Disease-relevant single cell photonic signatures identify S100β+ stem cells and their myogenic progeny in vascular lesions

Vascular lesions enriched with smooth muscle (SMC)-like cells contribute to the development of subclinical atherosclerosis yet the origin of these cells remains controversial. Herein, we combine single cell label-free photonics with lineage tracing analysis to examine the cellular heterogeneity of vascular cells within lesions. Single cell autofluorescence emissions from normal and injured vessels ex vivo were compared to aortic SMCs, cultured Movas SMCs, macrophages, B-cells, S100β+ resident vascular stem cells (mVSc), mesenchymal stem cells (MSC) and their myogenic progeny across five wavelengths 𝜆465 - 𝜆670 ±20 nm. Principal component analysis (PCA) revealed discrete photonic signatures, particularly at 𝜆565 nm, that reflect cell type and differentiation state. Supervised machine learning using multilayer perceptron (MLP) neural network analysis, linear discriminant analysis (LDA), and lineage tracing of S100β+ mVSc identified the presence of mVSc and their myogenic progeny within injured vessels. We conclude disease-relevant photonic signatures may have predictive value for vascular disease.