A combination of paclitaxel and siRNA-mediated silencing of Stathmin inhibits growth and promotes apoptosis of nasopharyngeal carcinoma cells

Abstract

Background

Stathmin, a microtubule associated protein (MAP), is an important molecular target for cancer therapy. Paclitaxel is one of the primary antitumor drugs targeting microtubules (MTs). We hypothesized that decreasing the expression level of Stathmin might improve the effectiveness of paclitaxel in the treatment of nasopharyngeal carcinoma (NPC).

Methods

NPC cell lines, CNE1-LMP1 and HNE2, and a CNE1-LMP1 tumor xenograft mouse model were used to test both in vitro and in vivo our siRNA-based Stathmin silencing strategy. The effects of Stathmin silencing on cell proliferation, apoptosis, and viability were investigated using MTT, AO/EB staining, TUNEL, caspase protein detection, and FCM assays. Cell migration and invasion were assayed using a Transwell assay. The combined effects of Stathmin silencing and paclitaxel were investigated using MTT, FCM, Western blot and indirect immunofluorescence assays. The effect of paclitaxel on Stathmin expression in NPC cells and, in addition, A375, MGC and HeLa cells was determined by RT-PCR and Western blotting.

Results

We found that siRNA-mediated silencing of Stathmin suppresses proliferation, induces apoptosis through the mitochondrial pathway, and causes G2/M-phase cell cycle arrest in the NPC cell lines CNE1-LMP1 and HNE2. Also, the migration and invasion of the respective NPC cells were found to be inhibited. In addition, we show that a combination of Stathmin silencing and paclitaxel is more effective than either agent alone in inhibiting proliferation and inducing apoptosis, cell cycle arrest, and MT polymerization. Furthermore, we found that Stathmin expression in the tumor cells is down-regulated by paclitaxel treatment.

Conclusion

siRNA-mediated silencing of Stathmin suppresses the proliferation, invasion and metastasis, and induces the apoptosis of NPC cells. Paclitaxel reduces the expression of Stathmin, and combining Stathmin silencing with paclitaxel treatment enhances MT polymerization. This combined strategy may provide a new approach for clinical NPC treatment.

Electronic supplementary material

The online version of this article (doi:10.1007/s13402-013-0163-3) contains supplementary material, which is available to authorized users.

Introduction

Cellular spindles mainly consist of microtubules (MTs). Stathmin, which is also called Op18 (oncoprotein 18), is confirmed to be a microtubule-associated protein (MAP) whose activity directly destabilizes microtubule dynamics. Stathmin is highly expressed in nearly all human carcinomas, including nasopharyngeal carcinoma (NPC) [1–9]. Using NPC cells expressing latent membrane protein 1 (LMP1), which is encoded by the Epstein-Barr virus (EBV), we previously showed that Stathmin is a downstream molecular target of EBV LMP1 [10]. EBV LMP1 regulates tubulin polymerization through the MAPK and Cdc2 pathways by controlling the phosphorylation of Stathmin during different cell cycle phases, which in turn affects cell cycle progression [11, 12]. The expression of Stathmin has been associated with tumor differentiation, invasion, migration, prognosis and recurrence [13, 14]. Because of its potential as a therapeutic target, many small molecules have been developed against Stathmin [15–19]. In addition, ribozyme [20] and small interfering RNA (siRNA) [21–24] targeting Stathmin were found to be effective alone, or in combination with chemotherapeutic drugs, as treatment modalities. One such chemotherapeutic drug is paclitaxel, an established primary antitumor drug targeting MTs. We hypothesized that by decreasing the Stathmin expression level, we might be able to improve the effectiveness of paclitaxel in inhibiting the growth of NPC cells.

Here we found that, using CNE1-LMP1 and HNE2 NPC cells and a CNE1-LMP1 xenograft mouse model as tools, siRNA-mediated silencing of Stathmin induces apoptosis and, thereby, suppresses the proliferation, invasion, and migration of NPC cells. We also found that a combination of siRNA-mediated Stathmin silencing and paclitaxel was even more effective than either agent alone, suggesting its efficacy for a combined NPC therapy.

Materials and methods

Reagents/antibodies

Paclitaxel powder (T7402; 25 mg) was purchased from Sigma/Aldrich (St. Louis, MO). Paclitaxel solution (30 mg/5 ml) was purchased from Haikou Pharmaceutical Manufacturing Ltd. (Haikou, China). The SuperScript III First-Strand Synthesis System for RT-PCR (18080-051), the PureLink^TM HQ Mini Plasmid Purification Kit (K2100-01), and the Lipofectamine^TM 2000 reagent (11668-019) were purchased from Invitrogen Corporation (Carlsbad, CA). The CytoSelect™ 24-Well Cell Migration and Invasion Assay kit (8 μm, Colorimetric Format) was obtained from Cell BioLabs Inc. (San Diego, CA). The Growth Factor Reduced Matrigel Matrix (356230) was purchased from Becton, Dickinson and Company (Franklin Lakes, NJ). The Annexin V-PE/7-AAD Apoptosis Detection Kit (KGA1017) was purchased from Keygen (Nanjing, China). The In Situ Cell Death Detection Kit, TMR red (12156792910), was purchased from F. Hoffmann-La Roche Ltd. (Basel, Switzerland). The primary antibodies used included mouse monoclonal anti-α-tubulin (5286, Santa Cruz, CA), rabbit polyclonal anti-stathmin (569391, Calbiochem, Darmstadt, Germany), and mouse monoclonal anti-β-actin (A5441, Sigma). The secondary antibodies used included horseradish peroxidase (HRP)-conjugated rabbit anti-mouse IgG (sc-2005, Santa Cruz, CA), Cy3-conjugated rabbit anti-mouse IgG (C2181, Sigma) and donkey anti-mouse-Alexa fluor 594 (A21203, Invitrogen).

