ORIGINAL
Glucocorticoids promote lung metastasis of pancreatic cancer cells through enhancing cell adhesion, migration and invasion
2023 Volume 70 Issue 7 Pages 731-743
Details
2023 Volume 70 Issue 7 Pages 731-743
Glucocorticoids (GCs) are the important stress hormones and widely prescribed as drugs. Although stress has been suggested as a promoter of tumor progression, the direct influence of GCs on metastasis of tumor is not fully understood. Metastasis is a major cause of death in pancreatic cancer patients. In the present study, we investigated the effect of GCs on progression of pancreatic cancer and elucidated the underlying mechanism. It was found that GCs significantly promote cell adhesion, migration, and invasion of pancreatic cancer cells in vitro and their lung metastasis in vivo. Further mechanistic studies showed that GCs notably up-regulate the expression of a trans-membrane glycoprotein, mucin 1 (MUC1) and increase the activation of AKT. Inhibiting MUC1 expression not only attenuates the activation of AKT, but also significantly reduces the promoting effects of GCs on cell adhesion, migration, invasion, and lung metastasis of pancreatic cancer cells. Moreover, GCs not only significantly up-regulate expression of Rho-associated kinase 1/2 (ROCK1/2) and matrix metalloproteinase 3 and 7 (MMP3/7), but also activate ROCK2, which are also involved in the pro-migratory and pro-invasive effects of GCs in pancreatic cancer cells. Taken together, our findings reveal that GCs promote metastasis of pancreatic cancer cells through complex mechanism. MUC1-PI3K/AKT pathway, ROCK1/2 and MMP3/7 are involved in the promoting effect of GCs on cell migration, invasion and metastasis in pancreatic cancer cells. These results suggest the importance of reducing stress and GCs administration in patients with pancreatic cancer to avoid an increased risk of cancer metastasis.
GROWING CLINICAL and epidemiological studies suggest that chronic stressful events, such as life pressure, cancer-related concerns and depression, are associated with poor cancer survival [1-3]. Animal-based preclinical studies also support that chronic stress alters the expression of cancer-related genes and promotes cancer progression [4-6]. Glucocorticoid (GC) is an important stress hormone, which is significantly increased in plasma under physical and psychological stress. In addition, the synthetic GCs, such as dexamethasone (DEX) are widely used as co-medication in the therapy of solid malignant tumors to relieve cancer-related complications (such as edema, inflammation, pain and nausea) and some of the side effects caused by chemotherapy (such as allergies, nausea and vomiting) [7-9]. Administration of GCs also leads to increased GCs levels in vivo. GCs not only regulate immune and inflammatory response, but also regulate metabolism, proliferation, and apoptosis of some tumor cells. Although it has been reported that GCs can indirectly affect tumor progression and metastasis through different ways, such as promoting tumor metastasis by suppressing immune reaction and enhancing the resistance of tumor cells to chemotherapy [10-14], or inhibiting tumor metastasis by inhibiting cell proliferation and angiogenesis in certain types of tumors [15-19], until now the direct effect of GCs on tumor metastasis is poorly understood. Our previous work found that GCs significantly promote melanoma metastasis through directly increasing migration and invasion of melanoma cells [11]. Since action of GC has tissue- and cell-type specificity, more studies on different tumors are needed to determine the effects of GCs on metastasis of tumors.
Pancreatic ductal adenocarcinoma (PDAC), which has become synonymous with “pancreas cancer,” is the most common type of pancreatic cancer and one of the deadliest types of cancer overall [20]. It is quite difficult for early diagnosis and therapy of pancreatic cancer, and the tumor is often found to have distant metastases by the time it is detected [21, 22]. There are some reports that diagnosis and treatment of patients with pancreatic cancer is associated with high levels of distress [23, 24], and chronic stress induced by subjecting mice to repeated daily restraint accelerates pancreatic cancer growth and invasion [5, 6, 25]. Although elevated catecholamine levels in chronic stress play an important role in promoting PDAC development at early stages via β-adrenergic signaling [6, 25, 26], direct effect of GCs on development and progress of pancreatic cancer remains unclear. It was reported that DEX inhibits invasion of a human PDAC cell line in vitro [19] and reduces tumor recurrence and metastasis after pancreatic tumor resection in SCID mice [18, 19]. However, other study has challenged the antitumor activity of GCs by demonstrating that DEX promotes PDAC proliferation, epithelial-mesenchymal transition and metastasis by induction of RAS/JNK and TGFβ signaling [27]. Therefore, it needs further study to clarify the effect and mechanism of GCs on pancreatic cancer.
It has been well known that most action of GC is mediated by glucocorticoid receptor (GR), which can directly modulate gene expression through binding to GC response elements in the promoter region of target genes or indirectly regulate gene expression by an interaction with other transcription factors and cross-talking with multiple trans-membrane signaling pathways [28]. In this study, we examined the effect of GCs on the pancreatic cancer cells, and found that GCs promoted pancreatic cancer cells adhesion, migration and invasion in vitro and lung metastasis in vivo. On this basis, we further explored the mechanism of pro-metastasis of GCs. Considering the formation of distant metastases constitutes a complex process with a variety of different genes and pathways involved, we choose several known metastasis-related molecules to investigate whether they are involved in the promoting effect of GCs on migration, invasion and metastasis in pancreatic cancer cells.
