The first two authors contributed equally to this work.
Cancer Cell Biology
Overexpression of PLK1 is associated with poor survival by inhibiting apoptosis via enhancement of survivin level in esophageal squamous cell carcinoma
Version of Record online: 9 SEP 2008
Copyright © 2008 Wiley-Liss, Inc.
International Journal of Cancer
Volume 124, Issue 3, pages 578–588, 1 February 2009
How to Cite
Feng, Y.-B., Lin, D.-C., Shi, Z.-Z., Wang, X.-C., Shen, X.-M., Zhang, Y., Du, X.-L., Luo, M.-L., Xu, X., Han, Y.-L., Cai, Y., Zhang, Z.-Q., Zhan, Q.-M. and Wang, M.-R. (2009), Overexpression of PLK1 is associated with poor survival by inhibiting apoptosis via enhancement of survivin level in esophageal squamous cell carcinoma. Int. J. Cancer, 124: 578–588. doi: 10.1002/ijc.23990
- Issue online: 18 NOV 2008
- Version of Record online: 9 SEP 2008
- Accepted manuscript online: 9 SEP 2008 12:00AM EST
- Manuscript Accepted: 21 AUG 2008
- Manuscript Received: 30 JAN 2008
- National Science Fund. Grant Numbers: 30630067, 30721001
- State Key Basic Research Grant of China. Grant Number: 2004CB518705
- National Key Technologies R&D Program of China. Grant Number: 2006BAI02A14
- Chinese Hi-Tech R&D Program Grant. Grant Number: 2006AA02A403
- esophageal squamous cell carcinoma;
PLK1 is essential for the maintenance of genomic stability during mitosis. In our study, we found that overexpression of PLK1 was an independent prognostic factor (RR = 4.253, p = 0.020) and significantly correlated with survivin, an antiapoptotic protein, in esophageal squamous cell carcinoma (ESCC). Reverse transcription-polymerase chain reaction and fluorescence in situ hybridization (FISH) revealed upregulation of PLK1 mRNA and amplification of PLK1 gene, respectively. Depletion of PLK1 activated the intrinsic apoptotic pathway, which was substantiated by loss of mitochondrial membrane potential, reduction of Mcl-1 and Bcl-2 as well as activation of caspase-9. Coimmunoprecipitation and confocal microscopy displayed that PLK1 was associated with survivin and PLK1 depletion led to downregulation of survivin. Cotransfection of survivin constructs could partially reverse PLK1-depletion-induced apoptosis. These data suggest that PLK1 might be a useful prognostic marker and a potential therapeutic target for ESCC. Survivin is probably involved in antiapoptotic function of PLK1. © 2008 Wiley-Liss, Inc.
Esophageal squamous cell carcinoma (ESCC) is one of the most frequent malignancies worldwide. Although the use of earlier detection and improved therapeutic strategies results in a moderate reduction in mortality rates, the risk to sustain a recurrence of disease remains high. Furthermore, it remains unclear which treatment strategies should be used to best improve an individual patients' survival time, as there are not good therapeutic and prognostic markers. With accumulated information about molecular changes in the carcinogenesis and tumor progression, molecular targeted modulation of signal transduction pathways that are functionally abnormal in carcinogenesis and malignant development is becoming a novel therapeutic strategy. Among them, cell cycle kinases have attracted special attention, given the relevance of cell proliferation to oncogenic processes. Increasing data indicate that the aberrance of cell cycle kinases might lead to 2 cancer-related errors: unlimiting cell proliferation and abnormal cell division leading to chromosomal instability.
Polo-like kinase 1 (PLK1) is a key regulator of DNA damage checkpoint, centrosome maturation, chromosome condensation, chromosome segregation and cytokinesis.1 PLK1 is able to physically associate with multiple important cellular proteins such as p53, Chk2 and cyclin B1.2–4 Ectopic expression of PLK1 in NIH 3T3 cells leads to transformation in vitro and in vivo.5 PLK1 was overexpressed in several solid tumors such as breast, bladder, gastric, ovarian and colorectal cancers.6–10 In our work, we found amplification and overexpression of PLK1 in ESCC when compared with normal esophageal tissues. Our results showed that alterations of PLK1 may be an independent prognostic factor in ESCC and found that PLK1 depletion induced apoptosis via mitochondria signaling pathways by possible interaction with survivin.
