Cancer progression involves the gradual loss of a differentiated phenotype and acquisition of progenitor and stem-cell-like features. Here, we provide novel stemness indices for assessing the degree of oncogenic dedifferentiation. We used an innovative one-class logistic regression (OCLR) machine-learning algorithm to extract transcriptomic and epigenetic feature sets derived from non-transformed pluripotent stem cells and their differentiated progeny. Using OCLR, we were able to identify previously undiscovered biological mechanisms associated with the dedifferentiated oncogenic state. Analyses of the tumor microenvironment revealed unanticipated correlation of cancer stemness with immune checkpoint expression and infiltrating immune cells. We found that the dedifferentiated oncogenic phenotype was generally most prominent in metastatic tumors. Application of our stemness indices to single-cell data revealed patterns of intra-tumor molecular heterogeneity. Finally, the indices allowed for the identification of novel targets and possible targeted therapies aimed at tumor differentiation.

Microglia are resident myeloid cells in the central nervous system (CNS) that control homeostasis and protect CNS from damage and infections. Microglia and peripheral myeloid cells accumulate and adapt tumor supporting roles in human glioblastomas that show prevalence in men. Cell heterogeneity and functional phenotypes of myeloid subpopulations in gliomas remain elusive. Here we show single-cell RNA sequencing (scRNA-seq) of CD11b⁺ myeloid cells in naïve and GL261 glioma-bearing mice that reveal distinct profiles of microglia, infiltrating monocytes/macrophages and CNS border-associated macrophages. We demonstrate an unforeseen molecular heterogeneity among myeloid cells in naïve and glioma-bearing brains, validate selected marker proteins and show distinct spatial distribution of identified subsets in experimental gliomas. We find higher expression of MHCII encoding genes in glioma-activated male microglia, which was corroborated in bulk and scRNA-seq data from human diffuse gliomas. Our data suggest that sex-specific gene expression in glioma-activated microglia may be relevant to the incidence and outcomes of glioma patients.

Diffuse infiltration of glioma cells into normal brain tissue is considered to be a main reason for the unfavorable outcomes of patients with malignant gliomas. Invasion of glioma cells into the brain parenchyma is facilitated by metalloprotease-mediated degradation of the extracellular matrix. Metalloproteases are released as inactive pro-forms and get activated upon cleavage by membrane bound metalloproteases. Here, we show that membrane type 1 metalloprotease (MT1-MMP) is up-regulated in glioma-associated microglia, but not in the glioma cells. Overexpression of MT1-MMP is even lethal for glioma cells. Glioma-released factors trigger the expression and activity of MT1-MMP via microglial toll-like receptors and the p38 MAPK pathway, as deletion of the toll-like receptor adapter protein MyD88 or p38 inhibition prevented MT1-MMP expression and activity in cultured microglial cells. Microglial MT1-MMP in turn activates glioma-derived pro-MMP-2 and promotes glioma expansion, as shown in an ex vivo model using MT1-MMP-deficient brain tissue and a microglia depletion paradigm. Finally, MyD88 deficiency or microglia depletion largely attenuated glioma expansion in 2 independent in vivo models.