Genome-wide association studies (GWAS) have identified more than thirty loci associated with Alzheimer's disease (AD), but the causal variants, regulatory elements, genes and pathways remain largely unknown thus impeding a mechanistic understanding of AD pathogenesis. Previously, we showed that AD risk alleles are enriched in myeloid-specific epigenomic annotations. Here, we show that they are specifically enriched in active enhancers of monocytes, macrophages and microglia. We integrated AD GWAS signals with myeloid epigenomic and transcriptomic datasets using novel analytical approaches to link myeloid enhancer activity to target gene expression regulation and AD risk modification. We nominate candidate AD risk enhancers and identify their target causal genes (including AP4E1, AP4M1 , APBB3 , BIN1 , CD2AP , MS4A4A , MS4A6A , PILRA , RABEP1 , SPI1 , SPPL2A , TP53INP1 , ZKSCAN1 , and ZYX ) in sixteen loci. Fine-mapping of these enhancers nominates candidate functional variants that likely modify disease susceptibility by regulating causal gene expression in myeloid cells. In the MS4A locus we identified a single candidate functional variant and validated it experimentally in human induced pluripotent stem cell (hiPSC)-derived microglia. Combined, these results strongly implicate dysfunction of the myeloid endolysosomal system in the etiology of AD. causal gene expression in eight loci, and experimentally validated one of these candidate causal variants in the MS4A locus in human induced pluripotent stem cell (hiPSC)-derived microglia.
Results
AD risk alleles are specifically enriched in active enhancers of monocytes, macrophages and microgliaOur earlier analyses showed a significant enrichment of AD risk alleles in various myeloid-specific epigenomic annotations, but not in brain or other tissues 15 . To further dissect this enrichment, we used ChIP-Seq profiles of histone modifications that define the chromatin signatures of regulatory elements (H3K27ac for active enhancers and promoters, H3K4me1 for enhancers, and H3K4me2 for active enhancers and promoters) from monocytes, macrophages and microglia to annotate the genome with myeloid active enhancers (AE), active promoters (AP), primed enhancers (PE) and primed promoters (PP) (see Methods) [18][19][20] . To identify which of these myeloid regulatory elements are enriched for AD risk alleles, we performed stratified LD score regression (LDSC) 21 of AD single nucleotide polymorphism (SNP) heritability partitioned by the aforementioned epigenomic annotations using the International Genomics of Alzheimer's Project (IGAP) AD GWAS dataset 22 . This analysis revealed selective enrichment of AD risk alleles in active enhancers of monocytes, macrophages and microglia (Figure 1a). In contrast, schizophrenia SNP heritability (using the Psychiatric Genomics Consortium SCZ GWAS dataset as control 23 ) was not enriched in any of these myeloid regulatory elements (Figure 1a). To identify TFs that likely regulate the activity of myeloid enhancers, we performed de novo motif an...