The murine pancreatic cancer cell line panc02 and the human pancreatic cancer cell line mia-paca2 were purchased from The Cell Bank of Type Culture Collection of the Chinese Academy of Sciences. Cells were cultured in DMEM (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% (v/v) FBS (Biological Industries), 100 units/mL penicillin and 100 units/mL streptomycin at 37°C, with 5% CO2. When cells were treated with DEX, the FBS in the medium was replaced with dextran coated carbon treated FBS, to avoid the possible interference of steroids in serum.
Cell proliferation assayCells (2.0 × 103 cells/well) were seeded in 96-well plates in triplicate and subjected to corresponding treatments for different intervals (0, 1, 2 and 3 days). MTT (500 μg/mL) was added to each well for 3 h and then dissolved in DMSO. The absorbance of each well was measured at 570 nm by a microtiter plate reader (Molecular Devices, LLC).
Cell migration and invasion assayCell migration and invasion were assessed using Transwell® chambers (Cat.No.3422; 24-well insert; pore size, 8 μm; Corning, Inc.) and BioCoat Matrigel Invasion Chambers (Cat.No.354480; BD Biosciences), respectively. After treatment with 100 nM DEX or vehicle for 24 h, cells at a density of 4 × 104 cells/well in a total volume of 200 μL were seeded in the upper chamber of each Transwell® insert in serum-free medium. In the lower chamber, media supplemented with 10% FBS was used as a chemoattractant; the media in both the upper and lower chambers contained 100 nM DEX. Subsequently, cells were incubated at 37°C for 24 h (migration) or 48 h (invasion). After the removal of the cells that had not migrated or invaded from the upper-side of the membrane using cotton swabs, cells were stained and counted in five randomly selected fields of view at 100× magnification using a light microscope to quantify cell migration and invasion.
Wound healing assayCells were seeded in a 6-well plate and allowed to reach confluence, then pre-treated with 100 nM DEX or vehicle for 24 h, and subsequently, a wound was created in the monolayer of cells using a sterile plastic 200 μL pipette tip. The monolayer of cells was washed three times with PBS to remove the detached cells and images were taken using a microscope (0 h). The remaining adherent cells were incubated with 100 nM DEX or vehicle for 12 h (panc02 cells) or 24 h (mia-paca2) and images were subsequently taken.
Cell adhesion assayAfter 48 h of treatment with 100 nM DEX or vehicle, the cells were digested into single cell suspensions and 8 × 104 cells/well were seeded into a 96-well plate in triplicate, and incubated at 37°C for 45 min in an incubator. Subsequently, the cells were washed thrice gently with PBS to remove the non-adherent cells. An MTT assay was used to quantify cell adhesion by measuring the number of remaining cells in the 96-well plate [29].
Small interfering (si) RNA transfectionMUC-1-siRNA was synthesized by Shanghai GenePharma Co., Ltd. and the sequences were as follows: MUC-1-siRNA (human), 5'-AAGTTCAGTGCCCAGCTCTAC-3'; control (Con) siRNA (human), 5'-CGCTTACCGATTCAGAATGG-3'; MUC-1-siRNA (mouse), 5'-GACUACAUCAGACUUAGCUTT-3'; Con siRNA (mouse), 5'-UUCUCCGAACGUGUCACGUTT-3'. Transient transfection was performed using the INTERFERinTM reagent (polyPLUS-transfection SA) according to the manufacturer’s protocol.
shRNA knockdownsThe lentiviral MUC1 shRNA (sc37266-SH) and control shRNA (sc10806) were purchased from Santa Cruz Biotechnology, Inc. and transfected to panc02 cells, respectively. The infected cells were selected with puromycin to construct panc02-MUC1-shRNA and panc02-control-shRNA cell lines. shRNA-based silencing was confirmed by western blotting.
Rho-GTP pull-down assayRho activity was measured using Rho-GTP pull-down assay established previously [11]. Briefly, cells were treated with 100 nM DEX and harvested to centrifuge to pellet insoluble materials. An equivalent amount of lysate from each sample was removed as an input control for detecting total Rho protein. The remaining lysate was combined with 60 μg Rhotekin-RBD protein beads (Cytoskeleton, Denver, CO, USA) and gently rotated for 1 h at 4°C. Precipitates were washed twice and resuspended with 30 μL SDS-PAGE loading buffer for detecting Rho-GTP protein by western blotting.