Material and methods
Patients and tissue specimens
One hundred and fifty pathologically and clinically confirmed ESCC and adjacent histologically normal tissues were obtained from patients who underwent single-stage curative esophagectomy at Cancer Hospital, Chinese Academy of Medical Sciences, Beijing, China. All surgical margins are negative for tumor involvement. Radiation therapy (33 cases) or chemotherapy (6 cases) followed the curative oesophagectomy in this cohort. Ninety cases were followed up for a mean period of 46.8 months (range: 4–73 months). Thirty-three of 90 patients died during the follow-up period. Each sample was classified according to the pathologic tumor-node-metastasis classification. Tissue specimens were kept in −80°C until used for immunohistochemistry, RNA isolation or protein extraction. The samples were used with the written informed consent from patient and the approval of the Institutional Review Board.
Tissue microarray and immunohistochemistry
The tissue samples were fixed in formalin and embedded in paraffin. For TMA construction, an H&E stained section was made from each paraffin block to define 3 representative epithelium regions or 5 representative tumor regions. The representative areas were carried into a recipient paraffin block using a tissue microarray machine.
For immunohistochemistry, 4-μm-thick sections of paraffin-embedded tissue arrays were deparaffinized in xyleme, rehydrated in graded ethanol and incubated in 3% hydrogen peroxide in methanol to block endogenous peroxidase. The sections were then heated for 20 min in 0.01 mol/l citrate-buffer (pH 6.0) in a microwave oven. After washing the sections with PBS for 3 times, they were incubated with an anti-human PLK1 monoclonal antibody (1:500 dilution; Upstate Biotechnology, Lake Placid, NY) or anti-survivin antibody (dilution 1:2,000; Cell Signaling Technology, Beverly, MA) at 4°C overnight. After washing in PBS, the sections were developed according to the manufacturer's instructions (PV-9000 Polymer Detection System, ZhongShan Golden Bridge, China) and counterstained with hematoxylin, dehydrated in graded ethanol and sealed with neutral resin.
Evaluation of immunohistochemical staining
PLK1 and survivin-positive samples were defined as those showing a nuclear or cytoplasmic pattern of tumor tissue. Stained slides were examined to identify the distribution and immunoreactivity of PLK1 and scored by 2 observers using the following scale according to the percentage of PLK1-positive cells: 0–1% (0), 1–5% (1), 6–29% (2), 30–59% (3) and more than 60% (4). PLK1 staining intensity was graded into 3 groups: no staining (0), weak staining (1), moderate staining (2) and intense staining (3). These values were multiplied together to provide a single PLK1 score for each case. For statistical analysis, cases were grouped as either PLK1 negative (score 0–3), PLK1 low positive (score 4–6) or PLK1 high positive (score 7–12). Samples were scored positive when more than 5% of the cells reacted with the anti-survivin antibody. Low or high extent of staining were defined when exhibited ≤20% or >20% of positive cells, respectively.11 For each case, 3 histologically normal esophageal tissues adjacent to tumors were examined as control and the same criteria were applied for evaluation.
Western blot analysis
Cells were washed with ice-cold PBS and lysed in a RIPA buffer [50 mM Tris (pH7.5), 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS] containing PMSF (1 mM) and protease inhibitors (2 μg/ml; Protease Inhibitor Cocktail Set III, Calbiochem) on ice for 30 min. The lysates were clarified by centrifugation at 13,000g for 30 min at 4°C. The total protein concentration was estimated using Protein Assay Kit (Bio-Rad, Richmond, CA). Protein samples (30–80 μg) were loaded on a 12% SDS-PAGE and subsequently transferred to polyvinylidene difluoride membranes. After being blocked with TBST [20 mM Tris (pH 7.5), 150 mM NaCl, 0.01% Tween-20] containing 5% non-fat dry milk for 1 hr at room temperature, membranes were probed with an appropriate antibody overnight at 4°C followed by a horseradish peroxidase (HRP)-linked goat anti-mouse or anti-rabbit antibodies at room temperature for 1 hr. The membranes were analyzed using super ECL detection reagent (Applygen, Beijing, China).
The following antibodies were used: anti-PLK1 antibody (dilution 1:1,000, Upstate), anti-cleaved PARP (dilution 1:1,000, Cell Signaling Technology, Beverly, MA), anti-caspase-3 (dilution 1:500, Santa Cruz Biotechnology, San Diego, CA), anti-Bcl-2 (dilution 1:500, Santa Cruz Biotechnology, San Diego, CA), anti-Mcl-1 (dilution 1:500, Santa Cruz Biotechnology, San Diego, CA), anti-procaspase-8 (dilution 1:1,000, Calbiochem-Novabiochem Corp, SanDiego, CA), anti-procaspase-9 (dilution 1:1,000, Cell Signaling Technology, Beverly, MA), anti-survivin (dilution 1:1,000, Cell Signaling Technology, Beverly, MA), anti-GFP (dilution 1:1,000, Cell Signaling Technology, Beverly, MA), anti-Cyclin B1 (dilution 1:1,000, MBL international corporation) and anti-β-actin (dilution 1:5,000, Sigma Chemical Company, St. Louis, MO).
