Tyrosine kinase inhibitors (TKI) are widely used to treat patients with leukemia driven by BCR-ABL11 and other oncogenic tyrosine kinases2,3. Recent efforts focused on the development of more potent TKI that also inhibit mutant tyrosine kinases4,5. However, even effective TKI typically fail to eradicate leukemia-initiating cells6–8, which often cause recurrence of leukemia after initially successful treatment. Here we report on the discovery of a novel mechanism of drug-resistance, which is based on protective feedback signaling of leukemia cells in response to TKI-treatment. We identified BCL6 as a central component of this drug-resistance pathway and demonstrate that targeted inhibition of BCL6 leads to eradication of drug-resistant and leukemia-initiating subclones. BCL6 is a known proto-oncogene that is often translocated in diffuse large B cell lymphoma (DLBCL)9. In response to TKI-treatment, BCR-ABL1 acute lymphoblastic leukemia (ALL) cells upregulate BCL6 protein levels by ~90-fold, i.e. to similar levels as in DLBCL (Fig. 1a). Upregulation of BCL6 in response to TKI-treatment represents a novel defense mechanism, which enables leukemia cells to survive TKI-treatment: Previous work suggested that TKI-mediated cell death is largely p53-independent. Here we demonstrate that BCL6 upregulation upon TKI-treatment leads to transcriptional inactivation of the p53 pathway. BCL6-deficient leukemia cells fail to inactivate p53 and are particularly sensitive to TKI-treatment. BCL6−/− leukemia cells are poised to undergo cellular senescence and fail to initiate leukemia in serial transplant recipients. A combination of TKI-treatment and a novel BCL6 peptide inhibitor markedly increased survival of NOD/SCID mice xenografted with patient-derived BCR-ABL1 ALL cells. We propose that dual targeting of oncogenic tyrosine kinases and BCL6-dependent feedback (Supplementary Fig. 1) represents a novel strategy to eradicate drug-resistant and leukemia-initiating subclones in tyrosine kinase-driven leukemia.

B-lymphoid transcription factors (e.g. PAX5, IKZF1) are critical for early B-cell development1–2, yet genetic lesions occur in >80% of cases of B-cell acute lymphoblastic leukemia (ALL)3–4. The significance of these lesions in ALL remained unclear. Combining ChIP-seq and RNA-seq studies, we identified a novel B-lymphoid program for transcriptional repression of glucose and energy supply. Our metabolic analyses revealed that PAX5 and IKZF1 enforce a state of chronic energy deprivation, resulting in constitutive activation of the energy-stress sensor AMPK5–7. Dominant-negative mutants of PAX5 and IKZF1 relieved glucose and energy restriction. Studying a transgenic pre-B ALL mouse model, heterozygous deletion of Pax5 increased glucose uptake and ATP-levels by >25-fold. Reconstitution of PAX5 and IKZF1 in pre-B ALL patient samples restored a non-permissive state and induced energy crisis and cell death. A CRISPR/Cas9-based screen of PAX5- and IKZF1- transcriptional targets identified NR3C1 (glucocorticoid receptor)8, TXNIP (glucose feedback sensor)9 and CNR2 (cannabinoid receptor)10 as central effectors of B-lymphoid restriction of glucose and energy supply. Interestingly, transport-independent lipophilic methyl-conjugates of pyruvate and TCA cycle metabolites bypassed the gatekeeper function of PAX5 and IKZF1 and readily enabled leukemic transformation. Conversely, pharmacological TXNIP- and CNR2-agonists and a small molecule AMPK-inhibitor strongly synergized with glucocorticoids, identifying TXNIP, CNR2 and AMPK as potential therapy-targets. Furthermore, our results provide a mechanistic explanation for the empiric finding that glucocorticoids are effective in the treatment of B-lymphoid but not myeloid malignancies. We conclude that B-lymphoid transcription factors function as metabolic gatekeepers by limiting the amount of cellular ATP to levels that are insufficient for malignant transformation.

SUMMARY Studying 830 pre-B ALL cases from four clinical trials, we found that human ALL can be divided into two fundamentally distinct subtypes based on pre-BCR function. While absent in the majority of ALL cases, tonic pre-BCR signaling was found in 112 cases (13.5%). In these cases, tonic pre-BCR signaling induced activation of BCL6, which in turn increased pre-BCR signaling output at the transcriptional level. Interestingly, inhibition of pre-BCR-related tyrosine kinases reduced constitutive BCL6 expression and selectively killed patient-derived pre-BCR+ ALL cells. These findings identify a genetically and phenotypically distinct subset of human ALL that critically depends on tonic pre-BCR signaling. In vivo treatment studies suggested that pre-BCR tyrosine kinase inhibitors are useful for the treatment of patients with pre-BCR+ ALL.

