Dual IKZF2 and CK1α Degrader Targets Acute Myeloid Leukemia Cells

Acute myeloid leukemia (AML) is an aggressive hematologic disease for which several epigenetic regulators have been identified as therapeutic targets. We previously found that a transcription factor and Ikaros family member, IKZF2 (HELIOS) is required for myeloid leukemic stem cell activity (Park et al. Cell Stem Cell 2019). Additionally, IKZF2 enhances self-renewal by increasing HOXA9 gene expression and in parallel blocks myeloid differentiation by reducing C/EBP expression. Thus, our studies suggested that targeting IKZF2 could be a new therapeutic strategy in AML.

Immunomodulatory drugs (IMiDs; lenalidomide and others) have been used clinically to specifically target and degrade other Ikaros family members. However, these have been unable to target IKZF2. Utilizing a structure-guided approach, we developed a small molecule with low nanomolar degrader activity against IKZF2 (DEG-35). DEG-35 treatment in different AML cell lines induce apoptosis and differentiation with IC₅₀ values of 5 ±0.4 nM in MOLM-13 cells to 27± 2.5 uM in NB4 cells.

To understand if DEG35 has activity against other substrates, we performed global quantitative proteomics and found casein kinase 1 alpha (CK1α), a serine/threonine kinase previously identified to be critical for myeloid leukemogenesis and a known IMiD substrate, to be the most depleted protein (Jaras et al. JEM). Also, we performed a PRISM (Profiling Relative Inhibition Simultaneously in Mixtures) screen assay which utilizes 770 barcoded cell lines from 20 lineages and identified multiple correlations with the p53 pathway. Previous studies found that CK1α suppresses the activity of p53 phosphorylation or binding MDM proteins. To further probe the mechanism behind DEG35 activity, we performed RNA sequencing at 24hrs in MOLM-13 cells. Gene set enrichment analysis (GSEA) using the rank list of differentially expressed genes from the MOLM-13 cells treated with DMSO and DEG-35 revealed enrichment for genes upregulated in CK1α KO keratinocytes and p53 targets. Additionally, we found enrichment for the MLL-AF9 IKZF2 KO LSC gene sets, myeloid differentiation, loss of MEIS1-HOXA9 and MYC targets. Immunoblotting validated the reduction of MYC and HOXA9 protein in DEG-35 treated MOLM-13 cells. To determine the contribution of CK1α and IKZF2 to the cellular phenotypes in MOLM13 cells, we developed CK1α and IKZF2 nondegradable mutants. CK1α, G40N was able to significantly reduce the cell killing activity of DEG-35 by 94% compared to vector cells (**p<0.01). Moreover, p53 deletion was also able to block DEG-35 induced apoptosis by 98% and differentiation by 88% compared to control. Additionally, expression of IKZF2 H141Q rescued DEG-35 mediated differentiation by 55% (***p<0.001), but not apoptosis. Thus, we demonstrate that the degradation of IKZF2 and CK1α by DEG-35 blocks cell growth and induces myeloid differentiation in human AML cell lines through CK1α-p53- and IKZF2-dependent pathways.

To further probe the efficacy and therapeutic index of targeting both IKZF2 and CK1α. We found that AML cells were more sensitive to DEG-35 compared to normal bone marrow cells (murine MLL-AF9 Crbn ^I1391V cells has 10 fold and AML PDX cells have 2-5 fold higher sensitivity to normal mouse or human bone marrow cells).Treatment of DEG-35 at 50 mg/kg in AML PDX-transplanted mice led to significant prolonged survival compared to vehicle treated mice (median survival of DMSO mice: 45 days vs DEG-35 mice: 61 days; ***p< 0.001) and reduced disease burden. To improve pharmacological properties, we developed analog DEG-77, which has 10-fold increased solubility than DEG-35 in DMSO. DEG77 exhibited better in vivo efficacy in the MLL-AF9 Crbn ^I1391V model (Increase in median survival by DEG-35: 2 days *p<0.05 vs DEG-77: 31 days ****p<0.0001). Importantly, the increase of median survival of 30 days (****p<0.0001) was also observed in the secondarily transplanted DEG-77 mice compared to DMSO mice, suggesting that DEG-35 can reduce LSC activity.

In summary, we report the development of nanomolar monofunctional cereblon-dependent degraders of IKZF2 and casein kinase 1 alpha (CK1α). DEG-35 or the analog DEG-77 delays leukemia progression in murine and human AML mice models. Thus, we provide a novel strategy for multi-targeted degradation of IKZF2/CK1α to enhance efficacy against AML that may be expanded to additional targets and indications.

No relevant conflicts of interest to declare.