KPT-330 inhibitor of CRM1 (XPO1)-mediated nuclear export has selective anti-leukaemic activity in preclinical models of T-cell acute lymphoblastic leukaemia and acute myeloid leukaemia

Authors

  • Julia Etchin,

    1. Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Takaomi Sanda,

    1. Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Marc R. Mansour,

    1. Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Alex Kentsis,

    1. Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Joan Montero,

    1. Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Bonnie T. Le,

    1. Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Amanda L. Christie,

    1. Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Dilara McCauley,

    1. Karyopharm Therapeutics, Natick, MA, USA
    Search for more papers by this author
  • Scott J. Rodig,

    1. Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Michael Kauffman,

    1. Karyopharm Therapeutics, Natick, MA, USA
    Search for more papers by this author
  • Sharon Shacham,

    1. Karyopharm Therapeutics, Natick, MA, USA
    Search for more papers by this author
  • Richard Stone,

    1. Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Anthony Letai,

    1. Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • Andrew L. Kung,

    1. Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    2. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    Search for more papers by this author
  • A. Thomas Look

    Corresponding author
    1. Division of Hematology/Oncology, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
    • Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
    Search for more papers by this author

Correspondence: Dr A. Thomas Look, Dana-Farber Cancer Institute, Mayer Bldg, Rm 630, 450 Brookline Ave, Boston, MA 02215, USA.

E-mail: thomas_look@dfci.harvard.edu

Summary

This study explored the anti-leukaemic efficacy of novel irreversible inhibitors of the major nuclear export receptor, chromosome region maintenance 1 (CRM1, also termed XPO1). We found that these novel CRM1 antagonists, termed SINE (Selective Inhibitors of Nuclear Export), induced rapid apoptosis at low nanomolar concentrations in a panel of 14 human T-cell acute lymphoblastic leukaemia (T-ALL) cell lines representing different molecular subtypes of the disease. To assess in vivo anti-leukaemia cell activity, we engrafted immunodeficient mice intravenously with the human T-ALL MOLT-4 cells, which harbour activating mutations of NOTCH1 and NRAS as well as loss of function of the CDKN2A, PTEN and TP53 tumour suppressors and express a high level of oncogenic transcription factor TAL1. Importantly, we examined the in vivo anti-leukaemic efficacy of the clinical SINE compound KPT-330 against T-ALL and acute myeloid leukaemia (AML) cells. These studies demonstrated striking in vivo activity of KPT-330 against T-ALL and AML cells, with little toxicity to normal murine haematopoietic cells. Taken together, our results show that SINE CRM1 antagonists represent promising ‘first-in-class’ drugs with a novel mechanism of action and wide therapeutic index, and imply that drugs of this class show promise for the targeted therapy of T-ALL and AML.

The treatment of acute lymphoblastic leukaemia (ALL) has improved over the last few decades as a result of the combination of intensive chemotherapy, radiotherapy and stem cell transplantation. However, T-cell acute lymphoblastic leukaemia (T-ALL) remains fatal in c. 25% of children and in 50–70% of adults, prompting the need to develop new therapies (Pui & Evans, 2006; Pui et al, 2008). In this study, we explored selective inhibition of nuclear-cytoplasmic trafficking as a new anti-T-ALL therapeutic strategy and demonstrate striking anti-leukaemic efficacy of novel inhibitors of nuclear exporter CRM1 (exportin 1 (CRM1 homolog, yeast); XPO1) in preclinical models of T-ALL.

Nuclear-cytoplasmic transport is a fundamental property of eukaryotic cells, mediated in part by the karyopherin family of proteins, which transport proteins and ribonucleic acids between the nucleus and the cytoplasm (Xu et al, 2010; Siddiqui & Borden, 2012). The major nuclear exporter protein CRM1, one of seven exportins, mediates the transport of c. 220 proteins (Xu et al, 2012a) and several mRNAs. Interestingly, CRM1 is the sole nuclear exporter of the major tumour suppressor and growth regulatory proteins p53 (TP53), p73 (TP73), FOXO (FOXO1; counteracts PI3K/AKT), NFkB1 (NFKB1), Rb (RB1), p21 (CDKN1A, and NPM (NPM1) (Fornerod et al, 1997; Fukuda et al, 1997; Ossareh-Nazari et al, 1997; Turner et al, 2012). CRM1 is upregulated in a range of solid tumours and haematological malignancies and its overexpression is correlated with poor prognosis, suggesting that alterations in nuclear-cytoplasmic trafficking, and hence mislocalization of tumour suppressor proteins, cell cycle regulators, and/or pro-apoptotic proteins, could lead to oncogenesis and resistance to chemotherapy (Noske et al, 2008; Huang et al, 2009; Shen et al, 2009; van der Watt et al, 2009; Yao et al, 2009).