Chronic myeloid leukemia (CML) is induced by the oncogenic BCR-ABL1 tyrosine kinase and can be effectively treated for many years with tyrosine kinase inhibitors (TKIs). However, unless CML patients receive life-long TKI treatment, leukemia will eventually recur; this is attributed to the failure of TKI treatment to eradicate leukemia-initiating cells (LICs). Recent work demonstrated that FoxO factors are critical for maintenance of CML-initiating cells; however, the mechanism of FoxO-dependent leukemia initiation remained elusive. Here, we identified the BCL6 protooncogene as a critical effector downstream of FoxO in self-renewal signaling of CML-initiating cells. BCL6 represses Arf and p53 in CML cells and is required for colony formation and initiation of leukemia. Importantly, peptide inhibition of BCL6 in human CML cells compromises colony formation and leukemia initiation in transplant recipients and selectively eradicates CD34+ CD38− LICs in patient-derived CML samples. These findings suggest that pharmacological inhibition of BCL6 may represent a novel strategy to eradicate LICs in CML. Clinical validation of this concept could limit the duration of TKI treatment in CML patients, which is currently life-long, and substantially decrease the risk of blast crisis transformation.

Summary Studying mechanisms of malignant transformation of human pre-B cells, we found that acute activation of oncogenes induced immediate cell death in the vast majority of cells. Few surviving pre-B cell clones had acquired permissiveness to oncogenic signaling by strong activation of negative feedback regulation of Erk signaling. Studying negative feedback regulation of Erk in genetic experiments at three different levels, we found that Spry2, Dusp6 and Etv5 were essential for oncogenic transformation in mouse models for pre-B acute lymphoblastic leukemia (ALL). Interestingly, a small molecule inhibitor of DUSP6 selectively induced cell death in patient-derived pre-B ALL cells and overcame conventional mechanisms of drug-resistance.

The B cell-specific transcription factor BACH2 is required for affinity maturation of mature B cells. Here, we show that Bach2-mediated activation of p53 is required for stringent elimination of pre-B cells that failed to productively rearrange immunoglobulin VH-DJH gene segments. Upon productive VH-DJH gene rearrangement, pre-B cell receptor signaling ends negative selection through BCL6-mediated repression of p53. In patients with pre-B ALL, BACH2-mediated checkpoint control is frequently compromised. Low levels of BACH2 expression represent a strong independent predictor of poor clinical outcome. Bach2+/+ pre-B cells resist leukemic transformation by Myc through Bach2-dependent upregulation of p53, and fail to initiate fatal leukemia in transplant recipient mice. ChIP-seq and gene expression analyses reveal that BACH2 competes with BCL6 for promoter binding and reverses BCL6-mediated repression of p53 and other checkpoint control genes. These findings identify Bach2 as a critical mediator negative selection at the pre-B cell receptor checkpoint and a safeguard against leukemogenesis.

Genetic lesions such as BCR-ABL1, E2A-PBX1 and MLL rearrangements (MLLr) are associated with unfavorable outcomes in adult B-acute lymphoblastic leukemia (B-ALL). Leukemia oncoproteins may directly or indirectly disrupt cytosine methylation patterning to mediate the malignant phenotype. We postulated that DNA methylation signatures in these aggressive B-ALLs would point towards disease mechanisms and useful biomarkers and therapeutic targets. We therefore performed DNA methylation and gene expression profiling on a cohort of 215 adult B-ALL patients enrolled in a single phase III clinical trial (ECOG E2993) and normal control B-cells. In BCR-ABL1-positive B-ALL, aberrant cytosine methylation patterning centered around a cytokine network defined by hypomethylation and overexpression of IL2RA(CD25). The E2993 trial clinical data showed that CD25 expression was strongly associated with a poor outcome in ALL patients regardless of BCR-ABL1 status, suggesting CD25 as a novel prognostic biomarker for risk stratification in B-ALL. In E2A-PBX1-positive B-ALL, aberrant DNA methylation patterning was strongly associated with direct fusion protein binding as shown by the E2A-PBX1 ChIP sequencing (ChIP-seq), suggesting that E2A-PBX1 fusion protein directly remodels the epigenome to impose an aggressive B-ALL phenotype. MLLr B-ALL featured prominent cytosine hypomethylation, which was linked with MLL fusion protein binding, H3K79 dimethylation and transcriptional upregulation, affecting a set of known and newly identified MLL fusion direct targets with oncogenic activity such as FLT3 and BCL6. Notably, BCL6 blockade or loss of function suppressed proliferation and survival of MLLr leukemia cells, suggesting BCL6 targeted therapy as a new therapeutic strategy for MLLr B-ALL.

BCL6 protects germinal center (GC) B cells against DNA damage–induced apoptosis during somatic hypermutation and class-switch recombination. Although expression of BCL6 was not found in early IL-7–dependent B cell precursors, we report that IL-7Rα–Stat5 signaling negatively regulates BCL6. Upon productive VH-DJH gene rearrangement and expression of a μ heavy chain, however, activation of pre–B cell receptor signaling strongly induces BCL6 expression, whereas IL-7Rα–Stat5 signaling is attenuated. At the transition from IL-7–dependent to –independent stages of B cell development, BCL6 is activated, reaches expression levels resembling those in GC B cells, and protects pre–B cells from DNA damage–induced apoptosis during immunoglobulin (Ig) light chain gene recombination. In the absence of BCL6, DNA breaks during Ig light chain gene rearrangement lead to excessive up-regulation of Arf and p53. As a consequence, the pool of new bone marrow immature B cells is markedly reduced in size and clonal diversity. We conclude that negative regulation of Arf by BCL6 is required for pre–B cell self-renewal and the formation of a diverse polyclonal B cell repertoire.