CC-90009, a novel cereblon E3 ligase modulator, targets acute myeloid leukemia blasts and leukemia stem cells

Abstract

A number of clinically validated drugs have been developed by repurposing the CUL4-DDB1-CRBN-RBX1 (CRL4CRBN) E3 ubiquitin ligase complex with molecular glue degraders to eliminate disease-driving proteins. Here, we present the identification of a first-in-class GSPT1-selective cereblon E3 ligase modulator, CC-90009. Biochemical, structural, and molecular characterization demonstrates that CC-90009 coopts the CRL4CRBN to selectively target GSPT1 for ubiquitination and proteasomal degradation. Depletion of GSPT1 by CC-90009 rapidly induces acute myeloid leukemia (AML) apoptosis, reducing leukemia engraftment and leukemia stem cells (LSCs) in large-scale primary patient xenografting of 35 independent AML samples, including those with adverse risk features. Using a genome-wide CRISPR-Cas9 screen for effectors of CC-90009 response, we uncovered the ILF2 and ILF3 heterodimeric complex as a novel regulator of cereblon expression. Knockout of ILF2/ILF3 decreases the production of full-length cereblon protein via modulating CRBN messenger RNA alternative splicing, leading to diminished response to CC-90009. The screen also revealed that the mTOR signaling and the integrated stress response specifically regulate the response to CC-90009 in contrast to other cereblon modulators. Hyperactivation of the mTOR pathway by inactivation of TSC1 and TSC2 protected against the growth inhibitory effect of CC-90009 by reducing CC-90009-induced binding of GSPT1 to cereblon and subsequent GSPT1 degradation. On the other hand, GSPT1 degradation promoted the activation of the GCN1/GCN2/ATF4 pathway and subsequent apoptosis in AML cells. Collectively, CC-90009 activity is mediated by multiple layers of signaling networks and pathways within AML blasts and LSCs, whose elucidation gives insight into further assessment of CC-90009s clinical utility. These trials were registered at www.clinicaltrials.gov as #NCT02848001 and #NCT04336982).

Conflict of interest statement

Conflict-of-interest disclosure: C.S., C.-C.L., I.S.J., E.R., D.M., T.C., E.T., C.F., A.C., M.M., K.W., P.C., B.C., J.C., J.H., J.F., D.W.P., M.P., M.R., A.L.-G., and G.L. are, or have been, employees and equity holders at BMS. J.E.D. served on the Science Advisory Board (SAB) at Trillium Therapeutics, and has ownership interest (including patents) in Trillium Therapeutics Inc.There is an existing license agreement between TTI and University Health Network, and L.J., J.C.Y.W., and J.E.D. may be entitled to receive financial benefits further to this license and in accordance with their institution’s intellectual property policies. M.D.M. serves on the SAB at Astellas Pharma Inc. J.E.D. received financial support by sponsored research agreements from Bristol-Myers Squibb. The remaining authors declare no competing financial interests.

© 2021 by The American Society of Hematology.

Figures

Graphical abstract

Figure 1.

CC-90009, a GSPT1-selective cereblon modulator, induces cereblon-and GSPT1-dependent anti-AML activity. (A) The chemical structures of CC-90009, CC-885, and lenalidomide (LEN) with the glutarimide ring shown in red. (B) The antiproliferative effect of CC-90009 in AML cell lines. Cells were incubated with DMSO or CC-90009 at the indicated concentrations. At day 3, cell proliferation was assessed by the Cell-Titer Glo (CTG) assay. (C-D) The effect of CC-90009 on cell viability of leukemic cells (C) and normal lymphocytes (D) in bone marrow aspirates of AML patients. Cells were treated with DMSO or increasing concentrations of CC-90009 for 48 hours. Cells were then stained with fluorescently labeled antibodies and Annexin-V, followed by flow cytometry to determine the cell number of live leukemic cells and lymphocytes. Data shown in panels C and D are presented as the percentage of viable cell counts in CC-90009-treated patient samples relative to DMSO controls. (E) Volcano plot of differentially abundant proteins in response to CC-90009 treatment relative to DMSO control. KG1 cells were treated with DMSO or 100 nM CC-90009 for 4 hours and subjected to tandem mass tag proteomics analysis. The x-axis indicates the log2–fold change of CC-90009 vs DMSO control for each protein. P values were corrected for multiple hypothesis testing using the Benjamini-Hochberg method to arrive at an adjusted P value (adj-P; also known as a false discovery rate). The y-axis is the -log10 (adj-P) values indicating statistical significance such that proteins lying above the dotted red line are statistically significant findings with adj-P <.05. (F) Immunoblot analysis of KG1 and U937 cells treated with DMSO or CC-90009 for 4 hours. Where indicated, cells were pretreated with mortezomib or MLN4924 or 30 minutes. (G-H) Immunoblot analysis (G) and cell proliferation (H) of U937-Cas9 parental cells or cells stably transduced with lentiviral vectors expressing a nontargeting sgRNA (sgNT-1), an sgRNA targeting a noncoding region (sgNC-8), or an sgRNA targeting CRBN (sgCRBN-8). Cells were treated with DMSO or CC-90009 at indicated concentrations. Crystal structure of GSPT1 in complex with cereblon, DDB1 and CC-90009. (I) Surface representation of the complex with DDB1 shown in purple, cereblon in blue, and GSPT1 in orange. The position of CC-90009 is shown with an arrow. (J) GSPT1 interaction with cereblon is mediated by a β-hairpin loop. Hydrogen bond interactions between cereblon and the GPST1 β-hairpin are shown as yellow dashes. (K-L) Immunoblot analysis (K) and cell proliferation (L) of U937 parental cells or cells stably expressing HA-GSPT1-G575N. Cells were treated with DMSO or CC-90009 at the indicated concentrations. Data in panels B-D, G, and L are shown as mean ± standard deviation (SD), n = 3 technical replicates. Result shown in all figure panels is representative of 3 biological replicates.

