Genetic-risk-informed integration of bulk and single-cell transcriptomes prioritizes ANPEP as a monocyte-associated candidate in aortic dissection

Abstract

Aortic dissection is characterized by vascular wall remodeling and inflammatory activation, but the cell-associated molecular programs linking transcriptional alterations to genetic susceptibility remain incompletely defined. This study aimed to identify reproducible monocyte-associated molecular candidates in aortic dissection. We integrated bulk transcriptomic discovery (GSE153434) with an exploratory secondary-cohort assessment (GSE52093), single-cell RNA sequencing (GSE189795), FinnGen genome-wide association study summary statistics, scPagwas, and cell-type-specific hdWGCNA. ANPEP and CCL20 transcriptional responses were subsequently assessed by droplet digital PCR in five independent paired experiments using angiotensin II-stimulated, PMA-differentiated THP-1 macrophage-like cells. Immune-context, cellular-state, compound-screening, and spatial analyses were performed as exploratory hypothesis-generating analyses. Genetic risk-informed analysis highlighted myeloid populations, particularly monocytes and neutrophils. Integration of bulk differential expression, scPagwas correlations, and a disease-associated monocyte hdWGCNA module prioritized ANPEP , CCL20 , and CD82 . ANPEP showed the most consistent support across the analyzed evidence layers. In the secondary cohort GSE52093, ANPEP and CD82 expression levels were higher in aortic dissection samples, whereas CCL20 showed weaker cross-cohort consistency; these findings were interpreted as exploratory secondary-cohort evidence rather than clinical validation. Angiotensin II stimulation increased ANPEP and CCL20 transcript abundance in THP-1 macrophage-like cells. Exploratory analyses further associated the prioritized genes with heterogeneous myeloid inflammatory states. Integrative genetic and transcriptomic analyses prioritized ANPEP as a comparatively consistent monocyte-associated candidate in aortic dissection. These results require confirmation in primary human cells, independent patient samples, and functional perturbation models.