To determine whether polymorphisms in the Toll-like receptor 4 ( TLR4 ) gene are associated with primary open-angle glaucoma (POAG), normal-tension glaucoma (NTG), and exfoliation glaucoma (XFG) in Japanese individuals.
Genetic association study.
Setting: Multicenter study. Study population: One hundred eighty-four unrelated Japanese patients with POAG, 365 unrelated patients with NTG, and 109 unrelated patients with XFG from 5 hospitals. Procedures: Genomic DNA was extracted from leukocytes of the peripheral blood, and 8 polymorphisms in the TLR4 genes were amplified by polymerase chain reaction (PCR) and directly sequenced. Allele and genotype frequencies and the inferred haplotypes were estimated. Main outcome measures: Differences in allele and genotype frequencies and haplotypes between subjects with POAG, NTG, and XFG.
The allele frequency of rs2149356 of the TLR4 gene in the POAG, NTG, and XFG groups was the most significantly different from that of the control group (minor allele frequency 0.446, 0.395, 0.404, vs 0.308; P = .000058, P = .0030, and P = .015). The allele frequencies of the 5 TLR4 SNPs were higher in all of the glaucoma groups than that in the control group. The statistics of genotypes of TLR4 were approximately the same for all allele frequencies. The haplotypic frequencies with Tag SNPs studied earlier showed that only POAG was statistically significant. Other haplotypes, such as rs10759930, rs1927914, rs1927911, and rs2149356, had higher statistical significance (overall P = .00078 in POAG, overall P = .018 in NTG, and overall P = .014 in XFG).
This study demonstrated that TLR4 polymorphisms are associated with NTG in the Japanese, and they also play a role in the pathogenesis of POAG and XFG.
Glaucoma is a complex, heterogeneous disease characterized by a progressive degeneration of the axons of the retinal ganglion cells (RGCs). It is the second-highest cause of blindness worldwide, affecting approximately 70 million people. Primary open-angle glaucoma (POAG), the most common type of glaucoma, is associated with an elevated intraocular pressure (IOP). However, there are some POAG patients who have normal IOPs of <22 mm Hg, and they are classified as having normal-tension glaucoma (NTG). The prevalence of NTG is higher among the Japanese than among whites. POAG and NTG appear to be a continuum of glaucoma with overlapping causative factors in addition to the IOPs. It is believed that the mechanism shifts from predominantly elevated IOP in POAG to that of independent factors in eyes with NTG. Although the precise molecular basis of POAG and NTG has not been determined, the glaucoma in patients with POAG and NTG is probably heterogeneous and is caused by the interaction of multiple genes and environmental factors.
Several genetic loci that contribute to the susceptibility of eyes to POAG/NTG have been identified, and at least 15 loci, from GLC1A to GLC1O, have been linked to POAG. Three genes have been identified worldwide: the myocilin ( MYOC ) gene, the optineurin ( OPTN ) gene, and the WD repeat domain 36 ( WDR36 ) gene, with a diverse mutation spectrum. Other studies have reported that the OPTN and WDR36 variants do not predispose individuals to POAG and NTG. The pseudoexfoliation syndrome (XFS; OMIM: 177650 ) is a generalized disorder of the extracellular matrix and is characterized by the pathologic accumulation of abnormal fibrillar material in the anterior segment of the eye. A recent genome-wide association study (GWAS) showed a strong association between single nucleotide polymorphisms (SNPs) in the lysyl oxidase–like 1 ( LOXL1 ) gene and XFS in the Swedish and Icelandic populations. The association between the LOXL1 gene and XFS and exfoliation glaucoma (XFG) has also been found in the Japanese population. XFG is a common identifiable cause of open-angle glaucoma worldwide, affecting an estimated 60 to 70 million people. Inflammation and oxidative stress may be a modifiable risk factor in the management of patients with XFS and XFG.