Chromosome region maintenance 1 recognises export cargos that contain short leucine–rich nuclear export signal (NES) consensus sequences (Dong et al, 2009; Monecke et al, 2009; Guttler et al, 2010; Xu et al, 2012b). Extensive studies with well-established natural product CRM1 inhibitors leptomycin B, ratjadone, anguinomycin, and goniothalamin, and recently developed small molecule inhibitors of CRM1, such as N-azolylacrylates, KOS-2464, and CBS9106 (Kudo et al, 1999; Daelemans et al, 2002; Meissner et al, 2004; Van Neck et al, 2008; Mutka et al, 2009; Bonazzi et al, 2010; Wach et al, 2010; Sakakibara et al, 2011) have clearly demonstrated the requirement of CRM1 nuclear export activity for the growth and survival of cancer cells. Blockade of CRM1 transport by these inhibitors has been shown to induce cancer cell death, possibly by promoting the forced nuclear retention of tumour suppressor proteins that are normally inactivated by cytoplasmic mislocalization. Moreover, interference with CRM1-directed nuclear export by these inhibitors has been shown to promote nuclear localization of topoisomerase IIα and to sensitise multiple myeloma cells to the topoisomerase II inhibitors etoposide and doxorubicin (Turner et al, 2009). However, despite the ability of existing CRM1 inhibitors to counteract the CRM1-mediated nuclear export and to promote anti-proliferative and apoptotic signalling pathways in cancer cells, these compounds exhibit extensive toxic effects against normal cells, apparently due to both on-target and possibly off-target activities (Mutka et al, 2009; Sakakibara et al, 2011). These caveats clearly emphasise the need for the development of CRM1 inhibitors with increased selectivity for cancer cells and reduced toxicity to normal cells as a prerequisite for their translation into clinical use. We and others have recently reported the striking anti-AML activity and high selectivity of a new class of drug-like, small molecule CRM1 antagonists called Selective Inhibitors of Nuclear Export, or SINE (Etchin et al, 2012; Ranganathan et al, 2012). SINE drugs were developed based on an in silico molecular modelling strategy, in which a structural model of the NES groove of CRM1 is used as a framework for selection and optimization of virtual library of irreversible CRM1 inhibitors (Etchin et al, 2012; Turner et al, 2012).

Recently, the first ever clinical trials of an oral SINE compound, KPT-330, were initiated, with two trials running in parallel: one includes patients with advanced solid tumours whose disease has progressed after at least one prior therapy for metastatic disease (NCT01607905); the second includes patients with advanced haematological malignancies including chronic lymphocytic leukaemia, non-Hodgkin lymphoma, multiple myeloma, and Waldenstrom macroglobulinaemia whose disease has relapsed after standard therapies (NCT01607892). Patients with AML will be eligible in future clinical trials once the tolerability profile of KPT-330 has been established.

The present study showed that the SINE compounds are highly active against human T-ALL cells carrying different genetic alterations. These compounds induce rapid apoptosis in T-ALL cells in vitro and promote striking growth suppression of T-ALL cells engrafted into immunodeficient mice. Importantly, our data demonstrate that KPT-330 is very active in preclinical models of T-ALL as well as AML, with minimal toxicity to normal blood cells both in the periphery and in the bone marrow. These data indicate that KPT-330 is a promising drug for the treatment of T-ALL as well as AML, and support the ongoing and future development of this novel class of agents.

Materials and methods

SINE CRM1 antagonists

KPT-185, KPT-251 (Etchin et al, 2012), and KPT-330 (molecular weight of 443·31, chemical formula: C17H11F6N7O) are structurally similar, selective CRM1 inhibitors with distinct pharmacokinetic (PK) properties (Karyopharm Therapeutics Inc., Natick, MA, USA). KPT-251 and -330 are suitable for in vivo use; KPT-185 is the most potent CRM1 inhibitor but has very poor PK properties making it unsuitable for in vivo use.