RNA isolation and reverse transcription-polymerase chain reaction
RNA isolation was performed using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacture's protocol. SuperScript Preamplification System (Gibco BRL, Gaithersburg, MD) was used for cDNA synthesis. Two micrograms of cDNA was used as a template for PCR reaction. The following primers were used: GAPDH: forward 5′-GTC AGT GGT GGA CCT GAC CT-3′ and reverse 5′-AGG GGT CTA CAT GGC AAC TG-3′, and PLK1: forward 5′-AAG AGA TCC CGG AGG TCC TA-3′ and reverse 5′- TCA TTC AGG AAA AGG TTG CC-3′.12 The cycling conditions were as follows: initial denaturation (5 min at 94°C), followed by the appropriate number of cycles (GAPDH, 20 and PLK1, 28) of denaturation (94°C; GAPDH, 30 sec and PLK1, 30 sec), annealing (GAPDH, 30 sec at 60°C and PLK1, 30 sec at 57°C) and elongation (30 sec at 72°C) and a final extension (10 min at 72°C). The samples were visualized by electrophoresis in 1.2% agarose gel and ethidium bromide.
Fluorescence in situ hybridization
Tumor tissues were cut, broken to pieces and digested with 0.2% type II collagenase (Sigma). Cell suspensions were incubated in hypotonic buffer (0.075 mol/l KCl) for 30 min at 37°C and fixed in methanol/acetic acid (3:1, v/v) for 30 min at 4°C. Interphase nuclei were prepared by dropping cell suspensions on slides, dried and aged overnight. The slides were digested using RNase (100 μg/ml) for 60 min at 37°C and rinsed with 2×SSC and then pepsin buffer (1 ml 0.1 N HCl with 0.5 μl 10% pepsin) for 8 min at 37°C. After rinsing (2×SSC), the slides were dehydrated by ethanol series (75, 85, 100%), air dried, denatured in 70% formamide in 2×SSC for 3 min at 75°C and then dehydrated in ice cold ethanol for 3 min each. The BAC RP11-141E3 probe was labeled using a random primer method with Cy3-dUTP (Amersham Biosciences, Piscataway, NJ). For denaturation, probes were incubated for 8 min at 72°C and then allowed to reanneal for 30 min at 37°C. A period of 48 hr was allowed for hybridization. FISH signals of at least 100 nuclei of each tumor case were evaluated using a cooled charged-coupled device camera (Princeton, Princeton, NJ) with a fluorescence optics microscope. Image processing was carried out with MetaMorph Imaging System (Universal Imaging Corp., West Chester, PA). Because there were 3 or 4 signals if normal cells were in mitosis, to make our data more evident, each case was classified as amplified for PLK1 gene locus if there were 10% cells in tumor tissues displaying more than 5 red signals.
Human ESCC cell lines KYSE 450, KYSE 410, KYSE 150, KYSE 180 and KYSE 510 were a gift from Dr. Yutaka Shimada (Kyoto University, Kyoto, Japan). The cell lines were maintained RPMI 1640 supplemented with 10% (v/v) fetal bovine serum (FBS) at 37°C under 5% CO2 in a humidified incubator. To investigate whether survivin degradation was proteasome dependent, cell cultures were treated for 24 hr with 10 μM proteasomal inhibitor MG-132 at 24 hr posttransfection of PLK1 siRNA. The cells were then harvested and subjected to Western blot.
Plasmids construction and transfections
To construct vectors for expressing PLK1-small interfering RNAs (siRNAs), the pGCsi-H1 plasmid (GeneChem, Shanghai, China) was digested with Bam HI and Hind III. Three chemically synthesized oligonucleotides encoding PLK1-short hairpin siRNAs that included a loop motif were inserted into downstream of the H1 promoter of the plasmid using DNA ligation kit (Takara) and cloned. The 3 targeting sequences of human PLK1 (GeneBank accession no. NM_005030) are AGA TCA CCC TCC TTA AAT ATT,13 ACC TCC GGA TCA AGA AGAA14 and CGG CAG CGT GCA GAT CAAC,15 corresponding to the coding regions of positions 1423–1443, 778–796 and 1581–1599, respectively. Nonsilencing siRNA was used as control. A cDNA for human survivin (GeneBank accession no. NC_000017.9) was prepared by reverse transcription-polymerase chain reaction (RT-PCR) from normal esophageal tissues with specific primers (sense, 5′-CCG CTC GAG GCA TGG GTG CCC CGA CGT TG-3′ and antisense, 5′-CCG GAA TTC CTC AAT CCA TGG CAG CCA GC-3′) designed to introduce EcoRI and XhoI restriction sequences at the 5_and 3_ ends, respectively. The resulting cDNA was cloned into pEGFP-C1 vector (Clontech). The mutant survivin expression vector, which contains Thr34 switched to Ala, was prepared by overlap PCR using the above plasmid as template with primers (sense, 5′-CTT CTT GGA GGG CTG CGC CTG CGC CCC GGA GCG GA TGG CC GAG-3′ and antisense, 5′-CTC GGC CAT CCG CTC CGG GGC GCA GGC GCA GCC CTC CAA GAA G-3′).
