Abstract

Abstract The development of a sensitive diagnostic method for medulloblastoma (MBL), a common malignant brain tumor in children, is still a challenge. Cerebral spinal fluid (CSF) contains various substances such as soluble proteins and extracellular vesicles (EVs) and may reflect pathological changes in the central nervous system. Especially, EVs within CSF have great potential in the dynamic monitoring and intervention strategies of MBL. This study attempted to determine whether the total CSF protein selected based on proteome information distinguished from the CSF in MBL is also considered in EVs cargo proteins. CSF samples from patients with MBL (N = 44) and hydrocephalus (HC, control, N = 18) were analyzed with LC-MS. A machine-learning approach was used to identify protein biomarkers discriminating MBL CSF. The biomarker potential of the selected proteins was verified through ELISA and ROC analysis. TEM, Nanosight, and Exoview systems were used to analyze the physiological and biological characteristics of proteins in EVs from MBL CSF. Proteomic analysis revealed four elevated soluble proteins (Transketolase (TKT), SPTBN1, HSP90AA1 and NME1-NME2) in MBL CSF. Verification of the selected proteins in whole CSF confirms that TKT was significantly elevated in MBL than in HC. Importantly, the AUC value of TKT was 0.94, which was recognized as an excellent diagnostic marker. In additional TKT analysis, when MBL was divided into subtypes, Group 4 was the most significant compared to HC, and when divided into metastasis group, metastasis was more significant than non-metastatic compared to HC. Consistently, TKT was more abundant in MBL CSF EVs compared to HC CSF EVs. Our results provide insights into the relevance of the proteins in CSF to the cargo proteins reflected in EVs, while also proposing TKT in CSF EVs as an MBL diagnostic marker.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call