Highlights d Genome-wide CRISPR-Cas9 screens in patient-derived glioblastoma stem cells d Identification of regulators of stemness governing glioblastoma stem cell growth d Multiple stress response pathways are genetic vulnerabilities in glioblastoma d Identification of modulators of sensitivity to standard of care chemotherapy

Developmental signal transduction pathways act diversely, with context-dependent roles across systems and disease types. Glioblastomas (GBMs), which are the poorest prognosis primary brain cancers, strongly resemble developmental systems, but these growth processes have not been exploited therapeutically, likely in part due to the extreme cellular and genetic heterogeneity observed in these tumors. The role of Wnt/βcatenin signaling in GBM stem cell (GSC) renewal and fate decisions remains controversial. Here, we report context-specific actions of Wnt/ βcatenin signaling in directing cellular fate specification and renewal. A subset of primary GBM-derived stem cells requires Wnt proteins for self-renewal, and this subset specifically relies on Wnt/βcatenin signaling for enhanced tumor burden in xenograft models. In an orthotopic Wnt reporter model, Wnt hi GBM cells (which exhibit high levels of βcatenin signaling) are a faster-cycling, highly self-renewing stem cell pool. In contrast, Wnt lo cells (with low levels of signaling) are slower cycling and have decreased self-renewing potential. Dual inhibition of Wnt/βcatenin and Notch signaling in GSCs that express high levels of the proneural transcription factor ASCL1 leads to robust neuronal differentiation and inhibits clonogenic potential. Our work identifies new contexts for Wnt modulation for targeting stem cell differentiation and self-renewal in GBM heterogeneity, which deserve further exploration therapeutically.

Chromatin accessibility discriminates stem from mature cell populations, enabling the identification of primitive stem-like cells in primary tumors, such as glioblastoma (GBM) where self-renewing cells driving cancer progression and recurrence are prime targets for therapeutic intervention. We show, using single-cell chromatin accessibility, that primary human GBMs harbor a heterogeneous self-renewing population whose diversity is captured in patient-derived glioblastoma stem cells (GSCs). In-depth characterization of chromatin accessibility in GSCs identifies three GSC states: Reactive, Constructive, and Invasive, each governed by uniquely essential transcription factors and present within GBMs in varying proportions. Orthotopic xenografts reveal that GSC states associate with survival, and identify an invasive GSC signature predictive of low patient survival, in line with the higher invasive properties of Invasive state GSCs compared to Reactive and Constructive GSCs as shown by in vitro and in vivo assays. Our chromatin-driven characterization of GSC states improves prognostic precision and identifies dependencies to guide combination therapies.

SummarySuccessful glioblastoma (GBM) therapies have remained elusive due to limitations in understanding mechanisms of growth and survival of the tumorigenic population. Using CRISPR-Cas9 approaches in patient-derived GBM stem cells to interrogate function of the coding genome, we identify diverse actionable pathways responsible for growth that reveal the gene-essential circuitry of GBM stemness. In particular, we describe the Sox developmental transcription factor family; H3K79 methylation by DOT1L; and ufmylation stress responsiveness programs as essential for GBM stemness. Additionally, we find mechanisms of temozolomide resistance and sensitivity that could lead to combination strategies with this standard of care treatment. By reaching beyond static genome analysis of bulk tumors, with a genome wide functional approach, we dive deep into a broad range of biological processes to provide new understanding of GBM growth and treatment resistance.SignificanceGlioblastoma (GBM) remains an incurable disease despite an increasingly thorough depth of knowledge of the genomic and epigenomic alterations of bulk tumors. Evidence from multiple approaches support that GBM reflects an aberrant developmental hierarchy, with GBM stem cells (GSCs), fueling tumor growth and invasion. The properties of this tumor subpopulation may also in part explain treatment resistance and disease recurrence. Unfortunately, we still have a limited knowledge of the molecular circuitry of these cells and progress has been slow as we have not been able, until recently, to interrogate function at the genome-wide scale. Here, using parallel genome-wide CRISPR-Cas9 screens, we identify the essential genes for GSC growth. Further, by screening in the presence of low and high dose temozolomide, we identify mechanisms of drug resistance and sensitivity. These functional screens in patient derived cells reveal new aspects of GBM biology and identify a diversity of actionable targets such as genes governing stem cell traits, epigenome regulation and the response to stress stimuli.

