Skip to main content
. 2019 Mar 19;216(4):966–981. doi: 10.1084/jem.20181276

Figure 2.

Figure 2.

Ezh2/EZH2 is a viable therapeutic target across multiple AML genotypes and in primary AML samples. (a) Representative cell proliferation assay plot demonstrating Ezh2+/+; MLL-AF9 tumor splenocytes cultured in liquid medium exhibit sensitivity to the EZH2 inhibitor GSK343 with an IC50 value of ∼10 µM. (b) Left: Clonogenic assay for Ezh2+/+; MLL-AF9 tumor splenocytes performed in the presence of DMSO (vehicle control) or 10 µM GSK343 (performed in triplicate, P = 0.0003, two-tailed t test). Right: Photomicrographs demonstrating reduction in colony size upon treatment with GSK343 compared with DMSO (bars, 100 µm). (c) Flow cytometry at 24 h for H3K27me3 and total H3 demonstrates that 10 µM GSK343 significantly reduces H3K27me3 over DMSO treatment in Ezh2+/+; MLL-AF9 tumor splenocytes cultured in liquid medium. (d) Liquid culture growth assay for human AM1-ETO–positive cell line Kasumi-1 over a 12-d time course in the presence of 10 µM GSK343 compared with DMSO (performed in duplicate). (e) 10 µM GSK343 treatment leads to a modest increase in apoptosis at late time points (96 h) in the Kasumi-1 cell line compared with DMSO (P = 0.037, two-tailed t test, performed in triplicate). (f) Representative cell cycle analysis plot at 96 h in Kasumi-1 also demonstrates mild G0/G1 cell cycle arrest following 10 µM GSK343 treatment. (g) Colony assays across a wide range of AML patient samples (n = 15) demonstrates a significant reduction in colony numbers following treatment with 10 µM GSK343 compared with DMSO. Cases include varying molecular subtypes, with variable karyotypic mutational and prognostic status. (P = 0.01, two-tailed t test). (h) Photomicrographs of two different human primary AML samples in a clonogenic assay cultured with either DMSO or 10 µM GSK343 (bars, 500 µm). (i) Volcano plot for DMSO- vs. 10 µM GSK343–treated Ezh2+/+; MLL-AF9 tumor splenocytes cultured in liquid medium demonstrating gene expression changes and showing the adjusted significance P value (log10) vs. fold change (log2; n = 2, biological replicates for DMSO and GSK343). Potential candidates that might underlie the antileukemic effects, Cdkn1a and Nfkbiz, are identified. (j) Kaplan–Meier graph of survival of secondary transplants of AML1-ETO9a AML generated from Ezh2+/+ mice comparing EPZ-6438–treated (n = 7) and vehicle control–treated (n = 6; log-rank [Mantel–Cox] test P = 0.0103). *, P < 0.05; **, P < 0.01. All error bars are ± SEM.