Inhibition of sphingosine kinase 2 downregulates the expression of c-Myc and Mcl-1 and induces apoptosis in multiple myeloma

Abstract

Sphingolipid metabolism is being increasingly recognized as a key pathway in regulating cancer cell survival and proliferation. However, very little is known about its role in multiple myeloma (MM). We investigated the potential of targeting sphingosine kinase 2 (SK2) for the treatment of MM. We found that SK2 was overexpressed in MM cell lines and in primary human bone marrow (BM) CD1381 myeloma cells. Inhibition of SK2 by SK2- specific short hairpin RNA or ABC294640 (a SK2 specific inhibitor) effectively inhibited myeloma cell proliferation and induced caspase 3–mediated apoptosis. ABC294640 inhibited primary human CD1381 myeloma cells with the same efficacy as with MM cell lines. ABC294640 effectively induced apoptosis of myeloma cells, even in the presence of BM stromal cells. Furthermore, we found that ABC294640 downregulated the expression of pS6 and directed c-Myc and myeloid cell leukemia 1 (Mcl-1) for proteasome degradation. In addition, ABC294640 increased Noxa gene transcription and protein expression. ABC294640, per se, did not affect the expression of B-cell lymphoma 2 (Bcl-2), but acted synergistically with ABT-737 (a Bcl-2 inhibitor) in inducing myeloma cell death. ABC294640 suppressed myeloma tumor growth in vivo in mouse myeloma xenograft models. Our data demonstrated that SK2 provides a novel therapeutic target for the treatment of MM.This trial was registered at www.clinicaltrials.gov as #NCT01410981.

Figures

Figure 1

SK2 was overexpressed in myeloma cells. (A) SK1 and SK2 gene expression in GSE6477 Affymetrix microarray dataset. Publicly available Affymetrix microarray data set GSE6477 was downloaded and the RMA normalized gene expression data were generated. The SK1 and SK2 expression levels between plasma cells from normal subjects (blue bar; n = 15) and purified CD138+ cells from newly diagnosed MM (red bar; n = 73) were compared. (B) SK1 and SK2 expression in myeloma cell lines and B-cell lines. RNA was extracted from 2 B-cell lines (EBV-immortalized B cells (IBC) and American Type Culture Collection (ATCC) CCL-156 B-lymphocytes) and 7 MM cell lines (NCI-H929, OPM1, U266, RPMI-8226, RPMI-8226-Dox40, MM.1R, and MM.1S) and RT-PCR was performed for SK1 and SK2. Relative SK1 and SK2 mRNA expression with respect to β-actin was shown (mean ± standard error of the mean of 3 separate sets of experiments). (C) SK2 expression in primary human BM CD138+ cells. Primary human CD138+ cells were isolated using CD138 enrichment kit from the BM aspirates of normal subjects (n = 5), monoclonal gammopathy of undetermined significance patients (n = 6), and myeloma patients (n = 34). SK2 gene expression was normalized against β-actin control. (Each dot represented 1 individual patient and the two solid circles represented amyloidosis patients). (D) Sphingosine level in myeloma cell lines and B-cell lines. Sphingosine was measured by high performance liquid chromatography (HPLC) in freshly prepared Epstein-Barr virus (EBV)-immortalized B cells, ATCC B-lymphocytes, and 6 MM cell lines (NCI-H929, OPM1, U266, RPMI-8226, RPMI-8226-Dox40, and MM.1S). Data represented the sphingosine concentration (pmol/1 × 106 cells) (mean ± standard error of the mean of 1 of 4 separate sets of experiments) (*P < .05; **P < .01).

Figure 2

SK2-specific shRNA inhibited cell proliferation and induced caspase 3 activation in myeloma cells. OPM1 cells were transduced with lentiviruses expressing SK2-shRNA-DsRFP or control shRNA-DsRFP for 4 hours. The cells were then washed and grew in regular culture medium for an additional 48 hours. (A) Fluorescent microscopy image showing DsRFP expression. (B) Expression of SK2 mRNA in SK2-shRNA- or control shRNA-transduced OPM1 cells. (C) Cell proliferation by MTT assay. Cell proliferation was measured using MTT assay at 0 hours and 48 hours after transduction. (D) Cell proliferation by flow cytometry. OPM1 cells transduced with SK2-shRNA or control shRNA were stained with CellTrace Violet Cell Proliferation dye and allowed to proliferate for 7 days. The dye fluorescence intensity was measured by flow cytometry. (E) Activation of caspase 3. OPM1 cells were transduced with SK2-shRNA viruses or control shRNA viruses. Forty-eight hours later, the cells were stained with Live/Dead Fixable cell dye, then fixed and permeabilized, and stained with caspase-3 antibody. Caspase 3 intensity was gated on the live cell population. Data were representative of 4 separate experiments.