Figure 2.

Identification of genes and pathways modulating the response to CC-90009 via CRIPSR/Cas9 screen. (A) Schematic showing the design of the genome-wide CRISPR screen to identify molecular determinants of CC-90009 response. At day 3 posttransduction, cells were treated with 10 µM CC-90009 or DMSO vehicle control for an additional 9 days, followed by amplification of sgRNA coding regions and next-generation sequencing. The log2 fold change (log2FC) in sgRNA read count in the CC-90009-treated sample as compared with DMSO control was designated as the enrichment score, and the average log2FC value of all sgRNAs for a gene of interest was used to quantify the effect of gene knockout on CC-90009 response. (B) Cell proliferation curve of U937-Cas9 cells transduced with the lentiviral sgRNA library and treated with DMSO or CC-90009. Three days posttransduction, cells were treated with DMSO or 10 µM CC-90009 for 9 days. (C) Pathway enrichment analysis of genes enriched by CC-90009 treatment with log2FC >2 and false discovery rate (FDR) <0.05 relative to DMSO control. The color and size of the dots represent adjusted significance level and gene ratio, respectively. Gene ratio refers to the number of input genes annotated to an individual pathway as a ratio of all input genes annotated to any Reactome pathway. (D) Scatter plot of 78 genes significantly enriched by CC-90009 (log2FC >2 and FDR <0.05). X-axis, CC-90009 enrichment score shown as log2FC (T12_CC-90009 vs T12_DMSO); y-axis, gene essentiality score shown as log2FC (T12_DMSO vs T3_DMSO). Some of these genes were grouped into 10 functional modules with different color coding. (E,F) Log2FC values of sgRNAs targeting CC-90009 enriched genes in the functional modules as indicated. Background shown in dark blue represents the log2FC values of all sgRNAs in the library. Each solid line with a color representing a functional module indicates the log2FC value of an individual sgRNA. Well-characterized genes known to be essential for the activity of the cereblon E3 ligase complex (E); novel genes that regulate the response to CC-90009 with no clear mechanistic understanding (F). Note that in panels E and F, multiple sgRNAs targeting each of these top-ranked genes were significantly enriched by CC-90009, strongly supporting the on-target gene knockout effect.

Figure 3.