No abstract

How human cells coordinate various metabolic processes, such as glycolysis and protein translation, remains unclear. One key insight is that various metabolic enzymes have been found to associate with mRNAs, however whether these enzymes regulate mRNA biology in response to changes in cellular metabolic state remains unknown. Here we report that the glycolytic enzyme, pyruvate kinase M (PKM), inhibits the translation of 7% of the transcriptome in response to elevated levels of glucose and pyruvate.Our data suggest that in the presence of glucose and pyruvate, PKM associates with ribosomes that are synthesizing stretches of polyacidic nascent polypeptides and stalls the elongation step of translation.PKM-regulated mRNAs encode proteins required for the cell cycle and may explain previous results linking PKM to cell cycle regulation. Our study uncovers an unappreciated link between glycolysis and the ribosome that likely coordinates the intake of glycolytic metabolites with the regulation of protein synthesis and the cell cycle. Results and Discussion Mass spectrometry analysis of ER and cytosolic polysomes and mRNPsOver the past decade, numerous proteins that lack RNA binding domains, such as metabolic enzymes, have been shown to exhibit mRNA and ribosome binding 1-12 . It is unclear whether these unconventional RNA binding proteins (RBP) associate with transcripts and ribosomes in a spatially defined manner, or if they link metabolic states to mRNA stability or translation. To determine the spatial distribution of RNA-, and ribosome-binding proteins, we isolated ER and cytosolic fractions from human osteosarcoma (U2OS) cells ( Figure 1A) and sedimented crude polysomes. The cytosol and ER represent the major division in cellular protein synthesis, each containing distinct pools of mRNAs and unique translational regulatory systems [13][14][15][16] . These isolated polysomes were then treated with RNase to liberate RNA-binding proteins ("RNA-bound fraction"), and resedimented to pellet ribosomes and associated proteins ("Ribosome-bound fraction"; Figure 1B). We then analyzed the composition of the RNA-bound and Ribosome-bound fractions ( Figure 1B-C) by mass spectrometry, as previously described 17 . In parallel we also isolated messenger ribonuclear protein (mRNP) complexes from the ER and cytosol using oligo-dT affinity chromatography ("mRNP-bound"; Figure 1B, D, "dT") and again analyzed these fractions by mass spectrometry. To control for non-specific binding to the oligo-dT resin we also performed the affinity chromatography with beads lacking any nucleic acid ( Figure 1B, "Mock Beads", 1D "B"). Our purification conditions, done in the absence of crosslinking, enabled recovery of proteins that are directly and indirectly bound to mRNAs and/or ribosomes.After statistical processing (see Methods), 370 proteins were present in the RNA-bound fraction, 414 were present in the ribosome-bound fraction, and 2690 proteins were enriched in the mRNPassociated fraction. Upon further manual curation (see methods), 496 protei...

41Chromatin accessibility discriminates stem from mature cell populations, enabling the 42 identification of primitive stem-like cells in primary tumors, such as Glioblastoma (GBM) where 43 self-renewing cells driving cancer progression and recurrence are prime targets for therapeutic 44 intervention. We show, using single-cell chromatin accessibility, that primary GBMs harbor a 45 heterogeneous self-renewing population whose diversity is captured in patient-derived 46 glioblastoma stem cells (GSCs). In depth characterization of chromatin accessibility in GSCs 47 identifies three GSC states: Reactive, Constructive, and Invasive, each governed by uniquely 48 essential transcription factors and present within GBMs in varying proportions. Orthotopic 49 xenografts reveal that GSC states associate with survival, and identify an invasive GSC signature 50 predictive of low patient survival. Our chromatin-driven characterization of GSC states improves 51 prognostic precision and identifies dependencies to guide combination therapies. 52 53 56only an additional 2.5 months in the small subset of responsive patients [2]. Despite extensive 57 characterization and stratification of the bulk primary tumors, no targeted therapies have been 58 successfully developed [1,3]. GBM tumors are rooted in self-renewing tumor-initiating cells 59 commonly referred to as glioblastoma stem cells (GSCs)[4] that drive disease progression in 60 vivo [5,6] and display resistance to chemo-and radiotherapy leading to disease recurrence [7]. 61 The promise of therapeutically targeting self-renewing tumor-initiating cancer cells depends on 62 3 our capacity to capture the full range of heterogeneity within this population from individual 63 tumors. Intratumoral heterogeneity within primary GBM has recently been documented 64 through single cell RNA-seq experiments and revealed a continuum between four cellular 65 states[8]: neural-progenitor-like (NPC), oligodendrocyte-progenitor-like (OPC), astrocyte-like 66 (AC), and mesenchymal-like (MES)[8]. A subsequent study[9] using single-cell gene-centric 67 enrichment analysis placed GBM cells along a single axis of variation from proneural to 68 mesenchymal transcriptional profiles, with cells expressing stem-associated genes lying at the 69 extremes of this axis. Hence, primary GBM consists of distinct states, across which stem-like 70 cells appear to be found. Whether these stem-like cells found across GBM states represent 71 functionally distinct GSC populations with tumor-initiating properties and unique dependencies 72 remains to be established to guide therapeutic progress. To address this issue, we combined 73 single-cell technologies to define GSC composition in primary GBM with functional assays to 74 reveal the unique dependencies across GSCs, reflective of invasive, constructive and reactive 75 states that relate to patient outcome. 76 77 RESULTS 78Chromatin accessibility readily discriminates stem from mature cell populations [10], 79 which can be resolved at the single cell level thro...

In light of the numerous studies identifying post-transcriptional regulators on the surface of the endoplasmic reticulum (ER), we asked whether there are factors that regulate compartment specific mRNA translation in human cells. Using a proteomic survey of spatially regulated polysome interacting proteins, we identified the glycolytic enzyme Pyruvate Kinase M (PKM) as a cytosolic (i.e. ER-excluded) polysome interactor and investigated how it influences mRNA translation. We discovered that the PKM-polysome interaction is directly regulated by ADP levels–providing a link between carbohydrate metabolism and mRNA translation. By performing enhanced crosslinking immunoprecipitation-sequencing (eCLIP-seq), we found that PKM crosslinks to mRNA sequences that are immediately downstream of regions that encode lysine- and glutamate-enriched tracts. Using ribosome footprint protection sequencing, we found that PKM binding to ribosomes causes translational stalling near lysine and glutamate encoding sequences. Lastly, we observed that PKM recruitment to polysomes is dependent on poly-ADP ribosylation activity (PARylation)—and may depend on co-translational PARylation of lysine and glutamate residues of nascent polypeptide chains. Overall, our study uncovers a novel role for PKM in post-transcriptional gene regulation, linking cellular metabolism and mRNA translation.