ESCC cells were transfected with the above plasmids (4 μg total) in 250 μl lipofectamine™ 2000 (Invitrogen Corporation, Carlsbad, CA) as described by the manufacturer.
Cell survival assays
The effects of PLK1 depletion on viability of KYSE 450 cells were assessed by cell counting kit-8 (CCK-8, Dojindo, Japan). Briefly, the cells were plated in 96-well plates. After transfection, CCK-8 (10 μl) was added to each well at different time points and incubated at 37°C for 1.5 hr. The absorbance (450 nm) was measured using a microplate spectrophotometer.
Cells growing on coverslips were washed with PBS, fixed with 4% paraformaldehyde at room temperature for 15 min and permeabilized with 0.2% Triton X-100 in PBS for 10 min. After washed with PBS, cells were subsequently incubated with a blocking solution (5% goat serum) for 30 min and incubated with primary antibody overnight at 4°C. The cells were washed 3 times and incubated with Cy3-conjugated anti-Rabbit or fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG (dilution 1:200, Jackson ImmunoResearch Laboratories, West Grove, PA) for 30 min at room temperature. DNA was counterstained using 4′,6′-diamidino-2-phenylinodole (DAPI). Rabbit anti-cleaved caspase-3 (Cell Signaling Technology) was used at a 1:500 dilution. Rabbit anticleaved caspase-9 (Cell Signaling Technology) was used at a 1:500 dilution for immunofluorescence and 1:100 for immunocytochemistry.
Cells double-stained with primary antibodies PLK1 (dilution 1:200, Upstate), survivin (dilution 1:100, Cell Signaling Technology), Bcl-2 or Mcl-1 (dilution 1:200, Santa Cruz Biotechnology) were observed under Fluoview laser scanning confocal microscope (Olympus, Tokyo, Japan).
Apoptosis and cell cycle analysis by flow cytometry
Cells in a 6-well culture slide were treated with plasmids as described earlier. After transfection, cells were gently harvested by trypsin digestion and added to the culture media and pelleted by centrifugation. Then the pellet was washed with ice-cold PBS and resuspended in 75% ethanol at 4°C overnight. The cells were collected by centrifugation and washed with PBS. Finally, the cells were resuspended in PBS containing 100 μg/ml RNase A and 50 μg/ml propidium iodide (PI). After incubation for 30 min at 37°C in dark, samples were subjected to flow cytometry (FCM) for cell cycle and apoptosis analysis.
Measurement of mitochondrial membrane potential
Fluorescence intensity of rhodamine 123 with a high affinity for the mitochondria can be used to reflect the change of MMP.16 Briefly, after the cells were treated as described earlier, the cells were trypsinized, collected and washed twice with PBS. Then cells were resuspended with 500 μl PBS and rhodamine 123 stock solution (1 mg/ml) was added at a final concentration of 5 μg/ml. After incubation at 37°C for 30 min, cells were washed twice with PBS and then analyzed using flow cytometer for ΔΨm.
Three milligrams of kYSE 450 cell lysates were incubated in a total volume of 1 ml with 50 μl of protein A-agarose suspension (Roche Applied Science, Mannheim, Germany) for 3 hr at 4°C on a rocking platform to reduce nonspecific binding. After removal of the beads, the supernatant was supplemented with 5 μl rabbit polyclonal antibody against survivin (Cell Signaling Technology), followed by incubation for an additional 1 hr at 4°C. Fifty microliters of protein A-agarose was then added to each immunoprecipitation mixture, and the incubation was continued overnight at 4°C on a rocking platform. Immunoprecipitates were collected by centrifugation, washed 3 times with the wash buffer. After loading buffer was added, 25 μl agarose was boiled and subjected to immunoblot analysis with a monoclonal antibody to PLK1.
