Characterization of extracellular vesicles derived from mesenchymal stromal cells by surface-enhanced raman spectroscopy

Extracellular vesicles (EVs) are secreted by all cells into bodily fluids and play an important role in intercellular communication through the transfer of proteins and RNA. There is evidence that EVs specifically released from mesenchymal stromal cells (MSCs) are potent cell-free regenerative agents. However, for MSC EVs to be used in therapeutic practices, there must be a standardized and reproducible method for their characterization. The detection and characterization of EVs are a challenge due to their nanoscale size as well as their molecular heterogeneity. To address this challenge, we have fabricated gold nanohole arrays of varying sizes and shapes by electron beam lithography. These platforms have the dual purpose of trapping single EVs and enhancing their vibrational signature in surface-enhanced Raman spectroscopy (SERS). In this paper, we report SERS spectra for MSC EVs derived from pancreatic tissue (Panc-MSC) and bone marrow (BM-MSC). Using principal component analysis (PCA), we determined that the main compositional differences between these two groups are found at 1236, 761, and 1528 cm(-1), corresponding to amide III, tryptophan, and an in-plane -C=C- vibration, respectively. We additionally explored several machine learning approaches to distinguish between BM- and Panc-MSC EVs and achieved 89 % accuracy, 89 % sensitivity, and 88 % specificity using logistic regression.

Automated summary

Extracellular vesicles (EVs) play a role in intercellular communication through the transfer of proteins and RNA. This study used gold nanohole arrays and Raman spectroscopy to detect EVs secreted by mesenchymal stromal cells (MSCs) and identified compositional differences between EVs derived from different sources, achieving high accuracy in distinguishing them through logistic regression.