Eye (2015) 29, 691–698 © 2015 Macmillan Publishers Limited All rights reserved 0950-222X/15 www.nature.com/eye 1,2,3,5 1,2,3,5 4,5 Gene–gene interaction L Huang , Q Meng , C Zhang , LABORATORY STUDY Y Sun1,2,3, Y Bai1,2,3,SLi1,2,3, X Deng1,2,3, of CFH, ARMS2, and B Wang1,2,3,WYu1,2,3, M Zhao1,2,3 and X Li1,2,3 ARMS2/HTRA1 on the risk of neovascular age-related macular degeneration and polypoidal choroidal vasculopathy in Chinese population Abstract rs1065489 did not show significant association with nAMD (P40.01), but was strongly Purpose To evaluate the association and associated with PCV in Chinese patients 1Department of interaction of five single-nucleotide (Po0.001). Ophthalmology, Peking polymorphisms (SNPs) in three genes (CFH, University People’s Conclusions In this study, we found that the ARMS2, and ARMS2/HTRA1) with Hospital, Beijing, China interaction of ARMS2 and ARMS2/HTRA1 is neovascular age-related macular degeneration significantly associated with nAMD, and the 2 (nAMD) and polypoidal choroidal Key Laboratory of Vision interaction of CFH and ARMS2 is pronounced vasculopathy (PCV) in Chinese population. Loss and Restoration, in PCV development in Chinese population. Ministry of Education, Methods A total of 300 nAMD and 300 PCV Eye 29, – Beijing, China patients and 301 normal subjects participated (2015) 691 698; doi:10.1038/eye.2015.32; published online 13 March 2015 in the present study. The allelic variants of 3Department of Abdominal rs800292, rs2274700, rs3750847, rs3793917, and Surgical Oncology, Cancer rs1065489 were determined by matrix-assisted Institute and Hospital, Introduction laser desorption/ionization time-of-flight mass Chinese Academy of Medical Sciences, Peking – Age-related macular degeneration (AMD) is the spectrometry (MALDI-TOF-MS). Gene gene Union Medical College, interactions were evaluated by the data most common cause of legal blindness in the Beijing, China mining approach multifactor-dimensionality elderly in developed countries. Advanced AMD, reduction (MDR) method. including neovascular AMD (wet AMD, or 4Department of Clinical Results The risk alleles of CFH rs800292, nAMD) and geographic atrophy (dry AMD), is Epidemiology, Peking ’ rs2274700, ARMS2 rs3057847, and ARMS2/ associated with substantial, progressive visual University People s Hospital, Beijing, China HTRA1 rs3793917 showed significant impairment.1 AMD is characterized as a chronic and progressive degeneration of photoreceptors, difference between nAMD or PCV patients 5These authors contributed and controls (all Po0.01). The homozygosity the underlying retinal pigment epithelium, equally to this work. of risk alleles for rs800292, rs2274700, Bruch’s membrane, and potentially, the rs3750847, and rs3793917 were significantly choriocapillaris in the macula.2 Polypoidal Correspondence: different between nAMD patients and choroidal vasculopathy (PCV) is associated with X Li, Department of o a decrease in vision in the elderly Asian Ophthalmology, Peking controls (all P 0.01), and predisposed to PCV ’ o University People s Hospital, patients (all P 0.01). After cross-validation population, and is characterized by a network of Xizhimen South Street 11, consistency (CVC) and permutation tests, the vessels with two distinct components: a complex Beijing 100044, China two-locus model rs2274700_rs3750847 has a of branching vessels and multiple, terminal, Tel/Fax: +86 010 88325413. – balanced accuracy of 64.37% in predicting reddish-orange polypoidal lesions.3 5 PCV has E-mail: [email protected] nAMD disease risk. The one-marker model, been described as a distinct clinical entity from rs3750847, and two-locus model AMD and the other diseases associated with Received: 21 April 2014 6 Accepted in revised form: rs2274700_rs3750847 has a balanced accuracy subretinal neovascularization. Nevertheless, 3 February 2015 of 66.07% and 65.89% in predicting PCV whether PCV represents a subtype of AMD Published online: disease risk, respectively. Furthermore, CFH remains controversial.7 Evidence suggests that 13 March 2015 Gene–gene interaction of CFH, ARMS2, and ARMS2/HTRA1 L Huang et al 692 nAMD and PCV, despite their different phenotypic at the Department of Ophthalmology in the Peking manifestations, may share common genetic risk University People’s Hospital, and the study was – factors.8 12 Susceptibility to nAMD is complex, involving approved by the Ethical Committee of Peking University genetic, lifestyle, and environmental factors, although the People’s Hospital. An informed consent process was specifics of the etiology remain unresolved. established following the guidelines of the Helsinki Complement factor H (CFH) is a major inhibitor of the Declaration, and consent forms were signed by all alternative complement pathway and is