The correlations between PLK1 and survivin expression levels and clinicopathologic characteristics were analyzed using X2 test. The correlation between these 2 proteins expression levels was analyzed using Spearman correlation test. The Kaplan–Meier method was used to determine the relationship between PLK1 staining and patient survival, and data were analyzed with the log-rank test. Multivariate analysis was done using the Cox regression model to study the effectors of the following variables (PLK1 expression, nodal status and tumor stage) on survival. The differences in apoptosis index between groups were compared using 1-way analysis of variance, and data were expressed as mean ± SEM. Statistical difference was accepted at p < 0.05.
PLK1 expression was an independent poor prognostic factor and correlated with survivin
We analyzed PLK1 expression using an immunohistochemical approach combined with tissue microarray. In histologically normal tissues adjacent to tumors, PLK1 was undetectable in most of epithelial cells, except weak expression in basal cells. However, tumor cells exhibited moderate to intense diffuse staining by anti-PLK1 antibody. Of the 150 patients with ESCC, 81 (54.0%) were classified as low and 26 (17.3%) as high positive expression of PLK1 according to the criteria described in “Material and methods” (Fig. 1). PLK1 expression was significantly associated with regional lymph node metastasis (p = 0.001) and tumor stage (p = 0.047) (Table I). Kaplan–Meier analysis of 90 cases showed that the overall survival rates of patients with negative PLK1 expression was significantly higher than that with positive PLK1 staining (p = 0.001) (Fig. 2). Mean survival time of the patients with PLK1-positive tumors and PLK1-negative tumors were 39.3 (n = 63) and 57.8 months (n = 27), respectively.
|Variables||PLK1 negative||PLK1 positive||Total|
|p value||p = 0.997|
|p value||p = 0.429|
|p value||p = 0.946|
|Regional lymph nodes|
|p value||p = 0.001|
|p value||p = 0.047|
|p value||p = 0.539|
To determine whether PLK1 has independent prognostic impact on survival, we performed multivariate Cox regression analysis. The following parameters significant for survival in univariate analysis were included: regional lymph node metastasis, tumor stage and PLK1 expression. The parameters were dichotomized as follows: regional lymph node metastasis N0 vs. N1, tumor stage I/II vs. III and PLK1 expression negative vs. positive. In the analysis, PLK1 expression (p = 0.020) and regional lymph node metastasis (p=0.045) were shown to be independent prognostic factors (Table II).
|Variables||Relative risk||95% confidence interval||p value|
|Regional lymph nodes metastasis||3.884||1.031–4.638||0.045|
Because of the limited tissue content, we picked 98 from the 150 cases mentioned earlier to examine the expression of another oncogene, survivin. Twenty-five (25.5%) and 48 (48.9%) patients presented low expression and high expression, respectively. There was no correlation between suvivin expression and clinicopathologic parameters (Supporting data 1). But interestingly, survivin expression significantly correlated with PLK1 (Table III and Fig. 3).
|Negative (%)||Low positive (%)||High positive (%)|
|Negative||7 (7.3%)||9 (9.5%)||8 (8.4%)||24 (25.2%)|
|Low positive||11 (11.5%)||14 (14.7%)||29 (30.5%)||54 (56.8%)|
|High positive||3 (3.15%)||1 (1.05%)||13 (13.7%)||17 (17.9%)|
|Total||21 (22.1%)||24 (25.3%)||50 (52.6%)||95 (100%)|
Gene amplification contributed partially to PLK1 overexpression in ESCC
To verify that PLK1 protein was upregulated in ESCC, 5 tumor samples and the corresponding normal tissues were subjected to Western blot analysis. As shown in Figure 4a, PLK1 overexpression was detected in all the 5 tumor samples. We next measured PLK1 mRNA expression in esophageal tumors and the corresponding normal tissues from other 30 ESCC patients by RT-PCR. PLK1 was upregulated in 16 (53.3%) cases (Figs. 4b and 4c). To investigate whether gene amplification contributed to overexpression of PLK1 protein and mRNA, we detected the change of gene copy number by FISH. Six of 16 cases (37.5%) showed an increase signals in PLK1-hybridizing loci (Fig. 4d).
We further examined PLK1 expression in 9 ESCC cell lines and found a high level of PLK1 mRNA and protein in all the cell lines tested (Supporting data 2). PLK1 amplification was also observed in the cell lines (Fig. 4d).