Cells and plasmids

The human CNE1-LMP1 and HNE2 cell lines were derived from a poorly differentiated NPC [25] and a CNE1 cell line stably expressing LMP1 [10], respectively. The cells were cultured in RPMI 1640 medium (GIBCO, Grand Island, NY) supplemented with 10 % v/v fetal bovine serum (FBS) and antibiotics at 37 °C in a 5 % CO₂ humidified atmosphere. A375 human melanoma cells (CRL-1619), MGC human gastric carcinoma cells (MGC-803), and HeLa human cervical cancer cells (CCL-2) were purchased from ATCC (Manassas, VA). The siRNA vector, pGCsilencer U6/Neo/GFP, was obtained from GENECHEM (Shanghai, China). The Stathmin siRNA (pGCsilencer U6/Neo/GFP-siRNA), targeting Stathmin fragment AGAGAAACTGACCCACAAA (374–393), was inserted into the BamH I/HindIII site of the pGCsilencer U6/Neo/GFP vector. The inserted fragment was verified by sequencing (GENECHEM, Shanghai, China). The sequences against Stathmin were: sense, gaAGAGAAACTGACCCACAAA and antisense, TTTGTGGGTCAGTTTCTCTtc. The siRNA mock vector comprised random DNA oligonucleotides inserted into the BamH I/HindIII site of the pGCsilencer U6/Neo/GFP vector.

Construction and transfection of plasmids

Using the pGCsilencer/U6/neo/GFP plasmid as a vector (“si-vector”), we constructed a siRNA plasmid targeting Stathmin, pGCsilencer/U6/neo/GFP-siRNA (“si-stathmin”), and a random oligonucleotide plasmid, pGCsilencer/U6/neo/GFP-NON (“si-mock”). We used the si-mock and empty si-vector plasmids as controls in all experiments. Each of the respective three plasmids was transfected into CNE1-LMP1 and HNE2 NPC cells and all exhibited similarly high transfection efficiencies. The siRNA plasmid (si-stathmin) substantially down-regulated Stathmin expression. The amplified plasmids were extracted using a PureLink^TM HQ 96 Plasmid DNA Purification kit, and then transiently transfected into CNE1-LMP1 NPC cells following the instructions included in the Lipofectamine^TM 2000 transfection kit. Cells were collected and cell cycle and apoptosis parameters were assessed using flow cytometry (FCM). Total RNA, total protein, and polymerized and solubilized tubulin were extracted for subsequent assays.

Semi-quantitative RT-PCR

The primer targeting Stathmin (GenBank No. X53305) was designed as follows: sense 5′-CTCGGACTGAGCAGGACTTTC-3′(6U); antisense 5′-ATTCTTTTGACCGAGGGCTG-3′ (172 L). The amplified product fragment length was 167 bp. β-Actin was used as a control with forward primer 5′-TTCCAGCCTTCCTTCCTGGG-3′ and reverse primer 5′-TTGCGCTCAGGAGGAGCAAT-3′. The amplified product fragment length was 250 bp. All primers were produced by Sangon Biotech Co. Ltd. (Shanghai, China). Total RNA was extracted from cells using a TRIzol® reagent kit (Invitrogen). Reverse transcription was performed following the instructions included in the ImProm-II™ Reverse Transcription System kit (Promega, Madison, WI). Following PCR, the products were resolved by electrophoresis in 2 % agarose gels and stained with ethidium bromide. The bands were photographed using a DNA gel graph analysis system (Pharmacia Biotech, Piscataway, NJ).

Western blotting

Cells (l × l0⁶) were disrupted with lysis buffer (50 mM Tris–HCl, 1 mM EDTA, 2 % SDS, 5 mM DTT, 10 mM PMSF) followed by a 30 sec ultrasonic dispersion, 10 min boiling in a water bath for protein denaturation, and 5 min of centrifugation at 13,000 × g for cell fragment clearance. The resulting proteins (100 μg) were separated by SDS-PAGE and electrotransferred to nitrocellulose membranes. After incubation with primary and secondary antibodies, the processed proteins were stained with a chemiluminescent substrate (Supersignal West Dura Extended Duration Substrate, PIERCE, Rockford, IL). Positive bands were exposed and their luminescent intensities normalized to β-actin as an internal loading control.

