TAp73 is a marker of glutamine addiction in medulloblastoma.

Medulloblastoma is the most common solid primary brain tumor in children. Remarkable advancements in the understanding of the genetic and epigenetic basis of these tumors have informed their recent molecular classification. However, the genotype/phenotype correlation of the subgroups remains largely uncharacterized. In particular, the metabolic phenotype is of great interest because of its druggability, which could lead to the development of novel and more tailored therapies for a subset of medulloblastoma. p73 plays a critical role in a range of cellular metabolic processes. We show overexpression of p73 in a proportion of non-WNT medulloblastoma. In these tumors, p73 sustains cell growth and proliferation via regulation of glutamine metabolism. We validated our results in a xenograft model in which we observed an increase in survival time in mice on a glutamine restriction diet. Notably, glutamine starvation has a synergistic effect with cisplatin, a component of the current medulloblastoma chemotherapy. These findings raise the possibility that glutamine depletion can be used as an adjuvant treatment for p73-expressing medulloblastoma.

Camptothecin and Fas receptor agonists synergistically induce medulloblastoma cell death: ROS-dependent mechanisms.

Medulloblastoma, a common malignant pediatric brain tumor, is highly resistant to death receptor-mediated apoptosis despite death receptor expression by tumor cells. Developing new strategies to overcome this resistance to death receptor activation could positively impact therapeutic outcomes. We explored the modulation of death receptor-induced medulloblastoma cell death by the topoisomerase I inhibitor camptothecin (CPT). CPT significantly increased the human medulloblastoma DAOY cell death response to agonistic anti-Fas antibody (CH-11). Cell death after CPT, CH-11, and CPT+CH-11 treatment was 9, 7, and 33%, respectively. Isobologram analysis showed that CH-11 and CPT act synergistically to induce cell death in DAOY cells. A similar pattern of synergism between CPT and CH-11 was found in ONS-76 medulloblastoma cells. Synergistic cell death was found to be predominantly apoptotic involving both extrinsic and intrinsic pathways as evidenced by annexin V staining, cleavage of caspases (3, 8, and 9), Bid and PARP, and cytoprotection by caspase inhibitors. Flow cytometric analyses showed that expression of cell surface Fas or Fas ligand did not change with drug treatment. Western blot analyses showed that the combination of CH-11+CPT induced a significant decrease in XIAP levels. Furthermore, reactive oxygen species, especially O2, were elevated after CPT treatment, and even more so by the CH-11+CPT treatment. The antioxidants glutathione and N-acetyl-cysteine prevented cell death induced by CPT+CH-11. Moreover, the mitochondrial respiratory chain complex I inhibitor rotenone potentiated CH-11-induced apoptosis in DAOY cells. Taken together, these findings show that CPT synergizes with Fas activation to induce medulloblastoma apoptosis through a mechanism involving reactive oxygen species and oxidative stress pathways.

Epigenetic silencing of the HIC-1 gene in human medulloblastomas.

The HIC-1 (hypermethylated in cancer) candidate tumor suppressor gene is located on chromosome 17p13.3, a region frequently deleted in medulloblastomas (MBs). MBs arising in the cerebellum represent the most common malignant brain tumors in children. In this study we have analyzed the sequence, methylation, and expression status of the HIC-1 gene in MBs. Hypermethylation of the 5'UTR and/or second exon of HIC-1 was detected in 33/39 (85%) of MB biopsies and in 7/8 (88%) of MB cell lines by methylation-specific PCR. There was a significant correlation (p < 0.001) between HIC-1 methylation and lack of transcription as determined by competitive RT-PCR. Treatment of the MB cell lines Daoy and MEB-MED-8A with 5-aza-2'deoxycytidine led to re-expression of HIC-1 transcripts, indicating a silencing of HIC-1 by CpG island methylation. Mutation analysis of the coding region of HIC-1 revealed a single deletion leading to an in-frame deletion of 4 amino acids in the second exon of HIC-1 (1/68, 1.5%). Our data indicate that a significant number of MBs exhibit strikingly reduced HIC-1 expression caused by altered CpG island methylation. These data suggest that epigenetic silencing of HIC-1 may well contribute to the pathogenesis in the majority of MBs.