Depletion of PLK1 inhibited tumors cells proliferation
We synthesized 3 shRNAs against PLK1 mRNA. Human H1 promoter was used to drive the expression of shRNA targeting PLK1 for the generation of a vector-based system (pGCsi-H1/PLK1). PGCsi-H1/Non was prepared as a control plasmid. PGCsi-H1/PLK1 and pGCsi-H1/Non were transfected into KYSE 450 cells. Cell lysates were assayed by Western blot analysis at 48 hr posttransfection with the plasmids. Target 1 and target 3 reduced endogenous PLK1 expression level by about 80 and 60%, respectively (Supporting data 3a) and appeared to be more effective than target 2 (Fig. 5a). Levels of PLK1 in cells transfected with pGCsi-H1/Non were unaffected (data not shown). Because Plk1 levels fluctuate in different cell cycle stages, we used an independent mitotic marker, cyclin B1, to demonstrate the specificity of PLK1 disappearance (Supporting data 3B). Target 1 was then selected for the subsequent study of PLK1 function.
Because PLK1 is one of the key components for the regulation of cell proliferation, we next assessed the antiproliferation impact of PLK1 inhibition on tumor cells. Compared to the control, transfections with shRNA 1 and shRNA 3 could evidently inhibit tumor cell proliferation (Fig. 5b), whereas shRNA 2 did not affect cell proliferation (data not shown).
Treatment with shRNAs induced apoptosis
We established the esophageal cancer cells model of downregulation of PLK1 by RNAi technology, which can specifically and effectively cause degradation of endogenous mRNA and lead to switching-off of synthesis of the target protein. Compared to the control, DNA staining by DAPI in PLK1 RNAi group showed an increase in cells with typical apoptotic morphology, including chromatin condensation and fragmentation. Cell cycle analysis revealed a sub-G1 peak representing the apoptotic cell population in a time-dependent manner (Fig. 5c)
We investigated the activation of the downstream effector caspase such as caspase-3. Western blot analysis of cells with PLK1 depletion showed that the level of 32 kDa pro-caspase-3 molecule grew downward upon PLK1 expression inhibition (Fig. 5d). As shown in Figure 5e, the proportion of cell showing active caspase-3 staining increased in PLK1-depleted cells at 48 hr posttransfection. At the same time, we can observe that DNA staining of the active caspase-3-positive cells was abnormal compared to control cells. The nuclei in cells with active caspase-3 were either relatively small, fragmented or chromatin condensation. Activated caspase-3 can cleave the death substrates, such as poly (ADP-ribose) polymerase (PARP). The nuclear enzyme PARP (116 kDa) is cleaved into fragments of 85 and 24 kDa by active caspase-3 during apoptosis. PARP cleavage has an active role in apoptosis. The expression of 85 kDa cleaved-PARP was increased after PLK1 inhibition (Fig. 5d). We also observed that PLK1 knockdown could induce KYSE 410 and KYSE 510 cell apoptosis followed the appearance of sub-G1 peaks.
PLK1 inhibition triggered the mitochondrial apoptosis pathway
To gain more insight into the mechanism of apoptotic signaling triggered by PLK1 inhibition, we explored the death receptor signal pathway. We tested the protein expression levels and activation of caspase-8, a key downstream molecule of the death receptor pathway. Western blot analysis showed that the expression of procaspase-8 was not affected (Fig. 6b).
To further delineate the role of the mitochondrial apoptotic pathway in PLK1 depletion-induced apoptosis, we investigated the change of the mitochondrial membrane potential (MMP), a characteristic of the mitochondrial apoptotic pathway, after PLK1 depletion. The rhodamine 123, a cationic mitochondrial fluorochrome, can be used to detect the ΔΨm combined with flow cytometer. When the MMP was cut down, the fluorescence got lower. After PLK depletion, cells displayed a significant loss of MMP when compared with the control cells (Fig. 6a).
To further examine the effect of PLK1 inhibition on mitochondrial pathway, we studied the activation of the downstream effector caspases, such as caspase-9. Contrasted with the control, pro-caspase-9 decreased and correspondingly, active caspse-9 staining intensity and proportion rose in PLK1 depletion group (Figs. 6b, 6d and 6e).
The mitochondrial apoptotic pathway can be negatively modulated by the antiapoptotic Bcl-2 family members, which play an important role in apoptotic control by suppressing modulating early events in cascade leading to cytochrome c release from mitochondria. Then, we detected whether PLK1 expression reduction can affect the protein levels of Bcl-2 family proteins such as Bcl-2 and Mcl-1. Western blot results showed that PLK1 inhibition caused the reduction of Bcl-2 and Mcl-1 (Fig. 6c).
Association between PLK1 and survivin
As PLK1 expression was correlated with an important anti-intrinsic apoptosis protein, survivin, PLK1 probably exerted its influence on apoptosis through survivin. We first investigated whether PLK1 might interact with survivin by immunofluorescence and coimmunoprecipitation.
