Two <i>APOE</i> splice sQTLs reduce Alzheimer’s disease risk in <i>APOE</i> 4/4 carriers

Abstract

Abstract Background Are there genetic variants in the APOE locus that alter APOE 4‐related risk for AD? Do these putative variants exert their effect through altering availability of the APOE protein? With these questions in mind, we investigated whether the APOE splice quantitative trait loci (sQTL; variants that alter post‐transcriptional splicing) affect risk for AD, age‐at‐AD‐onset, and APOE protein levels, in analyses stratified to APOE 2/4, 3/4 and 4/4 carriers, respectively. Method Genetic and phenotypic data was available from 35,760 eligible participants across twenty‐two independent AD‐related studies. Genetic data underwent standard quality control (Plink v1.9) and ancestry determination (SNPweights v.2.1). Included participants were ages 60+, non‐Hispanic, of European ancestry, had APOE genotype available, and were diagnosed as cases or controls (Table 1). From 17 APOE sQTL variants reported in GTEx, only rs157580 and rs439401 were genotyped in &gt;80% of the available sample and further investigated (Figure 1). The role of APOE sQTL SNPs on AD risk was evaluated through logistic regression analyses under a case‐control design, and on age‐at‐onset through multiple linear regression. Associations with APOE protein levels were evaluated in post‐mortem tissue from the dorsolateral prefrontal cortex (dlPFC, n = 655), in which rs157580 alters excision rates of an intron on the APOE ‐203 transcript variant. Given the relative paucity of subjects carrying APOE 2/4‐4/4 (n = 11‐11), protein level analyses were restricted to APOE 3/4 carriers (n = 161). Study design and model details/covariates are shown in Figure 1. Result rs157580 and rs439401 reduced AD risk in APOE 4/4 carriers (Table 2). rs157580 also showed a trending effect to increase age‐at‐AD‐onset in both APOE 4/4 and APOE 3/4 carriers (Table 3). Finally, rs157580 increased APOE protein levels in the dlPFC of APOE 3/4 carriers (Figure 2). Supplemental analyses suggest that the sQTLs and APOE 4 must be in phase in order to confer protection. Conclusion Our results bring new insights into (genetic) mechanisms that alter APOE expression and how this affects APOE 4‐related AD risk. This is crucially relevant for the development of AD drug therapies aimed at APOE 4 and genetic risk counseling. Future work is needed to more precisely evaluate how haplotypes of the common APOE* 2/3/4 polymorphisms and sQTLs affect AD risk.

Topics & Concepts