Therapeutic potential of GSK-J4, a histone demethylase KDM6B/JMJD3 inhibitor, for acute myeloid leukemia

Abstract

Purpose: Acute myeloid leukemia (AML) is a heterogeneous disease with poor outcomes. Despite increased evidence shows that dysregulation of histone modification contributes to AML, specific drugs targeting key histone modulators are not applied in the clinical treatment of AML. Here, we investigated whether targeting KDM6B, the demethylase of tri-methylated histone H3 lysine 27 (H3K27me3), has a therapeutic potential for AML.

Methods: A KDM6B-specific inhibitor, GSK-J4, was applied to treat the primary cells from AML patients and AML cell lines in vitro and in vivo. RNA-sequencing was performed to reveal the underlying mechanisms of inhibiting KDM6B for the treatment of AML.

Results: Here we observed that the mRNA expression of KDM6B was up-regulated in AML and positively correlated with poor survival. Treatment with GSK-J4 increased the global level of H3K27me3 and reduced the proliferation and colony-forming ability of primary AML cells and AML cell lines. GSK-J4 treatment significantly induced cell apoptosis and cell-cycle arrest in Kasumi-1 cells, and displayed a synergistic effect with cytosine arabinoside. Notably, injection of GSK-J4 attenuated the disease progression in a human AML xenograft mouse model in vivo. Treatment with GSK-J4 predominantly resulted in down-regulation of DNA replication and cell-cycle-related pathways, as well as abrogated the expression of critical cancer-promoting HOX genes. ChIP-qPCR validated an increased enrichment of H3K27me3 in the transcription start sites of these HOX genes.

Conclusions: In summary, our findings suggest that targeting KDM6B with GSK-J4 has a therapeutic potential for the treatment of AML.

Keywords: Acute myeloid leukemia; Histone demethylase; KDM6B; Small molecule inhibitor.

Fig. 1

KDM6B is up-regulated in human AML. a Analysis of KDM6B mRNA expression levels in primary AML samples with t(15;17), inv(16)/t(16;16), t(8;21), t(11q23) translocations, AML complex and normal human HSCs (Lin⁻ CD34⁺ CD38⁻ CD90⁺ CD45RA⁻) using the Bloodspot database. The horizontal bar indicates the median value of each group. **p < 0.01, ***p < 0.001, Mann–Whitney U test. b Comparison of KDM6B mRNA expression in bone marrow (BM) mononuclear cells (MNCs) between AML patients (n = 24) and healthy controls (n = 5, described as normal control). The expression levels of KDM6B were normalized to GAPDH. The horizontal bar indicates the median value of each group. *p < 0.05, Mann–Whitney U test. c Kaplan–Meier analysis showing the AML patients with high expression levels of KDM6B (red, n = 11) had a significant shorter overall survival rate than patients with low expression levels (blue, n = 47). **p < 0.01, Meta-analyses of microarray data are from the PrognoScan database

Fig. 2

GSK-J4 treatment inhibits the proliferation of AML cells. a Primary human AML MNCs were treated with 5.5 µM GSK-J4 or corresponding content of DMSO for 24 h. Cell numbers were counted by Trypan Blue exclusion assay and normalized with the control group. Data represents mean ± SEM. **p < 0.01, ***p < 0.001, NS no significance, two-tailed Student t test. b Total colony numbers of primary AML MNCs with 5.5 μM GSK-J4 treatment were counted after 14-day methylcellulose culture. The red asterisk represents that no colonies were observed. Data represents mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, two-tailed Student t test. c Cytotoxic analysis of GSK-J4 in different human AML cell lines. Cytotoxicity was determined by CCK-8 assay and displayed as inhibition curves with IC₅₀ values. KG-1, KG-1a, THP-1 and Kasumi-1 cells were treated with a range dose of GSK-J4 for 72 h, respectively. Numerical analysis data represents mean ± SEM. d Western blot showed the H3K27me3 levels in KG-1, KG-1a, THP-1 and Kasumi-1 cells. e Globally increased H3K27me3 levels were observed in KG-1, KG-1a and Kasumi-1 cells after 72-h treatment of 5.5 μM GSK-J4