Cell lines and cell viability assay

T-ALL cell lines (HPB-ALL, DU528, Jurkat, MOLT-4, SKW-3, HSB-2, KOPTK1, PF-382, CCRF-CEM, SUPT1, MOLT-16, P12-ICHIKAWA, LOUCY) were cultured in RPMI 1640 medium (GIBCO, Grand Island, NY, USA), supplemented with 10% fetal bovine serum and penicillin/streptomycin. Cell Titer Glo assay (Promega, Madison, WI, USA) was used to assess cell viability upon treatment with either dimethyl sulfoxide (DMSO) or KPT-185. Cells were plated at a density of 10 000 cells per well in a 96-well plate and incubated with DMSO or increasing concentrations of KPT-185. The cell viability was measured after 72 h exposure to KPT-185 and reported as a percentage of DMSO control cells. Jurkat cells that overexpress BCL2 were generated using MSCV-IRES-GFP retroviral expression system. Jurkat cells infected with BCL2 or control vector viruses were sorted by flow cytometry and the expression of BCL2 confirmed by Western blot analysis using BCL2 antibody (Cell Signaling, Danvers, MA, USA).

Apoptosis analysis

Jurkat and MOLT-4 cells were incubated with either DMSO control or KPT-185 for 6 or 13 h, washed with phosphate-buffered saline (PBS), and co-incubated with Annexin V- fluorescein isothiocynate (FITC) and propidium iodide (PI) from MEBCYTO Apoptosis Kit (MBL Co., Ltd., Nagoya, Japans). Cells were analysed by two-colour FACS cytometry (BD FACS Canto, BD Biosciences, San Jose, CA, USA) and the percentage of Annexin V and PI positive cells was determined based on the dot plots of FITC vs. PI.

Mitochondrial Sensitivity in permeabilized whole cells

2 × 104 cells/well of Jurkat cells were used. One volume of the 4 × single cell suspension was added to one volume of a 4 × dye solution (4 μmol/l JC-1, 40 μg/ml oligomycin, 0·02% digitonin, 20 mmol/l 2-mercaptoethanol) in (300 mmol/l Trehalose, 10 mmol/l HEPES-KOH pH 77, 80 mmol/l KCl, 1 mmol/l EGTA, 1 mmol/l EDTA, 0·1% bovine serum albumin, 5 mmol/l succinate) T-EB. This 2 × cell/dye solution was incubated for 5–10 min at room temperature to allow permeabilization and dye equilibration. 15 μl of the cell/dye mix was then added to each treatment well of the black 384-well plate (BD Falcon no. 353285) and the fluorescence at 590 nm monitored every 5 min at room temperature. Percentage loss of Ψm was calculated by normalisation to the solvent only control DMSO (0%) and the positive control FCCP (Ryan et al, 2010).

Cell cycle analysis

Jurkat and MOLT-4 cells were incubated with serial dilutions of KPT-185 for 24 h, washed with PBS, fixed with 70% ethanol, and incubated overnight at −20°C. The cells were then washed with PBS, stained with PI/RNase staining buffer (BD Biosciences), and analysed by flow cytometry using BD FACS Canto (BD Biosciences). The DNA histograms of Jurkat and MOLT-4 cells were analysed using FCS Express 4 Flow Cytometry cell cycle analysis software (De Novo Software, Los Angeles, CA, USA) and ModFit LT cell cycle analysis software (Verity Software House, Topsham, ME, USA).

Orthograft mouse models

T-ALL orthograft mouse model

MOLT-4 cells (3 × 106) expressing luciferase were injected into 7-week-old female NOD-SCID-IL2Rcγnull (NSG) mice (The Jackson Laboratory, Bar Harbor, ME, USA) via tail-vein injections. The leukaemia burden was established by bioluminescence imaging (BLI) using an IVIS Spectrum system (Caliper Life Sciences, Hopkinton, MA, USA) every 3–5 d. After onset of leukaemia, mice were divided into three groups (n = 8) and treated by oral gavage either with vehicle control (Pluronic F-68/PVP-K29/32), KPT-251 (50 mg/kg on days 1, 4, 6; 75 mg/kg on days 8, 11, 13, 15, 25, and 27 or until mice became moribund), or KPT-330 (20 mg/kg for days 1, 4, 6; and 25 mg/kg on days 8, 11, 13, 15, 25, 27, 29, 32, 34, and 36 or until mice became moribund) 3 times per week.

AML orthograft mouse model

Luciferase-expressing MV4-11 cells (2 × 106) were intravenously injected into 7-week-old female NSG mice. After leukaemia progression was established by BLI, mice were split into two groups of nine mice and treated with either vehicle (Pluronic F-68/PVP-K29/32) or KPT-330 three times per week at 20 mg/kg (days 1–7) and 25 mg/kg (days 8–35). Following 5 weeks of treatment, femur from one mouse from the treatment group was fixed in 10% formalin, sectioned, and paraffin-embedded. Slides were stained with haematoxylin and eosin and photographed using an Olympus BX41 microscope with Q-color5 digital camera (Olympus, Center Valley, PA, USA).

