Preclinical and clinical efficacy of XPO1/CRM1 inhibition by the karyopherin inhibitor KPT-330 in Ph+ leukemias

Abstract

As tyrosine kinase inhibitors (TKIs) fail to induce long-term response in blast crisis chronic myelogenous leukemia (CML-BC) and Philadelphia chromosome-positive (Ph(+)) acute lymphoblastic leukemia (ALL), novel therapies targeting leukemia-dysregulated pathways are necessary. Exportin-1 (XPO1), also known as chromosome maintenance protein 1, regulates cell growth and differentiation by controlling the nucleocytoplasmic trafficking of proteins and RNAs, some of which are aberrantly modulated in BCR-ABL1(+) leukemias. Using CD34(+) progenitors from CML, B-ALL, and healthy individuals, we found that XPO1 expression was markedly increased, mostly in a TKI-sensitive manner, in CML-BC and Ph(+) B-ALL. Notably, XPO1 was also elevated in Ph(-) B-ALL. Moreover, the clinically relevant XPO1 inhibitor KPT-330 strongly triggered apoptosis and impaired the clonogenic potential of leukemic, but not normal, CD34(+) progenitors, and increased survival of BCR-ABL1(+) mice, 50% of which remained alive and, mostly, became BCR-ABL1 negative. Moreover, KPT-330 compassionate use in a patient with TKI-resistant CML undergoing disease progression significantly reduced white blood cell count, blast cells, splenomegaly, lactate dehydrogenase levels, and bone pain. Mechanistically, KPT-330 altered the subcellular localization of leukemia-regulated factors including RNA-binding heterogeneous nuclear ribonucleoprotein A1 and the oncogene SET, thereby inducing reactivation of protein phosphatase 2A tumor suppressor and inhibition of BCR-ABL1 in CML-BC cells. Because XPO1 is important for leukemic cell survival, KPT-330 may represent an alternative therapy for TKI-refractory Ph(+) leukemias.

Figures

Figure 1

XPO1 expression is enhanced in Ph+ acute leukemia (CML-BC and B-ALL) progenitors. (A) Top right panel: XPO1 and BCR-ABL1 protein levels in BCR-ABL1- or empty vector–transduced 32Dcl3 myeloid cells were determined by immunoblot. Left panel: Protein levels of XPO1 expressed as mean ± standard error of the mean (SEM) of densitometric units after normalization with Grb2 levels, were determined by immunoblot of NBM (n = 7), CML-CP (n = 3), and CML-BC (n = 7) CD34+ progenitors, and Ph+ B-ALL (n = 5) and Ph− B-ALL (n = 4) CD34+/CD19+ progenitors. Bottom right panel: Sample of immunoblots used to determine XPO1 protein levels used for quantification. (B) Left panel: XPO1 protein levels expressed as mean ± SEM in vehicle- or imatinib-treated (1 µM, 12 hours) CML-BC CD34+ cells and Ph+ B-ALL CD34+/CD19+ cells. Right panel: Representative immunoblot of XPO1 protein levels and BCR-ABL1 activity (anti-PY) in vehicle- or imatinib-treated (1 µM, 12 hours) Ph+ B-ALL CD34+/CD19+ (lanes 1 and 2) and CML-BC CD34+ (lanes 3 and 4) cells. (C) Left panel: XPO1 protein levels in 32D-BCR/ABL cells treated (24 hours) with the indicated kinase inhibitors. Right panel: XPO1 mRNA levels assessed by quantitative reverse-transcription PCR in 32Dcl3 and untreated and kinase inhibitor–treated (24 hours) 32D-BCR/ABL cells. Asterisks indicate P values vs NBM; *P < .05, **P < .01.

Figure 2

KPT-330 decreased survival and clonogenic potential in CML-BC and Ph+ B-ALL cells. (A) Top panel: Representative (n = 3) western blot showing XPO1 protein levels in vehicle- or KPT-330–treated (1 µM, 72 hours) CD34+ CML-CP and CML-BC, and CD34+/CD19+ Ph+ ALL progenitor cells. Numbers above the blots indicate relative densitometric units. Bottom panel: Graph shows percentage of apoptosis (annexin V+) in vehicle- or KPT-330–treated (0-8 µM, 72 hours) NBM and CML-BC CD34+ cells. EC50 was calculated as described in “Methods.” (B) Graph shows percentage of annexin V+ cells (mean ± SEM) in vehicle- and KPT-330 (0.5-1 µM, 72 hours)–treated NBM (n = 3), CML-CP (n = 3) and CML-BC (n = 3) CD34+ BM cells, and CD34+/CD19+ Ph+ B-ALL (n = 3) and Ph− B-ALL (n = 3) BM cells. (C) Colony-forming ability of vehicle- or KPT-330–treated (0.5-1 µM, 72 hours) NBM (n = 3), CML-CP (n = 3), and CML-BC (n = 3) CD34+ BM cells. Clonogenic potential, shown as mean ± SEM, was normalized to the respective untreated sample. (D) Annexin V+ cells (mean ± SEM) in 32D-BCR/ABL cells treated with KPT-330 (1 µM, 24 hours) and imatinib (1 µM, 24 hours) used alone or in combination. Significance was determined using the Student t test of 3 identical experiments. Asterisks indicate P values vs NBM; *P < .05, **P < .01, ***P < .001.

