Targeting PIM kinase activity significantly augments the efficacy of cytarabine

Abstract

Drug resistance is a major cause of treatment failure for patients with acute myeloid leukaemia (AML) and novel strategies that circumvent resistance mechanisms are urgently needed (Swords et al, 2010). The PIM kinases (PIM1, PIM2, PIM3) are a small family of proto-oncogenes within the CAMK superfamily that are frequently overexpressed in many forms of cancer including AML. PIM kinases have essential roles in the regulation of signal transduction cascades that promote cell survival, proliferation, and drug resistance (Amaravadi & Thompson, 2005; Giles, 2005; Nawijn et al, 2011). However, the specific roles of PIM kinases as regulators of AML pathogenesis and of the sensitivity to standard agents utilized in AML therapy remain to be fully elucidated. SGI-1776 is novel small molecule inhibitor of PIM kinase activity that has demonstrated preclinical activity in cancer models and has entered Phase I clinical trials (Chen et al, 2009; Mumenthaler et al, 2009). Considering the roles of the PIM kinases in the regulation of cell survival and proliferation and their high basal expression in AML cells, we hypothesized that SGI-1776 would possess significant anti-leukaemic activity in AML models. We first investigated the in vitro efficacy of SGI-1776 in a panel of nine human AML cell lines (Fig 1A). Treatment of AML cells with SGI-1776 led to a dose-dependent reduction in viability, impaired clonogenic survival (Fig 1B), and apoptotic cell death (Fig 1C, D). These effects were associated with a significant reduction in the phosphorylation of the PIM kinase substrate and apoptotic regulator Bad (Ser112), an event that increases its pro-apoptotic function. The drug-related reduction in Bad phosphorylation did not appear to be due to alterations in AKT activity as SGI-1776 treatment did not significantly affect the phosphorylation of AKT (Thr308) in MV4-11 cells, which have constitutive AKT activity (Fig 1E). Inhibition of PIM kinase signalling disrupts AML cell survival. (A) SGI-1776 causes a dose-dependent reduction in AML cell viability. Nine human AML cell lines were treated with the indicated concentrations of SGI-1776 for 72 h. Cell viability was determined by MTT assay as previously described (Carew et al, 2007). n = 3 Mean ± SD. (B) SGI-1776 diminishes clonogenic survival. Human AML cell lines were treated with SGI-1776 for 24 h. The drug was washed away and cells were plated in MethoCult methylcellulose-containing medium. Colonies were scored 14 d later with the assistance of an Alpha Innotech imaging system (Alpha Innotech, Inc., San Leandro, CA, USA) as previously described (Carew et al, 2007). n = 3 Mean ± SD. (C) SGI-1776 induces apoptosis. Cells were treated with SGI-1776 for 48 h. Apoptosis was quantified by propidium iodide/fluorescence-activated cell sorting (PI/FACS) analysis as previously described (Carew et al, 2007). n = 3 Mean ± SD. (D) SGI-1776 activates caspase-3. Cells were exposed to SGI-1776 for 48 h. The percentages of cells expressing the active form of caspase-3 were determined using the BD Biosciences Active Caspase-3 Mab Apoptosis kit (BD Biosciences, Inc., San Jose, CA, USA) followed by flow cytometry. n = 3 Mean ± SD. (E) SGI-1776 abrogates phosphorylation of the BH3-only protein BAD (Ser112). MV4-11 cells were treated with SGI-1776 for 24 h. Protein lysates were subjected to sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) analysis and the impact of drug treatment on the relative expression levels of phosphorylated BAD, total BAD, phosphorylated AKT, and total AKT were evaluated by immunoblotting. Tubulin served as a loading control. (F) FLT3 inhibition contributes to the efficacy of SGI-1776 in AML. MV4-11 cells with and without stable shRNA-mediated FLT3 knockdown (Swords et al, 2010) were treated with the indicated doses of SGI-1776 for 72 h and the consequential impact on cell viability was determined by MTT assay. n = 3 Mean ± SD, *P < 0·05. Approximately 30% of patients with AML have constitutive fms-like tyrosine kinase-3 (FLT3) activity due to internal tandem duplication (ITD) or activating mutations. Considering that in vitro kinase activity screens with SGI-1776 and other PIM kinase inhibitors have demonstrated some off-target inhibition of FLT3, we