show Abstracthide AbstractAlzheimer's disease (AD) is a progressive neurodegenerative disorder and the most frequent cause of dementia. The disease has a substantial genetic component comprising both highly penetrant familial mutations (APP, PSEN1 and PSEN2) and sporadic cases with complex genetic etiology. Mutations in APP and PSEN1/2 alter the proteolytic processing of APP to its metabolites, including Ab and APP Intracellular Domain (AICD). In this study, we use transgenic porcine models carrying the human APPsw and PSEN1M146I transgenes to demonstrate the pathobiological relevance of transcriptional regulation facilitated by APP and its AICD domain. Through molecular characterization of hippocampal tissue, we describe the differential expression of gene sets that cluster in molecular pathways with translational relevance to AD. We further identify phosphorylated and unphosphorylated AICD in differential complexes with proteins implicated in signal transduction and transcriptional regulation. Integrative genomic analysis of transcriptional changes in somatic cell cultures derived from pigs treated with g-secretase inhibitor demonstrates the importance of g-secretase APP processing in transcriptional regulation. Collectively, our data supports a model in which APP, and in particular its AICD domain, modulates gene networks associated with AD pathobiology through interaction with signaling proteins. Overall design: Hippocampal tissue samples from wild type (WT) and transgenic Göttingen minipigs, carrying copies of human APP695 cDNA with the Lys670Asn/Met671Leu (APP) double-mutation and/or one copy of human PSEN1 cDNA with the Met146Ile (genotype PS1) mutation was included in study (4x APP/PS1, 4x PS1 and 4X WT). Library construction, sequencing and initial data filtering including adaptor removal were performed by BGI Europe Genome Center. Total RNA was subjected to oligo dT based mRNA enrichment. 100 bp paired-end read sequencing was performed on DNBseq platform (hippocampal tissues). More than 20 million clean reads were obtained per sample. Reads were aligned to the porcine genome build Sscrofa11.1 (Ensemble release 92) using HISAT2 aligner (v2.1.0). Transcript quantification was performed using htseq-count (v0.9.1) and the read counts were normalized for effective gene length, and sequencing depth to yield Transcripts Per Kilobase Million (TPM). Differentially expressed genes were determined from count tables using DEseq2 (v2.11.40.6)