Prognostic impact and targeting of CRM1 in acute myeloid leukemia

Abstract

Chromosomal region maintenance 1 (CRM1) is a nuclear export receptor recognizing proteins bearing a leucine-rich nuclear export signal. CRM1 is involved in nuclear export of tumor suppressors such as p53. We investigated the prognostic significance of CRM1 in acute myeloid leukemia (AML) and effects of a novel small-molecule selective inhibitor of CRM1. CRM1 protein expression was determined in 511 newly diagnosed AML patients and was correlated with mouse double minute 2 (MDM2) and p53 levels. High CRM1 expression was associated with short survival of patients and remained an adverse prognostic factor in multivariate analysis. CRM1 inhibitor KPT-185 induced mainly full-length p53 and apoptosis in a p53-dependent manner, whereas inhibition of proliferation was p53 independent. Patient samples with p53 mutations showed low sensitivity to KPT-185. Nuclear retention of p53 induced by CRM1 inhibition synergized with increased levels of p53 induced by MDM2 inhibition in apoptosis induction. KPT-185 and Nutlin-3a, alone and in combination, induced synergistic apoptosis in patient-derived CD34(+)/CD38(-) AML, but not in normal progenitor cells. Data suggest that CRM1 exerts an antiapoptotic function and is highly prognostic in AML. We propose a novel combinatorial approach for the therapy of AML, aimed at maximal activation of p53-mediated apoptosis by concomitant MDM2 and CRM1 inhibition.

Figures

Figure 1

Increased CRM1 expression predicts for poor survival in AML patients. (A) Three-dimensional surface plot of p53 against CRM1 and MDM2. There is a relationship between the 3 components (P = .0002), and the contour blots, projections of a three-dimensional surface onto a two-dimensional plane, demonstrate that CRM1 (positive correlation) and MDM2 level (negative correlation) interactively modulate p53 levels. The color scale refers to the expression (on the log 2 scale like the rest of the data) of the z-axis component (ie, p53). The scale of p53 ranges from <−0.5 to +2.5. The plot was generated using Statistica software. (B) Kaplan-Meier curves and results of multivariate analysis for overall survival in patients with AML. The final model of multivariate analysis includes 6 variables, and high CRM1 expression is an independent predictor of overall survival in AML.

Figure 2

The CRM1 inhibitor KPT-185 inhibits the growth of AML cell lines through cell cycle arrest and apoptosis induction. (A) AML cell lines with wild-type p53 (OCI-AML-3, MOLM-13, and MV4;11) or mutant p53 (NB4, HL-60, KG-1, THP-1, and U937) cells were incubated with 100, 200, or 500 nM for 72 hours, and the annexin V–positive fractions were measured by flow cytometry (black bars). Cells were treated in parallel with 5 µM Nutlin-3a (white bars). Results are expressed as mean ± SD. (B) Lentivirally transduced wild-type p53 AML cells (virus encoding either control shRNA [shC] or p53-specific shRNA [shp53]) were incubated with 0, 10, 20, 50, 100, 200, 500, or 1000 nM of KPT-185 for 72 hours, and the numbers of viable cells and annexin V–positive fractions were measured. *P < .05; **P < .01; ***P < .001.

Figure 3

KPT-185 induces full-length p53. OCI-AML-3 (OCI), MOLM-13 (MOLM), and MV4;11 (MV) cells were incubated with 100 nM KPT-185 for 10 hours, and expression of p53 isoforms was determined. KPT-185 increased the levels of full-length p53 (FL) in all cell lines. p53β/γ (β/γ) was barely detected in KPT-treated OCI-AML-3 and MV4;11 cells. Asterisk (*) indicates p53 fragments from full-length p53 or p53β/γ isoforms. Results are representative of 3 independent experiments. HI, high-intensity image.

Figure 4

KPT-185 and Nutlin-3a synergistically induce p53. (A) OCI-AML-3 cells were treated with 50 nM KPT-185 and/or 2.5 µM Nutlin-3a. KPT-185 synergizes with Nutlin-3a to induce p53. p53 expression levels were expressed as mean fluorescence intensity ratio (MFIR) calculated by the following formula: MFIR = (MFI for anti-p53 antibody)/(MFI for isotypic control). (B) MOLM-13 and MV4;11 cells expressing control shRNA (shC) or p53-specific shRNA (shp53) were treated with 50 nM KPT-185 and/or 2.5 µM Nutlin-3a for 6 hours. Results are expressed as fold change (mean ± SD) relative to the MFIR value in untreated negative control shRNA-expressing cells. Similar results were obtained in 2 other independent experiments. *P < .05; **P < .01; ***P < .001.

Figure 5

The KPT-185/Nutlin-3a combination induces p53-mediated cell death. (A) OCI-AML-3 cells expressing control shRNA (shC) or p53-specific shRNA (shp53) were treated with the indicated concentrations of KPT-185 or Nutlin-3a for 72 hours, either as individual agents or in combination. Cell viability was determined by trypan blue exclusion method. Annexin V–positive fractions were measured by flow cytometry. (B) MOLM-13 and MV4;11 cells expressing control shRNA (shC) or p53-specific shRNA (shp53) were treated with the indicated concentrations of KPT-185 or Nutlin-3a for 72 hours, either as individual agents or in combination. Cell viability was determined by trypan blue exclusion method. Comparable results were obtained in 2 other independent experiments.

Figure 6

The KPT-185/Nutlin-3a combination enhances nuclear function of p53. (A) The KPT-185/Nutlin-3a combination accumulates p53 into the nucleus of AML cells. OCI-AML-3 cells were treated for 6 hours with 100 nM KPT-185 and 2.5 µM Nutlin-3a. Cells were stained for p53 (green) and mitochondrial marker protein cytochrome c oxidase IV (COX-IV, red) and visualized by confocal microscopy. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole(blue). (B) Expression of p53 target proteins in OCI-AML-3 cells, which were treated with 50 nM KPT-185 and/or 2.5 µM Nutlin-3a for 12 hours. Intensity of the immunoblot signals was quantified, and the relative intensity compared with β-actin was calculated. Results are representative of 3 independent experiments.

Figure 7

p53 status affects KPT-185 sensitivity in primary AML cells. Primary AML samples were treated for 48 hours with 800 nM KPT-185 and 8 µM Nutlin-3a, and the annexin V–positive fractions were measured by flow cytometry. (A) AML samples with mutant p53 were less sensitive to KPT-185 compared with those with wild-type p53. (B) FLT3 mutations were associated with increased KPT-185 sensitivity in wild-type p53 samples. Results are expressed as mean ± standard error of the mean. (C) The extent of apoptosis induced by KPT-185 positively correlates with that induced by Nutlin-3a.