Project description:High-grade gliomas are malignant neuroepithelial tumors. The current WHO classification for adult-type diffuse gliomas is based on IDH1/2 mutational and 1p/19q-codeletion status, which can be refined by emerging genomics, transcriptomics, and methylomics approaches. Nevertheless, therapy has been dominated by radiochemotherapy for about 15 years and progressed little in efficacy or precision through advanced molecular patient stratification. Glioma proteome alterations remain undercharacterized despite their promise for a better stratification and, in particular, the identification of therapeutic targets. Here, we used mass spectrometry (MS) to characterize 42 formalin-fixed, paraffin-embedded (FFPE) samples from IDH-wildtype (IDHwt), IDH-mutant (IDHmut) gliomas with and without 1p/19q-codeletion, and non-neoplastic brain tissue controls. Based on more than 5500 quantified proteins and 5000 phospho-sites, gliomas separated according to IDH1/2 mutational status and reflected loss of proteins encoded on 1p/19q in IDHmut, gains on chromosome 7 and losses on chromosome 10 in IDHwt. Unexpectedly, the 1p/19q-codeletion resulted in minor proteome changes. Instead, two proteomic subtypes of IDHmut gliomas showed major perturbations of mitochondrial DNA-encoded proteins, aerobic/anaerobic energy metabolism, RNA metabolism, chromatin dynamics, the extracellular matrix as well as tumor suppressor and oncoproteins. Reanalysis of three glioma proteomics studies independently validate the observed hallmarks and associate these proteomic subtypes with the well-established proneural and classic/mesenchymal IDHwt glioma subtypes. These results suggest an alternative stratification of IDHmut gliomas as part of common phenotypic subtypes independent of the IDH status, with broad therapeutic implications for patients with IDHmut gliomas in the future. Altogether, our study provides a rich protein and phospho-site resource restratifying glioma subtypes and supporting future mechanism of action and target discovery investigations.

Project description:Intratumoral microglia and MΦ constitute up to 70% of the tumor mass of high-grade gliomas (HGG) with profound impact on hallmarks of malignancy such as angiogenesis and immunosuppression. The dynamics and functional states of intratumoral myeloid cells during tumor progression and the molecular mechanisms controlling them are poorly understood. Here we define homeostatic and antigen-presenting myeloid cellular states in experimental and human HGG by longitudinal single-cell RNA-sequencing and combined transcriptome and proteome profiling. During glioma progression, myeloid cells gradually shift from a homeostatic to a tumor-associated effector state. We show that these dynamics are under strict control by early changes in resident microglia and the tumor genotype: In gliomas with mutations in isocitrate dehydrogenase (IDH), a disease-defining driver mutation, differentiation of invaded myeloid cells was blocked resulting in an immature, immunosuppressive phenotype. In late-stage IDH-mutant gliomas, monocyte-derived MΦ drive a tolerogenic remodeling of the glioma microenvironment thus preventing T-cell response. We define the molecular mechanism responsible for the tumor genotype-dependent education of infiltrating MΦ to be causally related to an enzymatically enhanced tryptophan catabolism via TDO2, resulting in the production of kynurenine, an endogenous ligand of the aryl hydrocarbon receptor (AHR). TDO2 activation further induces an amino acid starvation response triggering the import of exogenous tryptophan by intratumoral MΦ via LAT1-CD98. We here provide evidence that paracrine R-2-HG and tryptophan are critically involved in the differentiation and activation of monocyte-derived MΦ and that the previously observed altered amino acid metabolism in IDHmut gliomas is also responsible for shaping an immunosuppressive tumor microenvironment through maintenance of this complex metabolic axis. We further show that this regulatory metabolic network is particularly active in infiltrating MΦ because of their distinct expression profile that constitutes an immune subset-specific metabolic vulnerability. Consequently, the immunosuppressive phenotype in IDH-mutant glioma models was reversed by pharmacological inhibition of LAT1-CD98 or AHR. Thus, we provide evidence for a tumor genotype-dependent, dynamic network of resident and recruited intratumoral myeloid cells that shape the immune microenvironment of IDH-mutant HGG through pleiotropic interaction with the tumor metabolome and identify tryptophan metabolism as a viable therapeutic target for the immunotherapy of IDH-mutant tumors.

Project description:Background: Molecular profiling of diffuse gliomas has provided significant insights into the pathogenesis, classification and prognostication of these malignancies. However, previous molecular studies of glioma have largely focused on genomic readouts and targeted proteomic profiling technologies. Consequently, proteomic and downstream functional landscape of gliomas in general, and molecular subgroups in particular, remains largely unexplored. Here, we utilize liquid chromatography tandem mass spectrometry to profile genomically-defined cohorts of gliomas spanning the full range of World Health Organization (WHO) grades. Methods: Bulk frozen tissue and purified micro-dissected regions from formalin-fixed paraffin-embedded (FFPE) tissues were assembled and utilized to define robust proteomic signatures of both low grade, infiltrative and high-grade tumors. As a final analysis, primary tumor tissue was compared with both IDH-mutated and IDH-wildtype glioblastoma stem cell (GSC) lines to further overcome tissue heterogeneity and pinpoint proteins differences likely arising in the relevant glial cellular drivers of tumor development. Results: In aggregate, 5,496 unique proteins over 3 glioma cohorts were identified, and span common molecular subclasses based on IDH and 1p19q co-deletion status and all four WHO grades. Supervised clustering highlights substantial proteome and systems-level pathway differences between different genetically defined glioma subtypes and WHO grades. By using bulk tumor statistical analysis, 833 proteins distinguish different WHO grade tumors, while FFPE tumor dissection reveals 287 proteins in GBMs with abundance changes according to IDH mutation status. Using our integrative approach, calcium signaling, proteins of the endoplasmic reticulum and extracellular integrin proteins are most conserved proteomic markers that distinguish aggressive, IDH-wt, from IDH-mut GBM tumors in primary and tissue culture models gliomagenesis. Conclusions: This proteomic survey provides the largest and most diverse unbiased protein-based brain tumor resource to date. Current treatments for glial tumors are largely non-specific and overlap between genomic subtypes and WHO grades. Our analysis provides early insight into the vast downstream and epigenetic protein-level differences within this molecular framework. Given the central position proteins occupy in driving biology and phenotype, further characterization of the substantial proteomic diversity that exist between the molecular subtypes and grades of gliomas, proteomics may help define more personalized prognostic and predictive biomarkers for precision care.

Dataset Information

The epigenetic evolution of glioma is determined by the IDH1 mutation status and treatment regimen

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure