Long-read sequencing reveals two common APOE ϵ3 and ϵ4 intragenic haplotypes in the Spanish population

Abstract

The APOE gene is a key genetic determinant of Alzheimers disease (AD) risk. While the pathogenic effects of the {varepsilon}4 allele are well established, the contribution of distinct haplotypic backgrounds within {varepsilon}3 and {varepsilon}4 remains poorly understood. Here, we used long-read Oxford Nanopore sequencing to identify and phase single nucleotide variants (SNVs) across a 3.9 kb region spanning APOE locus (3 to 5 UTR) in a large memory-clinic based cohort (N=1,266). We identified 48 SNVs, including nine novel variants, and reconstructed haplotypes through direct phasing. Our analysis revealed a structured haplotypic architecture, with two major configurations for {varepsilon}3 ({varepsilon}3a and {varepsilon}3b) and {varepsilon}4 ({varepsilon}4a and {varepsilon}4b), defined by the promoter variant rs405509 in phase with the canonical isoform-defining variants rs429358 and rs7412. The observed homoplasy in the rs405509, likely arising from a recombination or gene conversion event, complicates phasing by standard genetic approaches, highlighting the value of long-read sequencing for accurate characterization of APOE haplotypic diversity. Importantly, the {varepsilon}4a haplogenotype was associated with reduced risk of progression from mild cognitive impairment to AD dementia compared to {varepsilon}4b (APOE-stratified meta-analysis HR[95%CI]=0.564[0.400-0.794], p=0.001), while {varepsilon}3b showed a borderline protective effect towards AD biomarker positivity in {varepsilon}3{varepsilon}3 subjects (OR[95%CI]=0.76[0.57-1.00], p=0.053). Notably, haplogenotypes had no detectable effect on CSF APOE levels, indicating that haplotype-dependent effects are not mediated by differences in total APOE abundance. This study provides new insights into the intragenic allelic variability of the APOE gene, highlighting the importance of considering haplotype-specific effects when interpreting the functional impact of APOE and in designing targeted therapeutic strategies. Further research is needed to explore the broader regulatory network of the APOE locus and its interaction with neighboring loci in the 19q13 region.