Multiple Gene Polymorphisms Associated with Exfoliation Syndrome in the Uygur Population

Background Our previous data suggested that three single-nucleotide polymorphisms (SNPs), rs1048661, rs3825942, and rs2165241, of the lysyl oxidase-like 1 gene (LOXL1) are significantly associated with exfoliation syndrome (XFS) and exfoliation glaucoma (XFG). The following study investigated other SNPs that potentially effect XFS/XFG. Methods A total of 216 Uygur patients diagnosed with XFS/XFG, and 297 Uygur volunteers were admitted to the First Affiliated Hospital at Xinjiang Medical University between January 2015 and October 2017. Blood samples were collected by venipuncture. Alleles and genotypes of LOXL1, TBC1D21, ATXN2, APOE, CLU, AFAP1, TXNRD2, CACNA1A, ABCA1, GAS7, and CNTNAP2 were analyzed by direct sequencing. Results The allele G of rs41435250 of LOXL1 was a risk allele for XFS/XFG (P < 0.001), whereas the allele G of rs893818 of LOXL1 was a protective allele for XFS/XFG (P < 0.001). After adjusting all data for age and gender, the following results were obtained: the frequency of genotype CC for rs7137828 of ATXN2 was significantly higher in XFS/XFG patients than in controls (P = 0.027), while no significance was found with reference to the frequency of genotype TT. The frequency of genotype GG for rs893818 of LOXL1 (P < 0.001) and the frequency of genotype AA were both significantly higher in XFS/XFG groups compared to the control group (P < 0.001). In addition, the frequency of genotype TT for rs41435250 of LOXL1 was higher in XFS/XFG patients than in controls (P = 0.003), while no significant difference was found with reference to the frequency of genotype GG after adjusting for age and gender. In addition, the haplotypes G-A/T-G/G-G for rs41435250 and rs893818 were significantly associated with XFS/G. Conclusions With reference to LOXL1, the rs41435250 resulted as a risk factor and rs893818 as a protective factor for XFS/XFG in the Uygur populations. Meanwhile, the rs16958445 of TBC1D21 and the rs7137828 of ATXN2 have also shown to be associated with pathogenesis of XFS/XFG.


Introduction
Exfoliation syndrome (XFS) is an age-related, systemic, elastic microfibrillopathy characterized by deposition and progressive accumulation of a white, fibrillary, extracellular material affecting intraocular and extraocular tissues [1]. A recent study has suggested a high prevalence of XFS in the Uygur population [2,3]. XFS is characterized by rapid progression, high resistance to medical therapy, and poor prognosis and may lead to exfoliation glaucoma (XFG), open-angle glaucoma, angle-closure glaucoma, and acceleration of cataract insensibly [4]. In China, especially in Xinjiang, many XFS/XFG patients lost their visual acuity due to the lack of medical treatment.

Ethical Approval.
e Ethical Committee of the First Affiliated Hospital of Xinjiang Medical University, China, approved this study. In addition, the informed consent was obtained from all participants after explaining the objective and nature of the study. e study was conducted in accordance with the Declaration of Helsinki.

Study Population.
A total of 216 Uygur patients who were diagnosed as XFS/XFG and 297 normal Uygur volunteers who were admitted at the First Affiliated Hospital of Xinjiang Medical University, the First People's Hospital of Kashgar, and the Kuqa County Hospital between January 2015 and October 2017 were enrolled in this study. XFS was diagnosed based on the previously described approach [5]. In brief, XFS was diagnosed by exfoliation materials on the anterior lens capsule or on the pupil margin in either eye with dilation of the pupils. e inclusion criteria were the following: (1) IOP ≥22 mmHg in either eye; (2) glaucomatous changes on the optic disc, defined as a cup-to-disc ratio >0.7 in either eye or an asymmetric cup-to-disc ratio >0.2 or notching of the disc rim; and (3) characteristic glaucomatous visual field loss [14]. Patients with other causes of secondary glaucoma, such as uveitis, pigment dispersion syndrome, and iridocorneal endothelial syndrome, were excluded from the study. All study subjects were unrelated and received comprehensive ophthalmic examinations.
Peripheral blood samples (2-3 ml) were collected from each subject by venipuncture. Genomic deoxyribonucleic acid (DNA) was extracted from the whole blood using a Genomic DNA Extraction Kit ( e Beijing Genomics Institute, Beijing, China).
Genotypes of these SNPs were determined by direct DNA sequencing, using a BigDye Terminator v3.1 Kit (Applied Biosystems, Foster City, CA) in a 3730XL capillary sequencer (Applied Biosystems). e sequences were analyzed by sequencing analysis software Chromas (Technelysium Pty Ltd., Queensland, Australia).
Hardy-Weinberg equilibrium (HWE) analysis was tested by using the χ 2 test in SAS/Genetics v9.1 (SAS Institute Inc., Cary, NC, USA). e comparison of allelic and genotypic frequencies between the patient and control groups, as well as haplotype association analysis, was performed using a standard χ 2 test. A P value <0.05 was considered statistically significant. Relative risk association was estimated by calculating odds ratios (OR) along with 95% confidence intervals (CI).

