Cell type-specific histone acetylation profiling of Alzheimer’s Disease subjects and integration with genetics

Abstract

We profile genome-wide histone 3 lysine 27 acetylation (H3K27ac) of 3 major brain cell types from hippocampus and dorsolateral prefrontal cortex (dlPFC) of subjects with and without Alzheimer's Disease (AD). We confirm that single nucleotide polymorphisms (SNPs) associated with late onset AD (LOAD) prefer to reside in the microglial histone acetylome, which varies most strongly with age. We observe acetylation differences associated with AD pathology at 3,598 peaks, predominantly in an oligodendrocyte-enriched population. Strikingly, these differences occur at the promoters of known early onset AD (EOAD) risk genes (APP, PSEN1, PSEN2, BACE1), late onset AD (LOAD) risk genes (BIN1, PICALM, CLU, ADAM10, ADAMTS4, SORL1 and FERMT2), and putative enhancers annotated to other genes associated with AD pathology (MAPT). More broadly, acetylation differences in the oligodendrocyte-enriched population occur near genes in pathways for central nervous system myelination and oxidative phosphorylation. In most cases, these promoter acetylation differences are associated with differences in transcription in oligodendrocytes. Overall, we reveal deregulation of known and novel pathways in AD and highlight genomic regions as therapeutic targets in oligodendrocytes of hippocampus and dlPFC. INTRODUCTION:Alzheimer's Disease (AD) is the most common age-related neurodegenerative disorder 1 . The hallmarks of AD pathology are numerous and include neuronal loss, synaptic dysfunction, gliosis, and the accumulation of intercellular plaques of amyloid-β (Aβ) protein and intracellular neurofibrillary tangles (NFT) of phosphorylated tau protein (MAPT) 2 .Aβ plaques are formed by differential proteolytic cleavage of the amyloid β precursor protein (APP) 3-6 by the α-secretase, β-secretase and γ-secretase enzymes 7 . Studies of individuals affected by early onset (<60 yrs.) familial AD (EOAD) have identified causal autosomal dominant mutations primarily in Aβ processing proteins presenilin-1 (PSEN1) and presenilin-2 (PSEN2), which are part of the γ-secretase complex 8-10 , but also causal mutations or duplications in APP itself [11][12][13] . However, EOAD only accounts for a small minority of AD cases. Late onset sporadic AD (LOAD) is more frequent and accounts for up to 99% or more of AD cases. While increased age is the strongest risk factor and several environmental factors also confer risk for LOAD, its heritability has been estimated to be as high as 79% 14 .In contrast to EOAD, genetic risk for LOAD is less well understood. The ε4 allele comprising mutations in two codons in Apolipoprotein E (APOE) has been identified as the strongest genetic risk factor for LOAD [15][16][17][18][19] . More recently, genome wide association studies (GWAS) 20-27 have reproduced the APOE association and also identified 28 other unique loci harboring genetic variants which increase risk for developing LOAD 26-28 . Strikingly, from the set of most significant (or "sentinel") single nucleotide polymorphisms (SNPs) derived from GWAS and SNPs in strong linkage...

CHAS, a deconvolution tool, infers cell type-specific signatures in bulk brain histone acetylation studies of neurological and psychiatric disorders

“…MVB12B has previously been identified as an AD risk gene 20 and forms part of an oligodendrocyte-enriched gene network in the AD brain 21 . These results are consistent with a previous study in which oligodendrocytes were reported to show the most widespread acetylation differences in the AD brain 22 . The top-ranked AD-hypoacetylated peak was annotated to multiple cell types, and located near NKAIN3 ( Supplementary Table 1 ), an AD GWAS candidate gene that has been shown to be differentially expressed in astrocyte subclusters 23 .…”

“…Whereas epigenetic variation in late-stage AD predominantly points to oligodendrocytes, genetic risk points to microglia, suggesting independent biological mechanisms. Furthermore, our study corroborates the finding that genetic risk for AD is enriched at microglia-specific regulatory elements 4,22,48 and genes 35 . Additionally, we used CHAS to identify disease-associated pathways and transcription factors at cell type resolution.…”

“…Deconvolution of H3K27ac profiles from the AD brain using CHAS has highlighted that hypoacetylation in late-stage AD is enriched for both oligodendrocyte-specific H3K27ac and microglia-specific H3K27ac, whereas AD-associated hyperacetylation is only enriched for oligodendrocyte-specific H3K27ac. An independent study also found the largest H3K27ac changes in oligodendrocytes in the hippocampus and dorsolateral prefrontal cortex of individuals with AD 22 . Taken together, these data suggest that these oligodendrocyte-specific H3K27ac changes are not limited to a single brain region and warrant further investigation of the role of oligodendrocyte H3K27ac dysregulation in AD.…”

Addiction-Associated Genetic Variants Implicate Brain Cell Type- and Region-Specific Cis-Regulatory Elements in Addiction Neurobiology

“…The prioritized SNPs in the MAPT-CRHR1 locus are genome-wide significant for 5 of the 7 addiction-associated traits (Extended Data Fig. 7-1 ), and the locus has been implicated in other neuropsychiatric traits, such as Alzheimer's disease ( Hoffman et al, 2019 ; Corces et al, 2020 ; Ramamurthy et al, 2020 ). We provide the summary of CNN predictions in these reported loci across all 14,790 analyzed SNPs along with the accompanying annotations that we incorporated into our prioritization of candidate causal SNPs and their predicted cell types (Extended Data Fig.…”

Neurons burdened by DNA double strand breaks incite microglia activation through antiviral-like signaling in neurodegeneration

“…Substantial evidence supports a causal role for microglia and neuroinflammation in AD pathogenesis 29,60,61 . For example, many AD risk genes and genomic loci are most active in microglia 62–64 , and reactive microglia are thought to create a cytotoxic environment for neurons and exacerbate neurodegneration 65–68 . Here, we provide evidence that neurons participate in this inflammatory signaling as well.…”