BET bromodomain inhibition targets both c-Myc and IL7R in high-risk acute lymphoblastic leukemia

Abstract

We investigated the therapeutic potential of JQ1, an inhibitor of the BET class of human bromodomain proteins, in B-cell acute lymphoblastic leukemia (B-ALL). We show that JQ1 potently reduces the viability of B-ALL cell lines with high-risk cytogenetics. Among the most sensitive were lines with rearrangements of CRLF2, which is overexpressed in ~ 10% of B-ALL. CRLF2 heterodimerizes with the IL7 receptor (IL7R) and signals through JAK2, JAK1, and STAT5 to drive proliferation and suppress apoptosis. As previously observed, JQ1 induced the down-regulation of MYC transcription, the loss of BRD4 at the MYC promoter, and the reduced expression of c-Myc target genes. Strikingly, JQ1 also down-regulated IL7R transcription, depleted BRD4 from the IL7R promoter, and reduced JAK2 and STAT5 phosphorylation. Genome-wide expression profiling demonstrated a restricted effect of JQ1 on transcription, with MYC and IL7R being among the most down-regulated genes. Indeed, IL7R was the only cytokine receptor in CRLF2-rearranged B-ALL cells significantly down-regulated by JQ1 treatment. In mice xenografted with primary human CRLF2-rearranged B-ALL, JQ1 suppressed c-Myc expression and STAT5 phosphorylation and significantly prolonged survival. Thus, bromodomain inhibition is a promising therapeutic strategy for B-ALL as well as other conditions dependent on IL7R signaling.

Figure 1

Inhibition of proliferation and induction of apoptosis by JQ1 in B-ALL. (A) Dose-response of B-ALL cell line viability using a luminescent ATP detection assay with JQ1 treatment. Error bars represent SEM; n = 4. Inset: The chemical structure of JQ1. Table shows 50% growth inhibition values (GI₅₀) and maximum effected cells (E_max) with annotated characteristic translocations for each cell line. (B) MHH-CALL4 and MUTZ-5 cell numbers with 500nM JQ1 treatment, normalized to time = 0 levels. Error bars represent SEM; n = 4 counts. (C) Caspase-3 and -7 activity and cell viability with 500 nm JQ1 treatment; data shown relative to vehicle control values and normalized to baseline time = 0 levels. Error bars represent SEM; n = 3 measurements. (D) Cellular apoptosis after treatment with DMSO (Veh) or 500nM JQ1 for 48 hours by flow cytometry with PI and AV. (E) PARP immunoblotting with 500nM JQ1 treatment. *Cleaved PARP band; β-tubulin (TUB) shown as loading control. (F) Cell cycle analysis of total DNA content by PI staining with 500nM JQ1 treatment.

Figure 2

JQ1 decreases c-Myc expression in CRLF2-rearranged B-ALL. (A) Quantitative PCR analysis of MYC transcript levels in MHH-CALL4 and MUTZ-5 B-ALL cells treated with 500nM JQ1. β-actin expression shown as control. Data for each time point are normalized to vehicle control and presented as the ratio of expression compared with baseline expression at time = 0. Error bars represent ± SEM. *P < .1 (paired t test). **P < .01 (paired t test). (B) Immunoblotting for c-Myc in whole cell lysates of cells treated with 500nM JQ1. (C) ChIP with a BRD4 antibody at 2 sites within the MYC promoter region in cells treated with 500nM JQ1 for 4 hours. Enrichment is shown as the percentage of total input DNA. The negative control region primers amplify within a gene desert ∼ 1 Mb upstream of MYC. (D) Heatmaps of MYC and c-Myc target gene expression in MHH-CALL4 and MUTZ-5 cells treated with 500nM JQ1 for 4 hours. Scale bars represent fold expression changes of each gene relative to the average gene expression across the 4 samples. HK indicates housekeeping controls. Data for 2 biologic replicates are shown.^,

Figure 3

Bromodomain inhibition decreases IL7R expression and suppresses JAK2/STAT5 phosphorylation. (A) Immunoblotting of whole cell lysates from cells treated with 500nM JQ1 at indicated time points. (B) Quantitative PCR of CRLF2 transcript levels in MHH-CALL4 and MUTZ-5 cells treated with 500nM JQ1. Data for each time point are normalized to vehicle control and presented as a ratio of expression compared with baseline expression at time = 0. Error bars represent ± SEM. *P < .1 (paired t test). **P < .01 (paired t test). (C) Quantitative PCR analysis of IL7R transcripts levels, as in panel B. (D) Flow cytometry for CRLF2 and IL7R. Cells were treated with 500nM JQ1 or vehicle for 8 and 24 hours and stained with an antibody against the indicated target or an isotype IgG. (E) ChIP with a BRD4 antibody at 2 sites within the IL7R promoter region in cells treated with 500nM JQ1 for 4 hours. Enrichment is shown as the percentage of total input DNA. Negative control region primers amplify within a gene desert region ∼ 70 kb downstream of IL7R. *P < .1 (unpaired t test). **P < .01 (unpaired t test). (F) Quantitative PCR of MYC and IL7R expression and (G) flow cytometry of IL7R expression in B-ALL cell lines that lack CRLF2 rearrangements after 8-hour treatment with 500nM JQ1 or vehicle, as above.

Figure 4

Genome-wide transcriptome analysis of JQ1-treated cells. (A) Volcano plots of gene expression differences for MHH-CALL4 and MUTZ-5 cells treated with vehicle or 500nM JQ1 for 8 hours. (B) Correlation of gene expression changes between MHH-CALL4 and MUTZ-5 cells treated with JQ1. (C) Heatmap of the top 50 significantly up- and down-regulated genes. P < .01. Each gene row is normalized to mean expression level (M4 = MHH-CALL4; M5 = MUTZ-5). (D) RMA-normalized expression values for all expressed cytokine receptors. (E) Gene set enrichment analysis of 2 c-Myc signatures. (F) c-Myc and NF-κB binding site gene sets. (G) STAT5A binding site gene set. (H) An antiapoptosis gene set. NES indicates normalized enrichment score; and q = false discovery rate.

Figure 5

JQ1 decreases leukemic burden and improves survival in B-ALL primary human xenografts. (A) Immunohistochemistry for c-Myc and phospho-STAT5 (pSTAT5) in spleen samples obtained after 5-day treatment or at the moribund state. (B) Quantitative PCR of MYC and IL7R expression in leukemia cells harvested from spleens of mice treated for 5 days with JQ1 or vehicle (Veh). Error bars represent ± SEM; n = 3 PCR samples normalized to vehicle-treated sample. (C) Flow cytometry of splenic cells obtained after 5-day treatment with JQ1 or vehicle. Scatter plots on the left show overall IL7R expression versus hCD45. The histogram on the right is a quantification of total IL7R expression. (D) Peripheral blood cell counts after 5 days of treatment (n = 6 in each arm, mean levels for each sample marked). WBC indicates white blood cells. (E) Survival of engrafted mice treated with JQ1 (n = 9) or vehicle (n = 9).