Cell proliferation and apoptosis assays

(1) MTT assay: cells (6 × 10³ ~ 8 × 10³/well) were seeded in 96-well plates with antibiotic-free medium. After transfection, cells were incubated for 0, 24, 48, or 72 h. MTS reagent (20 μl) was added to each well and cells were incubated at 37 °C for another 1 ~ 4 h, at which time absorbance was read at 490 nm in an ELISA plate reader. The growth curves were expressed as relative % cell rate survival = (OD test/OD control) × 100%. (2) AO/EB staining: cells were seeded on glass cover slips in 12-well plates. After transfection, acridine orange/ethidium bromide (AO/EB) solution (10 μl; PBS solution of AO and EB 100 μg/ml) was added to each well. Cell morphologies were assessed under a fluorescent microscope and 300 cells were counted to calculate the number of cells undergoing apoptosis. (3) TUNEL assay: cell transfections were as for AO/EB staining. Cells were fixed with 4 % paraformaldehyde at 25 °C for 1 h. After washing, the penetrating fluid (0.1 % Triton-X100 dissolved in 0.1 % sodium citrate solution) was loaded onto the surface of cells on ice for 2 min. After washing, the reacting liquid (50 μl: 5 μl Enzyme Solution and 45 μl Labeling Solution) was added to each well, and the Labeling Solution (45 μL) was added to one well as a negative control. The slide was then incubated at 37 °C for 1 h, stained with DAPI (1 μg/μl) and assessed under a fluorescent microscope. (4) Apoptosis analysis by FCM: cells were seeded into 6-well plates for incubation after transfection. Cells (1 ~ 5 × 10⁵) were collected and Binding Buffer (500 μl) was added to the cell suspensions. Annexin V-PE (1 μl) and 7-AAD (5 μl) were mixed and allowed to react with rotation agitation for 5 ~ 15 min. Finally, cells were subjected to FCM analysis within an hour.

Cell cycle analysis

Cells were processed as for apoptosis analysis up to the cell suspension process. Then, the cells were suspended in 100 μl PBS and fixed with ethanol. After rinsing, the cells were suspended in 500 μl PBS buffer solution containing 100 units/ml RNaseA, and incubated at 37 °C for 30 min. Propidium iodide was added (10 μl of 2 mg/ml concentration) and cells were incubated for another 30 min, after which the cell cycle status was assessed by FCM.

Analysis of solubilized and polymeric tubulin

(1) Extraction and detection of tubulin by Western blotting. After transfection, cells were washed and an appropriate amount of microtubule stabilization buffer (MT-SB: 0.1 M pH 6.9 Pipes, 2 M glycerol, 1.5 mM MgCl₂, 2 mM EGTA, 0.5 % Triton X-100 and protease inhibitors supplemented with 4 μM paclitaxel to maintain MT stability during cell isolation) was added. Cells were left on ice for 20 min and centrifuged at 13,000 × g for 15 min. Then the supernatant fluid that contained solubilized tubulin was collected. A lysis buffer (50 mMTris-HCl, 1 mM EDTA, 2 % SDS, 5 mM DTT, 10 mM PMSF) was added to the cell pellet and left on ice for 30 min. Cell lysates and fragments were collected and placed in a boiling water bath for denaturation. Then they were cooled on ice and processed with a maximum power pulse of ultrasound for 30 sec. Following sonification, the fractions were centrifuged at 13,000 × g at 4 °C for 15 min to remove cellular fragments. The supernatant fractions contained the polymeric tubulin. The two types of tubulin were verified by Western blotting. (2) Detection of polymeric tubulin by indirect immunofluorescence. Glass cover slips were placed in a 6-well plate, and transfected cells were seeded into the individual wells. When the cells reached confluence, fluorescent staining was performed. The fluorescent staining steps included washing with cold PBS 3 times; fixing with 4 % paraformaldehyde for 30 min; processing with MT-SB for 20 min; washing again with PBS; incubating with anti-α-tubulin (1:200 dilution) as the primary antibody overnight at 4 °C; washing with PBS 3 times; incubating with a fluorescent Cy3 labeled secondary antibody (1:200) at 37 °C for 1 h; washing with PBS; removing the glass cover slip and mounting with blocking agent (including quenching agent); analysis by fluorescence microscopy.

Migration and invasion assays

(1) Migration assay. The “Cell Migration Assay Protocol” provided by the CytoSelect™ 24-Well Cell assay kit (Cell Biolabs, San Diego, CA) was used. Cells were disaggregated briefly by trypsinization. Then the cell suspensions were collected and the supernatant fractions were discarded after centrifugation. The cell pellets were washed and then re-suspended with serum-free media containing 0.1 % BSA. The cell suspensions were transferred to transwell chambers. FBS (10 %) was added to the lower chamber of a 24-well plate and the cells were incubated for 48 h. Next, the cells were fixed with methanol, stained with hematoxylin and counted under a microscope. The number of cells passing through the membrane on each mounted slide was counted (400×) in triplicate [26]. (2) Invasion assay. The “Cell Invasion Assay Protocol” provided by the CytoSelect™ 24-Well Cell assay kit (Cell Biolabs) was used. The transwell chamber on the bottom surface of the membrane was coated with 1:8 diluted Matrigel (50 mg/L) and allowed to dry at 4°C. The residual liquid was aspirated from the culture plate and serum-free medium (50 μl) containing 10 g/l BSA was added to each well and hydrated at 37 °C for 30 min. The remaining steps were the same as for the migration assay.