Figure 3

ABC294640 inhibited cell proliferation and induced apoptosis in MM cells. (A) Dose-dependent inhibition of cell proliferation by ABC294640. Six different MM cell lines were treated with various concentrations of ABC294640 for 48 hours and cell proliferation was measured by MTT assay (mean ± standard error of the mean [SEM] of 1 of 3 separate sets of experiments). (B) Time course of proliferation inhibition by ABC294640. Six different MM cells lines were treated with 30 µM of ABC294640 or dimethylsulfoxide (DMSO) for various durations. Cell proliferation was analyzed by MTT assay at the time-points indicated (mean ± SEM of 1 of 3 separate sets of experiments). (C) Cytotoxic effects of ABC294640 on MM cells. OPM1 cells were treated with 30 µM of ABC294640 or DMSO and total live cells were quantified at the time-points indicated (mean ± SEM of 3 separate sets of experiments). (D) Increased Annexin V+ cells by ABC294640. OPM1 cells were treated with 30 µM of ABC294640 or DMSO for 16 hours and cells were stained with Annexin V and 7-amino-actinomycin D (AAD) and analyzed by flow cytometry. Data shown were representative of 3 separate sets of experiments. (E) Caspase 3 activation. OPM1 cells were treated with 30 µM of ABC294640 or DMSO for 16 hours and cells were then fixed, permeabilized, and stained with caspase3 antibody. (F) Caspase 9 activation and PAPR cleavage. OPM1 cells were treated with 30 µM of ABC294640 (indicated as [A]) or DMSO control (indicated as [D]) for 3, 6, 9, and 12 hours, and analyzed for PARP and cleaved PARP, full length caspase-9, and cleaved caspase-9 by western blot analysis. β-actin was used as a loading control (data were representative of 3 separate sets of experiments). (G) Inhibition of primary human CD138+ myeloma cells by ABC294640. Primary human CD138+ cells were freshly isolated using CD138 enrichment kit from the BM aspirates of myeloma patients and cultured in triplicate at 1 × 104 cells in 100 μL of RPMI1640 medium supplemented with 2 mM Glutamax and 10% fetal calf serum containing DMSO control or various concentrations of ABC294640 for 24 hours at 37°C under 5% CO2. Cell proliferation was then measured by MTT assay. OPM1 cells were similarly treated for comparison (n = 3; Sample ID #7244, #7452, and #7476).

Figure 4

ABC294640 enhanced Mcl-1 proteasome degradation and increased Noxa expression. (A) ABC294640 downregulated Mcl-1 protein expression. MM cells (OPM1, RPMI-8226, MM.1S, JK6L, and U266) were treated with 30 µM of ABC294640 (A) or DMSO (D) for 16 hours, and whole cell lysates were prepared and analyzed for Mcl-1 expression by western blot analysis. β-actin was used as the loading control. Data were representative of 3 separate sets of experiments. (B) A294640 increased Mcl-1 protein degradation. OPM1 cells were treated with 30 µM of ABC294640 or DMSO for 3 hours and then cycloheximide (100 µg/mL) was added. Cells were collected at each hour after cycloheximide treatment and whole cell lysate was prepared and analyzed for Mcl-1 expression by western blot analysis. β-actin was used as the loading control. The graph represents the quantification of western blots. The western blots were quantified using ImageJ. Data were representative of 2 separate sets of experiments. (C) Proteasome inhibitor (MG132 and bortezomib) prevented the degradation of Mcl-1 by ABC294640. OPM1 cells were treated with DMSO control buffer, proteasome inhibitor MG132 (1 µM), or bortezomib (50 nM) for 1 hour, followed by treatment with DMSO or 30 µM of ABC294640 for an additional 6 hours. Whole cell lysate was prepared and analyzed for Mcl-1 expression by western blot analysis. Data were representative of 2 separate sets of experiments. (D) ABC294640 increased Noxa protein expression. OPM1, RPMI8226, MM.1S, and JK6L were treated with 30 µM of ABC294640 (A) or DMSO (D) for 16 hours, and whole cell lysates were prepared and analyzed for Noxa expression by western blot analysis. (E) ABC294640 induced Noxa gene expression. Eight MM cell lines were treated with 30 µM of ABC294640 (A) or DMSO (D) for 16 hours and RNA was isolated and analyzed for Noxa gene expression by RT-PCR. Gene expression was normalized against β-actin internal control. Graphs represented the fold change of Noxa mRNA in ABC294640-treated MM cells lines compared with DMSO-treated cells. Data shown in the figure were representative of at least 2 separate sets of experiments.

