Pim-1 acts as an oncogene in human salivary gland adenoid cystic carcinoma

Fifty-four patients with histopathologically proven salivary gland adenoid cystic carcinoma (ACC) in Zhejiang Cancer Hospital between July 2006 and July 2010 were included for this study. The study was approved by the Ethics Committee of Zhejiang Cancer Hospital, and patients have signed informed consent.

SACC-83 and SACC-LM were kind gifts from Prof. Li Shengling (Peking University School of Stomatology). SACC-83 is derived from human ACC tissue and SACC-LM is a lung metastasis cell line of SACC-83 [14],[15]. It is considered that SACC-LM is more malignant then SACC-83. Both cell lines were cultured in RPMI-1640 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (HyClone Laboratories, Logan, USA) in a humidified atmosphere of 5% CO2 at 37°C. Pim-1 siRNA and control siRNA were obtained from Santa Cruz Company (CA, USA). SACC-83 and SACC-LM cells were seeded in a 96-well culture plate with a density of 5 × 10³ cells/well or in a 6-well culture plate with a density of 1 × 10⁶ cells/well. After 24 h, siRNAs (0.1 μmol/L) were transfected with Oligofectamine TM Reagent (Invitrogen, Carlsbad, CA, USA) into the cells according to manufacturer's protocol.

SGI-1776 (Selleckchem, Houston, TX, USA) was dissolved in DMSO to obtain a stock solution at 10 mmol/L and stored at −80°C. After SACC-83 and SACC-LM were cultured, SGI-1776 was added to the culture system to achieve the final doses of 0–10 μmol/L. The highest concentration of DMSO in the culture was 0.1% and we tested that it had no effect on the cells viability.

According to manufacturer's instructions, Cell Counting Kit-8 (CCK-8) was used to determine the cell proliferation (Dojindo Biotechnology, China). Briefly, SACC-83 and SACC-LM cells were seeded in a 96-well cell plate for 24 h, and then transfected with Pim-1 siRNA/control siRNA, for 48 and 72 h and exposed to SGI-1776 for 24, 48 and 72 h respectively. Then the medium were discarded and replaced with 100 μl of fresh medium containing 10% CCK-8. After cells were incubated at 37°C for 4 h, the absorbance was detected at 450 nm on a microplate reader.

After transfected with Pim-1 siRNA and control siRNA for 72 h, SACC-83 and SACC-LM cells were washed with PBS (Phosphate Buffer Solution) and trypsined. Then the cells were seeded into a 12-well plate with a density of 500 cells/ well. After 7 days of incubation, colonies were stained by 0.5% crystal violet (Sigma-Aldrich, St. Louis, MO, USA) and counted directly.

In this study, Annexin V-FITC and propidium iodide (PI) (BD Biosciences, USA) were used to distinguish intact, dead and apoptotic cells by using the flow cytometric method. SACC-83 and SACC-LM cells were harvested and washed with cold PBS after transfected with Pim-1 siRNA and control siRNA for 72 h. Subsequently, the cells were resuspended in 100 μL binding buffer. 5 μL Annexin V-FITC and 1 μL PI were added to the cell suspension and incubated in darkness at room temperature for 15 min. Thereafter, 400 μl binding buffer was added to each sample and then the cells were analyzed by using the flow cytometer (BDLSR, Becton Dickinson, USA).

Cell cycle assay was performed using the Cycle Test Plus™ DNA Reagent Kit (340242, Becton Dickinson, USA) following the manufacturer’s instructions. SACC-83 and SACC-LM cells were harvested and washed with cold PBS after transfected with Pim-1 siRNA and control siRNA for 72 h. Then, the cells were fixed in pre-cooled 70% ethanol for 24 h at 4°C. After that, the cells were dyed with PI and detected by flow cytometry (BDLSR, Becton Dickinson, USA) to evaluate the cell cycle distribution.

After transfected with Pim-1 siRNA and control siRNA for 72 h, SACC-83 and SACC-LM cells were obtained and plated at 1.0 × 10⁵ cells/well in 0.5 mL of serum-free medium in 24-well matrigel-coated transwell units with polycarbonate filters (8 μm pore size; Costar Inc., Milpitas, CA, USA). The outer chambers were filled with 0.5 mL of RPMI 1640 medium supplemented with 10% FBS. After incubated for 24 h, the cells were fixed in methanol and stained with 2% crystal violet. The top surface of the membrane was gently removed and the invading cells were counted in five randomly selected fields under a microscope.

The mitochondrial membrane potential (MMP), which is recognized as a typical marker of early apoptosis was measured by the fluorescent probe JC-1 (MultiSciences Company, Hangzhou, China) in this study. As a cationic and lipophilic dye, JC-1 presents a fluorescence emission shift from green (525 nm) to red (590 nm) to indicate the potential-dependent accumulation in mitochondria. In healthy and normal cells with low membrane potentials, JC-1 appears as a monomer and produces a green fluorescence. At high membrane potentials, J aggregates is induced by JC-1 and the red fluorescence was emerged. In accordance with the manufacturer's protocol, after SACC-83 and SACC-LM cells were transfected by the Pim-1 siRNA for 72 h, the cells were harvested, centrifuged and re-suspended in 1 ml staining buffer. After stained with 1ul JC-1 staining solutions, the cells were incubated at 37°C in the dark for 30 min. Subsequently, the cells were washed twice with warm PBS buffer and re-suspended in 0.5 ml PBS buffer. Flow cytometry was performed to examine the red/green fluorescent signals.