utilized MV4-11 cells with stable FLT3 knockdown to investigate whether these potential off-target effects were a critical factor underlying the anti-leukaemic activity of SGI-1776 (Swords et al, 2010). FLT3 knockdown caused a modest reduction in sensitivity to SGI-1776 (Fig 1F), indicating that FLT3 inhibition contributes to the efficacy of SGI-1776, but is not its primary mechanism of action in AML. Several recent studies have suggested a mechanistic link between aberrant expression of PIM kinases and reduced sensitivity to certain anticancer agents (Xie et al, 2006, 2010). To address this issue in a manner relevant to AML therapy, we evaluated the expression levels of PIM1, PIM2, and PIM3 in paired HL-60 cells that are sensitive and resistant to cytarabine (ara-C). Our results showed that the levels of PIM1 and PIM3, but not PIM2, were significantly higher in ara-C-resistant HL-60 cells (Fig 2A, B). Consistent with this observation, ara-C treatment led to increased PIM1 and PIM3 expression as assessed by immunoblotting (MOLM-13 cells, Fig 2C) and quantitative reverse transcription polymerase chain reaction (RT-PCR) (MOLM-13 cells and primary AML blasts, Fig 2D). We next investigated whether inhibiting PIM kinase signalling with SGI-1776 could augment the efficacy of ara-C. Treatment of AML cells with the combination of ara-C and SGI-1776 led to significantly greater diminished viability and inhibition of clonogenic survival over what was achieved by either single agent (Fig 2E, F). Propidium iodide/fluorescence-activated cell sorting (PI/FACS) analysis of the effects of SGI-1776, ara-C, and the combination of these agents on cell cycle distribution showed that SGI-1776 promoted the accumulation of cells with G1 DNA content (Fig 2G). Targeting PIM kinase activity significantly increases the efficacy of cytarabine (ara-C). (A) Ara-C resistance is linked to overexpression of PIM1 and PIM3. The relative expression levels of PIM1, 2, and 3 were evaluated in paired HL-60 cells that are sensitive and resistant to ara-C by immunoblotting. Tubulin documented equal loading. (B) Relative expression levels of PIM kinases in HL-60 ara-C sensitive and –resistant cells. Quantitative RT-PCR was used to assess PIM1, 2, and 3 expression levels, n = 3 Mean ± SD *P < 0·05. (C) Treatment with ara-C induces PIM-1 and PIM-3 expression. Cells were treated with ara-C for 24 h. Protein lysates were subjected to SDS-PAGE. Immunoblotting was utilized to quantify the levels of PIM1, 2, and 3. Tubulin documented equal loading. (D) Impact of ara-C treatment on PIM expression. Primary human AML cells were isolated from the bone marrow of AML patients after obtaining informed consent. MOLM-13 cells and primary blasts from two patients with AML were treated with ara-C for 24 h. Quantitative RT-PCR was used to assess PIM1, 2, and 3 expression levels, n = 3 Mean ± SD *P < 0·05. (E) SGI-1776 significantly increases the in vitro anticancer activity of ara-C. Cells were treated with SGI-1776, ara-C, or both for 72 h. Cell viability was determined by MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. n = 3 Mean ± SD. *Indicates a significant difference compared to controls or **single agent treatment, P < 0·05. (F) SGI-1776 and ara-C cooperate to disrupt clonogenic survival. Cells were treated with SGI-1776, ara-C, or both for 24 h. Drug-containing medium was washed away and cells were plated in Methocult. Colonies were scored 14 d later with the assistance of an Alpha Innotech imaging system. Mean ± SD, n = 3. *Indicates a significant difference compared to controls or **single agent treatment, P < 0·05. (G) SGI-1776 and ara-C cooperate to disrupt cell cycle kinetics and induce apoptosis. MOLM-13 cells were treated with SGI-1776, ara-C, or the combination for 48 h. PI/FACS was used to evaluate drug-related effects on cell cycle distribution and apoptosis (cells with sub G0/G1 DNA content). Representative histograms are shown. (H) SGI-1776 partially restores the sensitivity of ara-C resistant cells to ara-C treatment. Paired HL-60 cells that are sensitive and resistant to ara-C were treated with SGI-1776, ara-C, or the combination for 72 h. The effects of drug treatment on cell viability were quantified. n = 3 Mean ± SD. (I) Effects of drug treatment on tumour growth. MOLM-13 AML cells were implanted