For T-ALL and AML in vivo studies, peripheral blood counts were analysed using Hemavet 950 F instrument (Drew Scientific, Dallas, TX, USA). Survival of the KPT-treated mice was measured as the time from the start of treatment until moribund state. Survival benefit was assessed by Kaplan–Meier survival analysis.

All animal studies were performed using protocols approved by the Dana–Farber Cancer Institute Institutional Care and Use Committee.

Results

Novel SINE CRM1 antagonists promote rapid apoptosis in T-ALL cell lines in vitro

To assess the anti-leukaemic activity of SINE compounds against T-ALL cells, we tested the effects on the cell viability of KPT-185, one of the three structurally highly related SINE compounds (KPT-185, KPT-251, and KPT-330). The growth of 13 T-ALL lines, which harbour different genetic aberrations, was dramatically reduced in response to treatment with KPT-185, with 50% inhibitory concentration (IC50) values of 16-395 nmol/l after 72 h of exposure (Fig 1A and Table 1). No particular genetic abnormality was associated with sensitivity to KPT-185. To determine whether the observed decrease in cell viability is due to apoptosis, we measured the effects of KPT-185 on two sensitive T-ALL lines, Jurkat and MOLT-4, with the early apoptosis marker, Annexin V. For these experiments, Jurkat and MOLT-4 cells were co-stained with propidium iodide (PI) to monitor for the appearance of the late-apoptotic and/or necrotic events. Analysis of Annexin V-staining showed a dose-dependent increase in the percentage of apoptotic Jurkat and MOLT-4 cells upon incubation of cells with 30, 60, or 120 nmol/l KPT-185 for 6 or 13 h when compared to DMSO-treated cells (Fig 2). Strikingly, both Jurkat and MOLT-4 cells demonstrated increased fractions of early apoptotic cells upon incubation with KPT-185 for only 6 h. For example, 69% of Jurkat cells exhibited positive Annexin V staining following incubation with 30 nmol/l of KPT-185 for 6 h (Fig 2B). Apoptotic MOLT-4 cells were detected in response to treatment with 120 nmol/l KPT-185 for 6 h with pronounced levels of apoptosis being observed after 13-h exposure. Specifically, 48% of MOLT-4 cells stained positive for Annexin V following treatment with 120 nmol/l KPT-185 for 13 h (Fig 2A).

Table 1. Sensitivity of 14 T-ALL cell lines to KPT-185
Cell lineOncogene groupMajor chromosomal rearrangement NOTCH1 FBXW7 MYB PTEN KRAS/NRAS TP53 Other genesIC50 at 72 h (nmol/l)
  1. References: Catalogue of Somatic Mutations in Cancer (Forbes et al, 2008, 2011), Maser et al (2007); O'Neil et al (2007); Palomero et al (2007); Weng et al (2004).

  2. IC50, 50% inhibitory concentration; MUT, mutated; WT, wild-type.

HPB-ALL TLX3 MUTWT WTJAK1 E966V16
DU.528 TAL1 TAL1-TRD@ WTMUTGAINWT17
JURKAT TAL1 MUTMUTNULLMUTCREBBP MUT19
MOLT-4 TAL1 MUTWTGAINNRAS G12CMUT 23
SKW-3/KE-37 MYC-TRA@ MUTGAIN29
HS-B2 TAL1 LCK-TRB@, STIL-TAL1 WTMUTRB1 deletion46
KOPT-K1 LMO2-TRG@ MUTWTWTMUT50
PF-382 TAL1 MUTWTWTNULLNRAS G12SMUT59
CCRF-CEM TAL1 STIL-TAL1 MUTMUTGAINNULLKRAS G12DMUT 74
SUPT1(t7;9)(q34;q34.3)WTWTWTMUT88
MOLT-16 TAL1 MYC-TRA@, STIL-TAL1 WTWTWTWTMUT112
P12-ICHIKAWA LMO2 MUTMUTNULLNRAS G12DMUT132
LOUCY ETP WTWTWTMUT395
Figure 1.

KPT-185 promotes cell death in T-ALL cells. (A) Cell viability of T-ALL cell lines after treatment with KPT-185 for 72 h (Cell Titer Glo assay, Promega). (B) Relative cell viability of Jurkat cells that either overexpress BCL2 or carry control MSCV vector. Cell viability was measured using Cell Titer Glo assay following treatment with KPT-185 for 24 h. (C) Mitochondrial sensitivity of Jurkat cells that either overexpress BCL2 (Jurkat BCL2) or carry control MSCV vector (Jurkat) in presence of 0·03 μmol/l Bim peptide following 24-h exposure with 1 or 10 nmol/l KPT-330. Mean +/− standard deviation values for three biological replicates are shown in A, B, and C.