Figure 5

ABC294640 enhanced c-Myc proteasome degradation. (A) ABC294640 downregulated c-Myc and pS6 protein expression. MM cells (OPM1, RPMI-8226, MM.1S, JK6L, and U266) were treated with 30 µM of ABC294640 (A) or DMSO (D) for 16 hours, and whole cell lysates were prepared and analyzed for c-Myc and pS6 expression by western blot analysis. β-actin was used as the loading control. (B) ABC294640 increased c-Myc protein degradation. OPM1 cells were treated with 30 µM of ABC294640 or DMSO for 3 hours and then cyclohexamide (100 µg/mL) was added. Cells were collected at each hour after cyclohexamide treatment, and whole cell lysate was prepared and analyzed for c-Myc expression by western blot analysis. β-actin was used as the loading control. The graph represents the quantification of western blots using ImageJ. (C) Proteasome inhibitor (MG132) prevented the degradation of c-Myc by ABC294640. OPM1 cells were treated with DMSO control buffer or MG132 (1 µM) for 1 hour, followed by treatment with DMSO or 30 µM of ABC294640 for an additional 6 hours. Whole cell lysate was prepared and analyzed for c-Myc expression by western blot analysis. Data shown in the figure were representative of at least 2 separate sets of experiments.

Figure 6

ABC294640 acted synergistically with Bcl-2 inhibitor in inhibiting myeloma cell growth and induced myeloma cell apoptosis in the presence of BM stromal cells. (A) ABC294640 did not affect Bcl-2 expression. MM cells (OPM1, RPMI-8226, MM.1S, and U266) were treated with 30 µM of ABC294640 (A) or DMSO (D) for 16 hours, and whole cell lysates were prepared and analyzed for Bcl-2 expression by western blot analysis. (B) Combination of ABC294640 and ABT-737 led to enhanced inhibition of cell proliferation. OPM1 cells were treated with various concentrations of ABT-737 in the absence or presence of ABC294640 (15 µM) for 48 hours and cell proliferation was measured by MTT assay. (C) FA-CI plots showing the synergistic effect of ABC294640 and ABT-737. Fa-CI plots for OPM1 cells revealed a synergistic inhibitory effect for ABC294640 15 μM and ABT-737 at 0.1 µM (indicated as 1), 0.3 µM (indicated as 2), and 1 µM (indicated as 3). In the Fa-CI plot, the line (combination index = 1) indicate an additive reaction between the 2 substances. Values below this line imply synergism. (D) ABC294640 induced myeloma apoptosis in the presence of BM stromal cells. GFP-expressing OPM1 cells were cultured on the monolayer of HS5 BM stromal cells and were treated for 8 hours with 30 µM of ABC294640 or DMSO. The cells were stained with Annexin V and 7-amino-actinomycin D (AAD) and Annexin V+ apoptotic cells were gated on GFP-positive OPM1 cells.

Figure 7

ABC294640 suppressed myeloma growth in vivo in mouse xenograft models. (A) ABC294640 inhibited myeloma growth in vivo in the intravenously administrated mouse xenograft model. NOD/SCID IL-2γ null (NSG) mice were sublethally irradiated (2.5 Gy) and injected via tail vein 0.5 × 106 MM.1S myeloma cells stably expressing luciferase. Two days later, the mice were divided randomly into 2 groups and treated with either ABC 294640 50 mg/kg intraperitoneally (A) or a control vehicle (C) (phosphate-buffered saline + 0.3% Tween-80) for 30 days. Every 10 days, the mice were imaged using the Perkin Elmer Ivis 200 imager and Live Image software. (B) ABC294640 inhibited myeloma growth in vivo in the subcutaneously inoculated mouse xenograft model. NSG mice were sublethally irradiated (2.5 Gy) and injected subcutaneously with 0.5 × 106 MM.1S myeloma cells stably expressing luciferase. Two days later, the mice were treated with either ABC294640 50 mg/kg intraperitoneally (mice indicated as “A”) or control vehicle (mouse “C”) for 30 days. Tumor growth was monitored by bioluminescence imaging. (C) ABC294640 inhibited myeloma growth in vivo. NSG mice were sublethally irradiated (2.5 Gy) and injected subcutaneously with 0.5 × 106 MM.1S myeloma cells stably expressing luciferase. Two weeks later, the mice were imaged using bioluminescence imaging and showed tumor engraftment (D0). The mice were then treated with ABC294640 50 mg/kg intraperitoneally or control daily for 27 days (from D0 to D27). Tumor growth was monitored by bioluminescence imaging at time-points indicated (up to 1 week after the discontinuation of injection [ie, D35]). (D) Schematic diagram of the mechanisms of action of ABC294640 in MM cells.