SACC-83 and SACC-LM cells were transfected with Pim-1 siRNA or control siRNA for 72 h. Total RNA was extracted by using Trizol (Invitrogen, Carlsbad, CA, USA) and reverse-transcription was done with PrimeScipt™ RT reagent Kit (TAKARA BIO INC., Otsu, Shiga, Japan) according to manufacturer’s instructions. Quantitative real-time PCR (qRT-PCR) was performed on an ABI Prism 7500 instrument (Applied Biosystems, Foster city, CA, USA) by using the SYBR Premix Ex Taq (TAKARA BIO INC., Dalian, China). Pim-1 and GAPDH mRNA levels were measured by their specific primers:

Pim-1 (F:CTGCTCAAGGACACCGTCTACA;

R: GATGGTAGCGGATCCACTCTG);

GAPDH (F: 5'-GAAGGTGAAGGTCGGAGTC-3';

R: 5'- GAAGATGGTGATGGGATTTC-3')

After PCR amplification, the dissociation of SYBR Green-labeled cDNA was carried out to affirm that there were no nonspecific PCR products. 2^-∆∆Ct method was performed to analyze the relative quantification of Pim-1 expression.

SACC-83 and SACC-LM cells were trypsinized and washed with cold PBS after transfected with Pim-1 siRNA or control siRNA for 72 h. Then cell pellet was lysed with RIPA Lysis Buffer (Beyotime, Nanjing, China). ACC tissue samples were ground with the Tissue Lyser-II (Qiagen, Germany) and lysed with RIPA Lysis Buffer. Protein concentrations were determined by the BCA Protein Assay Kit (Beyotime, Nanjing, China). All samples were stored at −70°C prior to electrophoresis. Each sample containing 50 μg of protein were subjected to 12% SDS-PAGE and separated by electrophoresis. Then the proteins were transferred electrophoretically from the gel to nitrocellulose membrane (Immobilon-P^SQ Transfer Membrane, Millipore Corporation, Bedford, U.S.A.). To prevent non-specific binding of reagents, the membranes were blocked in TBS buffer (50 mM Tris-Cl, 150 mM NaCl, pH 7.6) containing 5% non-fat dry milk at room temperature for 3 h. Then blotted with primary antibodies of Pim-1 (Novus biologicals, Cambridge, UK) (1:1000 dilution), RUNX3 (Abgent, San Diego, CA, USA) (1:100 dilution), Cyclin D1 (Proteintech, Whhan, China) (1:1000 dilution), CDK4 (Proteintech, Whhan, China) (1:1000 dilution), β-actin (Abcam, New Territories, HK) (1:200 dilution) and GAPDH (Huabio Technology, Hangzhou, China) (1:2000 dilution) at 4°C overnight and developed with peroxidase-labeled secondary antibodies (Huabio Technology, Hangzhou, China). After that, the membranes were extensive washed in TBST and exposed to 2 mL ECL chemiluminescence reagent. The images were captured and analyzed on the Bio-Rad GelDoc XR.

Section (4 μm) of paraffin-embedded tissues were cut, mounted on glass slides (MS-coated glass, Mats-unami, Oaska, Japan), and dried overnight at 37°C. After deparaffinization, antigen retrieval in 0.01 M citrate buffer, and inactivation of endogenous peroxidase activity in 3% H₂O₂/methanol, the slides were incubated with antibody for Pim-1 (Novus biologicals, Cambridge, UK) (1:200 dilution) and RUNX3 (Abcam, New Territories, HK) (1:200 dilution) at 4°C overnight. The immunoreactivity was visualized using a streptavidin–biotin peroxidase staining kit (Histofine Simple Stain Max PO Multi, Nichirei, Tokyo, Japan) and DAB solution (Simple Stain DAB, Nichirei). The results were presented as percentage of nucleus staining positive cells and the total cells. The scores of staining results were given as negative and positive. The criterion was consulted our previous study [16].

All data were analyzed using SPSS 10.0 software. Results from in vitro experiments were expressed as mean ± standard deviations and statistically analyzed by the one-way analysis variance (ANOVA). Associations between Pim-1, RUNX3 levels and clinicopathologic parameters were analyzed using the X² test or the Fisher exact test. Survival analysis was carried out by the Kaplan–Meier method and significant differences were assessed by means of the log-rank test. P values < 0.05 were considered to be statistical significance.

SACC-83 and SACC-LM cells were divided into three groups including blank, control siRNA, and Pim-1 siRNA groups. After siRNA transfection, Pim-1 transcript and protein levels were measured by real-time PCR and Western blot, respectively. Pim-1 transcript (Figure 1A) and protein levels (Figure 1B) were significantly decreased after Pim-1 siRNA transfection compared to those of blank and control siRNA, indicating that Pim-1 siRNA effectively inhibited Pim-1 expression in both cell lines.

