Computational Integration of Dietary Risk Factor Gene Sets with the Abdominal Aortic Aneurysm Transcriptome Identifies a Shared NLRP3–IL1β–MMP9 Inflammatory Program

Background: Dietary risk factors—high-fat, high-salt, high-sugar intake, chronic alcohol use, sugared carbonated beverages, and irregular eating—are epidemiologically associated with abdominal aortic aneurysm (AAA), but whether and through which molecular programs they converge on the aortic wall is poorly defined. No human aortic transcriptome with controlled dietary exposure exists in public repositories, which precludes direct case–control dietary differential-expression analysis in human tissue. Methods: We constructed six dietary-risk gene sets from animal-intervention transcriptomic studies in GEO, curated disease–gene association databases, and targeted literature mining. Human AAA differentially expressed genes (DEGs) were derived from GSE57691 (49 AAA vs. 10 non-aneurysmal donors) and GSE47472 (14 AAA vs. 8 donors) using limma (adjusted P<0.05, |log2FC|≥0.5); genes concordantly DE in both cohorts were retained as the consensus AAA set. Each dietary set was intersected with AAA DEGs; genes hit by ≥3 of six exposures were termed the strict core and genes hit by ≥2 were termed the extended core. Functional enrichment (GO/KEGG), STRING PPI analysis, and marker-based single-sample gene-set enrichment analysis (ssGSEA, GSVA) were performed. A per-sample multi-dietary score (DietScore) was computed in GSE57691 and its association with AAA status, the NLRP3–CASP1–IL1β–MMP9 axis, and marker-based immune signatures was tested by Wilcoxon and Spearman methods. Results: Thirty-six genes were shared between the six dietary sets and AAA DEGs. Seven strict-core genes were hit by ≥3 dietary exposures—IL1B (5 sets), NLRP3 (4), MMP9 (4), and IL6, CCL2, VCAM1, CAT (3 each)—and a further seven (CASP1, NFKB1, EDN1, NCF1, APOB, PPARG, MYH11) were hit by two. Enrichment converged on Lipid and atherosclerosis (P≈3×10⁻¹⁴), AGE–RAGE signaling in diabetic complications, NOD-like receptor (NLRP) signaling, and macrophage/foam-cell differentiation. STRING PPI centered on the NLRP3–CASP1–IL1β–IL6–MMP9 module. DietScore was higher in AAA than donor aorta (Wilcoxon P≈1.6×10⁻⁵) and correlated positively with CASP1 (ρ=0.59), IL1B (ρ=0.57), NLRP3 (ρ=0.47), MMP9 (ρ=0.46), and with macrophage, T-cell, and neutrophil signatures. Conclusions: This computational, hypothesis-generating integration of mostly animal-derived dietary gene sets with the human AAA transcriptome identifies a shared myeloid–inflammatory, NLRP3–IL1β–MMP9 program that is active in human AAA. The DietScore is an internally derived, descriptive computational index; it is not validated as a clinical AAA biomarker and must not be used for patient classification or diagnosis. This work does not establish a causal diet–AAA relationship; it cannot distinguish direct aortic-wall effects from effects mediated through hypertension, dyslipidemia, or diabetes; and the DietScore was derived and evaluated within a single cohort without an independently phenotyped external validation set. All findings require confirmation in dietary-intervention animal models and human AAA cohorts with matched dietary phenotypes.