The natural evolution of human atherosclerosis involves the progression to a vulnerable state and rupture, a pathologic fate that is exclusive to humans.[1–6] Atherosclerosis growth and progression are spatial processes influenced by disturbed patterns of blood flow. The consequent vascular tissue disorganization corresponds to histologic patterns with regional distributions that are an integral part of histopathologic plaque classification.[2]
A vulnerable and prone-to-rupture plaque is traditionally characterized by a large necrotic core, a thin fibrotic cap, an inflammatory phenotype, and intraplaque hemorrhage.[1,5] These pathologic features and their unique spatial distribution in the tissue context contribute to the highly heterogeneous nature of atherosclerotic tissue.
In this issue of the Journal of the American College of Cardiology, Sun et al[7] use an integrated approach to perform an in-depth regional mapping of a total of 163 human atherosclerotic lesions to establish the anatomic coordinates of plaque rupture and identify the corresponding biochemical, cellular, and transcriptional changes along the regional flow direction.
Leveraging the Carotid Plaque Imaging Project (CPIP),[8] the researchers first performed a systematic analysis of 41 carotid endarterectomy specimens from symptomatic patients, dividing them into 3 regions: proximal, most stenotic, and distal. The exact site of rupture was consistently identified proximally or near to the most stenotic region at a median distance of 9 mm from the proximal plaque end.
Histologic and bulk transcriptional analyses of the proximal, most stenotic, and distal segments identified the underlying molecular and cellular alterations. Overall, the results of this analysis suggest that inflammation and luminal endothelial damage are major spatial determinants of plaque rupture.
The researchers estimated immune cell type fractions from bulk RNA-sequencing data and found differences among the 3 segments that were not detectable by histology. The most stenotic segment, followed by the proximal region, was the most inflamed according to the increased macrophage, T cell, and natural killer cell proportions and corresponding transcriptional signature of macrophage activation. The vulnerable nature of the most stenotic segment was further confirmed by the lower proportions of vascular smooth muscle cells vs both the proximal and distal regions. The similar lipid and glycosaminoglycans content quantified by histology in the 3 segments suggests a less critical regional role of lipid accumulation and lipid tissue retention in determining plaque rupture.
Scanning electron microscopy of the luminal surface of plaques from symptomatic patients revealed profound regional differences in endothelial integrity between the proximal, most stenotic, and distal segments. Either the endothelium of the most stenotic segment was denuded, or, when present, endothelial cells had junctional abnormalities. The endothelial surface was covered by fibrin and platelets. Endothelial abnormalities, platelets, fibrin, and red blood cells were also described in the proximal region, whereas the endothelium of the distal region appeared intact.
The similarities between the proximal and most diseased segments of ruptured plaques were confirmed at the gene expression level, and the 2 rupture-associated regions were transcriptionally distinct from the distal region.
Spatial transcriptomics revealed that the expressions of the top differentially expressed genes—IGKC, which encodes immunoglobulin kappa constant, and PLN, which encodes phospholamban, a protein associated with the Ca2+-ATPase of the sarcoplasmic reticulum and regulatory roles in smooth muscle cell contraction[9]—were not specific to the site of rupture.
In contrast, matrix metalloproteinase (MMP9), which encodes MMP-9, an enzyme that degrades extracellular matrix and regulates tissue remodeling,[10] was largely expressed in the proximity of the area of plaque rupture. Here, MMP9 was coexpressed with macrophage and T cell markers. Spatial transcriptome analysis confirmed higher expression of MMP9 in the shoulder region from symptomatic vs asymptomatic patients.
Next, the researchers investigated whether the expression of IGKC, PLN, and MMP9 within the most stenotic segment was different between 27 asymptomatic and 51 symptomatic patients.
The most stenotic segment of plaques from symptomatic and asymptomatic patients expressed similar levels of IGKC. However, the most stenotic regions of symptomatic plaque expressed more MMP9 and less PLN than corresponding regions from asymptomatic patients. Although the observation of increased levels of MMP-9 in plaques from symptomatic patients is not surprising and is in line with previous observations from a proteomics and transcriptomics study of human plaques from symptomatic and asymptomatic patients,[11] these data add the spatial coordinates and position MMP9 expression at the exact location of rupture.
Furthermore, the researchers found that high MMP9 and low PLN expression levels in the maximal stenotic segment were associated with a greater risk for future cardiovascular events. Cardiovascular events adjudicated at follow-up for participants of the Carotid Plaque Imaging Project biobank included myocardial infarction, unstable angina, stroke, transient ischemic attack, amaurosis fugax, vascular interventions, and cardiovascular death.[8] Mendelian randomization analysis showed that higher levels of circulating MMP-9 contribute to increase risk of coronary atherosclerosis in this cohort.
The findings of Sun et al[7] highlight the importance of spatial resolution to mechanistically understand plaque rupture in human atherosclerosis. The authors provide a systematic histologic and transcriptional mapping of atherosclerosis plaques that revealed the exact point of vulnerabilities for plaque rupture, which was consistently the proximal side of the most stenotic segment, and identified the spatial localization of molecular and cellular alterations at the ruptured site. Not surprisingly, the integrity of the vascular endothelium was largely compromised at the site of rupture and proximally along the blood flow. The combination of bulk RNA sequencing and spatial transcriptomics identified MMP9 as a major candidate determinant of plaque rupture, which is consistent with its enzymatic function and the well-established role in atherosclerosis established using murine models of the disease.[10,12] The clinical relevance of the association between MMP-9 expressed at the maximal stenotic segment with future cardiovascular events is limited by the fact that the lesion was removed, and as such, a direct contribution to future clinical outcomes is unlikely. The possibility of similar spatial expression of MMP-9 in plaques at other vascular sites and systemically because of functionally significant sequence variations of the MMP9 gene will require further investigation.
The findings of this study take a first step toward incorporating spatial analysis into investigations of human plaque pathophysiology to determine how pathogenic mechanisms of plaque rupture operate at a discrete plaque site. The future implementation of spatial single-cell technologies will help refine human plaque biology and help decipher mechanisms of atherosclerosis and plaque rupture.
J Am Coll Cardiol. 2023;81(23):2228-2230. © 2023 American College of Cardiology Foundation