Results
A total of 216 Uygur XFS/XFG patients (case group) and 297 normal Uygur volunteers (control group) were included in the study. In the case group, there were 146 males and 70 females (average age: 68 years), while in the control group, there were 159 males and 138 females (average age: 62 year) ( Table 1).
All SNPs underwent the Hardy-Weinberg equilibrium test before further data analysis. Besides rs7137828 that deviated from HWE (P � 0.006) in the control group and rs35934224 that deviated from HWE (P � 0.005) in the case group, other SNPs were all in line with the HWE Table 2.
e genotype association analysis showed that the frequency of genotype AA for rs16958445 of TBC1D21 was higher in XFS/XFG patients than in controls (P � 0.033, OR � 5.481, 95% CI: 1.151-26.11), while the frequency of genotype GG was not significantly different between the two groups. After adjusting all data for age and gender, the following results were obtained: the frequency of genotype CC for rs7137828 of ATXN2 was significantly higher in XFS/XFG patients than in controls (P � 0.027, OR � 0.322, 95% CI: 0.118-0.879), while no significant differences were found with reference to the frequency of genotype TT. e frequency of genotype GG for rs893818 of LOXL1 (P < 0.001, OR � 0.511, 95% CI: 0.358-0.729) and the frequency of genotype AA were both significantly higher in the XFS/XFG group compared to the control group (P < 0.001, OR � 0.095, 95% CI: 0.033-0.272). In addition, the frequency of genotype TT for rs41435250 of LOXL1 was higher in XFS/XFG patients than in controls (P � 0.003, OR � 3.902, 95% CI: 1.580-9.640), while no significant difference was found with reference to the frequency of genotype GG after adjusting for age and gender. All data are shown in Table 4.
Moreover, our results indicated that all MAFs were greater than 0.05, which further suggested that all SNPs were statistically significant (Table 5).
After the study of alleles and genotypes, we screened out the LOXL1, APOE, and CNTNAP2 for the haplotype association analysis. e genotyping graphs for these SNPs are shown in Figure 1.
For the rs41435250 and rs893818 of LOXL1, three haplotypes were observed. As shown in Table 6, all   2 Journal of Ophthalmology

Discussion
So far, numerous studies have focused on the polymorphisms of LOXL1. Our previous studies have shown that there were polymorphisms of LOXL1 in different alleles and genotypes of different SNPs in XFS/XFG of different ethnic groups. In this study, we found two SNPs (rs41435250 and rs893818) of LOXL1 that were polymorphic and associated with XFS/XFG. Meanwhile, we also examined other genes which were previously affirmed to have polymorphisms in XFS/XFG. We found that rs16958445 of TBC1D21 and rs7137828 of ATXN2 were significantly associated with XFS/XFG. Yet, three haplotypes for rs429358 and rs7412 of APOE and rs2107856 and rs2141388 of CNTNAP2 had no connection between the case and control group. As a result, LOXL1 is still the susceptibility gene of XFS/ XFG in Uygur populations. e rs1048661, rs3825942, rs2165241, rs4886467, rs4558370, rs4461027, rs4886761, rs16958477 [5,6], and rs41435250 resulted to be risk factors, while rs893818 resulted to be a protective factor for XFS/ XFG in the Uygur population.
Genes, such as TBC1D21, ATXN2, APOE, CLU, AFAP1, TXNRD2, CACNA1A, ABCA1, GAS7, and CNTNAP2, have been associated with glaucoma. In this study, we discovered that SNPs, rs16958445 of TBC1D21 and the rs7137828 of ATXN2, had an important role in the pathogenesis of XFS/ XFG in the Uygur population. Nonetheless, it is necessary to   note that there may be other factors affecting the pathogenesis of XFS/G, which should be addressed by future studies.
In this research, we gathered a number of genes to study the polymorphisms of the special ethnic groups, thus providing valuable information and expanding the knowledge on the gene mechanism of XFS/XFG. Nonetheless, the current study has some limitations that should be pointed out. Although the patients were recruited from the three largest areas of Xinjiang, the sample representativeness may be somewhat inaccurate, which could be addressed by expanding the sample size and thus improving the accuracy. We found that multiple gene polymorphisms had an important role in the pathogenesis of the disorder in Uygur patients, but we cannot exclude the possibility that other additional genetic or environmental factors also participate in modifying the development of this disorder.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.

Authors' Contributions
Yi-Nu Ma was in charge of statistical analysis and manuscript writing; Ting-Yu Xie was involved in the diagnosis and screening of patients; and Xue-Yi Chen was the instructor.