Evaluation of drug combinations

Coefficients of drug interaction (CDI) were calculated as follows [27]: CDI = AB/(A + B), where A and B are the survival values of the single agents and AB is the survival value of the two-drug combination. An effect was considered to be a synergistic effect of a two-drug combination when CDI < 0.85, and a significant synergistic effect of a two-drug combination when CDI < 0.7 (p < 0.05, combination vs. antitumor drug alone).

Animal studies

(1) Animal ethics statement. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of the Laboratory Animals of the National Health and Medical Research Council of China. The experimental protocol was approved by the Third Xiangya Hospital of Central South University Animal Ethics Committee (SYXK-2010-0001), China. (2) Animal experiments. Female Balb/c nu/nu mice (age, 4–6 weeks) were obtained from the Hunan SJA Laboratory Animal Co. Ltd., China. Animals were housed in a standard pathogen-free environment with unlimited access to food and water. CNE1-LMP1 cells were transfected with respectively the si-vector, si-mock or si-stathmin plasmids as described for the in vitro studies above. Twenty-four mice were randomly divided into 3 groups (n = 8 each group) and CNE1-LMP1 cells (2 × 10⁶) transfected with the si-vector, si-mock or si-stathmin plasmids (100 μl) were inoculated subcutaneously into the right axilla. One day after inoculation, each group of mice was randomized further into two smaller groups (n = 4 each). One group was injected with 0.9 % sodium chloride solution (NS) and the other was injected with paclitaxel (Taxol) solution (Table 1). The Taxol groups each received an intraperitoneal injection of paclitaxel at a dose of 10 mg/kg and each NS group received 0.9 % sodium chloride solution at the same volume of paclitaxel injection solution. Injections were administered on day 1, 7, and 13. After the first injection, animals were monitored for tumor volume using a digital caliper every 2 days up to the 20^th day. The tumor volume (TV) was calculated using the following equation: TV = longest dimension × (shortest dimension)²/2. After completing the treatment schedule, tumor-bearing mice were euthanized and solid tumors were photographed, weighed and collected for further examination. The tumor inhibitory rate (TIR) was calculated using the following equation: TIR = (the average tumor weight of the control group - the average tumor weight of the experimental group)/the average tumor weight of the control group × 100 %. (3) Detection of polymeric tubulin by indirect immunofluorescence ex vivo. For immunofluorescence, formalin fixed, paraffin embedded 5 μm sections of the xenograft tumors were deparaffinized and hydrated. Fluorescent staining was performed as described for the in vitro studies above. The primary antibody used was anti-α-tubulin and the secondary antibody used was donkey anti-mouse-Alexa fluor 594. Staining was visualized using a confocal laser scanning microscope (CLSM).

Table 1.

A total of 24 mice were randomized into 6 groups (n = 4)

	si-vector	si-Mock	si-stathim
0.9 % sodium chloride solution (NS)	si-vector/NS	si-Mock/NS	si-stathim/NS
Paclitaxel solution (Taxol)	si-vector/Taxol	si-Mock/Taxol	si-stathim/Taxol

Statistics

All statistical calculations were performed using the statistical software program SPSS12.0. Differences with p-values less than 0.05 were considered statistically significant. Data are expressed as means ± S.D.

Results

Silencing of Stathmin inhibits proliferation and induces apoptosis in NPC cells through the mitochondrial pathway

Using the pGCsilencer/U6/neo/GFP plasmid as a vector (si-vector), we generated a siRNA plasmid targeting Stathmin, pGCsilencer/U6/neo/GFP si-RNA, (si-stathmin) and a random oligonucleotide plasmid, pGCsilencer/U6/neo/GFP-NON, (si-mock). Throughout the experiments, si-vector and si-mock were used as controls. After transfection of each of the three types of plasmids into CNE1-LMP1 and HNE2 NPC cells, RT-PCR and Western blot analyses were used to confirm that si-stathmin significantly down-regulated Stathmin expression (p < 0.05) (Supplementary Fig. S1).

The subsequent MTS assay showed that si-stathmin significantly suppresses the proliferation of CNE1-LMP1 and HNE2 cells at 24, 48 and 72 h after transfection (Fig. 1) compared to control cells transfected with si-vector or si-mock. Furthermore, the results of the TUNEL assay revealed that si-stathmin transfected CNE1-LMP1 (Fig. 2 a1) and HNE2 (Fig. 2 a2) cells exhibited substantially more apoptosis compared to the two control groups. AO/EB staining revealed that 22.1 % of the si-stathmin tranfected CNE1-LMP1 cells underwent apoptosis compared to 12.5 % or 10.2 % of si-vector or si-mock transfected cells, respectively (p < 0.05; Fig. 2b).