An IOP elevation is considered a major risk factor for glaucoma, but an elevated IOP is not associated with glaucomatous characteristics in all glaucoma patients. Other possible pathogenetic factors, such as autoimmune mechanisms including apoptosis, may be involved in some patients with glaucoma. Wax and associates were the first to report an elevation of antibody titers in patients with NTG (eg, an increase in the level of heat shock protein 60 [HSP60] antibodies) and also higher levels of antibodies against small HSPs (eg, [alpha] A-crystalline, [alpha] B-crystalline, and HSP27) in NTG patients. A number of other autoantibodies against retinal or optic nerve proteins have been identified in many NTG patients. Because some glaucoma patients have increased titers of serum antibodies against these proteins, the degeneration of the RGCs in glaucoma may be attributable to a failure of immune regulation of both pro-apoptotic and protective pathways.
The Toll-like receptor ( TLR ) family, an anchor of innate immunity system, recognizes external ligands and differentiates self from nonself proteins. The ability of a tissue to recognize pathogens is mediated by a set of receptors that are referred to as pattern-recognition receptors (PRRs). To date, 13 members of the TLR family have been identified in mammals. TLR4 is a transmembrane receptor that mediates immune responses to exogenous and endogenous ligands, and not only recognizes bacterial lipopolysaccharides (LPSs) but is also activated by endogenous ligands such as heat shock proteins (HSPs). Toll-like receptors (TLRs) can also recognize endogenous ligands that are induced during inflammatory responses. Recently, the TLR4 (OMIM 603030 ) gene was implicated in NTG in the Japanese population, but not in the South Korean population.
Glaucoma is a neurodegenerative disease, but the mechanisms causing the RGC loss are still undetermined. Several studies have pointed to a possible involvement of autoimmune mechanisms in the pathogenesis of glaucoma, especially NTG. On the other hand, it is believed that the mechanisms shift from predominantly elevated IOP in the POAG and XFG to other factors such as autoimmune reactions in NTG.
Thus, the purpose of this study was to determine whether mutations in the TLR4 gene contributed to POAG, NTG, and XFG in unrelated Japanese patients.
Patients and Methods
One hundred eighty-four unrelated Japanese patients with POAG (119 men and 65 women; mean age 64.6 ± 14.3 years), 365 unrelated Japanese patients with NTG (171 men and 194 women; mean age 58.6 ± 13.1 years), and 109 unrelated Japanese patients with XFG (57 men and 52 women; mean age 77.6 ± 6.2 years) were studied. They were diagnosed with glaucoma in the ophthalmological clinic of the Tohoku University Hospital, Sendai; Niigata University Hospital, Niigata; Tokyo Metropolitan Police Hospital, Tokyo; Ideta Eye Hospital, Kumamoto; and Ehime University Hospital, Ehime, Japan. All of the subjects were enrolled from 2004 through 2010.
Routine ophthalmic examinations were performed on all patients. The criteria for classifying a patient as having POAG were: applanation IOP >22 mm Hg in each eye; glaucomatous cupping including cup-to-disc ratio >0.7 in each eye; visual field defects determined by Goldmann perimetry and/or Humphrey visual field analysis consistent with the glaucomatous cupping in at least 1 eye; and an open anterior chamber angle. Patients with glaucoma of secondary causes (eg, trauma-, uveitis-, or steroid-induced) were excluded. The criteria for NTG were applanation IOP <22 mm Hg in both eyes at each examination and the same characteristics as that of the POAG group. The IOP used for the statistical analyses was the clinic-based value. We checked the IOP in at least 3 visits and the measurements were made during the daylight hours. Patients were excluded if the IOP was 22 mm Hg or more for any of the measurements. The criteria for XFG were an open anterior chamber angle with accumulation of abnormal fibrillar material in the anterior segment of the eye and the same characteristics as the POAG group.
The control subjects (116 men and 100 women; age, 69.7 ± 11.3 years) had the following characteristics: IOP <22 mm Hg, normal optic discs, and no family history of glaucoma. To decrease the chance of studying individuals with presymptomatic glaucoma, we studied individuals who were older than 60 years in this group.