After transfection with Pim-1 or control siRNA for 48 h and 72 h, cell viabilities were determined by using the CCK-8 assay. Results represented that viabilities of both cells showed a decreased trend along blank, control siRNA and Pim-1 siRNA. Moreover, there were significant differences between the control siRNA and Pim-1 siRNA groups in the SACC-LM cell (Figure 1C and D). Meanwhile, the colony formation for both SACC cell lines after transfected with Pim-1 siRNA was evaluated. As shown in Figure 2, colonies in SACC-83 and SACC-LM cells which tranfected with Pim-1 siRNA were significant reduction compared to those of blank and control siRNA. These suggested that Pim-1 siRNA could reduce the proliferation activity in SACC cells.

Figure 3 showed that after exposed to 0, 0.5, 1, 2.5, 5 μmol/L SGI-1776 for 24 h, 48 h and 72 h, the proliferation rates of SACC cells were decreased in a dose and time-dependent manner. In both cells, there are significant differences between the 5, 10 μmol/L groups and control groups at all the SGI-1776 treatment time points.

In this study, both early-apoptosis rate (Annexin-V+/PI-) and total apoptosis rate (Annexin-V+/PI- and Annexin-V+/PI+) were analyzed to describe the effect of Pim-1 knockdown on apoptosis. Flow cytometry analysis showed that a majority of cells in the blank group were intact live cells. After siRNA transfection, apoptosis and dead cells increased measureably (Figure 4A). The data demonstrated that both SACC-83 and SACC-LM cells showed a significant increase in early-apoptosis rate in Pim-1 siRNA group, compared to control siRNA group after 72 h transfection (Figure 4B and C). A consistent trend of total cell apoptosis rates was also observed in this study. These data demonstrate that Pim-1 siRNA can dramatically induce apoptosis in SACC cells.

After transfection with Pim-1 or control siRNA for 72 h, cell cycles were determined by flow cytometer (Figure 5A). In SACC-LM cells, the percentage of G0-G1 phase in Pim-1 siRNA transfected cells was significantly higher than that in control siRNA transfected cells (73.52% vs 64.99%, p < 0.05), whereas the percentages of G2-M phase and S phase in Pim-1 siRNA transfected cells were significantly lower than those in control siRNA transfected cells (2.86% vs 8.09%, p < 0.05; 23.62% vs 26.91%, p < 0.05) (Figure 5C). The same tendency was observed in SACC-83 cells except that there was no significant difference in the percentage of S phase between Pim-1 siRNA and control siRNA groups (Figure 5B). The results indicated that the down-regulation of Pim-1 led to cell cycle arrest and reduced the proliferation of SACC cells, which corroborates results from the CCK-8 assay.

Figure 6 showed that in both SACC-83 and SACC-LM cells, the invasion potentials in blank and control siRNA were not significantly different, while the invasion potential of the Pim-1 siRNA transfected cells were significantly decreased compared to the control group. In contrast, the invasion inhibitory effect by Pim-1 siRNA was more obvious in SACC-83 cells than in SACC-LM cells.

After transfection by Pim-1 siRNA in both SACC cells, the JC-1 red fluorescence was reduced while JC-1 green fluorescence was increased relative to control siRNA (Figure 7A). Therefore, the data presented a significant decrease in the red/green fluorescence intensity ratio, indicating a decreased ΔΨm (Figure 7B). This result suggests that the reduction of Pim-1 expression could drive mitochondrial depolarization.

After transfection with Pim-1 siRNA for 72 h, RUNX3, Cyclin D1 and CDK4 protein expression of both SACC cells were measured by Western blot. It was showed that there was a slight increase in RUNX3 level in SACC-83 cell (Figure 8A). Figure 8B illustrated that Pim-1 knockdown may downregulate the Cyclin D1 and CDK4 protein expression and contribute to the cellular effects we identify in this study.

Immunohistochemical (IHC) staining of Pim-1 and RUNX3 in ACC tissues is shown in Figure 9. Positive ratios of Pim-1 and RUNX3 were 83.33% (45/54) and 20.37% (11/54), respectively. Table 1 showed there is a significant inverse correlation between the expression of Pim-1 and RUNX3.

To identify the role of Pim-1 and RUNX3 expression in ACC, we analyzed the correlation between Pim-1 and RUNX3 expression in ACC tissues and clinicopathologic indexes. The IHC staining results are displayed in Figure 7. Table 2 illustrates that both Pim-1 and RUNX3 levels were closely associated with T-status and nerve invasion (p < 0.05), but not with gender, age, tumor location, histological grade type, lymph node involvement or distant metastasis (p > 0.05). It can be deduced that patients with more malignant tumors tend to have higher Pim-1 and lower RUNX3 level.

Kaplan–Meier survival curves (Figure 10) showed that Pim-1 level had a weak association with overall survival of ACC patients (p = 0.091). Patients with lower Pim-1 levels had a better outcome than that with higher Pim-1 levels.