Fig. 1.

siRNA targeting of Stathmin inhibits proliferation of CNE1-LMP1 cells (a) and HNE2 cells (b). Each cell line was transfected with the respective plasmids si-vector, si-Mock (controls) or si-stathmin. Proliferation was assessed by MTS assay at 24, 48 and 72 h after transfection. Absorbance was read at 490 nm. The data are shown as means ± S.D. of at least 3 independent experiments performed in duplicate

Fig. 2.

siRNA targeting of Stathmin induces apoptosis of NPC cells. Cells were transfected for 24 h with each of the same respective three plasmids as in for Fig. 1. a Visualization of apoptotic cells using the TUNEL assay. CNE1-LMP1 cells (a1) and HNE2 cells (a2) were stained with DAB and DAPI after transfection. Cells not undergoing apoptosis are blue and cells undergoing apoptosis exhibit orange nuclear fragments. b CNE1-LMP1 cells were stained with acridine orange and ethidium bromide after transfection and viewed under a fluorescent microscope. In this case, nonapoptotic cells are shown in green, and cells undergoing apoptosis are bright green and orange in the nucleus. The histogram (bottom) shows the percentage of apoptotic cells in each transfected cell type (500 cells were counted). c Expression of the caspase 3, 8 and 9 proteins in CNE1-LMP1 cells was detected by Western blotting. d Dot plots show each transfected CNE1-LMP1 cell type (see Fig. 5). The ratio of red to green fluorescence is shown in the histogram (bottom). Each experiment was performed at least 3 times in duplicate. Data are shown as means ± S.D. and asterisks (*) indicate a significant effect of si-stathmin for both (b) and (d)

Caspase expression in the three transfected CNE1-LMP1 cell types was assayed by Western blotting. The results obtained showed that cleavage of caspase 3 and caspase 9 in si-stathmin transfected cells was clearly increased compared to the two control cell groups, but no change in caspase 8 was noted (Fig. 2c). Mitochondrial depolarization occurs at the initial stage of apoptosis through the mitochondrial pathway, and can be detected trough the application of JC-1dye. The results of the flow cytometry (FCM) assay indicated a significant (p < 0.05) decrease in mitochondrial membrane potential in si-stathmin transfected CNE1-LMP1 cells compared to si-vector or si-mock transfected cells, as indicated by a reduced ratio of red/green fluorescence (Fig. 2d). Overall, these results imply that the mitochondrial pathway is the primary pathway through which down-regulation of stathmin by siRNA induces apoptosis of NPC cells.

Si-stathmin suppresses migration and invasion of NPC cells

Microtubules (MTs) play an important role in cellular dynamics, and siRNA-mediated suppression of Stathmin may alter the distribution of cellular MTs. This altered distribution may, in turn, influence cellular migration and invasion. The results obtained using a transwell assay revealed that the number of si-stathmin transfected CNE1-LMP1 (Fig. 3a1) or HNE2 (Fig. 3a2) NPC cells crossing the transwell membranes was significantly (p < 0.05) decreased compared to either control group, indicating that down-regulation of Stathmin inhibits the migratory capacity of these cells in a 3D matrix. Similarly, this assay allows only cells with an invasive capacity to cross the basement membrane layer. Our results revealed that the number of si-stathmin transfected CNE1-LMP1 NPC cells that crossed the membrane layer was also significantly (p < 0.05) decreased compared to the number of control cells crossing the membrane (Fig. 3b). This latter result indicates that down-regulation of Stathmin expression also inhibits the invasive capacity of NPC cells.

Fig. 3.

si-Stathmin reduces motility and invasion of NPC cells. a Representative photographs of CNE1-LMP1 (a1) and HNE2 (a2) cells migrating on a membrane on the bottom of a transwell plate. Cells are stained as pink and small holes in the membranes are displayed in black. The histogram (bottom) represents quantitative data of transwell migration of each transfected cell type. b Representative photographs of CNE1-LMP1 cells invading on a membrane coated with “Matrigel” in the bottom of the well. The histogram (bottom) represents quantitative data of transwell invasion of each transfected cell type. For (a) and (b), the data are shown as means ± S.D. of at least 3 independent experiments performed in duplicate and the asterisks (*) indicate a significant inhibition by si-stathmin

Synergy between si-stathmin and paclitaxel in NPC cells

Paclitaxel is currently used as a chemotherapeutic agent to treat NPC. Our MTT results indicated that paclitaxel effectively suppresses the proliferation of si-vector, si-mock and si-stathmin transfected CNE1-LMP1 or HNE2 NPC cells, all in a dose-dependent manner. This inhibition was more dramatic in si-stathmin transfected cells (Fig. 4a). The assessment of drug combinations showed that the coefficients of drug interaction (CDI) values of si-stathmin and paclitaxel with concentrations ranging from 0.1 to 100 nM were 0.28, 0.30, 0.21 and 0.24, respectively, in CNE1-LMP1 cells, and 0.41, 0.50, 0.56 and 0.61, respectively, in HNE2 cells. All values were < 0.7, indicating a synergistic interaction between si-stathmin and paclitaxel (Fig. 4b).

