Using an adeno-associated virus–mediated, direct in vivo CRISPR screen, the authors mapped a quantitative landscape of glioblastoma suppressors. Their study revealed gene combinations that functionally drive gliomagenesis from normal glia in native mouse brains. The authors further demonstrate that mutational background can differentially influence gene expression and chemotherapeutic resistance. A causative understanding of genetic factors that regulate glioblastoma pathogenesis is of central importance. Here we developed an adeno-associated virus–mediated, autochthonous genetic CRISPR screen in glioblastoma. Stereotaxic delivery of a virus library targeting genes commonly mutated in human cancers into the brains of conditional-Cas9 mice resulted in tumors that recapitulate human glioblastoma. Capture sequencing revealed diverse mutational profiles across tumors. The mutation frequencies in mice correlated with those in two independent patient cohorts. Co-mutation analysis identified co-occurring driver combinations such as B2m–Nf1, Mll3–Nf1 and Zc3h13–Rb1, which were subsequently validated using AAV minipools. Distinct from Nf1-mutant tumors, Rb1-mutant tumors are undifferentiated and aberrantly express homeobox gene clusters. The addition of Zc3h13 or Pten mutations altered the gene expression profiles of Rb1 mutants, rendering them more resistant to temozolomide. Our study provides a functional landscape of gliomagenesis suppressors in vivo.

The role of NF-κB in the expression of inflammatory genes and its participation in the overall inflammatory process of chronic diseases and acute tissue injury are well established. We and others have demonstrated a critical involvement of poly(ADP-ribose) polymerase (PARP)-1 during inflammation, in part, through its relationship with NF-κB. However, the mechanism by which PARP-1 affects NF-κB activation has been elusive. In this study, we show that PARP-1 inhibition by gene knockout, knockdown, or pharmacologic blockade prevented p65 NF-κB nuclear translocation in smooth muscle cells upon TLR4 stimulation, NF-κB DNA-binding activity, and subsequent inducible NO synthase and ICAM-1 expression. Such defects were reversed by reconstitution of PARP-1 expression. PARP-1 was dispensable for LPS-induced IκBα phosphorylation and subsequent degradation but was required for p65 NF-κB phosphorylation. A perinuclear p65 NF-κB localization in LPS-treated PARP-1^−/− cells was associated with an export rather an import defect. Indeed, whereas PARP-1 deficiency did not alter expression of importin α3 and importin α4 and their cytosolic localization, the cytosolic levels of exportin (Crm)-1 were increased. Crm1 inhibition promoted p65 NF-κB nuclear accumulation as well as reversed LPS-induced p65 NF-κB phosphorylation and inducible NO synthase and ICAM-1 expression. Interestingly, p65 NF-κB poly(ADP-ribosyl)ation decreased its interaction with Crm1 in vitro. Pharmacologic inhibition of PARP-1 increased p65 NF-κB–Crm1 interaction in LPS-treated smooth muscle cells. These results suggest that p65 NF-κB poly(ADP-ribosyl)ation may be a critical determinant for the interaction with Crm1 and its nuclear retention upon TLR4 stimulation. These results provide novel insights into the mechanism by which PARP-1 promotes NF-κB nuclear retention, which ultimately can influence NF-κB–dependent gene regulation.