The diversity of molecular states and cellular plasticity of immune cells within the glioblastoma (GBM) environment remain poorly investigated. Here, we conduct deep transcriptional profiling of lymphoid and myeloid cell populations by scRNA-sequencing, map potential cellular interactions and cytokine responses that lead to the dysfunctional and exhausted phenotype of T cells. We identified Interleukin 10 (IL-10) response during T cell activation, which lead to a dysfunctional state of T cells. By the use of a novel method: The nearest functionally connected neighbor (NFCN), an in-silico model to explore cell-cell interaction, the dysfunctional/exhausted phenotype was found to be driven by subset of myeloid cells defined by high expression of HMOX1. By using spatial transcriptomic RNA-sequencing, we identified a correlation between T cell exhaustion and colocalized mesenchymal gene expression. We found that HMOX1 expressing myeloid cells occupying regions marked by T cell exhaustion. Using a human neocortical slice model with myeloid cell depletion we confirmed the functional interaction of myeloid and lymphoid cell leading to the dysfunctional state of T cells. A comprehensive understanding of cellular states and plasticity of lymphoid cells in GBM aids in providing successful immunotherapeutic approaches.