For dual immunofluorescent labeling, KYSE 450 cells were fixed and incubated with rabbit polyclonal anti-survivin and mouse monoclonal anti-PLK1 antibodies that were revealed by Cy3-conjugated anti-rabbit IgG (red) and fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG (green), respectively. Merge analysis (yellow) showed the colocalization of PLK1 and survivin. By immunofluorescence and confocal microscopy, endogenous PLK1 and survivin colocalized during all cell cycle stages (Fig. 7). The same colocalization was observed at other ESCC cell lines such as KSYE 180, KYSE 150 and KYSE 410.
To perform coimmunoprecipitation, whole cell lysates prepared from KYSE 450 cells were immunoprecipitated with rabbit IgG or with a rabbit polyclonal anti-survivin antibody, and the immunoprecipitates were analyzed by immunoblotting with a mouse monoclonal anti-PLK1 antibody. As shown in Figure 8a, PLK1 was coimmunoprecipitated with the endogenous survivin, but not present in the control immunoprecipitates obtained with the rabbit IgG.
Reversal of apoptosis by cotransfection of survivin constructs
As survivin is an inhibitor of apoptosis proteins and its inhibition can induce cell apoptosis via mitochondrial pathway, survivin might participate in antiapoptotic function of PLK1. To evaluate this hypothesis, we first examined the effect of PLK1 knockdown on the endogenous survivin protein level. As shown in Figure 8b, PLK1 depletion treatment reduced the protein level of survivin. Because ubiquitin-proteasome pathway mediates the degradation of survivin,17 as a first step to investigate the mechanisms underlying survivin decrease following PLK1 depletion, we treated cell cultures with the proteasomal inhibitor MG-132. Figure 8b indicated that survivin downregulation upon PLK1 depletion was proteasome dependent. We then assessed whether PLK1 depletion-induced apoptosis could be rescued by survivin. To perform this experiment, wild-type and mutant survivin (Thr34Ala, which acts in a dominant-negative manner by binding to cdc2 to prevent phosphorylation of endogenous survivin18) expression plasmids were constructed and cotransfected with pGCsi-H1/PLK1. PI staining demonstrated that wild-type survivin can partially reverse the apoptosis phenotype induced by depletion of PLK1. In contrast, mutant survivin caused a small increase in sub-G1 cell population (Figs. 8c–8e). The data suggest that PLK1 may inhibit apoptosis partly by interacting with survivin and enhancing its protein level.
The development and progression of ESCC is thought to mainly arise from changes in some key genes that are related to cell proliferation, apoptosis and genomic stability. Alterations in several oncogenes and tumor suppressor genes have been described such as p53, Aurora-A, cyclin D1, fascin and Rb.19–22 In our study, we found that elevated protein level of PLK1 was detected in 71.3% of tumor samples examined. These data indicate that alteration of PLK1 expression level is a frequent event in human ESCC. PLK1 expression level was correlated to tumor stage, node status and overall survival of patients with esophageal cancer. Multivariate analysis showed that PLK1 was an independent prognostic factor. Thus, PLK1 could play an important role in esophageal carcinogensis and progression. Upexpression of PLK1 might be a useful prognostic marker.
What might be the reasons for the overexpression of PLK1 in a series of carcinomas? Gain of chromosome 16p might contribute to aberrant PLK1 expression at least in certain cancers. A common amplification of region 16p12, situated at the locus of the human PLK1 gene, has been demonstrated in gastric, breast, colorectal, oral and ESCCs.23–27 We observed amplification of the locus RP11-141E3 containing PLK1 gene in 37.5% of tumor samples by fluorescence in situ hybridization (FISH), suggesting that gene amplification might be the mechanism underlying PLK1 overexpression in a part of esophageal carcinomas. Upregulation rate of PLK1 mRNA was 53.3%, higher than gene amplification rate. Given the widespread PLK1 protein overexpression in ESCC, it is likely that multiple oncogenic signaling pathways might converge on the abnormal expression of PLK1. It is interesting to notice that 2 cases displayed downregulation of PLK1 mRNA level. Because similar results have never been reported before, further investigations are needed. Tokumitsu et al. previously detected overexpression of PLK mRNA in 97% (45/47) esophageal carcinomas, which is higher than our results. Their research also suggested that PLK was an independent prognostic factor.28
After PLK1 expression was downregulated, we found cell apoptosis in esophageal carcinoma in a time-dependent manner, confirmed by appearance of a sub-G1 population in FACS profiles. To further confirm the induction of apoptosis after PLK1 depletion, we detected the activation of caspase-3 and its substrate. Our examination revealed a sharp decrease in pro-caspase-3 and the cleavage of PARP. Cell apoptosis could be led by either the extrinsic/death receptor pathway or the intrinsic/mitochondrial pathway. In the case of mitochondrial pathway, we found that MMP was loss and the initiator caspase-9 was activated. Several groups demonstrated that PLK1 depletion decreased cancer cell viability.29–32 However, our results suggest for the first time that activation of the caspase cascade via the mitochondrial apoptotic pathway is one of the essential mechanisms for PLK1 inhibition-induced apoptosis.