Figure 2.

KPT-185 induces rapid apoptotic cell death in T-ALL cells. Dot plot analysis of Annexin V and propidium iodide (PI) staining of the MOLT-4 (A) and Jurkat (B) cells upon treatment with either DMSO or different concentrations of KPT-185 for 6 or 13 h. Mean +/− standard deviation percentages of Annexin/PI positive cells are graphed in A and B. FITC, fluorescein isothiocyanate.

KPT-330 has exhibited similar effects on the viability of T-ALL cells. Treatment with clinical compound KPT-330 produced results very similar to KPT-185 and also reduced cell growth in MOLT-4, Jurkat, HBP-ALL, KOPTK-1, SKW-3, and DND-41 cell lines, with IC50 values of 34-203 nmol/l after 72 h of exposure (Fig S1A). Like KPT-185, KPT-330 elicited rapid apoptotic response in T-ALL cells (Fig S2). These data clearly indicate that the KPT-SINE compounds promote rapid apoptosis at low nanomolar concentrations in T-ALL cells treated in vitro. Furthermore, overexpression of the anti-apoptotic protein BCL2 in Jurkat cells markedly decreased the sensitivity of cells to KPT-185 and KPT-330, pointing to the involvement of intrinsic (mitochondrial) signalling pathway in the SINE-induced apoptosis (Fig 1B and Fig S1B). BCL2 levels remained unchanged upon treatment of Jurkat cells with KPT-330 for 12 h (Fig S1C).

We have further explored the involvement of intrinsic apoptosis in triggering cell death in response to KPT-330 by measuring the population of cells that undergo mitochondrial outer membrane permeabilization (MOMP) upon treatment with KPT-330. For this experiment, Jurkat cells were treated with KPT-330 and the mitochondrial sensitivity was assessed using Bim peptide (Del Gaizo Moore & Letai, 2012). Treatment by KPT-330 of Jurkat cells increased the priming of cells that ultimately were killed by the drug, consistent with KPT-330 killing via the mitochondrial pathway of apoptosis (Fig 1C). Importantly, overexpression of BCL2 protects against apoptosis, as demonstrated by the dramatic decrease in priming in BCL2-overexpressing Jurkat cells in response to treatment with KPT-330 (Fig 1C).

SINE promote cell cycle arrest in G1 phase

We next performed cell cycle analysis to examine whether KPT SINE compounds alter cell cycle progression of T-ALL cells. Cell cycle distribution of Jurkat and MOLT-4 was determined by PI staining after treatment of cells with either DMSO control or increasing concentrations of KPT-185 for 24 h (Fig 3). Our data clearly demonstrated cell cycle arrest at G1 phase in the MOLT-4 cell line in response to treatment with KPT-185 as shown by an increase in the G1 fraction from 54% in the control DMSO-treated cells to 81% in KPT-treated cells (Figs 3A, B). Interestingly, the treatment with KPT-185 induced a much less prominent increase in cells in the G1 phase of the cell cycle in Jurkat cells (Figs 3C, D). Similar effects on cell cycle progression were obtained upon treatment of MOLT-4 and Jurkat cells with KPT-330 compound (Fig S3).

Figure 3.

Treatment with KPT-185 leads to cell cycle arrest in G1 phase. (A and B) PI (DNA content) histograms of MOLT-4 cells treated with DMSO or KPT-185 (30 and 60 nmol/l) for 24 h. (B) Mean +/− standard deviation percentages of MOLT-4 cells in each cell cycle stage are shown. (C) PI histograms of Jurkat cells upon incubation with either DMSO or KPT-185 (15 and 30 nmol/l) for 24 h. (D) Mean +/− standard deviation percentages of Jurkat cells in each cell cycle stage are shown. The DNA histograms were analysed using FCS Express software (De Novo Software).

