Back

Chronic AMPK inactivation slows SHH medulloblastoma progression by inhibiting mTORC1 signaling and depleting tumor stem cell populations

Dismuke, T.; Malawsky, D. S.; Liu, H.; Brenman, J.; Tikunov, A.; Gershon, T. R.

2021-12-09 cancer biology
10.1101/2021.12.09.471978 bioRxiv
Show abstract

We show that inactivating AMPK in vivo in a genetic model of medulloblastoma depletes tumor stem cell populations and slows tumor progression. Medulloblastoma, the most common malignant pediatric brain tumor, grows as heterogenous communities comprising diverse types of tumor and stromal cells. Previously, we showed that different populations in medulloblastomas show different sensitivities to specific targeted therapies. To determine if specific populations depend on AMPK, we analyzed mice with AMPK-inactivated medulloblastomas. We engineered mice with conditional deletion of the AMPK catalytic subunits Prkaa1 and Prkaa2 and conditional expression SmoM2, an oncogenic Smo allele that hyperactivates Sonic Hedgehog (SHH) signaling. We compared these medulloblastomas to SmoM2-driven medulloblastomas in AMPK-intact mice. AMPK-inactivation slowed tumor growth and progression, allowing longer event-free survival (EFS). scRNA-seq showed that AMPK inactivation altered cellular heterogeneity, increasing differentiation, decreasing tumor stem cell populations and reducing glio-neuronal multipotency. Surprisingly, AMPK-inactivated tumors showed decreased mTORC1 activation and Hk2 expression. Genetic Hk2 deletion in SmoM2-medulloblastomas similarly decreased stem cell populations, implicating reduced aerobic glycolysis in the tumor-suppressive effect of AMPK inactivation. Our results show that AMPK inactivation impairs tumor growth through mechanisms that disproportionately affect tumor stem cell populations that have proved refractory to conventional therapies.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.