Abstract
Abstract 285
Activating mutations in NOTCH1 are common in T-cell lymphoblastic leukemias (T-ALL), making this receptor a promising target for drugs such as γ-secretase inhibitors (GSIs), which block a proteolytic cleavage required for NOTCH1 activation. Aberrant activation of the PI3K-AKT pathway due to mutational loss of PTEN is found in 20% of human T-ALLs and has been linked with in vitro resistance to GSIs therapy in T-ALL cell lines, suggesting that in the absence of PTEN, constitutive activation of the PI3K pathway may render T-cell lymphoblasts insensitive to NOTCH1 inhibition with GSIs. Still, cell lines frequently fail to recapitulate the biology of primary tumor cells, and in vitro studies fall short of addressing the role of the NOTCH-PI3K interaction in clinical resistance to GSI therapy, which is best defined as disease progression under treatment in vivo. Moreover, the specific role and mechanisms of GSI resistance downstream of PTEN loss in T-ALL remain to be elucidated. To address this question, and to analyze the actual significance of the interaction between NOTCH1 signaling and PTEN loss in T-cell transformation and therapy response we analyzed the response of NOTCH1 induced PTEN-positive and PTEN-deleted isogenic tumors to GSI therapy in vivo. Towards this goal we first generated NOTCH1 induced T-ALLs via bone marrow transplantation of tamoxifen-inducible conditional PTEN knockout (Rosa26TMCre PTEN flox/flox) hematopoietic progenitors infected with retroviruses expressing a mutant constitutively active form of the NOTCH1 receptor (NOTCH1 L1601P Δ-PEST). NOTCH1 L1601P Δ-PEST Rosa26TMCre PTEN flox/flox tumor cells injected into secondary recipients were treated with vehicle only or tamoxifen in order to generate PTEN-non-deleted and PTEN-deleted isogenic tumors, respectively. Treatment of PTEN-positive tumor bearing mice with DBZ, a highly active GSI, demonstrated marked responses to therapy by in vivo bioimaging compared with vehicle only treated controls, which translated into a significant improvement in survival (P < 0.005). In contrast, all mice harboring PTEN-deleted tumors failed to respond to DBZ treatment, showed overt progression under treatment and died of their disease demonstrating a direct role of PTEN loss in the development of resistance to inhibition of NOTCH1 signaling with GSIs in vivo. Moreover, limiting dilution analyses demonstrated that secondary loss of PTEN increased the leukemia initiating cell potential of NOTCH1 L1601P Δ-PEST induced tumors. Analysis of NOTCH1 signaling showed complete clearance of activated NOTCH1 protein and marked downregulation of Hes1 in both in PTEN-positive and PTEN-deleted NOTCH1 L1601P Δ-PEST induced tumors treated with DBZ compared with controls. However, global analysis of gene expression profiling with oligonucleotide microarrays showed that while NOTCH1 direct target genes are downregulated in both PTEN-positive and PTEN deleted tumors, there is a global reversal of much of the transcriptional effects of NOTCH inhibition, consisting of downregulation of genes involved in anabolic pathways and upregulation of genes involved in catabolic pathways and autophagy, upon PTEN loss. Consistently, electron microscopy analysis demonstrated increased autophagy in NOTCH1 induced tumors upon NOTCH1 inhibition, which was reversed upon PTEN deletion. Moreover, global metabolomic analyses of PTEN-non-deleted and PTEN-deleted NOTCH1 L1601P Δ-PEST induced tumors treated with DBZ compared with controls demonstrated that NOTCH inactivation induces a global anabolic shutdown in T-ALL with a marked block of glycolysis and glutaminolysis which renders NOTCH induced tumors dependent on branched amino acid catabolism to sustain their carbon metabolism. Notably these effects are globally rescued in PTEN deleted tumor cells, which show high basal levels of glycolysis and sustained glycolysis and glutaminolysis despite effective NOTCH1 inhibition with DBZ.
Overall, these results formally demonstrate that loss of PTEN induces in vivo drug resistance to NOTCH inhibition in T-ALL; highlight the fundamental importance of NOTCH1 in the control of tumor cell metabolism; strongly suggest that increased glycolysis and sustained carbon metabolism can induce resistance to GSI therapy and provide the basis for the design of new therapeutic strategies targeting these metabolic pathways in T-ALL.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.