Sample Preparation and Mutation Screening
Genomic DNA was extracted from leukocytes of peripheral blood and purified with the Qiagen QIAamp DNA Blood Kit (Qiagen, Valencia, California, USA). Eight SNPs were amplified by polymerase chain reaction (PCR) using 0.5 μM intronic primers, 0.2 mM dNTPs, and 0.5 U Ex Taq polymerase (Takara, Shiga, Japan) with 30 ng template DNA in the amplification mixture (25 μL). The annealing temperature and sequence of primer set are given in the Supplemental Table (available at AJO.com ).
|Forward Primer||Reverse Primer||Annealing Temperature(C)|
Oligonucleotides for the amplification and sequencing were selected using Primer3 software ( frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi/ , provided in the public domain by the Massachusetts Institute of Technology, Cambridge, Massachusetts, USA). The PCR fragments were purified with ExoSAP-IT (USB, Cleveland, Ohio, USA), sequenced by the BigDye Terminator v3.1 Cycle Sequencing Kit (Perkin-Elmer, Foster City, California, USA) on an automated DNA sequencer (ABI PRISM 3100 Genetic Analyzer, Perkin-Elmer).
Differences in the genotype frequencies among the cases and controls were tested by Fisher exact test or χ depending on the cell counts. The inferred haplotypes and LD (linkage dysequilibrium), expressed as D ′, quantified between all pairs of biallelic loci, were estimated using the SNPAlyze program version 5.0.3 (Dynacom, Yokohama, Japan). The significance of an association was determined by contingency table analysis using χ or Fisher exact tests. The Hardy-Weinberg equilibrium was analyzed using gene frequencies obtained by simple gene counting and the χ test with Yates’ correction for comparing observed and expected values.
AII of the 8 SNPs in the TLR4 gene were genotyped, and all were in Hardy-Weinberg equilibrium in the glaucoma cases and control subjects. All SNPs were located in 1 haplotype block, and the magnitude of the LD between each SNP was very high, with a pairwise D ′ of more than 0.90. However, rs11536889 had a pairwise D ′ less than 0.80.
Allele and Genotype Frequencies in TLR4 Variants Detected in Subjects
The allele frequencies of the 8 SNPs in the glaucoma cases and control subjects are shown in Table 1 . The frequencies of the minor alleles of all SNPs were higher in the glaucoma cases than in control subjects. In the POAG subjects, the allele frequencies of 6 SNPs (rs10759930, rs1927914, rs1927911, rs12377632, rs2149356, and rs7037117) were significantly different from the control group ( P < .05). In addition, 5 SNPs (rs10759930, rs1927914, rs1927911, rs2149356, and rs7037117) in NTG subjects and 4 SNPs (rs1927914, rs1927911, rs12377632, and rs2149356) in XFG subjects were significantly different from that in the control group ( P < .05; Table 1 ). Three SNPs, rs1927914, rs1927911, and rs2149356, were identical for the POAG, NTG, and XFG groups. Among these 3 SNPs, the minor allele of rs2149356, located in intron 2 of TLR4 , conferred the highest increased risk of POAG ( P = .000058, OR = 1.77, 95% CI = 1.31–2.39), NTG ( P = .0030, OR = 1.51, 95% CI = 1.17–1.95), and XFG ( P = .015, OR = 1.56, 95% CI = 1.11–2.20).
|SNP||This Study||Previous Study|
|POAG (n = 184)||P Value||NTG (n = 365)||P Value||XFG (n = 109)||P Value||Control (n = 216)||NTG (n = 250)||Control (n = 318)||P Value|
The genotype frequencies of 8 SNPs are shown in Table 2 . The genotype frequency of 5 SNPs was significantly higher in the POAG and NTG subjects than in the controls, and none of the SNPs was significantly higher in the XFG subjects than in the control group ( P = .16, P = .059, P = .080, P = .13, P = .062, P = .95, P = .12, P = .69, respectively; χ test). Considering the dominant model, 4 SNPs in the XFG group were significant compared with the genotype frequencies of the control group. In POAG, NTG and XFG individuals bearing the minor allele of rs2149356 had the most significantly increased risk for glaucoma over that of control subjects ( P = .00014, P = .015, P = .062, respectively).