SINE exhibit remarkable growth suppression of T-ALL cells in vivo

To assess in vivo efficacy of KPT-SINE against human T-ALL cells, we engrafted MOLT-4 cells expressing luciferase into NOD-SCID-IL2Rgnull (NSG) mice, enabling quantification of leukaemia cells by serial BLI. For this experiment, 3 × 106 MOLT-4 cells were engrafted into mice and monitored for leukaemia burden using BLI of the mice following D-luciferin injection. Once leukaemia development and progression were established (Fig S4), mice were divided into control and treatment groups and administered orally either vehicle control, KPT-251 (50 mg/kg on days 1, 4, 6; 75 mg/kg on days 8, 11, 13, 15, 25, and 27 or until mice became moribund), or KPT-330 (20 mg/kg for days 1, 4, 6; and 25 mg/kg on days 8, 11, 13, 15, 25, 27, 29, 32, 34, and 36 or until mice became moribund). After 15 d from the start of treatment, the vehicle-treated mice demonstrated logarithmic expansion of the leukaemia burden, became moribund and were sacrificed (Figs 4 and 5A, B). Remarkably, the KPT-treated mice showed striking suppression of the leukaemia cell growth, as indicated by low and relatively unchanged BLI values (Figs 4 and 5A, B). These mice were taken off treatment for 1 week to allow them to regain body weight, after which the treatment with SINE was reinstated (Figs 5A, C). The Kaplan–Meier survival analysis showed a significant survival benefit for the mice treated with either KPT-251 or KPT-330 as compared to vehicle-treated animals (Fig 5B). These findings establish the efficacy of novel SINE CRM1 antagonists against T-ALL cells in vivo.

Figure 4.

SINE compounds dramatically suppress the growth of MOLT-4 cells engrafted into NSG mice. Bioluminescent images of two representative mice treated with either vehicle control, or SINE compounds, KPT-251 and KPT-330, for the indicated number of days. Scale bar shows the bioluminescence intensities. Mice were not dosed on days 18, 20, and 22 to allow for weight recovery.

Figure 5.

SINE compounds inhibit growth of MOLT-4 cells in vivo and provide a significant survival benefit. (A) Mean bioluminescence of mice treated with vehicle, KPT-251, and KPT-330 during the 36 d of treatment or until mice became moribund and were sacrificed (n = 8) All treatments significantly delayed leukaemia growth by two-way anova analysis (P < 0·0001). (B) Kaplan–Meier survival analysis of mice treated with vehicle control, KPT-251 and KPT-330 (n = 8; P < 0·0001). (C) Weights of mice treated with vehicle, KPT-251, or KPT-330 from the start of treatment until day 21 are displayed.

KPT-330 demonstrates high activity against AML cells in vivo

We next examined the anti-leukaemic activity of the clinical compound KPT-330 against AML cells in vivo, following up on our previous studies of the tool compound KPT-251 in this model (Etchin et al, 2012). For these studies, we intravenously injected AML MV4-11 cells that express luciferase into NSG mice and monitored for the development of leukaemia. Following leukaemia onset and progression, mice were orally administered KPT-330 at 20 mg/kg (days 1–7) and 25 mg/kg (days 8–35) three times a week for 5 weeks. As shown in Fig 6, treatment with KPT-330 dramatically suppressed the growth of MV4-11 cells with minimal net growth of leukaemia cells during the treatment period (Fig 6A). Survival analysis of KPT-330 demonstrated significant survival benefit in animals treated with KPT-330 as compared to mice treated with vehicle (Fig 6B; vehicle data reference from our previous study (Etchin et al, 2012)).

Figure 6.

KPT-330 suppresses growth of MV4–11 cells in vivo. (A) Mean bioluminescence of mice treated with KPT-330 for 35 d (n = 9). Representations of the mean and standard error bioluminescence of vehicle-treated mice that we reported previously are shown for comparison (Etchin et al, 2012) (shaded area). (B) Kaplan–Meier survival analysis of mice treated with either vehicle control or KPT-330 (n = 8; P < 0·0001). Vehicle-treated mice were moribund and were all sacrificed at the end of treatment on day 35.

SINE spare normal haematopoietic cells

To determine the selectivity of SINE against T-ALL cells, we assessed the effect of KPT-SINE on normal mouse haematopoietic cells. We obtained peripheral counts of white blood cells (WBC), neutrophils, and platelets, and measurement of haematocrit for animals treated with KPT-251 or KPT-330 after 26 d from the start of therapy. Strikingly, the treatment of mice with these compounds resulted in only minimal toxicity to circulating blood cells (Fig 7). The lack of toxicity was also observed in circulating blood counts of mice engrafted with AML MV4-11 cells after treatment with KPT-330 for 31 d (Fig 7). Furthermore, the bone marrow biopsy of mice engrafted with MV4-11 and treated with KPT-330 for 35 d demonstrated normal haematopoietic cell morphology and cellularity within the bone marrow (Fig 8), as we have previously shown for the tool compound KPT-251 (Etchin et al, 2012).

Figure 7.

