In systemic biology one uses high throughput data from various study types to derive information on better ways to diagnose, treat, and prevent complex diseases like exfoliation syndrome (XFS) (Figure). Here we focus on how analytical epidemiological studies contribute to an improved understanding of XFS and its attendant unwanted consequences (such as glaucoma,1 cataract,2 retinal vein occlusion,3 etc.). Specifically, we concentrate on how the environment may modify the association between genotypes and exfoliation syndrome (XFS). The discovery of gene-environment (GxE) interactions for XFS may form the basis for novel XFS primary prevention strategies. It should be stated at the outset that GxE interactions for XFS have not been confirmed but several are suspected. The discovery of the robust association between LOXL1 genotypes and XFS4 has led directly to the discovery of environmental risk factors or the disease, setting the stage to assess GxE interactions for this condition. This review will cover how the association between LOXL1 genotypes and XFS was discovered and confirmed. The impact of the geographical distribution of LOXL1 gene variants on the identification of environmental risk factors will be reviewed. The strongest candidate GxE interactions for XFS will be presented, and the implications these have for understanding the disease and for targeted strategies to reduce disease burden will be described. Finally, the necessary steps required for testing putative interactions between LOXL1 gene variants and environmental factors will be discussed. Figure A systems biology approach is required to assimilate how genes and environment contribute to the formation of exfoliation material in the human eye as illustrated in the clinical photograph with the white arrow that points to a sheet of exfoliation material ... The discovery of LOXL1 variants in relation to exfoliation syndrome XFS is a strongly age-related disease,5 making it difficult to collect multigenerational families with the disorder, although some evidence of familial aggregation has been reported.6 Using a large Finnish family for genome wide linkage study, several genomic regions were suggested to contain genes contributing to XFS.7 However, further study has not identified disease-associated genes in these regions, suggesting loci in these regions are not common causes of XFS. Candidate gene studies using selected genes of interest in case control studies of unrelated individuals have been performed. However, these studies have been mostly underpowered and the results have been inconsistently replicated across populations.8 The results from Phase I of The Human Genome Project suggested that an agnostic search for gene variants related to any trait is possible.9 First, the human genome contains DNA blocks that tend to travel together during cell reproduction. The size of such blocks differs by ancestry. Second, roughly every 300th base, there is a nucleotide change that occurs commonly (>5% of the time) in the general population, referred to as a common single nucleotide polymorphism (SNP). These realizations combined with improvements in microfabrication and nucleotide chemistry assays allowed for the creation of technologies that can simultaneously genotype several hundred thousand common polymorphic sites strategically located throughout the genome. Thus, these genotyping technologies enabled genome wide association studies (GWAS), which are agnostic searches for common polymorphisms that have associations with complex diseases, like XFS. An adequately powered GWAS using a case control design performed in Iceland demonstrated that LOXL1 variants were associated with XFS.4 The researchers used a simple chi square test to compare the genotype frequency between cases and controls at 304,250 loci throughout the genome. The association between each marker in cases and control was adjusted for inherent population structure that might confound the relation between a genotype and affected status. The typical p-value for statistical significance (p=0.05) was adjusted to 1.6 × 10−7 using the Bonferroni method by dividing 0.05 by the number of tests performed (304,250). Using this approach, polymorphic variants located in the genomic region that includes LOXL1 were found to be statistically associated with XFS. These results were confirmed using a candidate approach in another population of subjects as part of the same publication. There were two rather remarkable features of the XFS GWAS performed by the Icelandic group. First, the effect size for the rs3825942 LOXL1 variant found in association with XFS was among the highest seen in a GWAS for common complex disease (odds ratio = 20.1); typically, odds ratios for SNPs in association with common complex diseases are in the 1.2-1.5 range and rarely are they above 2.0. This suggests that the rs3825942 allelic variant represents a locus of major functional significance in XFS. Second, while 99% of cases harbored the at-risk allele, 85% of controls also harbored this allele. This suggests that while the rs3825942 LOXL1 variant is necessary for disease, it was not sufficient to produce it. Additional factors that could produce XFS include other genes, environmental factors or epigenetic factors.
Read full abstract
Jun 4, 2024
Karen N Mcfarland + 14
Open Access