|This Study||Previous Study|
|POAG (n = 184)||NTG (n = 365)||XFG (n = 109)||Control (n = 216)||NTG (n = 250)||Control (n = 318)|
|T/T||49 (26.6%)||141 (38.6%)||40 (36.7%)||103 (47.7%)||81 (32.4%)||137 (43.1)|
|T/C||103 (56.0%)||159 (43.6%)||50 (45.9%)||85 (39.4%)||127 (50.8%)||141 (44.3%)|
|C/C||32 (17.4%)||65 (17.8%)||19 (17.4%)||28 (12.9%)||42 (16.8%)||40 (12.6%)|
|P value b||.000085||.074||.16||.028|
|P value c (dominant)||.000015||.032||.060|
|A/A||47 (25.5%)||137 (37.5%)||38 (34.9%)||105 (48.6%)||82 (32.8%)||137 (43.1%)|
|A/G||106 (57.6%)||164 (44.9%)||51 (46.8%)||82 (38.0%)||126 (50.4%)||141 (44.3%)|
|G/G||31 (16.9%)||64 (17.5%)||20 (18.3%)||29 (13.4%)||42 (16.8%)||40 (12.6%)|
|P value b||.000011||.030||.059||.036|
|P value c (dominant)||.0000022||.0089||.018|
|G/G||51 (27.7%)||139 (38.1%)||40 (36.7%)||106 (49.1%)||87 (34.8%)||141 (44.3%)|
|G/A||101 (54.9%)||166 (45.5%)||50 (45.9%)||85 (39.4%)||122 (48.8%)||135 (42.5%)|
|A/A||32 (17.4%)||60 (16.4%)||19 (17.4%)||25 (11.5%)||41 (16.4%)||42 (13.2%)|
|P value b||.000072||.027||.080||.067|
|P value c (dominant)||.000013||.0095||.034|
|C/C||53 (28.8%)||137 (37.5%)||41 (37.6%)||104 (48.1%)||86 (34.4%)||140 (44.0%)|
|C/T||90 (48.9%)||190 (52.1%)||49 (45.0%)||87 (40.3%)||122 (48.8%)||138 (43.4%)|
|T/T||41 (22.3%)||38 (10.4%)||19 (17.4%)||25 (11.6%)||42 (16.8%)||40 (12.6%)|
|P value b||.00012||.020||.13||.053|
|P value c (dominant)||.000079||.012||.071|
|G/G||53 (28.8%)||139 (38.1%)||40 (36.7%)||107 (49.5%)||87 (34.8%)||140 (44.0%)|
|G/T||98 (53.3%)||164 (44.9%)||50 (45.9%)||85 (39.4%)||122 (48.8%)||138 (43.4%)|
|T/T||33 (17.9%)||62 (17.0%)||19 (17.4%)||24 (11.1%)||41 (16.4%)||40 (12.6%)|
|P value b||.00012||.015||.062||.070|
|P value c (dominant)||.000025||.0069||.028|
|G/G||95 (51.6%)||196 (53.7%)||62 (56.9%)||127 (58.8%)||146 (58.4%)||177 (55.6%)|
|G/C||83 (45.1%)||145 (39.7%)||40 (36.7%)||76 (35.2%)||93 (37.2%)||119 (37.4%)|
|C/C||6 (3.3%)||24 (6.6%)||7 (6.4%)||13 (6.0%)||11 (4.4%)||22 (6.9%)|
|P value b||.083||.49||.95||.42|
|P value c (dominant)||.15||.23||.74|
|A/A||111 (60.3%)||222 (60.8%)||65 (59.6%)||153 (70.8%)||138 (55.2%)||213 (67.0%)|
|A/G||64 (34.8%)||125 (34.2%)||39 (35.8%)||54 (25.0%)||98 (39.2%)||94 (29.6%)|
|G/G||9 (4.9%)||18 (4.9%)||5 (4.6%)||9 (4.2%)||14 (5.6%)||11 (3.5%)|
|P value b||.082||.049||.12||.015|
|P value c (dominant)||.027||.015||.043|
|A/A||152 (82.6%)||313 (85.8%)||93 (85.3%)||191 (88.4%)||203 (81.2%)||269 (84.6%)|
|A/G||31 (16.8%)||50 (13.7%)||15 (13.8%)||24 (11.1%)||45 (18.0%)||49 (15.4%)|
|G/G||1 (0.6%)||2 (0.5%)||1 (0.9%)||1 (0.5%)||2 (0.8%)||0 (0.0%)|
|P value b||.25||.66||.69||.19|
|P value c (dominant)||.097||.36||.43|