Whole exome sequencing analysis identifies genes and pathways in sporadic early‐onset Alzheimer’s disease

Abstract

Background Sporadic early‐onset Alzheimer’s disease (EOAD) patients (onset age < 65 years) do not harbor a known causative variant and have a heritability estimated to be 92‐100% (Barber et al. 2016). Here we performed whole exome sequencing (WES) to identify functional exonic variants in sporadic EOAD. Method DNA was extracted and sequenced from blood samples of 109 Korean patients with EOAD or Mild Cognitive Impairment (MCI) younger than 65 years of age. Short‐read sequences from WES were mapped to the NCBI reference human genome (build 37) using BWA and GATK. All the variants were annotated with ANNOVAR. PolyPhen‐2 and SIFT were used to predict potential impact on protein structure or function of missense variants. After quality control procedures for samples and variants, we performed a case‐control association using PLINK on 109 Korean patients and 405 East Asia population from the 1000 Genomes Project. Result The genetic association analysis in a case‐control study design identified several sporadic EOAD susceptibility genes with missense variants with minor allele frequency (MAF) greater than 1%. A common exonic variant in LILRB2 had a significant p‐value of 1.036 × 10e ‐11 with MAF 0.11 compared with MAF = 0.395 in East Asian and MAF = 0.160 in cognitively normal Korean population (source: KRGDB genome browser). A common exonic variant in HLA‐DQB2 was also identified with a p‐value of 2.252 × 10e ‐08 with MAF 0.197 compared with MAF = 0.413 in East Asian and MAF = 0.34 in KRGDB. A common exonic variant in LIPC showed a marginal significance with a p‐value of 4.537 × 10e ‐05 with MAF = 0.0688 compared to MAF = 0.208 in East Asian and MAF = 0.12 in KRGDB. Other common exonic variants were found in TNR, ITGAX (p‐values < 5.00xe ‐056. The protein‐protein interaction network analysis using STRING showed that LILRB2 has a direct interaction with ADAM10 and ITGAX, HLA‐DQB2 has a direct interaction with ITGAX that has a direct interaction with ADAM10, and LIPC has interactions with APP, SORL1 and CLU, which are known AD risk genes. Conclusion Whole‐exome sequencing analysis identified exonic variants associated with sporadic EOAD. Studying these genes and their pathways might lead us to a better understanding of the genetic architecture of sporadic EOAD.