Fig. 4.

si-Stathmin and paclitaxel effectively inhibit proliferation of NPC cells. a A combination of si-stathmin and paclitaxel suppresses proliferation of CNE1-LMP1 (a1) and HNE2 (a2) cells. b Evaluation of si-stathmin and paclitaxel by CDI calculation in CNE1-LMP1 (b1) and HNE2 (b2) cells. NPC cells were transfected as indicated and cultured in media with or without paclitaxel for 48 h. Proliferation was measured using the MTT assay and data are shown as means ± S.D. of at least 3 independent experiments performed in duplicate. Asterisks (*) indicate a significant decrease in proliferation induced by si-stathmin and paclitaxel

Next, NPC cells transfected with each of the respective plasmids were treated with paclitaxel and their apoptotic status was analyzed by FCM. Compared to either si-stathmin or paclitaxel treatment alone, a combination of both agents more markedly increased apoptosis (Fig. 5a). The apoptotic rate of CNE1-LMP1 NPC cells that were treated with both siRNA (si-stathmin) and paclitaxel was 51.86 %, which was significantly higher than that of cells treated with si-stathmin alone (30.91 %) or cells treated with paclitaxel alone (si-vector, 34.17 % or si-mock, 36.09 %; p < 0.05; Fig. 5a bottom). These data suggest that a combination of si-stathmin and paclitaxel promotes apoptosis of CNE1-LMP1 cells.

Fig. 5.

A combination of si-stathmin and paclitaxel induces apoptosis and cell cycle arrest in NPC cells. a Cells were collected for assessment of apoptosis by flow cytometry (FCM). R2 in the dot plot represents cells in the late stages of apoptosis and R5 represents early apoptotic cells. R3 represents dead cells and R4 represents live cells. b CNE1-LMP1 (b1) and HNE2 (b2) cells were collected and stained with propidium iodide for cell cycle determination by FCM. Each experiment was performed at least 3 times in duplicate

The effect of combining si-stathmin and paclitaxel on cell cycle progression was analyzed using FCM. The results revealed that both si-stathmin or paclitaxel induced a delay in progression through the G2/M phase of the cell cycle of CNE1-LMP1 and HNE2 cells, and that its combination enhanced this effect (Fig. 5b). The percentage of CNE1-LMP1 cells treated with both si-stathmin and paclitaxel that were arrested at the G2/M phase was 74.14 %, which was significantly higher than that of cells treated with either si-stathmin (29.43 %) or paclitaxel alone (si-vector, 53.50 % or si-mock, 55.64 %; p < 0.05; Fig. 5b1). Similarly, the percentage of HNE2 cells treated with both si-stathmin and paclitaxel arrested at the G2/M phase was 73.32 %, which was significantly higher than that of cells treated with either si-stathmin (8.53 %) or paclitaxel alone (si-vector, 47.50 % or si-mock, 51.46 %; p < 0.05; Fig. 5b2). These data indicate that a combination of si-stathmin and paclitaxel treatment substantially enhances a G2/M arrest of CNE1-LMP1 and HNE2 cells compared to treatment with either agent alone.

The effect of Stathmin on tumor cell cycle progression and apoptosis has been reported to be closely associated with changes in MTs, which are also known to be a target of paclitaxel. Thus, we hypothesized that a combination of si-stathmin and paclitaxel could be more effective in inducing changes in MTs than either agent alone. Indeed, CEN1-LMP1 cells treated with si-stathmin showed stronger fluorescence signals (see Section 2) and longer MTs than either control group with or without paclitaxel treatment, indicating that si-stathmin increased the amount and length of polymeric tubulin within the cells. Paclitaxel had a stronger effect on increasing polymeric tubulin in cells transfected with si-stathmin than in cells only transfected with si-stathmin. Overall, NPC cells treated with si-stathmin and paclitaxel showed the strongest fluorescence signals and the longest MTs (Fig. 6a).

Fig. 6.

A combination of si-stathmin and paclitaxel increases polymerization of microtubulins in CNE1-LMP1 cells. Transfected cells were cultured in media with 0 or 100nM paclitaxel. a Microtubulin was stained with Cy3 and observed under a fluorescent microscope. b Respective polymerized and solubilized microtubulins were extracted and their expression detected by Western blotting with anti-tubulin (loading control not shown). The histogram (bottom) represents the quantitative luminescence ratio of polymerized microtubulin and solubilized microtubulin. Each experiment was performed at least 3 time in duplicate. Data are shown as means ± S.D. and asterisks (*) indicate a significant increase in polymerization of microtubulins

Similarly, the ratio of polymeric to solubilized tubulin (P/S) in cells transfected with si-stathmin was higher than that observed in the two control groups, whether or not treated with paclitaxel. Without paclitaxel, the P/S ratio of the cells treated with si-stathmin was 1.41, compared to 1.10 and 1.12 in the si-vector and si-mock groups, respectively. In cells treated with paclitaxel the P/S ratio of the si-stathmin transfected cells was 2.05, compared to 1.32 and 1.45 in the si-vector and si-mock groups, respectively. These results indicate that, after siRNA-mediated silencing of Stathmin, the solubilization of tubulin decreases and the polymerization of tubulin increases. Similarly, the P/S ratio of the two control groups treated with paclitaxel was higher than that of the untreated cells, suggesting that paclitaxel can also reduce solubilized tubulin and increase polymeric tubulin. Also here, the P/S ratio of NPC cells treated with paclitaxel combined with si-stathmin showed synergism (P/S = 2.05; Fig. 6b). Thus, down-regulation of Stathmin could be combined with paclitaxel to decrease solubilized tubulin and to increase polymeric tubulin in CNE1-LMP1 cells (p < 0.05).

