Abstract
Introduction: Oncogenic WNT/β-catenin signaling and activating mutations that increase stability and transcriptional activity of β-catenin are among the most frequent lesions throughout all main types of cancer (Morin et al. 1997; Rubinfeld et al. 1997).
Results: Strikingly, pan-cancer analysis of activating mutations of β-catenin in patient samples and global proteomic analysis of human cancer cell lines revealed that B-cell malignancies are exempt from oncogenic activation of β-catenin. While nuclear accumulation of β-catenin was observed in >80% of cancer samples studied, nuclear β-catenin was consistently absent in normal and malignant B-lymphoid cells.
To model the effects of inducible β-catenin activation in B-ALL, pre-B cells from Ctnnb1 ex3fl/+ mice were transformed with BCR-ABL1 or NRAS G12D oncogenes. Cre-mediated excision of exon 3 removes GSK3β and CK1 phosphorylation sites (S 33, S 37, T 41 and T 45) required for β-catenin degradation, therefore resulting in accumulation of β-catenin. Stabilization of β-catenin rapidly impaired competitive fitness of B-ALL cells, abolished colony forming ability and induced cell cycle arrest. Likewise, expression of constitutively active form of β-catenin in patient derived B-ALL cells (n=6) or B-cell lymphoma (n=4) compromised proliferation, clonal fitness, and induced cell death. Interestingly, activation of β-catenin signaling in myeloid leukemia (n=4) accelerated proliferation. Corroborating the lineage specific effects of β-catenin, CEBPα-mediated reprogramming of B-ALL cells into myeloid lineage leukemia cells, reversed the deleterious effect of inducible β-catenin-activation.
In epithelial cell types, β-catenin/TCF complexes drive transcriptional activation of MYC (He et. al.1998). In striking contrast to epithelial cells, however, our transcriptomic analysis showed that inducible activation of β-catenin in B-cell malignancies resulted in transcriptional repression of Myc and its target genes. Surprisingly, proximity-based labeling and interactome studies in B-ALL cells revealed Ikzf1 and Ikzf3 as top-ranking interaction partners for β-catenin. Furthermore, several components of the Nucleosome Remodeling and Deacetylase (NuRD) complex including Mta1/2, Gatad2a/b, Chd4, Hdac1/2 were among the most significantly enriched proteins within the β-catenin interactome. Interaction of Ikzf1 and Ikzf3 with NuRD complex induced loss of histone H3 lysine 27 acetylation (H3K27Ac) and suppressed Myc expression. To test whether repressive complexes of β-catenin with Ikzf1, Ikzf3:NuRD subvert proliferation and survival by transcriptional repression of Myc, we introduced Cas9-RNP for genetic deletion of Ikzf1 and Ikzf3 in B-ALL cells. Interestingly, deletion of both Ikzf1 and Ikzf3 was required to restore clonal fitness, proliferation and Myc-driven transcriptional programs. Conversely, genetic deletion of β-catenin was sufficient to abrogate Ikzf1/Ikzf3-mediated tumor suppression and transcriptional repression of Myc. Together, these results suggest that engagement of b-catenin set the threshold for Ikzf1/Ikzf3-mediated tumor suppression in B-ALL and mature B-cell malignancies. Together with B-lineage-specific expression of Ikzf1 and Ikzf3, these findings provide a mechanistic explanation as to why activating lesions of β-catenin are not detected in B-cell malignancies.
Therapeutic implication: To leverage the unique sensitivity of B-cell malignancies to β-catenin-activation, we tested pharmacological activation of β-catenin based on small molecule-inhibition of GSK3β, a central negative regulator of β-catenin. Testing six different GSK3β inhibitors, we identified LY2090314 as the most potent inhibitor in killing B-ALL cells and suppressing Myc (IC50=4.5 nM). Treatment of NSG mice bearing patient derived B-ALL xenografts with LY2090314 substantially reduced leukemia burden and significantly extended overall survival compared to vehicle treated mice (n=9, P= 6.5E-05). These findings suggest that β-catenin activation is a unique vulnerability and small molecule GSK3β inhibition represents a novel opportunity to overcome drug-resistance in refractory B-ALL.
Disclosures
No relevant conflicts of interest to declare.
Discovery of small molecule antagonists of human Retinoblastoma Binding Protein 4 (RBBP4)