subcutaneously into the flanks of nude mice as previously described (Carew et al, 2008). Mice with palpable tumours were randomized into groups of 10 and treated with vehicle, SGI-1776 (100 mg/kg orally administered 5 d per week for 3 weeks), ara-C (75 mg/kg intraperitoneal injection 3 d per week for 3 weeks), or both drugs for 21 d. Tumour volume was monitored with calliper measurements, *Indicates a significant difference compared to vehicle or **single agent treatment, P < 0·05. (J) Treatment with SGI-1776 and ara-C is well tolerated. Mouse weight (g) was monitored throughout the 21-d treatment regimen. (K) SGI-1776 and ara-C diminish BAD phosphorylation (Ser112), inhibit tumour cell proliferation, and activate apoptosis. Immunohistochemistry was utilized to assess the relative levels of phospho-BAD (p-BAD), total BAD, PCNA, and active caspase-3 (C-3) in tumour sections obtained from animals in all treatment groups. Images were captured using an Olympus microscope with a DP71 camera (Olympus America, Inc., Center Valley, PA, USA) and a 20× objective. (L) Quantification of the relative intra-tumoural expression levels of phospho-BAD, PCNA, and active caspase-3. The relative intensity of phospho-BAD was quantified using image-pro plus software 6.2.1. (Media Cybernetics, Inc., Bethesda, MD, USA) Mean ± SD, n = 5. The percentage of PCNA and active caspase-3 positive cells were scored manually. Mean ± SD, n = 5. *Indicates a significant difference compared to controls or **single agent treatment, P < 0·05. HL-60 ara-C sensitive and resistant cells were utilized to investigate whether targeting PIM kinase activity with SGI-1776 could be used as a strategy to overcome intrinsic ara-C resistance. Our results showed that SGI-1776 partially restored the sensitivity of ara-C resistant cells to ara-C (Fig 2H), indicating that ara-C resistance is a multifaceted problem with multiple underlying mechanisms including PIM overexpression (Fig 2A, B). Additionally, our findings show that abrogating PIM kinase activity could possibly be utilized as a novel approach to improve the therapeutic efficacy of ara-C including in circumstances of de novo ara-C resistance. In order to further investigate the therapeutic utility of this combination, we established AML xenografts in nude mice using the MOLM-13 AML cell line. Mice were randomized into groups of 10 and were administered vehicle, ara-C, SGI-1776, or ara-C and SGI-1776 for 21 d. Treatment with the combination of these two agents was well tolerated and significantly increased the efficacy of single agent ara-C therapy (Fig 2I, J). Immunohistochemical analyzes of tumours from mice revealed that SGI-1776 significantly diminished Bad phosphorylation and cooperated with ara-C in vivo to promote activation of caspase-3 and inhibit tumour cell proliferation (PCNA expression) (Fig 2K, L). Collectively, our data demonstrate that antagonizing PIM kinase activity is an effective treatment approach that represents a new strategy to augment the therapeutic efficacy of ara-C in AML. Further investigations aimed to define the role(s) of PIM kinases in AML pathogenesis and evaluate the therapeutic potential of PIM kinase inhibition as a strategy to circumvent drug resistance are warranted. The authors would like to thank Dr Kapil Bhalla for kindly providing cytarabine-resistant HL-60 cells. This work was supported by grants from LeukemiaTexas, the AT&T Foundation and the J.C. & Irene H. Heyser Myeloma Endowment. KRK was involved in all aspects of the study including experimental design, performing research, data analysis, and manuscript preparation; CME provided intellectual input regarding experimental design and data interpretation, performed research, and was involved in the preparation of the manuscript; PT, GC, and FJG provided intellectual input regarding experimental design and data interpretation; SP obtained primary patient specimens and provided intellectual input regarding experimental design, data interpretation, and manuscript preparation; STN provided intellectual input regarding experimental design, data interpretation, and manuscript preparation; JSC directed the study and was involved in all aspects of experimental design, data analysis/interpretation, and manuscript preparation. Pietro Taverna and Gavin Choy are employees of Supergen Inc.