Figure 3

KPT-330 treatment alters the subcellular localization of tumor suppressors and negative regulators of PP2A. (A) Single-channel and merged confocal micrographs of 32D-BCR/ABL cells treated with vehicle or KPT-330 (1 µM, 12 hours) and stained with anti-SET, anti-PP2Ac, anti-hnRNP A1, anti-CIP2A, anti-IkBα, or anti-FoxO3a antibody (left panels; green or red), 4,6 diamidino-2-phenylindole (middle; blue), and merged (right). SET and hnRNP A1 (B), and p21 and p53 (C) protein levels in nuclear and cytoplasmic subcellular fractionated extracts from vehicle- and KPT-330–treated (1 µM, 12 hours) 32D-BCR/ABL cells. Histone H1 and Grb2 levels were used as a control for purity of nuclear and cytoplasmic fractions, respectively. (D) KPT-330–mediated XPO1 inhibition abrogates leukemogenesis by altering nuclear/cytoplasmic shuttling. In Ph+ acute leukemia progenitors, XPO1 expression is increased at least in part through a BCR-ABL1 kinase-dependent mechanism, and is responsible for nuclear export of the SET oncogene and CIP2A, and for the nucleocytoplasmic shuttling activity of hnRNP A1, a regulator of SET mRNA metabolism. SET and CIP2A are BCR-ABL1/Jak2- and BCR-ABL1-regulated inhibitors of the PP2A tumor suppressor, respectively. In these cells, XPO1 activity also controls the subcellular localization of important regulators of cell survival as p53, p21, IκBα, and FoxO3a. Bottom panel: On inhibition of XPO1 activity with the SINE KPT-330, the SET and CIP2A proteins are sequestered in the nucleus, which leads to activation of PP2A that, in turn, triggers inhibition/degradation of BCR-ABL1 contributing to cell death. Cytoplasmic accumulation of hnRNP A1 also contributes to the decreased SET levels. In addition, nuclear accumulation of p53, p21, IκBα, and FoxO3a also likely contribute to impair leukemogenesis of Ph+ acute leukemia progenitors.

Figure 4

KPT-330 treatment increases PP2A activity and downregulates BCR-ABL1 expression and activity. (A) Representative (n = 3) western blot showing BCR-ABL1 activity (anti-PY) and expression (anti-ABL) in vehicle- and KPT-330–treated CD34+ CML-BC cells and CD34+/CD19+ Ph+ ALL cells. (B) PP2A activity in 32Dcl3 (positive control), vehicle- and KPT-330–treated (250 nM; 48 hours) 32D-BCR/ABL cells. PP2A activity was normalized to 32Dcl3 cells. (C) Left panel: Graph shows percentage of apoptotic cells (annexin V+) in vehicle- or KPT-330–treated (1 µM, 36 hours) parental and small-t–expressing 32D-BCR/ABL cells. Significance was determined using the Student t test of 3 identical experiments. Asterisks indicate P values vs untreated; *P < .05. Right panel: BCR-ABL1 activity (anti-PY) and expression (anti-ABL) in vehicle- and KPT-330 (1 µM, 24 hours)–treated parental and small-t–expressing 32D-BCR/ABL cells. (D) Quantitative reverse-transcription PCR shows BCR-ABL1 mRNA levels in untreated, imatinib (1 µM, 24 hours)–treated, and KPT-330 (1 µM, 24 hours)–treated 32D-BCR/ABL1 cells. (E) Western blot shows effect of KPT-330 (1 µM, 16 hours) on the activity of BCR-ABL1 (anti-PY), STAT5 (anti-pSTAT5Y694), Akt (anti-pAktS473), and p42/44 MAPK (anti-pMAPKT202/Y204). Heat shock protein 90 was used as a control for equal loading.

Figure 5

KPT-330 treatment increases survival time of leukemic mice. (A) Nested RT-PCR for BCR-ABL1 mRNA in the PB measured 1, 10, and 16 weeks after injection. PB of age-matched mice (NC) and a 1:106 dilution of 32D-BCR/ABL cells with 32Dcl3 cells were used as negative and positive controls, respectively. GAPDH mRNA levels were used as a control. (B) Gross anatomy of spleens isolated from vehicle- and KPT-330–treated (15 mg/kg, 2 times weekly) mice injected with 32D-BCR/ABL cells, age-matched controls, and/or KPT-330 only–treated mice at 5 weeks (top panel) and 16 weeks (bottom panel) after cell-injection. (C) Kaplan-Meier curve shows effect of KPT-330 treatment (15 mg/kg, 2 times weekly) on survival of SCID mice injected with 32D-BCR/ABL cells (n = 10, red line). Untreated mice injected with cells (n = 8, blue line) or KPT-330–treated mice that did not receive cells (n = 7, green line) were used as controls. Survival was calculated by the Kaplan-Meier method, and the log-rank test evaluated the differences among survival distributions: P = .002 (32D-BCR/ABL–untreated vs 32D-BCR/ABL KPT-330–treated mice). (D) Wright/Giemsa staining of PB and H&E staining of sections from the BM, spleen, and liver of untreated and KPT-330–treated control and cell-injected mice.