Paclitaxel down-regulates Stathmin expression in tumor cells

We next treated cells with paclitaxel in order to determine whether this drug affects Stathmin expression. Our results show that the expression of both Stathmin mRNA (Fig. 7a) and Stathmin protein (Fig. 7b) decreased substantially in a dose-dependent manner in CNE1-LMP1 cells. Moreover, paclitaxel treatment of A375, MGC or HeLa cells, which each express high levels of Stathmin, significantly decreased Stathmin mRNA and protein levels (Fig. 7c, d).

Fig. 7.

Paclitaxel down-regulates stathmin expression in tumor cells. Cells were cultured in media with various concentrations of paclitaxel. Paclitaxel regulates the expression of Stathmin at the mRNA (a) and protein (b) levels in CNE1-LMP1 cells. The accompanying histograms (lower panels) show the quantitative luminescence values of each lane individually. Data are shown as means ± S.D. and asterisks (*) indicate a significant decrease in Stathmin expression. Paclitaxel also regulates expression of Stathmin at the mRNA (c) and protein (d) levels in A375, MGC, and HeLa cells. Each experiment was performed at least 3 times in duplicate

Synergistic antitumor effects of si-stathmin and paclitaxel in vivo

To evaluate the synergistic antitumor activity of si-stathmin and paclitaxel in vivo, we used a human NPC tumor xenograft mouse model. Twenty-four nu/nu mice were randomly divided into six groups (Table 1). After inoculation and treatment of the mice, we found that tumor development was suppressed in the si-stathmin treated groups (si-stathmin/NS) compared to the control groups (si-vector/NS and si-mock/NS). Tumor development was also suppressed in the si-vector/Taxol and si-mock/Taxol groups compared to the untreated control si-vector/NS and si-mock/NS groups. Tumor development was most severely suppressed in the combination si-stathmin/Taxol groups compared to the control groups. Tumor development was significantly suppressed from day 12 to day 20 after injection (p < 0.01; Fig. 8a and b).

Fig. 8.

The Antitumor effects of si-stathmin and paclitaxel in a NPC CNE1-LMP1 mouse model. a CNE1-LMP1 cells transfected as indicated were injected into mice and tumor-bearing nu/nu mice were administered with 0.9 % sodium choride solution (NS) or paclitaxel (Taxol). Each group contains 4 mice and data are shown as mean values ± S.D. The asterisk indicates a significant decrease in tumor volume (p < 0.01). b Photgraphs of the tumors extracted from the mice. c Effects of si-stathmin and paclitaxel on tumor weight. Data are shown as means ± S.D. and the asterisk (*) indicates a significant decrease in tumor weight (p < 0.01). d A combination of si-stathmin and paclitaxel increases polymerization of microtubulins in CNE1-LMP1 cells. CNE1-LMP1 cells were injected as a tumor xenograft and mice were treated with NS or paclitaxel. Microtubulin was stained with Alexa Fluor 594 and observed under a confocal laser scanning microscope. Each experiment was performed at least 3 times in duplicate

After completion of the study, mice were euthanized and solid tumors were weighed (Fig. 8c). The tumor inhibition rate (TIR) was calculated using the equation described in Materials and Methods. The TIR of in mice injected with si-stathmin transfected CNE1-LMP1 cells was 52.10 %, the TIR of mice treated with paclitaxel was 75.19 %, and the TIR of mice injected with si-stathmin transfected cells treated with paclitaxel was 80.94 %. Although we found that si-stathmin and paclitaxel can each suppress NPC tumor growth in vivo, the effect after treatment with both si-stathmin and paclitaxel was significantly greater than that after treatment with either si-stathmin or paclitaxel alone (p < 0.01).

Similar to the in vitro results, CEN1-LMP1 cells transfected with si-stathmin showed stronger fluorescence signals and longer MTs than control cells treated or not treated with paclitaxel. This result indicates that si-stathmin increases the amount and length of polymeric tubulin within the cells. In general, cells treated with paclitaxel showed a higher increase in polymeric tubulin compared to untreated cells, whether or not transfected with si-stathmin. Overall, we found that NPC cells treated with si-stathmin and paclitaxel showed the strongest fluorescence signals and the longest MTs (Fig. 8d).

Discussion

Microtubules (MTs) represent the main components of the cellular cytoskeleton. The different microtubule-associated structures and functions are determined by the microtubule-associated proteins (MAPs). Stathmin is an important MAP. Previously, Stathmin expression was found to be significantly up-regulated in NPC samples [9]. In the present study, NPC cells containing either si-vector, si-mock, or si-stathmin, were established using the pGCsilencer/U6/neo/GFP vector construct. The si-stathmin, targeting Stathmin, was found to substantially silence the expression of Stathmin. This silencing led to a decrease in the proliferation of CNE1-LMP1 and HNE2 cells and, in addition, an induction of apoptosis through the mitochondrial pathway mediated by caspase 3 and caspase 9.

Previously, over-expression of Stathmin has been shown to enhance hyperplasia, adhesion, and migration of colorectal cancer cells, and to be closely associated with tumor differentiation, invasion, lymphatic metastasis, Dukes classification, and tumor-node-metastasis (TNM) classification [28]. After subsequent Stathmin silencing, cell morphology, dynamics, microtubule network construction and distribution were all restored [29]. In our current study, the migration and invasion properties of cells transfected with si-stathmin were significantly suppressed, supporting the idea that si-stathmin may reduce the invasion and migration of NPC cells by repressing Stathmin expression and, hence, by stabilizing microtubules. The suppression of proliferation, migration and invasiveness, and the induction of apoptosis, were typically found in NPC cells transfected with si-stathmin, which showed a reduced Stathmin protein expression.

