Lenalidomide derivatives and proteolysis-targeting chimeras for controlling neosubstrate degradation

Abstract

Lenalidomide, an immunomodulatory drug (IMiD), is commonly used as a first-line therapy in many haematological cancers, such as multiple myeloma (MM) and 5q myelodysplastic syndromes (5q MDS), and it functions as a molecular glue for the protein degradation of neosubstrates by CRL4^CRBN. Proteolysis-targeting chimeras (PROTACs) using IMiDs with a target protein binder also induce the degradation of target proteins. The targeted protein degradation (TPD) of neosubstrates is crucial for IMiD therapy. However, current IMiDs and IMiD-based PROTACs also break down neosubstrates involved in embryonic development and disease progression. Here, we show that 6-position modifications of lenalidomide are essential for controlling neosubstrate selectivity; 6-fluoro lenalidomide induced the selective degradation of IKZF1, IKZF3, and CK1α, which are involved in anti-haematological cancer activity, and showed stronger anti-proliferative effects on MM and 5q MDS cell lines than lenalidomide. PROTACs using these lenalidomide derivatives for BET proteins induce the selective degradation of BET proteins with the same neosubstrate selectivity. PROTACs also exert anti-proliferative effects in all examined cell lines. Thus, 6-position-modified lenalidomide is a key molecule for selective TPD using thalidomide derivatives and PROTACs.

Fig. 1. Identification of critical positions on thalidomide derivatives for enhancing IKZF1 selectivity over SALL4.

a Chemical structures of thalidomide (Th), lenalidomide (Le), pomalidomide (Po), 5-hydroxythalidomide (5HT), 4-hydroxythalidomide (4HT), and 10 thalidomide derivatives (NE-001–NE-010). b Schematic diagram of the AlphaScreen (AS)-based biochemical interaction assay using recombinant proteins for detecting thalidomide-derivative-dependent complex formation between CRBN and neosubstrates. c Thalidomide-derivative-dependent biochemical interaction assay. The CRBN–IKZF1, CRBN–SALL4, and CRBN–PLZF complex formation was analysed using AS technology. The relative AS signals are expressed as the luminescence signal relative to the luminescence signal of dimethylsulfoxide (DMSO), which is considered 1. Error bars denote standard deviations (independent experiments, n = 3). d Immunoblot analysis of thalidomide-derivative-dependent exogenous neosubstrate degradation. HEK293T cells were transfected with pcDNA3.1-AGIA-SALL4 and pcDNA3.1-Myc-IKZF1 and treated with DMSO, 10-µM Th, or 10-µM thalidomide derivatives for 16 h. The experiment was independently repeated thrice, with similar results. e–f Immunoblot analysis of dose-dependent endogenous neosubstrate degradation in (e) HuH7 cells or (f) MM1.S cells. Each cell line was treated with DMSO, Po, Le, or NE-005 for 24 h. The experiment was independently repeated thrice, with similar results. g Chemical structures of 6-halogenated pomalidomides (NE-003, NE-011, and NE-012) and 6-halogenated lenalidomides (NE-005, NE-013, and NE-014). h Thalidomide-derivative-dependent biochemical interaction assay. The CRBN–IKZF1, CRBN–SALL4, and CRBN–PLZF complex formation was analysed using AS technology. The relative AS signals are expressed as the luminescence signal relative to the luminescence signal of DMSO, which is considered 1. Error bars denote standard deviations (independent experiments, n = 3). Source data are provided as a Source data file.