SINE compounds spare normal haematopoietic cells. White blood cell (WBC; A), neutrophil (B), and platelet (C) counts, and haematocrit reading (D) after treatment of mice engrafted with MOLT-4 with either KPT-251 or KPT-330 for 26 d and mice engrafted with MV4–11 for 31 d. Mean and standard error ranges for vehicle-treated mice from our previous work (Etchin et al, 2012) are shown for reference.

Figure 8.

Bone marrow of mice treated with KPT-330 show normal haematopoietic cell morphology. 4-μm sections of bone marrow isolated from mice engrafted with MV4–11 and treated with KPT-330 for 35 d. Haematoxylin and eosin staining of bone marrow at ×400 (A) and ×1000 (B) magnification are shown.

Discussion

CRM1 coordinates the nuclear-cytoplasmic export of c. 220 proteins (Xu et al, Xu et al, 2012a) and several RNAs, including mediators of proliferative and prosurvival signalling pathways, and has been shown to be required for the survival of cancer cells (Turner et al, 2012). The present study established the antileukaemic effects of SINE inhibitors in T-ALL cells, both in vitro and in pre-clinical orthograft models of the disease. Our data show that low nanomolar concentrations of SINE induce rapid apoptosis of T-ALL cell lines. These compounds dramatically reduce viability in 13 T-ALL cell lines in vitro and suppress growth of human MOLT-4 T-ALL cells engrafted into immunocompromised mice. We also show that orally administered KPT-330, the clinical compound of this class, is very active in vivo against AML cells as well as T-ALL cells, with little toxicity to normal haematopoietic cells.

T-ALL lines tested in the KPT-SINE sensitivity assays represent different T-ALL subsets. These include T-ALL lines characterised by aberrant expression of TAL1 (DU528, Jurkat, MOLT-4, HSB-2, PF-382, CCRF-CEM, and MOLT-16), TLX3 (HPB-ALL), and MYC-TRA@ (SKW3), which define different molecular pathways leading to T-ALL (Ferrando & Look, 2003; Armstrong & Look, 2005). These lines also harbour a variety of genetic alterations in NOTCH1, NRAS, and FLT3, and the tumour suppressor genes, PTEN and TP53. Interestingly, T-ALL lines that were most sensitive to treatment with KPT-SINE, such as MOLT-4, Jurkat, HPB-ALL, DU.528, and SKW-3, do not cluster into the same oncogene group or display a similar pattern of genetic abnormalities (Table 1). For example, MOLT-4, HPB-ALL, and SKW-3 lines belong to different oncogenic subgroups, due to their expression of TAL1, TLX3, and MYC-TRA@, respectively, and yet the three lines show similar IC50 values in response to treatment with KPT-SINE. This observation suggests that common vulnerabilities exist in different oncogenic subclasses of T-ALL as well as AML, presumably based on the shared dependence of the transformed cells on the nuclear-cytoplasmic balance of CRM1 cargos that are broadly required for leukaemia cell survival.

Our findings demonstrated remarkable selectivity of the KPT-SINE compounds in inducing apoptosis in T-ALL as well as AML cells without causing toxicity to normal haematopoietic cells. This is probably due to the advanced design of SINE based on structure-based modelling and the in silico screening of a virtual inhibitor library, which greatly reduced the off- and on-target toxicity that has characterized leptomycin B and other CRM1 inhibitors. In this molecular modelling strategy, the structure of the NES groove of CRM1 was used as a framework for a docking-and-binding mode analysis of a small virtual library of compounds. The crystal structure of the CRM1-Ran-RanBP1 complex bound to KPT-251 shows that KPT-251 penetrates deep into the NES groove to outcompete CRM1 protein cargo and block nuclear export (Etchin et al, 2012).

The three KPT-SINE compounds (KPT-185, KPT-251, and KPT-330) presented in our study inhibit CRM1 nuclear export activity by covalent modification of the essential Cys528 residue of the NES groove of CRM1. These three compounds are structurally similar, but differ in the pharmacokinetic parameters. KPT-251 and KPT-330 display high Tmax, area under the curve, and bioavailability as compared to in vitro tool compound, KPT-185. In the toxicology studies, the primary effects of orally administered KPT-SINE compounds were dose-dependent reductions in food intake with consequent reductions in body weights, with minimal clinical gastrointestinal (GI) symptoms (no or mild non-bloody diarrhoea, no or minimal vomiting, etc.), and associated with relatively modest or no GI atrophy. These side effects can be minimised by reducing the dosing frequency of SINE to 2–3 times/week with at least 48 h between dosing and by providing the animals with food supplements. This schedule maintained in vivo activity of the KPT-SINE compounds (see Figs 5 and 6).

