FOXM1 contributes to treatment failure in acute myeloid leukemia
Irum Khan, … , Olga Frankfurt, Andrei L. Gartel
Irum Khan, … , Olga Frankfurt, Andrei L. Gartel
Published August 9, 2018
Citation Information: JCI Insight. 2018;3(15):e121583. https://doi.org/10.1172/jci.insight.121583.
View: Text | PDF
Research Article Hematology Oncology

FOXM1 contributes to treatment failure in acute myeloid leukemia

Abstract

Acute myeloid leukemia (AML) patients with NPM1 mutations demonstrate a superior response to standard chemotherapy treatment. Our previous work has shown that these favorable outcomes are linked to the cytoplasmic relocalization and inactivation of FOXM1 driven by mutated NPM1. Here, we went on to confirm the important role of FOXM1 in increased chemoresistance in AML. A multiinstitution retrospective study was conducted to link FOXM1 expression to clinical outcomes in AML. We establish nuclear FOXM1 as an independent clinical predictor of chemotherapeutic resistance in intermediate-risk AML in a multivariate analysis incorporating standard clinicopathologic risk factors. Using colony assays, we show a dramatic decrease in colony size and numbers in AML cell lines with knockdown of FOXM1, suggesting an important role for FOXM1 in the clonogenic activity of AML cells. In order to further prove a potential role for FOXM1 in AML chemoresistance, we induced an FLT3-ITD–driven myeloid neoplasm in a FOXM1-overexpressing transgenic mouse model and demonstrated significantly higher residual disease after standard chemotherapy. This suggests that constitutive overexpression of FOXM1 in this model induces chemoresistance. Finally, we performed proof-of-principle experiments using a currently approved proteasome inhibitor, ixazomib, to target FOXM1 and demonstrated a therapeutic response in AML patient samples and animal models of AML that correlates with the suppression of FOXM1 and its transcriptional targets. Addition of low doses of ixazomib increases sensitization of AML cells to chemotherapy backbone drugs cytarabine and the hypomethylator 5-azacitidine. Our results underscore the importance of FOXM1 in AML progression and treatment, and they suggest that targeting it may have therapeutic benefit in combination with standard AML therapies.

Authors

Irum Khan, Marianna Halasi, Anand Patel, Rachael Schultz, Nandini Kalakota, Yi-Hua Chen, Nathan Aardsma, Li Liu, John D. Crispino, Nadim Mahmud, Olga Frankfurt, Andrei L. Gartel

×

Figure 3

Ixazomib inhibits FOXM1 activity.

Options: View larger image (or click on image) Download as PowerPoint
Ixazomib inhibits FOXM1 activity. (A) KG-1 FOXM1-knockdown cells show sm...
(A) KG-1 FOXM1-knockdown cells show smaller colony size and reduced colony numbers compared with cells transduced with empty vector in a methylcellulose assay. Colonies were imaged with the EVOS XL Core Imaging System using the 4× objective. (B) Plot represents the mean ± SEM of 3 independent colony assay experiments, each plated in duplicate. P < 0.05 by unpaired 2-tailed t test. (C) Ixazomib inhibits FOXM1 transcriptional activity in an inducible luciferase cell line. The luciferase activity was determined by using the Luciferase Assay System (Promega). We show significant dose-dependent inhibition of FOXM1 transcriptional activity. Plot shows quantification as fold induction of firefly luciferase activity compared with control cells, mean ±SD of a representative triplicate experiment. P < 0.05 by 1-way ANOVA followed by Tukey’s multiple comparison post test. (D) Ixazomib-treated and untreated KG-1 leukemia cells were collected for RNA extraction. Quantitative PCR was carried out with FOXM1, AurkB, Cdc25B, and Plk1 primers. Graph shows quantification as percentage of mRNA expression levels in treated cells compared with control cells; mean ± SEM of 3 independent experiments. P < 0.05 by 1-way ANOVA followed by Tukey’s multiple comparison post test. (E and F) In a panel of AML cell lines (KG-1 and HL-60), FOXM1 protein expression was suppressed by treatment with ixazomib, as detected by immunoblotting. This also correlated with stabilization of HSP-70 and caspase-3 cleavage. (G and H) FLT3-ITD–transformed primary murine BM cells (generated as described in Figure 2, B and C) were sorted and treated as indicated and studied in colony and apoptotic assays, as assessed by flow cytometry after annexin V–PE staining. Plots show resistance to ixazomib treatment in FOXM1-overexpressing BM cells compared with their treated control counterparts. Data are expressed as the mean ± SEM (n = 2/group); P < 0.05 by 1-way ANOVA followed by Tukey’s multiple comparison post test.