gms | German Medical Science

Objective: In malignant brain tumours, crosstalk and adaptation of brain resident and infiltrating myeloid cells form an immunosuppressive environment that causes dysfunctional activation of the T cells and leads to an inadequate immune response. Most recently, checkpoint inhibition (PD-1) failed to rescue the lost T cell response, highlighting that dysfunctional/exhausted T-cell activation in glioma is insufficiently understood. Here, we used single-cell RNA-sequencing analysis and cell-cell coactivation models to map cell fate decisions in T-cell activation.

Methods: The diversity of different myeloid cell-types was explored by multi-channel FACS analysis. Single-cell RNA-sequencing of 4 patients was performed by magnetic-bead enrichment of CD3 positive cells and removal of dead-cells by negative selection (MACS dead-cell removal) resulting in a stable vital cell population between 90-98%. The single-cell emulsion was formed by a microfluidic device and cells were barcoded by encapsulated gel beads containing unique molecular identifiers. After library construction and next-generation sequencing, we used a novel machine learning approach to model lineage development and cell fate decisions as well as cell-cell interactions.

Results: After quality control, 17.705 cells remained with an average of ~20.000 reads per cell. After batch removal and adaptation of stress-induced pathway activation we identified 15 clusters highly associated to T-cell lineages (CD4+/CD8+/γδT,Treg), macrophages (CD163+), microglia (TMEM119+), monocytes (LYZ+), dendritic cells and glial lineages (OPC and oligodendrocytes). Integrative data analysis identified cell fate decisions based on IL10 induced transformation in effector T-cells causing the dysfunctional state in T cells. Prolonged proliferation of a subset of dysfunctional T-cells induced apoptosis at a later stage. Dysfunctional activation was traced back to a distinct subpopulation within the microglia and macrophage cluster marked by high expression of SPP1.

Conclusion: We identified a novel subpopulation of the tumour microenvironment which aids the evolution of dysfunctional T-cells by IL10 release. Targeting of the tumour microenvironment may provide promising therapy strategies in the future.