Peripheral Blood Brain Microvascular Endothelial Cells As a Novel Marker in the Detection of Blood-Brain Barrier Injury in EV71 Severe Encephalopathy

Enterovirus 71 (EV71), the main causative agent of hand, foot, and mouth disease (HFMD), is also considered a neuroinvasive virus. EV71 causes sporadic epidemics and outbreaks with more than 13 million people worldwide, predominantly occurring in the Asia-Pacific, and elicited more than 3000 deaths, including 2457 deaths in China from 2008 to 2014. However, there are no available noninvasive biomarkers in peripheral blood for neuro-EV71 infection and optimizing therapeutic options. Brain microvascular endothelial cells (BMECs) are a special type of cells that constitute the blood-brain barrier (BBB), which dynamically shed associated with BBB injury. A mouse model of BBB injury induced by microbial and non-microbial agents was performed, using magnetic bead extraction (MBE) to identify circulating endothelial cells (CECs,CD146+/DAPI+), circulating BMECs (cBMECs, CD146+/S100B+/DAPI+) and endothelial progenitor cells (EPCs, CD146+/CD133+/DAPI+), which are identified based on their S100B (brain marker)+/ CD146 (EC marker) +/CD133+ (PC marker) /DAPI (nuclei)+phenotype with significantly higher levels in the experimental group than in the control group. These results indicate that it is feasible to evaluate BBB injury by quantifying BMECs and EPCs with MBE.

The whole cell proteins were prepared by us from EV71 infected BMECs, and among the samples infected with EV71, MALDI TOF / TOF mass spectrometry successfully identified 28 proteins, 4 proteins including vimentin (VIM) were significantly up-regulated, 6 proteins were down regulated, and the other 18 proteins were expressed to various degrees at different incubation times. Furthermore, we performed a glioblastoma model with low vimentin expression and found that EV71 may activate the NLRP3 inflammasome via the VIM-ERK-NFκB pathway during EV71 CNS infection, which promotes caspase-1 activation and the inflammatory factor IL-1β release, causing an inflammatory response that ultimately leads to BBB/CNS disorders and neuronal death. The same results for those experiments with neonatal mice of EV71 infection were also observed. The results showed that cBMECs may be used as potential cell-based biomarkers for indexing of BBB injury.

Then peripheral blood samples were obtained from the children with severe HFMD infection from a medical center for women and children, and examination showed that the counts of cBMECs were significantly higher in severe HFMD children than in the control children, indicating that the detection of cBMECs may be used for the early diagnosis of severe HFMD. In addition, Mfsd2 antibody specifically on BMECs were attached to magnetic beads and used directly to isolate BMECs from blood, followed by immunofluorescence staining with CD146 antibody, and the two corresponding fluorophores, CD146 and DPAI, appeared microscopically to preliminarily determine BMECs. The use of mfsd2 antibody to isolate BMECs both simplifies experimental procedures and improves detection precision and efficiency.

In conclusion, in this study, cBMECs are demonstrated to serve as cellular biomarkers for BBB injury caused by microbial and non-microbial factors. Proteomics combined with mechanistic studies indicate that cBMECs are important indicators of BBB/CNS disorders in severe HFMD. Ultimately, this study successfully established a method to detect EV71 severe encephalopathy in human peripheral blood, which is safer and noninvasive than conventional cerebrospinal fluid puncture and pathological collection methods, causes minimal physical trauma to patients, and is more intuitive and accurate than conventional imaging examination. Therefore, this assay has good clinical application and may provide a new means for noninvasive and visual detection of EV71 severe encephalopathy. However, further cases were also collected to verify the accuracy of peripheral blood BMEC count for diagnosing severe HFMD, while establishing suitable cut-off values for application in early diagnosis. (Acknowledgements: Corresponding author: Hong Cao, gzhcao@smu.edu.cn; The Basic Research Project of Key Laboratory of Guangzhou, China (No.202102100001) and Grant from School of Public Health of Southern Medical University, China, No.GW202222 to H.C.)

No relevant conflicts of interest to declare.