Fig. 2. Biochemical and cell-based analyses of 6-position-modified lenalidomides.

a Chemical structures of thalidomide (Th), pomalidomide (Po), lenalidomide (Le), 6-fluoro lenalidomide (F-Le), 6-chloro lenalidomide (Cl-Le) and 6-bromo lenalidomide (Br-Le). b Dose-dependent interaction assay using AlphaScreen (AS) technology. The CRBN–IKZF1, CRBN–SALL4, and CRBN–PLZF complex formation was analysed. The relative AS signals are expressed as the luminescence signal relative to the luminescence signal of dimethylsulfoxide (DMSO), which is considered 1. Error bars denote standard deviations (independent experiments, n = 3). c In vitro ubiquitination assay of SALL4 and IKZF1 by CRL4^CRBN. Purified FLAG-GST-IKZF1 or -SALL4 were mixed with recombinant CRL4^CRBN, E1, E2, and HA-Ub, and ubiquitination reactions were performed in the presence of DMSO, 20-µM pomalidomide (Po), 20-µM Le, 20-µM F-Le, 20-µM Cl-Le, or 20-µM Br-Le. Ubiquitinated SALL4 and IKZF1 were immunoprecipitated using an anti-FLAG antibody. The experiment was repeated twice independently, with similar results. d In the cell ubiquitination assay of SALL4 and IKZF1 by CRL4^CRBN. HEK293T cells were transfected with pcDNA3-HA-ubiquitin and pcDNA3.1-FLAG-CRBN and pcDNA3.1-AGIA-SALL4 or -IKZF1 and treated with DMSO, 1-µM Po, 10-µM Le, 10-µM F-Le, 10-µM Cl-Le, or 10-µM Br-Le in the presence of 10-µM MG132 for 8 h. Ubiquitinated SALL4 and IKZF1 were immunoprecipitated using an anti-AGIA antibody. The experiment was repeated twice independently, with similar results. e–g Immunoblot analysis of dose-dependent neosubstrate degradation in (e) MM1.S, (f) HuH7, or (g) NTERA-2 cells. Each cell line was treated with DMSO, Po, Le, F-Le, Cl-Le, or Br-Le for 24 h. The experiment was independently repeated thrice, with similar results. Source data are provided as a Source data file.

Fig. 3. Anti-proliferative effect of 6-position-modified lenalidomide on multiple myeloma and 5q myelodysplastic syndromes.

a Immunoblot analysis of IRF4 and c-Myc. MM1.S or H929 cells were treated with dimethylsulfoxide (DMSO), Po, Le, or lenalidomide derivatives for 72 h. The experiment was independently repeated thrice, with similar results. b Immunoblot analysis of RUNX1. Myelodysplastic syndrome (MDS)-L cells were treated with DMSO, Po, Le, or lenalidomide derivatives for 24 h. The experiment was repeated twice independently, with similar results. c Expression of SELP and ITGB3, which are upregulated in 5q MDS cells treated with lenalidomide. MDS-L cells were treated with DMSO, 10-µM Po, 10-µM Le, or 10-µM lenalidomide derivatives for 3 or 6 days, and mRNA expression was measured using quantitative RT-PCR. Relative mRNA expression was determined using the expression level following DMSO treatment. Error bars denote standard deviation (biological replicates; n = 3). d Dose–response curve of the anti-proliferative effect of 6-position-modified lenalidomides on MM cell lines. MM1.S and H929 cells were treated with DMSO, Po, Le, F-Le, or Cl-Le for 10 days, and cell viability was analysed using the CellTiter-Glo assay kit. Cell viability was expressed as the luminescence signal relative to the luminescence signal of DMSO, which was considered 100. Error bars denote standard deviation (biological replicates; n = 3). e Dose–response curve of anti-proliferative effect by 6-position-modified lenalidomides on 5q MDS cell line. MDS-L cells were treated with DMSO, Po, Le, F-Le, or Cl-Le for 12 days, and cell viability was analysed using the CellTiter-Glo assay kit. Cell viability was expressed as the luminescence signal relative to the luminescence signal of DMSO, which was considered 100. Error bars denote standard deviation (biological replicates; n = 4). f The half-maximal growth inhibition (GI₅₀) and the maximal growth inhibition (GI_max) values were calculated using the dose–response curve in (d, e). N/A means not applicable. Source data are provided as a Source data file.

