistic regression model was used to study the effects of the rs3213787 and rs735860 alleles when comparing the POAG patients with control subjects. The predictor variables were age, sex, maximum IOP, refractive error, rs3213787 A allele, and rs735860 C allele. The odds ratios of age, maximum IOP, and refractive error are determined per year, mm Hg, and diopter, respectively. Demographic and clinical features, including age, sex, refractive error, maximum IOP, family history of glaucoma, and history of glaucoma surgery, were compared between the POAG patients with and without the rs3213787 G allele, by using Student's t-test for continuous variables and Fisher's exact test for a comparison of proportions. Demographic and clinical features in patients with POAG were also compared between rs735860 genotypes, by using an analysis of variance (ANOVA) for continuous variables and the χ² test for a comparison of proportions. IOP was not adjusted by the central corneal thickness, because the central corneal thickness had not been measured. A value of P < 0.05 was considered to be statistically significant (all analyses: SAS statistical software, ver. 9.1; SAS Institute Inc., Cary, NC). 

Results

A different set of participants from those in our previous study, ⁸ including 370 Japanese patients with POAG (158 patients with NTG and 212 patients with HTG) and 191 control subjects were enrolled in the present study. The demographic and clinical data in patients with POAG and the control subjects are shown in Table 1. The mean age was 65.3 ± 13.9 years (SD) in patients with POAG and 65.7 ± 11.4 years in the control subjects. The mean of maximum known IOP was 24.2 ± 8.5 mm Hg in patients with POAG and 15.0 ± 2.7 mm Hg in the control subjects. 

Table 1.

 

Demographic and Clinical Features of Patients with POAG and Control Subjects

	Control (n = 191)	POAG
		NTG (n = 158)	HTG (n = 212)
Age, y	65.7 ± 11.4	68.6 ± 11.8	62.9 ± 14.8
Male, n (%)	70 (36.6)	65 (41.1)	129 (60.8)
Maximum known IOP, mm Hg	15.0 ± 2.7	18.5 ± 2.0	28.5 ± 9.0
Refractive error, D	−0.4 ± 2.2	−1.4 ± 3.4	−2.1 ± 3.0
Familial history of glaucoma, n (%)	0 (0)	36 (22.8)	62 (29.2)

 Continuous variables are expressed as the mean ± SD.

The genotype and allele frequencies of the SRBD1 (rs3213787) and ELOVL5 (rs735860) gene polymorphisms in patients with POAG and the control subjects are shown in Table 2. The genotype and allele frequencies were in Hardy-Weinberg equilibrium in patients with NTG and HTG and the control subjects. There was a significant difference (P = 0.0004, χ² test) in the rs3213787 genotype frequencies between the NTG patients and control subjects, and the frequency of the A allele was significantly higher (P = 0.0003, Fisher's exact test) in patients with NTG in comparison to the control subjects (98.4% vs. 92.7%). The genotype and allele frequencies were still significantly different (P = 0.014, χ² test, and P = 0.014, Fisher's exact test, respectively), even though the NTG patients were limited to those diagnosed at age 60 years or older by excluding patients with early-onset (<60 years old) NTG. Similarly, a significant difference (P = 0.0018, χ² test) was found in the rs3213787 genotype frequencies between the HTG patients and control subjects, and the frequency of the A allele was significantly higher (P = 0.0013, Fisher's exact test) in patients with HTG in comparison to the control subjects (97.6% vs. 92.7%). No significant differences were observed regarding the rs735860 allele and genotype frequencies between the NTG or HTG patients and the control subjects. After adjusting for age, sex, maximum IOP, refractive error, and the rs735860 C allele, a nearly 5.8-fold increased risk of POAG (P = 0.0027, odds ratio 5.78, 95% confidence interval 1.84–18.2) was found for the rs3213787 A allele (Table 3). 

Table 2.