Fig. 3

GSK-J4 inhibits the proliferation and colony formation of Kasumi-1 cells by inducing apoptosis and cell-cycle arrest. a The number and size of colonies decreased in Kasumi-1 cells treated with GSK-J4. ***p < 0.001, two-tailed Student t test. b Growth curve of Kasumi-1 cells treated with GSK-J4 at indicated concentrations. Error bars indicate the SEM of triplicates. c Dose-dependent apoptosis was induced by GSK-J4 in Kasumi-1 cells, as determined by the specific fluorescent probe of mitochondrial membrane potential (MMP), JC-1. In Kasumi-1 cells treated with GSK-J4, increase in the emitted light of JC-1 Green (FITC channel) and decrease in JC-1 Red (PE channel) indicate depolarization of mitochondria during apoptosis, as compared to DMSO treated cells. Data represents mean ± SEM, **p < 0.01, ***p < 0.001, two-tailed Student t test. d Cell-cycle arrest was induced by GSK-J4 in Kasumi-1 cells. After a 24-h GSK-J4 treatment, the percentage of cells at G₀/G₁ phase increased while the S phase and G₂/M phase decreased significantly. Data represents mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001, two-tailed Student t test. e GSK-J4 synergized with Ara-C to inhibit proliferation and colony-forming capability of Kasumi-1 cells in vitro. Combination Index (CI) versus Fractional Effect (FE) plot was derived from CalcuSyn Software and CI values < 1.0 indicate synergy, whereas CI values > 1.0 indicate antagonism. The following table described the dose combinations shown on the CI versus FE plot. f The colony numbers of Kasumi-1 cells treated with GSK-J4, Ara-C and a combination of these two drugs were shown. Data represents mean ± SEM, ***p < 0.001, two-tailed Student t test

Fig. 4

GSK-J4 reduces tumor burden in the Kasumi-1 cells transplanted xenograft mouse model in vivo. a Schema depicted the strategy of NCG mice transplantation with human Kasumi-1 cells. GSK-J4 was applied at a dose of 50 mg/kg during 4–6 weeks by intraperitoneal injection. b The percentage of hCD45⁺ mCD45⁻ human leukemic cells in the BM of mice was analyzed 4 weeks after GSK-J4 or DMSO treatment (week 10). The horizontal bar indicates the median value of each group. *p < 0.05, two-tailed Student t test. Representative scatter plots of flow cytometry showed the engrafted human leukemic cell populations after DMSO or GSK-J4 treatment. c The proportions of hCD45⁺ mCD45⁻ cell engraftments in the BM were divided into three groups (0–20%, 20–50% and 50–100%) after mice treated with DMSO or GSK-J4. d Representative photographs of histological BM sections by H&E staining from mice treated with DMSO or GSK-J4 were shown. Scale bars represent 50 μm. Lightly stained, small nuclei cells with rich cytoplasm are human leukemia cells (red arrows), while the deeply stained, relatively smaller cells are normal mouse hematopoietic cells (blue arrows)

Fig. 5

GSK-J4 treatment alters the transcriptional profiles of Kasumi-1 cells. a Heatmap of the significantly differentially expressed genes between DMSO and GSK-J4 treated Kasumi-1 cells (n = 3 per group, adjusted p < 0.05, fold change > 1.6). b Dot plots showed the top-20 significantly enriched GO terms for down-regulated genes (adjusted p < 0.05). c Histogram of top-20 significantly enriched KEGG pathways for down-regulated genes (adjusted p < 0.05). d The expression levels of key down-regulated genes related to DNA replication and cell-cycle progression in Kasumi-1 cells treated with GSK-J4 were verified by qPCR. Gene expression was normalized to GAPDH. ***p < 0.001, two-tailed Student t test. e HOX genes were down-regulated in Kasumi-1 cells treated with GSK-J4. *p < 0.05, **p < 0.01, ***p < 0.001, two-tailed Student t test. f ChIP-qPCR revealed the increased H3K27me3 enrichment in the promoter regions of HOXA5, HOXA7, HOXA9 and HOXA11 genes in GSK-J4 treated Kasumi-1 cells. **p < 0.01, ***p < 0.001, two-tailed Student t test