We further investigated some apoptosis-related proteins and found that the expression of PLK1 and survivin proteins were correlated with each other in ESCC tissues. Moreover, PLK1 could colocalize with survivin and endogenous PLK1 can be coimmunoprecipitated with this molecule. To rule out nonspecific interaction, a rabbit IgG was included in the immunoprecipitation assays. These results demonstrated that PLK1 interacted with survivin in ESCC cells. Survivin is involved in both regulation of apoptosis and cell cycle progression. Cell death induced by interfering survivin function using siRNA or a phosphorylation-defective survivin Thr34Ala mutant had the hallmarks of mitochondrial-dependent apoptosis with loss of MMP,33 release of cytochrome c34, 35 and activation of caspase-9.18, 36, 37 Phosphorylation on Thr34 by mitotic kinase complex p34cdc2-Cyclin B (Cdk1) is crucial to maintain the stability and function of survivin.18 Ablation of Cdk1 results in loss of survivin expression level and enhances cell apoptosis.37–39 Inhibition of survivin phosphorylation on Thr34 led to mitochondrial-dependent apoptosis, cytochrome c release and caspase-9 activation.18, 37 Transgenic mouse expressing survivin could be protected against ultraviolet-B (UVB) irradiation-induced mitochondrial apoptosis, but no Fas-induced death receptor apoptosis.40 We found that PLK1 depletion treatment could cause the reduction of survivin protein level. In addition, wild-type but not mutant survivin showed a capacity to reverse the apoptosis induced by PLK1 knockdown. Bioinformatics analysis shows survivin protein contains a sequence (S-T-P) matching the consensus phosphorylation site for PLK1.41 Therefore, PLK1 may regulate cytoprotection through modulation of survivin stability and Cdk1 activity by phosphorylation. Taking together, these data indicate that survivin is an integral component and a downstream signaling molecule within the PLK1 inhibition-triggered cell death pathway. The detailed molecular mechanism through which PLK1 affects survivin function remains to be clarified in further studies.
In our study, we identified a decrease of Bcl-2 and Mcl-1 as a correlative link in the initiation of mitochondrial apoptosis induced by PLK1 depletion. Members of the Bcl-2 family of proteins are key messengers for delivering the apoptotic signal to the mitochondria. Decrease of Bcl-2 and Mcl-1 protein level was followed by loss of mitochondrial transmembrane potential and activation of caspase-9. We did not find PLK1 could interact with Bcl-2 and Mcl-1 by coimmunoprecipitation and colocalization assays. Thus, PLK1 inhibition could indirectly lead to degradation of the antiapoptotic Bcl-2 family member Bcl-2 and Mcl-1.
PLK1 inhibit induces apoptosis independently of the death receptor pathway, but it activates the intrinsic/mitochondrial apoptotic signaling cascade. At the mitochondrial level, PLK1 depletion-triggered signals cause the change of various mitochondrial antiapoptotic molecules that either directly propagate the apoptotic process or activate downstream molecules in the apoptotic signaling pathway.
Data obtained in our work and previous studies allow us to propose a model for the regulation of cell apoptosis by PLK1. PLK1 is involved in the inhibition of the mitochondrial-mediated apoptosis pathway. PLK1 inhibits apoptosis by maintaining the stability of several antiapoptotic proteins acting on mitochondria such as survivin, Bcl-2 and Mcl-1. Among them, PLK1 can be directly associated with survivin and affect its stability and function, thereby subsequently activating initiator caspase-9 and downstream effector caspases.
In summary, PLK1 protein expression was frequently upregulation in ESCC when compared with normal esophageal epithelia, and overexpression of PLK1 was significantly and independently linked to a poor prognosis. PLK1 depletion induced cell apoptosis via mitochondrial pathway. Survivin may be involved in this process. These data point to a potential role of PLK1 in the esophageal carcinogenesis and malignant progression. PLK1 might be an attractive potential target in ESCC therapy.
Additional Supporting Information may be found in the online version of this article.
|IJC_23990_sm_suppinfotable.doc||51K||Table 1. Relationship between survivin expression and clinicopathologic parameters in 98 ESCC patients|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.