Divergent MYC- and BCL6-Driven Metabolic Programs Enable Dynamic Regulation of Cell Biomass in B-Cell Malignancies

Background: Inhibition of oncogenic tyrosine kinases in pre-B acute lymphoblastic leukemia (ALL) results in phenotypic changes that resemble the differentiation from IL7-dependent large proliferative pre-B cells to quiescent small resting pre-B cells. Both oncogenic signaling in pre-B ALL and IL7 signaling in large pre-B cells induce STAT5 activation, which is regulated by negative feedback control via SOCS family proteins.

Results: Despite the role of STAT5 signaling as a driver of pre-B ALL, we found that the SOCS proteins CISH and SOCS2 are highly expressed and predictive of poor patient outcome in pre-B ALL. Genetic ablation of Cish, Socs2, or Socs3 resulted in Stat5 hyperactivation, reactive oxygen species production, cell cycle checkpoint molecules accumulation, and subsequent leukemia cell death. These results suggest that modulation of STAT5 signaling strength by negative feedback regulators is crucial to pre-B ALL.

To investigate the dynamic regulation of STAT5 activity, we designed a doxycycline inducible vector system for a constitutively active form of Stat5 (Stat5-CA) and a non-phosphorylatable form of Stat5 where tyrosine 694 is replaced with phenylalanine (Stat5-Y694F). Induction of Stat5-CA caused accumulation of biomass indicated by increased cell size, whereas Stat5-Y694F caused rapid cell shrinkage. mTOR regulates cell biomass by stimulating protein synthesis, so we next assessed the rescue effect of mTOR or protein synthesis inhibition on Stat5-CA phenotypes. While rapamycin had no effect on empty vector (EV) control cells, it balanced cell size and partially rescued death of Stat5-CA cells. Interestingly, treatment with cycloheximide almost fully rescued Stat5-CA cells, suggesting that Stat5-CA phenotype caused by excessive protein synthesis. Indeed, Stat5-CA cells had higher protein synthesis rates than EV control cells, and it was associated with ER stress/unfolded protein response (UPR) indicated by expanded ER.

Consistent with the role of mTOR in promoting glycolysis via HIF1α, metabolomic analysis revealed increased glycolytic intermediates in Stat5-CA cells, while the opposite effect was seen in Stat5-Y694F cells. Instead, Stat5-Y694F cells showed increased levels of phosphatidylethanolamine (PE) intermediates compared to its EV and Stat5-CA counterparts. As a previous work has shown that surface PE regulates T-cell differentiation, we tested if surface PE levels are increased in Stat5-Y694F cells. Although surface PE levels in the entire population were decreased, surface PE⁺ small differentiated cells were emerged after Stat5-Y694F induction. These results suggest that balance between the divergent glycolytic and lipid metabolic programs is key determinant of cell survival in B-cells.

Time-course RNA-seq and western blot analyses revealed that induction of Stat5-CA increases Myc in parallel with glycolytic and UPR gene expression, whereas Stat5-Y694F causes the opposite effect. Instead, Stat5-Y694F increases Bcl6 in parallel with autophagic and lipid metabolic genes. Interestingly, only partial knockdown of Myc, resulting in approximately 80% mRNA levels compared to scramble control, led to marked cell proliferation in Stat5-CA pre-B ALL cells, whereas both deletion and more effective knockdown were counterproductive. Similar effects were observed with Bcl6 deletion and knockdown in Stat5a-Y694F pre-B ALL cells, suggesting that the balance between Myc and Bcl6 is a key determinant of cell survival in pre-B-ALL. Flow cytometry analysis of Myc-eGFP/Bcl6-mCherry dual reporter expression demonstrated increased levels of Myc but decreased levels of Bcl6 in Stat5-CA cells, and vice versa in Stat5-Y694F cells. Similar expression patterns of Myc and Bcl6 were also seen in vivo settings including the large versus small pre-B cells and the marginal zone versus follicular B cells. ChIP-seq analysis showed enrichment for glycolytic, mTOR signaling, ER stress/UPR, and lipid metabolic genes that were bound by both MYC and BCL6.

Conclusion: Our results suggest that MYC and BCL6 enable to maintain sufficient cell biomass in B-cells by alternating the divergent metabolic programs. This sleep-wake cycle of cell biomass production can also explain alternative survival pathways driving drug resistance in B-cell malignancies.

No relevant conflicts of interest to declare.