Regulation of CRBN splicing and CC-90009 response by the ILF2 and ILF3 complex. (A-C) Assessment of the effect of ILF3 knockout on CC-90009 response by a flow cytometry-based CRISPR competition assay. U937 cells stably expressing Cas9 were infected with lentiviral vectors coexpressing RFP and a nontargeting sgRNA (sgNT-1), an sgRNA targeting a noncoding region (sgNC-1), or an sgRNA targeting ILF3 (sgILF3-2 or sgILF3-4). The expression of RFP is driven by an EF1a-HTLV hybrid promoter, whereas the expression of sgRNAs was under the control of a doxycycline-inducible H1/TO promoter. Three days after sgRNA induction with 1 μg/mL doxycycline, the cells were mixed at a 1:1 ratio with U937 Cas9 cells infected with a lentiviral vector constitutively expressing GFP and a nontargeting sgRNA (sgNT-1) and treated with DMSO or 10 μM CC-90009. The change of RFP+/GFP+ ratio was monitored by flow cytometry every 2 days thereafter. (A) Schematic design of the flow cytometry-based CRISPR competition assay. (B) Immunoblot analysis of U937-Cas9 cells inducibly expressing sgNT-1, sgNC-1, sgILF3-2, or sgILF3-4. Cells were treated with doxycycline (DOX) for 6 days. (C) The RFP+/GFP+ ratios of U937-Cas9 cells coexpressing RFP and sgNT-1, sgNC-1, sgILF3-2, or sgILF3-4 mixed with cells coexpressing GFP and sgNT-1 at each indicated timepoint were normalized to the RFP+/GFP+ ratio of the cell mixtures on day 0. (D) Immunoblot analysis of U937-Cas9 cells stably expressing sgNT-1 or sgILF3-2 under the control of the H1/TO promoter. Cells were treated with or without DOX for 4 days, followed by incubation with DMSO or an increasing concentration of CC-90009 for 6 hours. (E) RNA sequencing analysis of U937-Cas9 cells with inducible expression sgNT-1 or sgILF3-2 for 7 days. Evidence of differential splicing was observed in a total of 967 unique genes by up- and/or downregulated exon usage with ILF3 knockout in U937 cells, reaching a corrected significance level (FDR) <0.05. At the gene level, 791 genes were found to be significantly (FDR <0.05) up- or downregulated with ILF3 knockout. Top: Venn diagram showing the overlap of genes with significant differential exon usage (DEU; LHS) and genes with differential expression at the gene-level (DEG; RHS). Bottom: pathway enrichment analysis of DEU and DEG genes. The color and size of the dots represent adjusted significance level and gene ratio respectively. Gene ratio refers to the number of input genes annotated to an individual pathway as a ratio of all input genes annotated to any Reactome pathway. (F) DEU analysis revealed significant differential splicing of individual CRBN exons (red bars) with knockout of ILF3. An exon, which annotates to the truncated transcript, CRBN.213 (exon bin no.13), is significantly elevated (FDR, 0.02) with ILF3 knockout relative to NT controls. Conversely, exons downstream of this isoform are significantly underrepresented (FDR, 0.05; exon bin no. 14) in the ILF3 knockout cells relative to parental. (G) Quantitative PCR analysis of the expression levels of CRBN transcripts as indicated in U937-Cas9 cells with inducible expression sgNT-1 or sgILF3-2 for 7 days. Data in panel G are shown as mean ± standard deviation (SD), n = 4 technical replicates. Result shown in all figure panels is representative of 3 biological replicates.

Figure 4.

Loss of TSC1 or TSC2 attenuates the response to CC-90009. A-C) Assessment of the effect of TSC1 or TSC2 knockout on CC-90009 response by a flow cytometry based CRISPR competition assay. U937 cells stably expressing Cas9 were infected with a lentiviral vector constitutively co-expressing GFP and sgNT-1, or with lentiviral vectors constitutively co-expressing RFP and sgNT-1, sgNC-8, or 1 of the 3 sgRNAs targeting TSC1 or TSC2 as indicated. Three days after infection, RFP and GFP cells were mixed at a 1:1 ratio and treated with DMSO or 10 μM CC-90009. The change of RFP+/GFP+ ratio was monitored by flow cytometry every 2 days thereafter. A) Left, schematic design of the flow cytometry based CRISPR competition assay. Right, confirmation of TSC1 or TSC2 knockout by immunoblot analysis. B) and C) The RFP+/GFP+ ratios of U937-Cas9 cells co-expressing RFP and sgNT-1, sgNC-8, or 1 of 3 sgRNAs targeting TSC1 (B) or TSC2 (C) mixed with cells co-expressing GFP and sgNT-1 at each indicated timepoint were normalized to the RFP+/GFP+ ratio of the cell mixtures on “Day 0”. D) and E) Immunoblot analysis of U937-Cas9 parental cells or cells stably expressing sgRNAs as indicated. Cells were treated with DMSO or CC-90009 in the absence (D) or presence (E) of cycloheximide. (F) Immunoblot analysis of anti-HA immunoprecipitates (top) or whole cell extracts (bottom) of U937-Cas9 parental cells or cells stably expressing the indicated sgRNAs. Cells were treated with MLN4924 and DMSO or CC-90009. Result shown in all figure panels is representative of 3 biological replicates.

Figure 5.

