Andres Kriete

Background : An increasing amount of evidence demonstrates that immune dysfunction and neuroinflammation contribute to the pathogenesis and progression of late-onset Alzheimer's disease (LOAD). To identify and characterize the transcriptomic and genetic contributors to immune dysfunction in LOAD, we have designed a comprehensive analysis framework for performing clinical segregation analysis, co-expression modeling, functional enrichment, and gene perturbation testing by integrating clinical neuropathological, gene expression, and genotype data.
Methods : We applied our analysis framework to 503 human subjects from the Religious Orders Study and Memory and Aging Project (ROSMAP) to map and compare functional gene networks, transcription factors, and genomic loci studies to decipher their contributions to disease progression.
Results : We identified deregulation in gene networks involving immune response, synaptic function, and Hippo signaling. One module contains 191 genes, including ABI3, CD33, HLA-DRA, HLA-DMB, IL10RA, MS4A4A, TREM2, and TYROBP and is strongly associated with microglial function. Segregation analysis revealed an association between network organization and cognitive decline in these functionally enriched modules. A reduction in co-regulated genes indicates changes in several key signaling nodes associated with microglia activation and inflammatory response. Finally, we identified HLA and MAPT alleles in association with LOAD.
Conclusions : Our results corroborate findings of the involvement of the adaptive and innate immune systems in LOAD pathology and can inform immunotherapies aimed at treating this condition. Finally, this study serves at a paradigm for integrating detailed clinical phenotypes with various -omics data to further resolve molecular endophenotypes in LOAD and other complex diseases.