Protective Effect of Factor B and Complement Component 2 in Age-Related Macular Degeneration

Abstract

Purpose:: Age-related macular degeneration (AMD) renders affected individuals unable to recognize faces, read, or drive, and may progress to legal blindness for the 7.5 million elderly Americans afflicted. Abnormal regulation of the complement system caused by the genetic variant Y402H in the complement factor H gene on chromosome (chr) 1 is a well-known risk factor for AMD, as is the rs10490924 SNP in the LOC387715 gene on chr 10. Recently, polymorphisms in the factor B (BF) and complement component 2 (C2) genes on chr 6 were associated with decreased susceptibility to AMD (Gold et al. 2006).

Methods:: To validate this association, we genotyped 2 SNPs in C2 and 4 SNPs in BF in two independent datasets: a family-based dataset with 451 individuals including 130 multiplex families and a case-control dataset of 492 cases and 208 controls. In the family-based dataset, we calculated multipoint nonparametric LOD scores and tested SNPs for association using Association in the Presence of Linkage (APL). In the case-control dataset, allelic association was assessed by a chi-square test, and odds ratios were estimated using logistic regression, after controlling for age, Y402H, rs10490924, and smoking.

Results:: No evidence for an effect of C2 or BF was found in the family-based dataset (max LOD score for the region 0.10, smallest APL p-value 0.21 at rs641153 in BF). In contrast, in the independent case-control dataset rs547154 in C2 and rs641153 were very strongly associated with decreased risk for AMD (p-value <10^-6 for both), with the minor alleles at these loci present in 4% of cases vs. 10% of controls. As these SNPs are in strong linkage disequilibrium (r² 0.95), we believe that these results represent the same protective signal. After controlling for known AMD risk factors, this effect remained strong in a reduced dataset of 311 cases and 176 controls with complete covariate data (OR 0.17, 95% confidence interval (0.08 to 0.35) p-value <10^-4).

Conclusions:: Though this association was not replicated in the family-based dataset, strong evidence for a protective effect was found in the case-control dataset. Future efforts will determine whether variants in C2 or BF or both genes are the source of the decreased AMD susceptibility.

Keywords: age-related macular degeneration • genetics • gene mapping