Genetics-Informed Mapping Identifies a CRIM1-Associated Endocardial Inflammatory Remodeling State in Acute Myocardial Infarction

Abstract

Background:Acute myocardial infarction (AMI) reflects inherited susceptibility and inflammatory remodeling, but the cellular contexts linking genetic risk to disease remain unclear. Material/Methods:We integrated a meta-transcriptome-wide association study (TWAS) with a human cardiac single-nucleus RNAsequencing atlas contained 11 individuals (5 AMI and 6 donor) to identify genetics-informed cellular programs.Composite program states were defined by global score quartiles.A fixed 5-gene panel was evaluated for nucleus-level endocardial low-transcriptional-state (Endo_LTS) vs endocardial high-transcriptional-state (Endo_HTS) discrimination within the AMI endocardium using 5-fold leave-1-patient-out cross-validation.Functional followup used CRIM1 silencing in hypoxia-treated human induced pluripotent stem cell (hiPSC)-derived endocardial endothelial-like cells and complementary peripheral blood analyses. Results:The endocardium exhibited the most prominent infarction-associated increase in TWAS-anchored program activity, with expansion of program-high states and higher CytoTRACE scores.A consensus 5-gene panel (RPS8, PLEC, CFDP1, CRIM1, TNS2) was identified.Among 2163 AMI endocardial nuclei from 5 patients, the state classifier included 364 Endo_LTS and 751 Endo_HTS nuclei; 1048 Endo_MTS nuclei were excluded.Pooled out-offold ROC-AUCs ranged from 0.665 to 0.831.The panel also showed discriminatory value in an independent peripheral-blood AMI-vs-control cohort.CRIM1 was prioritized as a candidate linked to the remodeling program.CRIM1 silencing attenuated ACTA2/a-SMA, vimentin, LDHA, CCL2, and VEGFA and partially restored CD31, whereas TGF-b remained elevated. Conclusions:These findings identify a genetics-informed endocardial inflammatory remodeling state in AMI and define a 5-gene surrogate of its activated state.CRIM1 is prioritized as a candidate linked to selected inflammatory, metabolic, and structural outputs.Persistent TGF-b elevation after CRIM1 silencing argues against a simple linear regulatory model and indicates that further mechanistic validation is required.