Fig. 2: ENL degradation by rac-dHTC1 proceeds through CRL4^CRBN despite minimal CRBN affinity.

a, FACS-based CRISPR screens for UPS components affecting ENL stability in KBM7 reporter cells (ENL–BFP–P2A–mCherry) treated with DMSO or rac-dHTC1 (10 µM) for 8 h (6 sgRNAs targeting 1,301 UPS-associated genes). Gene-level fold changes and P values were determined by one-sided MAGeCK analysis¹¹⁴. b, Immunoblot analysis of WT treated with rac-dHTC1 (10 µM, 6 h) or DMSO vehicle control. Data are representative of two independent experiments. c, SPR analysis of CRBN–DDB1 binding to immobilized ENL YEATS in the presence of rac-dHTC1 or buffer control (mean; n = 2 technical replicates). The same experiment is presented in Extended Data Fig. 3g; rac-dHTC1 treatment was repeated in both panels. d, Rescue of dBET6-mediated BRD4 degradation to assess CRBN target engagement. MV4;11 cells expressing BRD4–HiBiT were pretreated with compounds in a dose–response series for 2 h before a 1-h treatment with dBET6 (500 nM). HiBiT luminescence signal is normalized to DMSO-treated cells (mean ± s.e.m.; n = 3 biological replicates). Data for pomalidomide are repeated in Extended Data Fig. 4e, which originated from the same experiment. Conc., concentration. e, SPR analysis of pomalidomide binding to immobilized CRBN^midi (mean ± s.d.; n = 3 technical replicates). f, Same as e, but for rac-dHTC1. g, Heat map depicting CRBN FP signal (background-subtracted and normalized to DMSO) following treatment with the indicated compounds, in the presence of ENL (3 µM) or buffer control (mean; n = 3 biological replicates). Cooperativity values are calculated from the ratio of IC₅₀ values with and without ENL. h, Structures of dHTC1 enantiomers and their time-dependent and dose-responsive loss of HiBiT–ENL signal in MV4;11 cells treated with rac-dHTC1. The HiBiT luminescence signal was normalized to the DMSO vehicle control (mean ± s.e.m.; n = 4 biological replicates).

Source data