 

Genotype and Allele Frequencies of rs3213787 and rs735860 in Patients with POAG and in Control Subjects

Allele/Genotype	Control (n = 191)	POAG
		NTG (n = 158)	P	HTG (n = 212)	P *
rs3213787
GG	1 (0.5)	1 (0.6)		1 (0.5)
AG	26 (13.6)	3 (1.9)	0.0004*	8 (3.8)	0.0018*
AA	164 (85.9)	154 (97.5)		203 (95.7)
Allele
G	28 (7.3)	5 (1.6)	0.0003†	10 (2.4)	0.0013†
A	354 (92.7)	311 (98.4)		414 (97.6)
rs735860
CC	31 (16.2)	23 (14.5)		43 (20.3)
CT	89 (46.6)	84 (53.2)	0.47*	106 (50.0)	0.24*
TT	71 (37.2)	51 (32.3)		63 (29.7)
Allele
C	151 (39.5)	130 (41.1)	0.70†	192 (45.3)	0.10†
T	231 (60.5)	186 (58.9)		232 (54.7)

 Data are given as numbers (percentage).

*  χ² test.

†  Fisher's exact test.

Table 3.

 

Results of Logistic Regression Analysis between POAG Patients and Control Subjects

Variable	P	Odds Ratio (95% CI)
Age, y	0.0012	1.04 (1.02–1.07)
Male	0.69	1.12 (0.63–1.99)
IOP, mm Hg	<0.0001	1.97 (1.74–2.24)
Refractive error, D	0.0034	0.85 (0.76–0.95)
rs3213787 A allele	0.0027	5.78 (1.84–18.2)
rs735860 C allele	0.083	1.69 (0.93–3.05)

 CI, confidence interval.

When the demographic and clinical features of patients with POAG were compared between the rs735860 genotypes, significant differences in the age (P = 0.032, ANOVA, Fig. 1) and the existence of a family history of glaucoma (P = 0.048, χ² test, Fig. 2) were noted. The POAG patients with the CC or CT genotypes were significantly older (P = 0.043 and P = 0.015 respectively; Fig. 1) than those in the POAG patients with the TT genotype. The frequency of a family history of glaucoma (the existence of relatives with glaucoma in the first to fourth degrees) in the POAG patients with the CC genotype was significantly higher (P = 0.015, Fig. 2) than that in the POAG patients with the TT genotype. There were no significant associations between the maximum IOP and these two SNPs in the patients with POAG. 

Figure 1.

 

Association between the rs735860 genotype of the ELOVL5 gene and age in patients with POAG. The POAG patients with the CC or CT risk genotypes were significantly older (analysis of variance, P = 0.032) than were the POAG patients with the TT genotype.

Figure 2.

 

Association between the rs735860 genotype of the ELOVL5 gene and a family history of glaucoma in patients with POAG. A significant difference (P = 0.048, χ² test) in the frequency of a family history of glaucoma was noted between the rs735860 genotypes in patients with POAG.