CC-90009 activates the GCN2-mediated integrated stress response and subsequent apoptosis in AML. Characterization of the role of (A) GCN2, (B) GCN1, (C) ATF4, and (D) DDIT4 in mediating CC-90009 response using a flow cytometry-based CRISPR competition assay. U937 cells stably expressing Cas9 were infected with a lentiviral vector constitutively coexpressing GFP and sgNT-1, or with lentiviral vectors constitutively coexpressing RFP and sgNT-1, sgNC-8, or 1 of the gene-specific sgRNAs as indicated. Three days after infection, RFP and GFP cells were mixed at a 1:1 ratio and treated with DMSO or 10 μM CC-90009. The change of RFP+/GFP+ ratio was monitored by flow cytometry every 2 days thereafter. The RFP+/GFP+ ratios of U937-Cas9 cells coexpressing RFP and sgNT-1, sgNC-8, or 1 of 3 sgRNAs targeting GCN2 (A), GCN1 (B), ATF4 (C), or DDIT4 (D) mixed with cells coexpressing GFP and sgNT-1 at each indicated timepoint were normalized to the RFP+/GFP+ ratio of the cell mixtures on day 0. (E-F) Immunoblot (E) or quantitative RT-PCR analysis (F) of U937 parental and GCN2−/− cells treated with DMSO or CC-90009 at the indicated concentrations for 24 hours. The U937 GCN2−/− cell line is derived from a single clone of U937 parental cells stably infected with a lentiviral CRISPR vector targeting GCN2 (see “Methods and materials”). (G) Immunoblot analysis of U937 parental cells and GCN2−/− cells with or without a stably transduced lentiviral vector expressing HA-tagged GCN2 wild-type or mutants as indicated. Cells were treated with DMSO or CC-90009 for 24 hours. (H) The effect of CC-90009 on proliferation of cells shown in panel G. On day 3 after CC-90009 treatment, cell proliferation was assessed by CTG. Data in panels F and G are shown as mean ± standard deviation (SD), n = 3 or 4 technical replicates. Result shown in all figure panels is representative of 3 biological replicates.

Figure 6.

CC-90009 treatment reduces leukemic engraftment in AML xenograft models. (A) GSPT1 expression at 24 hours in culture with CC-90009 at indicated concentrations. Each line indicates data relative to 0 nM CC-90009 for an individual patient sample (n = 23). Lines in black, cyan, and red represent samples with >70%, 50% to 70%, and <50% GSPT1 reduction, respectively, at the 100-nM treatment dose. (B) Schematic experimental design for in vivo CC-90009 treatment. IF, intrafemoral transplantation; LDA, limiting dose assay; LSC17, a 17-gene leukemia stem cell signature. (C) Relative GSPT1 degradation to controls in the RF and BM of NOD.SCID mice engrafted with AML and treated with CC-90009 (n = 2-4 mice per treatment group). PR, partial responder; R, responder. (D) CD45+CD33+ AML engraftment levels in RF and BM of xenotransplanted mice treated with vehicle (dark green bars) or CC-90009 (purple bars) (n = 5 per treatment group). Bars represent mean ± standard error of the mean (SEM) of an individual patient sample. (E) Relative reduction in leukemic engraftment in mice treated with CC-90009 compared with controls, and transplanted with de novo or secondary or relapsed AML, normal or abnormal cytogenetics, MRC adverse or intermediate risk group, and high or low LS17 scores. Each dot represents an individual patient sample. Bars are the median values for each group. For panels C and E, relative change for each patient sample was calculated as (median of CC90009 – median of vehicle)/median of vehicle × 100. *P < .05, **P < .01, ***P < .001, ****P < .0001. ns, not significant.

Figure 7.

Self-renewal LSCs are targeted by CC-90009. (A) Flow cytometric analysis of 2 representative samples (AML3 and AML25) showing CD34+CD14− cells in the AML graft in mice treated with CC-90009 or vehicle control. (B-C) Relative change in %CD34+ population in AML graft (B) or in the mouse bone marrow (C) from CC-90009-treated mice compared with controls. (D) LSC frequency in CC-90009- or control-treated mice determined by secondary transplantation at limiting doses for representative sample AML3 (left) and summary graph showing fold change in LSC frequency for all samples tested by limiting dilution analysis (right). (E) Number of LSCs in the primary mouse bone marrow following treatment with CC-90009 (purple bars) or vehicle control (dark green bars) (5 mice per treatment group). Each bar indicates the mean ± SEM of an individual patient sample. (F) Relative reduction in LSC numbers in mice treated with CC-90009 compared with vehicle control. *P < .05, **P < .01, ***P < .001, ****P < .0001. NR, nonresponder; ns, not significant.