EBF1 is expressed in pericytes and contributes to pericyte cell commitment.

Early B-cell factor-1 (EBF1) is a transcription factor with an important role in cell lineage specification and commitment during the early stage of cell maturation. Originally described during B-cell maturation, EBF1 was subsequently identified as a crucial molecule for proper cell fate commitment of mesenchymal stem cells into adipocytes, osteoblasts and muscle cells. In vessels, EBF1 expression and function have never been documented. Our data indicate that EBF1 is highly expressed in peri-endothelial cells in both tumor vessels and in physiological conditions. Immunohistochemistry, quantitative reverse transcription polymerase chain reaction (RT-qPCR) and fluorescence-activated cell sorting (FACS) analysis suggest that EBF1-expressing peri-endothelial cells represent bona fide pericytes and selectively express well-recognized markers employed in the identification of the pericyte phenotype (SMA, PDGFRß, CD146, NG2). This observation was also confirmed in vitro in human placenta-derived pericytes and in human brain vascular pericytes (HBVP). Of note, in accord with the key role of EBF1 in the cell lineage commitment of mesenchymal stem cells, EBF1-silenced HBVP cells showed a significant reduction in PDGFRß and CD146, but not CD90, a marker mostly associated with a prominent mesenchymal phenotype. Moreover, the expression levels of VEGF, angiopoietin-1, NG2 and TGF-ß, cytokines produced by pericytes during angiogenesis and linked to their differentiation and activation, were also significantly reduced. Overall, the data suggest a functional role of EBF1 in the cell fate commitment toward the pericyte phenotype.

The proneural gene ASCL1 governs the transcriptional subgroup affiliation in glioblastoma stem cells by directly repressing the mesenchymal gene NDRG1.

Achaete-scute homolog 1 gene (ASCL1) is a gene classifier for the proneural (PN) transcriptional subgroup of glioblastoma (GBM) that has a relevant role in the neuronal-like differentiation of GBM cancer stem cells (CSCs) through the activation of a PN gene signature. Besides prototypical ASCL1 PN target genes, the molecular effectors mediating ASCL1 function in regulating GBM differentiation and, most relevantly, subgroup specification are currently unknown. Here we report that ASCL1 not only promotes the acquisition of a PN phenotype in CSCs by inducing a glial-to-neuronal lineage switch but also concomitantly represses mesenchymal (MES) features by directly downregulating the expression of N-Myc downstream-regulated gene 1 (NDRG1), which we propose as a novel gene classifier of MES GBMs. Increasing the expression of ASCL1 in PN CSCs results in suppression of self-renewal, promotion of differentiation and, most significantly, decrease in tumorigenesis, which is also reproduced by NDRG1 silencing. Conversely, both abrogation of ASCL1 expression in PN CSCs and enforcement of NDRG1 expression in either PN or MES CSCs induce proneural-to-mesenchymal transition (PMT) and enhanced mesenchymal features. Surprisingly, ASCL1 overexpression in MES CSCs increases malignant features and gives rise to a neuroendocrine-like secretory phenotype. Altogether, our results propose that the fine interplay between ASCL1 and its target NDRG1 might serve as potential subgroup-specific targetable vulnerability in GBM; enhancing ASCL1 expression in PN GBMs might reduce tumorigenesis, whereas repressing NDRG1 expression might be actionable to hamper the malignancy of GBM belonging to the MES subgroup.

Tuberous sclerosis complex-associated CNS abnormalities depend on hyperactivation of mTORC1 and Akt.

Tuberous sclerosis complex (TSC) is a dominantly inherited disease caused by hyperactivation of the mTORC1 pathway and characterized by the development of hamartomas and benign tumors, including in the brain. Among the neurological manifestations associated with TSC, the tumor progression of static subependymal nodules (SENs) into subependymal giant cell astrocytomas (SEGAs) is one of the major causes of morbidity and shortened life expectancy. To date, mouse modeling has failed in reproducing these 2 lesions. Here we report that simultaneous hyperactivation of mTORC1 and Akt pathways by codeletion of Tsc1 and Pten, selectively in postnatal neural stem cells (pNSCs), is required for the formation of bona fide SENs and SEGAs. Notably, both lesions closely recapitulate the pathognomonic morphological and molecular features of the corresponding human abnormalities. The establishment of long-term expanding pNSC lines from mouse SENs and SEGAs made possible the identification of mTORC2 as one of the mediators conferring tumorigenic potential to SEGA pNSCs. Notably, in spite of concurrent Akt hyperactivation in mouse brain lesions, single mTOR inhibition by rapamycin was sufficient to strongly impair mouse SEGA growth. This study provides evidence that, concomitant with mTORC1 hyperactivation, sustained activation of Akt and mTORC2 in pNSCs is a mandatory step for the induction of SENs and SEGAs, and, at the same time, makes available an unprecedented NSC-based in vivo/in vitro model to be exploited for identifying actionable targets in TSC.