The mechanism underlying the ability of CRM1 inhibition to selectively eliminate cancer cells and not normal cells remains to be elucidated. Nearly all of the major tumour suppressor proteins (TSPs) such as TP53, TP73, FOXO1, −3 and −4, p21, p27, RB, BRCA1 and −2, APC, IkB, NPM, PAR4, and others are exported from the nucleus exclusively by CRM1, despite the fact that there are six other known nuclear export proteins. Therefore, CRM1 inhibition leads to the forced nuclear retention, upregulation, and activation of multiple TSPs. Restoration/reactivation of TSPs is known to selectively kill tumour cells (Martins et al, 2006; Ventura et al, 2007). Because multiple TSPs are activated at once, tumouricidal activity is largely independent of the underlying oncogenic growth drivers responsible for maintaining the neoplastic cell. One hypothesis is that TSPs initiate a ‘genome fidelity survey’, a survey which cancer cells will fail, leading to their selective elimination. CRM1 inhibition by SINE compounds in normal cells leads to transient cell cycle arrest without cytotoxicity, followed by fast recovery once the drug is removed. This was shown recently in vitro for normal B-cells, normal T and NK cells (Lapalombella et al, 2012). Our group has shown no toxicity of normal haematopoietic cells in mice treated with KPT-330. These results were confirmed in Good Laboratory Practise toxicology studies that formed the basis for the ongoing Phase 1 clinical trials with KPT-330 in haematological and solid tumour malignancies.

Given the rapidity of apoptosis induction observed in neoplastic leucocytes, it is possible that cancer cells are reliant on the proactive maintenance of growth-promoting and/or anti-apoptotic signalling pathways that are mediated by CRM1 nuclear export. Presumably normal cells do not share the same level of pathway dependence, leading to a greater level of resistance to inhibition of CRM1 by KPT-330. Therefore, forced nuclear sequestration of CRM1 cargos following treatment with SINEs results in the shift of balance of the pro- and anti- survival signals leading to cancer cell death, but is tolerated by normal cells, implying a differential dependence on one or more CRM1-mediated pathways. Future investigations will explore the unique mechanisms underlying the hyperactive apoptotic signalling in response to treatment with SINE compounds.

Our study findings demonstrate the clinical relevance of targeting T-ALL cells with SINE compounds, as has already been shown for AML (Etchin et al, 2012; Ranganathan et al, 2012). Importantly, we show that SINE drugs induce the rapid suppression of T-ALL cells as well as AML cells in vivo in orthograft models established in NSG immunosuppressed mice. These and other preclinical studies have provided the basis for the recent initiation of Phase I clinical trials testing the oral SINE KPT-330 in both solid tumours and haematological malignancies. These trials will reveal the maximum tolerated dose, which, based on our study, is more likely to involve anorexia, which can be mitigated by food supplements and that is highly reversible upon cessation of dosing, than it is to be based on toxicities affecting normal haematopoietic cells. Because many drugs currently used to treat leukaemia have profound haematological toxicity, if KPT-330 shows similar selective antineoplastic activity in early trials against human haematological malignancies, then it may work well in combination with existing chemotherapy regimens.

Acknowledgements

The research was supported by William Lawrence and Hughes Blanche Foundation (J.E. and T.S.), Karyopharm Therapeutics Incorporated, Alex's Lemonade Stand (J.E.), the Leukemia and Lymphoma Society Translational grant (A.T.L.), National Cancer Institute (1K99CA157951; T.S.), the Children's Leukemia Research Association (T.S.) and the Japan Society for the Promotion of Science (T.S.), the Kay Kendall Leukaemia Fund, UK (M.R.M.), and NIH-K08CA160660 (A.K.).

Author Contributions

J.E. designed experiments, analysed data, and wrote the paper. T.S., M.R.M, A.K., and R.S. helped design experiments, analyse data, and edit the manuscript. J.M. and A.L. designed mitochondrial sensitivity assays and analysed data. B.T.L. designed and performed experiments and analysed data. A.L.C. helped design and perform xenograft mouse experiments. S.J.R. carried out histological analysis. S.S., M.K., and D.M. designed KPT-SINE and analysed data. A.L.K. designed mouse xenograft studies and analysed the results of in vivo mouse experiments. A.T.L. guided the research presented in the paper, analysed data, and wrote the paper.

Conflict of interest

S.S, M.K., and D.M. are employees of Karyopharm Therapeutics Incorporated and receive compensation and hold equity in the Company.

Ancillary