Single-cell RNA Sequencing Analysis Of Glioma And Identification Of Immunotherapy Target S100A4

Glioma is the most common malignant tumor in the central nervous system,with high heterogeneity,high invasive ability and high recurrence rate.Among all gliomas,glioblastoma（GBM）is the most malignant,accounting for 40%.Surgery,chemotherapy and radiotherapy are the main treatment methods for glioma.However,GBM patients have the worst overall prognosis compared to other glioma patients with these therapies.The median survival time of GBM is about 12-15 months with high recurrent rate whereas the 5-year survival rate is less than 5%.A considerable number of clinical trials have been conducted around the world to explore whether immunotherapy can be a more effective treatment.Unlike successful experience in other tumors,the available results showed that most glioma patients were not sensitive to immunotherapies such as immune checkpoint inhibitors.Some possible reasons for these failures are the tumor heterogeneity and immunosuppression features.However,the heterogeneity of tumor tissue in glioma has not been fully revealed,the role of immune cells in the glioma tumor microenvironment and the mechanism of immune regulation remains unclear.This project is aimed to reveal the cellular and molecular heterogeneity of tumor cells in glioma at the single-cell level,explore the cell phenotypic characteristics and composition of T cells as well as myeloid cells,and find novel therapy targets to develop immunotherapy strategies.Fresh tumor samples were collected from 44 different sites of 18glioma patients,including 11 new diagnosed glioblastomas,5 recurrent glioblastomas,and 2 low-grade gliomas.Tumor tissue was digested and separated to generate single cell suspension.About 6000 cells were captured per sample and 44 single cell sequencing gene libraries were generated and sequenced by Hi Seq 4000 or BGI DNBseq platform.Single-cell sequencing data were analyzed by Cell Ranger and Seurat4.0.0 software.The S100A4^-/-mouse host model was established by replacing the exon 2 of the S100A4 gene with the GFP gene in mice.Two S100A4^-/-and B6 mouse models of similar sex ratio and age were injected with 2 primary glioma cell spheres 5459 and 2808 by intracranial injection separately.The survival curves and immune cells in tumor-bearing mouse were analyzed.In present study,sc RNA-seq was performed on more than 200,000single cells from 18 glioma patients.Through analysis,it was found that the gene expression characteristics of tumor cells in gliomas can be summarized into four different types:OPC-like cells,NPC-like cells,AC-like cells and Mes-like cells.Mes-like cells were enriched with gene sets related to EMT,hypoxia and IFNγresponse.The expression level of PDCD1/PD1 in T lymphocytes was low.Nine subtypes were identified in myeloid cells.Among these subtypes,activated,homeostatic and inflammatory microglia were associated with good prognosis,whereas antigen-presenting microglia,immunosuppressive macrophages and MDSCs were associated with poor prognosis.Nine subtypes were validated in another published independent database.S100A4 was highly expressed in immunosuppressive T cells and myeloid cells.Glioma patients with high S100A4 expression had a worse prognosis compared to low expression patients.The survival time of S100A4^-/-tumor-burdened mice of either 5459 or 2808 tumor models was significantly prolonged,with increased lymphocyte infiltration including increased CD4⁺and CD8⁺T cells.In conclusion,there is obvious heterogeneity of tumor cells and immune cells in glioma.Each sample contained a mixture of glioma cell in different cell states.The nine subtypes found in myeloid cells can predict the prognosis of glioma patients.The expression of S100A4 is high in suppressive immune cells.S100A4 knockout in host mice can prolong the survival of glioma mice and increase lymphocyte infiltration in the tumor microenvironment,which indicates S100A4 is a new potential target for glioma immunotherapy.