The tissue inhibitor of metalloproteinases-1 (TIMP-1) protein can regulate the expression of certain proteases and microRNAs in cancer cells, and it is highly possible to diagnose cancers through analyzing the expression of TIMP-1 on exosomes. However, it is still a great challenge to obtain reliable physiological information on TIMP-1 by label-free method from exosomes in plasma. Here, we designed a porous-plasmonic SERS chip functionalized with synthesized CP05 polypeptide, which can specifically capture and distinguish exosomes from diverse origins. The SERS chip can accurately locate the plasmon in TIMP-1 protein to analyze the discrepancy of related fingerprint peaks of different exosomes. Based on the designed SERS chip, we successfully distinguished the lung and colon cancer cell-derived exosomes from normal exosomes at the single vesicle level by unique Raman spectroscopy and machine learning methods. This work not only provides a practical SERS chip for the application of Raman technology in human tumor monitoring and prognosis, but also provides a new idea for analyzing the feature of exosomes at the spectral level.
Keywords: Exosomes; Machine learning; Plasmon; SERS; TIMP-1; Tumor.
Copyright © 2022 Elsevier B.V. All rights reserved.