Proteomic signatures of carotid plaque vulnerability: Proteolysis, inflammation, metabolic reprogramming, and lipid...

Carotid plaque vulnerability is a current feature that aids in the decision-making for ischemic stroke risk. Proteomic analysis of plaque tissue can reveal molecular indicators of instability that complement imaging findings. We sought to identify a proteomic signature distinguishing vulnerable from stable carotid plaques in patients undergoing endarterectomy, with the aim of uncovering candidate biomarkers for potential diagnostic and therapeutic targets.

Twenty-eight carotid plaque specimens were collected from 27 patients (including 1 patient with bilateral endarterectomy). Samples were classified as vulnerable (n = 14) or nonvulnerable (n = 14) based on preoperative magnetic resonance angiography with vessel wall imaging. A tandem mass tag-based multiplexing strategy followed by mass spectrometric analysis was used to profile the proteomes of all samples. Normalized and log₂-transformed protein intensities were compared using two-sample t tests with unequal variances, and P values were adjusted for multiple testing with the Benjamini-Hochberg method to obtain false discovery rate q values. Proteins with a q value of ≤0.25 were designated high-confidence candidates, and those with a P value of <.05 but a q value of >0.25 were considered exploratory.

From 3267 proteins identified, 398 reached nominal significance (P < .05). From those, 29 reached at least log₂(fold-change) of +1, whereas 3 of -1 log₂(fold change), yielding a total of 32 proteins. Fifteen were significant for a q value ≤0.25. All were upregulated in vulnerable lesions and these included: matrix-degrading enzymes (matrix metalloproteinase [MMP]7, MMP9, MMP1, and ADAM-like decysin-1), neutrophil-derived effectors (azurocidin, cathelicidin antimicrobial peptide, lactotransferrin, and myeloperoxidase), inflammatory regulators (interleukin-1 receptor antagonist and interleukin-4-induced protein), glycolytic enzymes (hexokinase-3 and hexokinase-2), and lipid-handling proteins (lipoprotein-associated phospholipase A₂, apolipoprotein B, and paraoxonase-1). An additional 17 exploratory proteins showed nominal significance (P < .05, q > 0.25) with at least log₂(fold-change) of 1, and 366 proteins with nominal significance but with a log₂(fold-change) of <1.

Our proteomic profiling delineates a robust vulnerability signature marked by enhanced proteolysis, neutrophil activation, inflammatory signaling, metabolic reprogramming, and lipid dysregulation. High-confidence proteins emerged as tissue biomarkers of plaque instability. Validating their association with future cerebrovascular events is the next step toward clinically actionable stroke prediction. Exploratory candidates warrant further analysis.