found within subjects. All subjects received a comprehensive drusen,13 which is an important sign in early AMD. CFH ophthalmic examination, including visual acuity has been identified by various studies as a major AMD measurements, slit-lamp biomicroscopy, and dilated susceptible gene in the Chinese, Japanese, and Caucasian fundus examination performed by a retinal specialist. – populations.14 17 Previous studies have proved that All cases with nAMD and PCV underwent fluorescein multiple variants of CFH gene, such as I62V (rs800292), angiography, optic coherence tomography (OCT), and are associated with the risk of nAMD in different ethnic indocyanine green angiograms with HRA2 (Heidelberg – groups.18 21 Engineering, Heidelberg, Germany). The diagnosis of Beyond the complement pathway, the Age-Related nAMD or ARM was defined by the International Maculopathy Susceptibility 2 (ARMS2) locus at Classification System for ARM.29 The diagnosis of PCV chromosome 10q26 has been implicated as another major was based on indocyanine green angiography (ICGA) genetic contributor to the nAMD disease process. ARMS2 results that showed a branching vascular network rs3750847 has been demonstrated to have a strong terminating in aneurysmal enlargements, which typify association with both nAMD and PCV in Chinese polypoidal lesions. Exclusion criteria included any eye population.22 HTRA1 (High-Temperature Requirement with any other macular abnormalities, such as pathologic factor A 1), another major locus at chromosome 10q26, myopia, idiopathic choroidal neovascularization (CNV), was recently shown to have an important role in the presumed ocular histoplasmosis, angioid streaks, and any fi pathogenesis of nAMD or PCV.23 Previous studies proved other secondary CNV. Normal controls were de ned as that gene variants like rs3793917 localized to the having no clinical evidence of nAMD or PCV in either eye intergenic region between ARMS2 and HTRA1 were or any other eye diseases, excluding mild age-related high-risk factors for nAMD.24,25 cataracts. Subjects with severe cataracts were excluded To identify multilocus interactions in the pathogenesis from the study. of nAMD in Chinese population, we applied the multifactor-dimensionality reduction (MDR) analysis, Genetic analysis which has been used to characterize gene–gene – interactions in nAMD.26 28 Blood samples were collected from all participants and − In this study, we aimed to evaluate the association of stored at 80 °C before DNA was extracted. Genomic three genes (CFH, ARMS2, and ARMS2/HTRA1) with DNA was extracted from venous blood leukocytes using nAMD and PCV, and to determine their possibility as a genomic extraction kit (Beijing eBios Biotechnology, biomarkers for genetic factors predisposing to nAMD and Beijing, China), and genotyping was performed with matrix-assisted laser desorption/ionization time-of-flight PCV in Chinese population. We analyzed the genetic mass spectrometry, as previously described.30 Briefly, association of the reported major risk SNPs (rs800292, ~ 30 ng of genomic DNA was used to genotype each rs2274700, rs3750847, rs3793917, and rs1065489) in CFH, sample. The DNA samples were amplified, and the PCR ARMS2, and ARMS2/HTRA1 genes, and examined gene– products were used for locus-specific single-base gene interactions using the MDR method in unrelated extension reactions. The resulting products were desalted Chinese patients with nAMD and PCV. and transferred to a 384 SpectroCHIP array (Sequenom, San Diego, CA, USA). Allele detection was performed Methods using matrix-assisted laser desorption/ionization time-of- flight mass spectrometry. The mass spectrograms were Subjects analyzed using MassARRAY Typer software version 4.0 A total of 901 unrelated Chinese subjects were studied in (Sequenom, San Diego, CA, USA). this case–control cohort. Three hundred patients had nAMD, and three hundred patients had PCV. Three Statistical analysis hundred and one individuals without age-related maculopathy (ARM) were studied as controls. The The data were analyzed using SPSS (version 16.0; SPSS genders and ages of the controls and cases are given in Science, Chicago, IL, USA). All identified polymorphisms Supplementary 1. The study participants were recruited were assessed for Hardy–Weinberg equilibrium using Eye Gene–gene interaction of CFH, ARMS2, and ARMS2/HTRA1 L Huang et al 693 χ2-tests. We also used χ2-test to determine significant SNPs showed no significant deviation from Hardy– difference in risk allele of SNPs, and single-marker Weinberg equilibrium in the control group (P40.05), association analysis was performed using logistic except rs3750847 and rs3793917 in the nAMD group regression incorporating an additive genetic model as (P = 0.03, P = 0.01,respectively; shown