A diverse array of cancer-associated MTOR mutations are hyperactivating and can predict rapamycin sensitivity

Abstract

Genes encoding components of the PI3K-AKT-mTOR signaling axis are frequently mutated in cancer, but few mutations have been characterized in MTOR, the gene encoding the mTOR kinase. Using publicly available tumor genome sequencing data, we generated a comprehensive catalog of mTOR pathway mutations in cancer, identifying 33 MTOR mutations that confer pathway hyperactivation. The mutations cluster in six distinct regions in the C-terminal half of mTOR and occur in multiple cancer types, with one cluster particularly prominent in kidney cancer. The activating mutations do not affect mTOR complex assembly, but a subset reduces binding to the mTOR inhibitor DEPTOR. mTOR complex 1 (mTORC1) signaling in cells expressing various activating mutations remains sensitive to pharmacologic mTOR inhibition, but is partially resistant to nutrient deprivation. Finally, cancer cell lines with hyperactivating MTOR mutations display heightened sensitivity to rapamycin both in culture and in vivo xenografts, suggesting that such mutations confer mTOR pathway dependency.

Figure 1. Cancer-associated MTOR mutations

1A – The domain structure of MTOR is shown. 1B – The graph indicates the number of samples with MTOR mutations at the indicated amino acid positions. Red colored samples were tested for pathway activation in Figure 2. 1C – A density plot of the same MTOR mutations from 1B identifies six distinct mutational clusters with a p-value <0.05. 1D – The number of samples with MTOR mutations from a given tissue types varies across the length of the gene; the intensity of color reflects the number of samples from a given cancer subtype in that region. 1E – Cancer sequencing studies with 5% or more of samples harboring MTOR point mutations is shown.

Figure 2. mTORC1 and mTORC2 activating mutations in MTOR and RHEB

2A – Several distinct mTOR mutants induce S6K1 phosphorylation. HEK-293T cells were co-transfected with HA-GST-S6K1 and mTOR WT or mutant cDNAs in expression vectors followed by whole cell lysis 48 hours after transfection. The lysates were then immunoblotted for the indicated proteins and the phosphorylation state of S6K1. Densitometry of pS6K1 vs. HA-GST-S6K1 from three separate experiments is shown on the right, the error bars represent SEM. 2B – Several distinct mTOR mutants induce 4EBP1 phosphorylation. HEK-293T cells were cotransfected with HA-GST-4EBP1 and mTOR WT or mutant cDNAs in expression vectors followed by whole cell lysis 48 hours after transfection. The lysates were then immunoblotted for the indicated proteins and the phosphorylation state of 4EBP1. Densitometry of p4EBP1 vs. HA-GST-4EBP1 from three separate experiments is shown on the right, the error bars represent SEM. 2C – Several mTOR mutants induce Akt1 phosphorylation. HEK-293E cells were co-transfected with HAGST-Akt1 and mTOR WT or mutant cDNAs in expression vectors followed by whole cell lysis 48 hours after transfection and an overnight 0.5% serum starvation. The lysates were then immunoblotted for the indicated proteins and the phosphorylation state of Akt1. Densitometry of pAkt1 vs. HA-GST-Akt1 from three separate experiments is shown on the right, the error bars represent SEM. 2D – The C1483 cluster harbors non-recurrent mTORC1-activating mutations from kidney cancer patients. HEK-293T cells were co-transfected with HA-GST-S6K1 and mTOR WT or mutant cDNAs in expression vectors followed by whole cell lysis 48 hours after transfection. The lysates were then immunoblotted as above. Densitometry of pS6K1 vs. HA-GST-S6K1 from three separate experiments is shown on the right, the error bars represent SEM. 2E – Rheb1 mutants induce S6K1 phosphorylation. HEK-293T cells were transfected with Rheb1 WT or mutant cDNAs in expression vectors followed by whole cell lysis 48 hours after transfection. The lysates were then immunoblotted as above. Densitometry of pS6K1 vs. total S6K1 from three separate experiments is shown on the right, the error bars represent SEM. In all cases, one asterisk indicates a p value less than 0.1, two asterisks indicates a p value less than 0.01.

Figure 3. mTOR mutants bind less Deptor and cells expressing the mTOR S2215Y mutation are resistant to nutrient deprivation

3A – Exogenously expressed WT and mutant mTOR co-immunoprecipitate equally well endogenous Raptor and Rictor, but have reduced Deptor binding. HEK-293T cells were co-transfected with mTOR WT or mutant cDNAs in expression vectors, lysates were prepared 48 hours after transfection, and the anti-FLAG immunoprecipitates or lysates were immunoblotted for the indicated proteins. 3B – mTOR WT and S2215Y are equally sensitive to pharmacological mTOR inhibition. Parental TREX cells or those with stable expression of mTOR WT or S2215Y were treated with rapamycin or Torin1 as described in the methods. Whole cell lysates were then immunoblotted as in Figure 2. 3C – The S2215Y mTOR mutant confers resistance to nutrient deprivation on S6K1 phosphorylation. Parental TREX cells or those with stable expression of mTOR WT or S2215Y were starved and re-fed amino acids or glucose as described in the methods. Whole cell lysates were analyzed with immunoblotting as above.

Figure 4. Rapamycin inhibits the proliferation of cancer cell lines with hyperactivating MTOR mutations

4A – Proliferation of multiple cancer cell lines with WT or mutant mTOR is differentially sensitive to rapamycin treatment. Cancer cell lines were seeded onto 96 well plates, treated with rapamycin 24 hours later, and cell viability was measured using Cell Titer Glo as described in the methods. Error bars indicate SD. 4B – The tissue of origin, MTOR mutational status, commonly mutated oncogenes/tumor suppressors and rapamycin IC50 from Figure 4A are shown for each cancer cell line. 4C – HEC59 cell xenografts are exquisitely sensitive to rapamycin treatment as compared to HeLa cell xenografts. Nine injections each of 3×10⁶ HeLa or HEC59 cells were implanted subcutaneously into the flanks of eight-week-old nude mice. One week after injection, the mice were treated daily with vehicle or rapamycin and tumor size was measured for an additional three weeks. Error bars indicate SEM, two asterisks indicate p-values less than 0.05, and three asterisks indicate p-values less than 0.01.