Fig. 4. Neosubstrate selectivity for IKZF1, CK1α, SALL4, and PLZF of 6-position-modified lenalidomides.

a Chemical structures of NE-015 (F₃C-Le) and NE-016 (F₃CO-P). b Thalidomide-derivative-dependent biochemical interaction assay. The CRBN–IKZF1, CRBN–SALL4, and CRBN–PLZF complex formation was analysed using AlphaScreen (AS) technology. The relative AS signals are expressed as the luminescence signal relative to the luminescence signal of dimethylsulfoxide (DMSO), which is considered 1. Error bars denote standard deviations (independent experiments, n = 3). c In-cell proximity-dependent biotinylation of neosubstrates by AirID-CRBN. HuH7 and MM1.S cells stably expressing AGIA-AirID-CRBN were treated with DMSO, Po, Le, or lenalidomide derivatives in the presence of 10-µM biotin and 5-µM MG132 for 6 h. The experiment was repeated twice independently, with similar results. d–f Immunoblot analysis of degradations of five neosubstrates in (d) MM1.S cells, (e) NTERA-2 cells, or (f) HEK293T cells. Each cell line was treated with DMSO, Po, Le, or lenalidomide derivatives for 24 h. The experiment was independently repeated thrice, with similar results. g Dose–response curves of degradation of four neosubstrates by 6-position-modified lenalidomides using HiBiT system. HEK293T cells stably expressing IKZF1, CK1α, SALL4, or PLZF with C-terminal HiBiT-tag were treated with DMSO, Po, Le, or lenalidomide derivatives for 16 h. The protein expression level was expressed as the luminescence signal relative to the luminescence signal of DMSO, which was considered 100. Error bars denote standard deviation (biological replicates; n = 3). h, The half-maximal degradation concentration (DC₅₀) and the maximal degradation (D_max) values were calculated using dose–response curves in (g). N/A means not applicable. Source data are provided as a Source data file.

Fig. 5. Evaluation of global neosubstrate selectivity using 6-position-modified lenalidomides.

a Neosubstrate selectivity of 6-position-modified lenalidomides in HEK293T cells. HEK293T cells were treated with dimethylsulfoxide (DMSO), Po, Le, lenalidomide derivatives, or CC-122 for 48 h. The experiment was independently repeated thrice, with similar results. b–d Neosubstrate selectivity of 6-position-modified lenalidomides in HuH7, Jurkat, and HaCaT cells. (b) HuH7, (c) Jurkat, or (d) HaCaT cells were treated with DMSO, Po, Le, lenalidomide derivatives, CC-122, or CC-885 for 24 h. The experiment was independently repeated twice, with similar results. e Effect of 6-position-modified lenalidomides on MEIS2 protein expression level. HEK293T cells were treated with DMSO, Po, Le, or lenalidomide derivatives in the presence of cycloheximide (CHX) for 4 h. All relative MEIS2 band intensities are expressed as band intensity relative to that of DMSO. Error bars denote the standard deviation (biological replicates; n = 3), and P-values were calculated using one-way ANOVA with Tukey’s post-hoc test (*P < 0.05, **P < 0.01, and ****P < 0.0001). f Heatmap showing the ratio of global neosubstrates in whole-proteome quantification. MM1.S cells were treated with DMSO, 10-µM Th, 10-µM Po, 10-µM Le, or 10-µM lenalidomide derivatives for 5 h; quantitative proteomics analysis was performed using a tandem mass tag (biological replicates, n = 2). Source data are provided as a Source data file.

