Abstract
Significance Phosphoinositide-3 kinase (PI3K) and MYC are recognized as central components of a shared signaling network that regulates metabolism to drive cell proliferation. Here we show genetically increasing MYC expression in B-ALL creates permissiveness for PI3K hyperactivation to generate a more aggressive malignant state accounts for relapsed B-ALL. Our study suggests that combining the FDA-approved PI3Kγ/δ inhibitor, Duvelisib, with CAR-T therapy holds significant promise for the treatment of R/R B-ALL. Furthermore, we identified NADH accumulation as a key metabolic vulnerability in R/R B-ALL, rendering these cells susceptible to reductive stress enhancing treatment and glutamine derivation. Notably, Pegaspargase is particularly important for treating R/R B-ALL, and here we report that combining Pegaspargase with a potent complex I inhibitor IACS, which impairs NADH oxidation process, exhibits significant therapeutic effects on mice with PI3KhiMYCamp B-ALL.
Results By analyzing cancer cohorts based on transcription profiles indicative of PI3K and MYC activity, we found unlike their established roles as individual prognostic indicators in other cancers, neither PI3K nor MYC activity alone robustly predicts B-ALL prognosis. However, patients with concurrent activation of both PI3K and MYC had an unfavorable clinical outcome compared to the others, emphasizing the synergistic role of PI3K and MYC in driving B-ALL progression. Our findings also reveal that excessive PI3K activation reduces c-MYC protein levels in B-ALL through translational inhibition. Furthermore, c-MYC amplification effectively enables B-ALL cell to tolerate PI3K hyperactivation. Given that relapsed B-ALL exhibits concurrent upregulation of both PI3K and MYC, a finding not seen in newly diagnosed samples, targeting PI3K presents a logical therapeutic strategy for relapsed disease. Our results indicated that combining duvelisib with CAR-T therapy significantly extends the survival of recipients using PI3KhiMYCamp B-ALL xenograft models, providing a promising therapeutic approach for patients with R/R B-ALL.
Transmission electron microscopy revealed that PI3K hyperactivation induced significant alterations in the mitochondrial ultrastructure, characterized by cristae loss and mitochondrial swelling in B-ALL cells. Furthermore, our results demonstrate that PI3K hyperactivation severely reduces OCR levels, glutamine consumption, and ATP production in B-ALL cells, all of which can be fully rescued by ectopic expression of MYC. Mechanistically, PI3KhiMYCamp B-ALL cells rely on glutaminolysis to fuel TCA cycle, together with fully activated mTORC1 to support aggressive malignant growth.
Untargeted metabolomic profiling revealed a significant increased amount of various amino acids in PI3KhiMYCamp mouse B-ALL cells, except for aspartate (Asp), which was markedly reduced compared to basal B-ALL cells. By using 13C5-glutamine and 13C6-glucose tracing, we found that Asp synthesized from both glutamine and glucose was reduced in PI3KhiMYCamp B-ALL cells. Additionally, we demonstrated that human B-ALL cells with elevated PI3K and MYC activities have lower intracellular Asp level. Asp facilitates NAD+ regeneration in the cytoplasm through malate-aspartate shuttle (Wu et al., Nat Immunol, 2021). Notably, we found that the NADH/NAD+ ratio is at a critical threshold for redox balance in PI3KhiMYCamp B-ALL cells, and even a slight increase in this ratio-through LDH or Complex I inhibition-disrupts redox homeostasis, inducing reductive stress and inhibiting cell growth. These results provide rationale of targeting NAD+ regeneration and glutamine availability. Finally, we proved that the combination of Complex I inhibitors IACS and FDA-approved Pegaspargase could effectively eradicated PI3KhiMYCamp B-ALL cells in vivo.
Conclusion In summary, our study revealed that PI3K hyperactivation disrupts mitochondrial structure and bioenergetics, impairing the metabolic adaptation necessary for oncogenic proliferation in B-ALL cells. Amplification of MYC effectively counteract cell death induced by excessive PI3K signaling. Additionally, PI3KhiMYCamp serves as a prognostic marker, identifying B-ALL patients with poor outcomes. We have demonstrated that PI3KhiMYCamp B-ALL cells are more sensitive to reductive stress driven by NADH accumulation and the potential of targeting NADH homeostasis as a promising therapeutic strategy for R/R B-ALL patients.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal