Single Cell-type Integrative Network Modeling Identified Novel Microglial-specific Targets for the Phagosome in Alzheimer’s disease

Abstract

Late-Onset Alzheimer's Disease (LOAD) results from a complex pathological process influenced by genetic variation, aging and environment factors. Genetic susceptibility factors indicate that myeloid cells such as microglia play a significant role in the onset of LOAD. Here, we developed a computational systems biology approach to construct probabilistic causal and predictive network models of genetic regulatory programs of microglial cells under LOAD diagnosis by integrating two independent brain transcriptome and genome-wide genotype datasets from the Religious Orders Study and Rush Memory and Aging Project (ROSMAP) and Mayo Clinic (MAYO) studies in the AMP-AD consortium. From this network model, we identified and replicated novel microglialspecific master regulators predicted to modulate network states associated with LOAD.We experimentally validated three microglial master regulators (FCER1G, HCK and LAPTM5) in primary human microglia-like cells (MDMi) by demonstrating the molecular impact these master regulators have on modulating downstream genomic targets identified by our top-down/bottom-up method and the causal relations among the three key drivers. These master regulators are involved in phagocytosis, a process associated with LOAD. Thus, we propose three new master regulator (key driver) genes that emerged from our network analyses as robust candidates for further evaluation in LOAD therapeutic development efforts.

Genome-wide meta-analysis, fine-mapping and integrative prioritization implicate new Alzheimer’s disease risk genes

“…Recently proposed AD candidate genes supported by our analyses include RIN3 , HS3ST1 , and FCER1G . As noted above, FCER1G is a microglial master regulator 81 – 83 ; RIN3 interacts with both BIN1 and CD2AP in the early endocytic pathway 93 ; HS3ST1 is involved in cellular uptake of tau 94 and was recently been associated with AD in an independent Norwegian sample 62 .…”

“…While our prioritization further supports many established AD candidate genes, it also implicates novel genes. Among these are FCER1G , which has been reported as a hub gene in microglial gene modules associated with neurodegeneration 81 , 82 , and has been experimentally shown to influence microglial phagocytosis 83 . Another candidate is ZYX , which receives a top network score, is highly expressed in microglia, and which was recently nominated as an AD risk gene based on chromatin interactions between the ZYX promoter and AD risk variants in a ZYX enhancer 84 .…”

Rejuvenation of the aged brain immune cell landscape in mice through p16-positive senescent cell clearance

“…Fifteen conserved genes increased with age across regions and sexes (Fig. 1b), and 13 of 15 (B2m, C1qa, C1qb, Ctss, Fcer1g, Grn, Lag3, Ly9, Man2b1, Mpeg1, Slamf9, Tyrobp, and Vav1) were linked to brain myeloid cell activation, including DAM/ WAM profiles 5,8,9,[30][31][32] . Next, we studied the age-and region-specific expression patterns that were conserved and distinct between sexes and identified genes exclusively differentially expressed in a single brain region and sex (Fig.…”

Brain cross-omics integration in Alzheimer’s disease

“…Knight-C4 contains multiple pathways related to different aspects of neurodegeneration ( Fig 4G and 4H ). For instance, KEGG pathways identified multiple intracellular trafficking and signaling pathways ( Fig 4G ), including (e.g., VPS29 and VPS35 ) endocytosis (p=6.5×10 −04 ), (e.g., ATP6V1A and RAC1 ) phagosome (p=1.2×10 −06 ), and (e.g., GSK3B , and IGF1) mTOR signaling pathway (p=3.3×10 −02 ), all of which have been shown to play major roles in the pathological processes of AD [38–40]. Genes ( PPT1 , GBA , HEXB , and LGMN) from lysosomal pathway, which is critical for microglia, and neurons, were also identified (p=4.7×10 −02 , Fig 4G ).…”