Fig. 6. Neosubstrate selectivity of PROTACs using 6-position-modified lenalidomides.

a Chemical structures of ARV-825 (Po-P), NE-017 (Le-P), NE-018 (F-P), NE-019 (Cl-P), and NE-020 (F₃C-P). b Binding ability of proteolysis-targeting chimeras (PROTACs) for CRBN was analysed using AlphaScreen (AS) technology. The half-maximal inhibition concentration (IC₅₀) values were calculated using the dose–response curves in (b). c, The ternary complex formation was analysed using AS technology. d Dose–esponse curves of degradation of four neosubstrates using HiBiT system. HEK293T cells stably expressing BRD3-HiBiT were treated with dimethylsulfoxide (DMSO) or PRTOACs for 6 h. Protein expression was expressed as the luminescence signal relative to the luminescence signal of DMSO, which was considered 100. Error bars denote standard deviation (biological replicates; n = 3). The DC₅₀ values were calculated using the dose–response curves in (d). e In-cell proximity-dependent biotinylation of BET proteins and neosubstrates through AirID-CRBN. HuH7 and MM1.S cells expressing AGIA-AirID-CRBN were treated with DMSO or 0.1-µM PROTACs in the presence of 10-µM biotin and 5-µM MG132 for 6 h. The experiment was repeated twice independently, with similar results. f Neosubstrate selectivity of PROTACs based on 6-position-modified lenalidomides. NTERA-2 cells were treated with DMSO, 0.1-µM Po, 0.1-µM Le, 0.1-µM lenalidomide derivatives, or 0.1-µM PROTACs for 24 h. The experiment was independently repeated thrice, with similar results. g–h Immunoblot analysis of BET proteins and neosubstrates. (g) HuH7 or (h) MM.1S cells were treated with DMSO or PROTACs for 24 h. The experiment was independently repeated thrice, with similar results. i Heatmap showing the ratio of BET proteins to neosubstrates in whole-proteome quantification. NTERA-2 and MM1.S cells were treated with DMSO or 0.3-µM PROTACs for 16 h; quantitative proteomics analysis was performed using a tandem mass tag (biological replicates, n = 3). All relative AS signals are expressed as the luminescence signal relative to the luminescence signal of DMSO, which is considered 1. Error bars denote standard deviations (independent experiments, n = 3). Source data are provided as a Source data file.

Fig. 7. Anti-proliferative effect of proteolysis-targeting chimeras (PROTACs) using 6-position-modified lenalidomides on diverse cancer cell lines.

a Dose–response curves of anti-proliferative effects of PROTACs on MM cell lines. MM1.S and H929 cells were treated with dimethylsulfoxide (DMSO), PROTACs, or OTX-015 for 5 days, and cell viability was analysed using the CellTiter-Glo assay kit. Cell viability was expressed as the luminescence signal relative to the luminescence signal of DMSO, which was considered 100. Error bars denote standard deviation (biological replicates; n = 4). b Degradation of BET proteins by PROTACs in a neuroblastoma cell line. IMR32 cells were treated with DMSO or PROTACs for 24 h. The experiment was independently repeated thrice, with similar results. c Dose–response curves of anti-proliferative effects of PROTACs on the neuroblastoma cell line. IMR32 cells were treated with DMSO, PROTAC, or OTX-015 for 4 days, and cell viability was analysed using the CellTiter-Glo assay kit. Cell viability was expressed as the luminescence signal relative to the luminescence signal of DMSO, which was considered 100. Error bars denote standard deviation (biological replicates; n = 4). d Degradation of BET proteins by PROTACs in a colon cancer cell line. HCT116 cells were treated with DMSO or PROTACs for 24 h. The experiment was independently repeated thrice, with similar results. e Dose–response curves of anti-proliferative effects of PROTACs on a colon cancer cell line. HCT116 cells were treated with DMSO, PROTACs, or OTX-015 for 4 days, and cell viability was analysed using the CellTiter-Glo assay kit. Cell viability was expressed as the luminescence signal relative to the luminescence signal of DMSO, which was considered 100. Error bars denote standard deviation (biological replicates; n = 4). f The half-maximal growth inhibition (GI₅₀) and the maximal growth inhibition (GI_max) values were calculated using dose–response curves in (a, c, e). Source data are provided as a Source data file.