Discussion

In our previous study, ⁸ we selected the NTG patients with a comparatively early onset (<60 years old), because an early onset of disease suggests a stronger involvement of genetic factors, and reported that the SRBD1 and ELOVL5 genes were associated with NTG. Although the present study included NTG patients with a wide range of ages (27–88 years) compared with those in our previous study, ⁸ an association between the SRBD1 gene and NTG was found. Even though the NTG patients were limited to those in whom the disease was diagnosed at 60 years of age or older, this association was still confirmed, suggesting that the SRBD1 gene was associated with not only early-onset but also late-onset NTG. A variety of genetic factors contributes to optic neuropathy in POAG and can be classified into two types of genetic factors. One is a non–IOP-related genetic factor, and the other is a high-IOP–related genetic factor, and it is presumed that non–IOP-related genetic factors would predominate in patients with NTG. Furthermore, high-IOP–related genetic factors are predicted to predominate in patients with HTG. The SRBD1 gene polymorphism appears to contribute to glaucomatous optic neuropathy as a non–IOP-related genetic factor, because this gene polymorphism is associated with NTG, in which the IOPs are consistently within the statistically normal population range, and there was no statistically significant difference in the maximum IOP between the POAG patients with and without the rs3213787 G allele in the present study. However, interestingly, it was found that the SRBD1 gene, considered to be a non–IOP-related genetic factor, was also associated with HTG, thus indicating that the factors without high IOP, such as the vulnerability of the optic nerve, may play an important role in the development of glaucomatous optic neuropathy in patients with high IOP. The present study excluded patients with ocular hypertension who did not have POAG but had high IOP. If ocular hypertension was included in patients with HTG, an association between the SRBD1 gene and HTG may not be found, because of the low prevalence of the risk A allele in patients with ocular hypertension. Further studies should therefore be performed to confirm the association between this gene and ocular hypertension. We reported that the risk A allele correlated with the enhanced expression of SRBD1. ⁸ If the expression of SRBD1 in patients with the AA genotype is considered to be normal, because the A allele is a major allele, the restricted SRBD1 expression caused by the minor G allele may reduce the expression of RNA binding proteins with the S1 domain. This decrease may indirectly lead to the promotion of cell growth ^9,10 and general protein synthesis, ¹¹ and inhibit the apoptosis ¹² of retinal ganglion cells, resulting in the prevention of POAG (both NTG and HTG). 

As for ELOVL5, an association between the rs735860 genotypes and a phenotype feature in patients with POAG was found, although no significant difference was observed regarding the allele and genotype frequencies between the NTG or HTG patients and the control subjects. The age of the POAG patients with the risk C allele of rs735860 was older than that of POAG patients without the risk C allele. There was an increased risk (odds ratio) of early-onset NTG in the patients with the risk C allele, but this was not high, as reported in our previous study, ⁸ whereas the risk C allele frequencies in the Japanese population and the control subjects of the present study were high (49% and 40%, respectively). These data suggest that the effect of the ELOVL5 gene polymorphism for POAG is not strong and that typical phenotype in patients with POAG associated with the ELOVL5 gene polymorphism may show a late onset. Although ELOVL5 was identified as a susceptibility gene for early-onset NTG, other genetic factors in addition to ELOVL5 would be necessary to induce an early onset. There was no statistically significant difference in the maximum IOP between the POAG patients with and without the risk C allele. This result supports that the ELOVL5 gene polymorphism appears to contribute to glaucomatous optic neuropathy as a non–IOP-related genetic factor. ELOVL5 is one of the mammalian fatty acid condensing enzymes involved in microsomal long-chain polyunsaturated fatty acid (LCPUFA) elongation. We reported that the risk C allele correlated with the enhanced expression of ELOVL5. ⁸ Alterations in LCPUFA expression levels caused by the ELOVL5 gene polymorphism may contribute to glaucomatous optic neuropathy by inducing the apoptosis ¹³ in retinal ganglion cells. The frequency of a family history of glaucoma in POAG patients with the risk C allele was higher than that in POAG patients without the risk C allele. This result supports the association of the ELOVL5 gene polymorphism with POAG. 

In conclusion, the present study demonstrated that the SRBD1 gene polymorphism was associated with the development of HTG as well as NTG, including late-onset NTG. Although the ELOVL5 gene polymorphism alone may not be sufficient to cause POAG, the association of the ELOVL5 gene polymorphism with the phenotypic features and a family history of glaucoma in patients with POAG suggests that the ELOVL5 gene polymorphism can contribute to POAG and that typical POAG associated with the ELOVL5 gene polymorphism may have a late onset rather than an early onset. Further studies should be performed to evaluate whether these genes are associated with POAG in other ethnic populations. 

Footnotes

 Supported in part by Grant-in-Aid 21592221 from the Ministry of Education, Science, Sports, and Culture, Japan.

Footnotes

 Disclosure: F. Mabuchi, None; Y. Sakurada, None; K. Kashiwagi, None; Z. Yamagata, None; H. Iijima, None; S. Tsukahara, None