Two CFDE datasets · one mechanistic story
Atherosclerosis progresses silently until a plaque ruptures. This visualization bridges local plaque EV biology with circulating exRNA in a large epidemiologic cohort.
Human carotid endarterectomy · plaque vs marginal zones · symptomatic (n=16) vs asymptomatic (n=13) · EV enrichment by SEC · miRNA-seq + MS proteomics + scRNA-seq integration + HUVEC angiogenesis validation
Framingham Heart Study Third Generation · n≈4,095 · 665 plasma exRNA species (miRNA, tRNA frags, lncRNA, piRNA) · coronary artery calcium (CT) as atherosclerosis readout · 7-year follow-up
EV-miRNA cargo discriminates plaque zone from marginal zone
371 miRNAs enriched in plaque-zone EVs; 290 enriched in marginal-zone EVs (DESeq2, FDR < 0.05, n=20 donor-matched pairs). Top species highlighted in inflammation are shown in red on the volcano.
661 proteins differentially enriched in plaque-zone EVs
MS-based proteomics across 23 donor-matched pairs. Top categories include inflammation, ECM/adhesion, angiogenesis, complement, and cholesterol metabolism.
Plaque EVs are dominated by endothelial-cell origin
EV cellular origin inferred via TISSUES database + Tabula Sapiens enrichment scoring (AddModuleScore, Seurat v5). Plaque-zone EVs are predominantly endothelial; marginal-zone EVs are VSMC-dominant. This endothelial signature drives an angiogenesis/neovascularization mechanism validated in HUVEC sprouting assays.
Unlike prior assumptions that macrophages are the primary EV source in plaque, this study shows endothelial cells dominate plaque-zone EV biogenesis (82% enrichment score vs 34% in marginal zones).
Conversely, vascular smooth muscle cells (VSMCs) predominate in marginal-zone EVs (78%), reflecting the quiescent, contractile phenotype of non-diseased vessel wall.
Functional validation: Plaque EVs from symptomatic patients significantly promoted HUVEC spheroid sprouting relative to marginal-zone EVs and asymptomatic controls — directly confirming an EV-driven angiogenesis/neovascularization mechanism in vulnerable plaque.
35 KEGG pathways shared between EV-miRNA targets and EV-proteins
KEGG enrichment via Enrichr on predicted EV-miRNA mRNA targets (miRTarBase) and differentially enriched EV-proteins. 60 pathways unique to EV-miRNA, 60 unique to EV-proteins, 35 shared (FDR < 0.05).
FHS plasma exRNA links to subclinical coronary atherosclerosis
The Framingham Heart Study exRNA–CAC cohort (NCT03225196) is a bona fide NIH Common Fund deposit in the exRNA Atlas (ERCC). It profiles 665 plasma exRNAs in ~4,095 adults and relates them longitudinally to coronary artery calcium (CAC).
Plasma oxylipins stratify CAD severity and 5-year survival
Targeted HPLC-MS/MS profiling of 39 plasma oxylipins in 74 symptomatic CAD patients (≥70% stenosis, 1–3 diseased arteries) vs 23 asymptomatic low-risk controls. Key findings: 6 oxylipins fall with increasing vessel burden; a 5-oxylipin panel identifies 3-vessel disease at 100% sensitivity; a 2-oxylipin survival panel achieves 86% sensitivity / 91% specificity.
Knight Cardiovascular Institute, OHSU · n=97 (74 CAD, 23 controls) · Targeted HPLC-MS/MS · 39 oxylipins across COX, LOX, CYP450, and non-enzymatic pathways · 5-year longitudinal outcome follow-up
Convergent miRNAs bridge plaque EVs and FHS circulating exRNA
miRNA species enriched in carotid plaque EVs (Raju et al.) overlap with the FHS 665-species plasma exRNA panel. The bubble chart maps plaque EV fold-change against FHS–CAC association, revealing a convergent pro-atherogenic exRNA signature.
Key findings and access points
| Finding | Evidence | Source |
|---|---|---|
| Plaque-zone EVs contain 661 DE proteins + 371 DE miRNA vs marginal zones | FDR < 0.05, donor-corrected PCA | Raju et al. 2025, Fig 2 |
| Endothelial cells are the dominant predicted EV source in plaque (not VSMC or macrophage) | TISSUES + Tabula Sapiens AddModuleScore | Raju et al. 2025, Fig 3–4 |
| Symptomatic plaques have a distinct EV-miRNA signature (miR-21, miR-155 ↑; miR-143, miR-145 ↓) | DESeq2, FDR < 0.05 | Raju et al. 2025, Fig 5 |
| Plaque EVs functionally promote endothelial angiogenesis/sprouting ex vivo | HUVEC spheroid sprouting assay | Raju et al. 2025, Fig 6 |
| 665 plasma exRNAs profiled in 4,095 FHS adults; 45% have CAC > 0 | CFDE exRNA Atlas (ERCC) | NCT03225196; exrna-atlas.org |
| miR-21-5p, miR-146a, miR-92a detected in both plaque EVs and FHS plasma panel | Cross-dataset convergence | Integration analysis |
| 6 plasma oxylipins decrease with CAD vessel burden; 5-oxylipin panel diagnoses 3-vessel disease at 100% sensitivity | Targeted HPLC-MS/MS, n=97 | Kaul et al. Front Cardiovasc Med 2021 |
| 9-HODE + 10,11-EpDPA panel predicts 5-year survival (AUC 0.93; 86% sensitivity, 91% specificity) | 5-year follow-up, OHSU cohort | Kaul et al. Front Cardiovasc Med 2021 |
| EV-miRNA cargo (miR-21, miR-155, miR-92a, miR-143/145) mechanistically links to COX/LOX/CYP oxylipin shifts | miRNA→enzyme→oxylipin axis | Cross-dataset mechanistic inference |
Current atherosclerosis risk stratification relies primarily on lipid panels and clinical risk scores, which fail to capture the biological heterogeneity of plaque vulnerability.
This multiomics visualization supports a model in which circulating EVs carry a plaque-stage-specific molecular fingerprint that is detectable in plasma and correlates with imaging-confirmed coronary atherosclerosis burden (CAC).
The convergent EV miRNA panel (miR-21-5p, miR-146a-5p, miR-155-5p, miR-92a-3p) represents a candidate liquid biopsy signature for non-invasive atherosclerosis risk stratification beyond conventional cholesterol-based models.