Paclitaxel is a primary antitumor agent that exerts its effects on MTs. By enhancing polymerization of MTs and inhibiting its depolymerization, paclitaxel maintains the stability of MTs and, thereby, suppresses mitosis. In vitro experiments have shown that paclitaxel blocks the cell cycle at the G2/M phase. Stathmin destabilizes MTs by promoting MT depolymerization, which subsequently enhances tumor cell proliferation, migration, and invasion. Stathmin also stabilizes the mitotic spindle [30], blocks the cell cycle at the G2/M phase, interferes with the binding of paclitaxel to tubulin, increases resistance to paclitaxel, and lowers a patient’s sensitivity to paclitaxel. Some or all of these phenomena have been shown to occur in osteosarcoma cells [23], breast cancer cells [14, 24], ovarian cancer cells [31] and ovarian cancer patients [32]. As of yet, paclitaxel has been used primarily for the treatment of breast and ovarian cancers.

Our results show that either si-stathmin or paclitaxel can increase the level of polymeric tubulin, repress proliferation, enhance apoptosis, and block the cell cycle at the G2/M phase in CNE1-LMP1 and HNE2 cells. In all cases, paclitaxel showed stronger effects than the siRNA. However, the strongest effects on microtubule polymerization, cell proliferation, apoptosis, and cell cycle G2/M delay were observed in NPC cells treated with both paclitaxel and si-stathmin. For the first time we statistically confirm that in NPC cells the combined effects of si-stathmin and paclitaxel synergistically enhance the anticancer effects of either treatment alone. A similar phenomenon was observed in in vivo experiments, i.e., a combination of si-stathmin and paclitaxel treatment substantially suppressed NPC tumorigenesis compared to either agent alone. Therefore, a combination of si-stathmin and paclitaxel is proposed for the treatment of epithelial NPC.

Although exact the mechanism underlying the combined effect of si-stathmin and paclitaxel on NPC cells is as yet unknown, Stathmin over-expression has been reported to reduce the rate of microtubule depolymerization, to decrease microtubule dynamics, to reduce spindle damage caused by paclitaxel and, overall, to decrease the cellular sensitivity to paclitaxel. Stathmin over-expression also protects cells against paclitaxel-induced mitotic abnormalities, suggesting that high levels of Stathmin are closely related to the formation of paclitaxel-resistant tumor cells [33]. Mechanistically, Forkhead box M1 (FoxM1) is likely to be involved. Stathmin is a direct transcription target of FoxM1 and reduced FoxM1 expression is known to enhance the sensitivity to paclitaxel [34]. Moreover, paclitaxel is known to suppress the dissolution of lysosomes and to increase the transfection efficiency of lipidosome-mediated Stathmin siRNA [35, 36]. Although the ultimate target of both anti-stathmin-siRNA and paclitaxel is the microtubule, paclitaxel is known to act primarily through the death receptor pathway, triggered by Fas/FasL. Accordingly, anti-stathmin-siRNA induces apoptosis of NPC cells through the mitochondrial pathway, and this information forms the rationale for a combination therapy in NPC. Martello et al. hypothesized that the changes in tubulin/MAPs that result in increased microtubule instability might be related to an α-tubulin mutation and might be compensated for by the stabilizing properties of Taxol [37]. Balasubramani et al. suggested that Stathmin over-expression may also protect against paclitaxel-induced mitotic abnormalities, and that high levels of Stathmin are closely related to the formation of paclitaxel-resistant tumor cells [33].

More interestingly, our data show that paclitaxel significantly decreases Stathmin expression in CNE1-LMP1 cells. Similar results were observed in additional cell lines expressing high levels of Stathmin, including A375, MGC, and HeLa. Paclitaxel is known to activate ERKs, which are MAPK family members [38]. Our group previously reported that the role of Stathmin in the ERK-mediated Stathmin signaling pathway in association with cell cycle progression [12]. We subsequently presumed that paclitaxel down-regulates Stathmin in tumor cells through the ERK-mediated signaling pathway, but the exact mechanism remained to be established. Our current findings revealed a synergy between si-stathmin and paclitaxel, and may provide a mechanistic explanation for the synergy, converging at their effects on microtubules.

In summary, siRNA-mediated silencing of Stathmin expression was shown to 1) increase microtubule polymerization, 2) repress proliferation, 3) stimulate apoptosis, 4) block the cell cycle at the G2/M phase, and 5) decrease migration and invasion of NPC cells. Over-expression of Stathmin in tumor cells appears to weaken the binding of paclitaxel to MTs whereas, conversely, paclitaxel inhibits Stathmin expression. Thus, the combined application of anti-stathmin-siRNA and paclitaxel reinforces the stimulatory effect of both agents on apoptosis and their inhibitory effect on proliferation, migration, and invasion. The application of siRNA strategies targeting Stathmin may provide a new approach for the treatment of NPC.

Electronic supplementary material

Below is the link to the electronic supplementary material.