mTORC1 promotes malignant large cell/anaplastic histology and is a targetable vulnerability in SHH-TP53 mutant medulloblastoma
Valentina Conti, … , Pietro L. Poliani, Rossella Galli
Valentina Conti, … , Pietro L. Poliani, Rossella Galli
Published October 21, 2021
Citation Information: JCI Insight. 2021;6(23):e153462. https://doi.org/10.1172/jci.insight.153462.
View: Text | PDF
Research Article Neuroscience Oncology

mTORC1 promotes malignant large cell/anaplastic histology and is a targetable vulnerability in SHH-TP53 mutant medulloblastoma

Abstract

Medulloblastoma (MB), one of the most malignant brain tumors of childhood, comprises distinct molecular subgroups, with p53 mutant sonic hedgehog–activated (SHH-activated) MB patients having a very severe outcome that is associated with unfavorable histological large cell/anaplastic (LC/A) features. To identify the molecular underpinnings of this phenotype, we analyzed a large cohort of MB developing in p53-deficient Ptch+/– SHH mice that, unexpectedly, showed LC/A traits that correlated with mTORC1 hyperactivation. Mechanistically, mTORC1 hyperactivation was mediated by a decrease in the p53-dependent expression of mTORC1 negative regulator Tsc2. Ectopic mTORC1 activation in mouse MB cancer stem cells (CSCs) promoted the in vivo acquisition of LC/A features and increased malignancy; accordingly, mTORC1 inhibition in p53-mutant Ptch+/– SHH MB and CSC-derived MB resulted in reduced tumor burden and aggressiveness. Most remarkably, mTORC1 hyperactivation was detected only in p53-mutant SHH MB patient samples, and treatment with rapamycin of a human preclinical model phenocopying this subgroup decreased tumor growth and malignancy. Thus, mTORC1 may act as a specific druggable target for this subset of SHH MB, resulting in the implementation of a stringent risk stratification and in the potentially rapid translation of this precision medicine approach into the clinical setting.

Authors

Valentina Conti, Manuela Cominelli, Valentina Pieri, Alberto L. Gallotti, Ilaria Pagano, Matteo Zanella, Stefania Mazzoleni, Flavia Pivetta, Monica Patanè, Giulia M. Scotti, Ignazio S. Piras, Bianca Pollo, Andrea Falini, Alessio Zippo, Antonella Castellano, Roberta Maestro, Pietro L. Poliani, Rossella Galli

×

Figure 8

mTORC1 activation is specifically found in human p53 mutant SHH MB with LC/A component and may be a subgroup-specific therapeutic vulnerability.

Options: View larger image (or click on image) Download as PowerPoint
mTORC1 activation is specifically found in human p53 mutant SHH MB with ...
(A) High numbers of tumor cells positive for the mTORC1 surrogate markers pS6 and p4EBP1 (cytoplasmic staining, brown) are observed only in human LC/A MB belonging to the SHH subgroup with p53 mutation and are not detected in human desmoplastic/nodular (D/N) SHH p53wt, classic WNT, and LC/A non-WNT/non-SHH MB. Scale bars: 50 μm. (B) Quantification of pS6 and p4EBP1 expression in the different subgroups. Only statistically significant pairwise comparisons are shown. (C) GSEA indicates that the genes qualifying ‘human p53 mutant SHHα MBs with LC/A histology’ are significantly enriched in two different mTOR-related gene signatures. (D) Tumors from human DAOY cells after RhebQ64L transduction show increased frequency of pS6- and p4EBP1-IR tumor cells, which are very few in mock tumors and are reduced in number after treatment with rapamycin. RhebQ64L transduction promotes the acquisition of LC/A traits, such as the presence of large cells with prominent nucleoli (white arrowheads), which are not observed in controls and are strongly diminished by rapamycin (H&E). The activation of pNDRG1 and the expression of GPNMB are both increased in RhebQ64L-transduced MB and turned off by rapamycin. Scale bars: 50 μm. Quantification of the level of marker expression is shown in the graphs. (E) The expression of YAP1, N-MYC, GLI2, SOX2, and c-MYC is enhanced by mTORC1 hyperactivation and decreased by rapamycin administration. Scale bars: 50 μm. Quantification of the level of marker expression is shown in the graphs. Quantitative data are represented as a box-and-whisker plot, with bounds from 25th to 75th percentile, median line, and whiskers ranging from minimum to maximum values. One-way ANOVA followed by Tukey’s multiple comparison test. *P < 0.05; **P < 0.01; ***P < 0.005; ****P < 0.001.