AnimalsThe animal protocol reported in the present paper was approved by the Committee on Ethics of Medicine of Naval Medical University (Shanghai, China) and performed in compliance with the University’s Guidelines for the Care and Use of Laboratory Animals. Female BALB/C nude mice (4–6 weeks old) were purchased from Shanghai SLAC Laboratory Animal Company and were fed in a specific pathogen-free environment at a standard temperature and maintained with a 12-hour light/dark cycle, with ad libitum access to food and water. All the mice were fed under the above condition for 1–2 weeks before the experiment to make them adapt to the new environment. Every cage housed no more than 5 mice. Individual mice were identified by toe-clipping and cages were identified by label paper.
In vivo pancreatic cancer metastasis assayThe mice were randomly assigned to different groups by body weight and a total of 1 × 106 pan02 cells or panc02-MUC1-shRNA cells or panc02-control-shRNA cells were injected into the tail vein of the mice. After 1 day, corticosterone (50 μg/mL CORT; Sigma-Aldrich; Merck KGaA) or an equivalent volume of the vehicle (alcohol) was added to the drinking water (actual daily consumption of about 3.5mL per mouse) given to the mice for 28 days [11, 30]. At the end of the experiment, the thoracic and abdominal cavities of mice anesthetized using isoflurane were opened and examined carefully. The lungs with metastatic nodules were removed and the macro-metastases on the surface of the lungs were counted. The lungs were fixed in 4% paraformaldehyde, embedded in paraffin and stained with hematoxylin and eosin (H&E) for histopathological examination. All the mice were killed by cervical dislocation upon anesthia. No mice died before the end of experiment.
Determination of CORT levels in serumPrior to dissection of the mice, blood was collected from the eye venous plexus of anesthetized mice and placed in an anticoagulant tube at dusk to avoid the influence of physiological GC fluctuations in vivo as much as possible. Blood was left to stand at 4°C for 30 min, and then centrifuged for 5 min at 10,000 rpm to obtain the plasma. CORT concentrations were determined using a radioimmunoassay and are expressed as ng/mL plasma.
Western blottingProteins were resolved using SDS-PAGE and transferred to nitrocellulose membranes. After blocking with 5% non-fat milk, the membranes were incubated with specific antibodies against MUC1 (Abcam), p-AKT1/2/3 (Ser473) (Santa Cruz Biotechnology, Inc.), AKT1/2/3 (Santa Cruz Biotechnology, Inc.), ROCK1 (BD Biosciences), ROCK2 (BD Biosciences), phospho- (p-)ROCK2 (S1366) (Abcam), MMP2, MMP9 (Santa Cruz Biotechnology, Inc.), MMP3 or MMP7 (Cell Signaling Technology, Inc.), followed by horseradish peroxidase-conjugated secondary antibodies (Rockland Immunochemicals, Inc.). Signals were visualized using ECL reagent (Pierce; Thermo Fisher Scientific, Inc.) and imaged using a Luminescent Image Analyzer LAS-4000 mini (FUJIFILM). β-actin (Merck KGaA) was used as the internal control for protein detection. ImageJ was used for densitometry analysis (National Institutes of Health).
Reverse transcription-quantitative PCR (RT-qPCR)Total RNA was extracted from cells treated with 100 nM DEX for different time intervals using TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc.), and 2 μg of total RNA was reverse transcribed using Reverse Transcription Reagents (Fermentas; Thermo Fisher Scientific, Inc.) according to the manufacturer’s protocol. qPCR was performed using SYBR Green PCR Master mix (Toyobo Life Science) on a Mastercycler ep realplex (Eppendorf). The primer sequences are shown in Table 1. The mRNA levels of MUC1 were normalized to GAPDH or β-actin (internal control) and relatively quantified using the 2ΔΔCT formula. Changes in gene expression were expressed as a relative fold-increase in mRNA compared with that of control. The experiments were performed in triplicate.
| genes | forward primers | reverse primers |
|---|---|---|
| MUC1 (human) | TCAGTGCCGCCGAAAGAAC | GCTCATAGGATGGTAGGTATCCC |
| MUC1 (mouse) | GGCATTCGGGCTCCTTTCTT | TGGAGTGGTAGTCGATGCTAAG |
| GAPDH(human) | CATGAGAAGTATGACAACAGCCT | AGTCCTTCCACGATACCAAAGT |
| β-actin (mouse) | CTGTATGCCTCTGGTCGTAC | TGATGTCACGCACGATTTCC |
Statistical analysis was performed using GraphPad Prism (GraphPad Software, Inc.). To compare differences between groups, a two-tailed t-test or an ANOVA was used. Unless otherwise specified, the quantitative data were presented as mean ± standard deviation. p < 0.05 was considered to indicate a statistically significant difference.
First, the effects of DEX, a synthetic GC, on mouse pancreatic cancer panc02 cells and human pancreatic cancer mia-paca2 cells were assessed. Treatment with DEX (100 nM) did not alter cell proliferation (Fig. 1A) and morphology (data not shown), but significantly increased cell adhesion (Fig. 1B), wound healing (Fig. 1C), Transwell migration (Fig. 1D) and invasion (Fig. 1E). Compared with the vehicle group, the adhesion, migration and invasion of panc02 cells treated with DEX were increased 1.8× (p < 0.01), 8× (p < 0.01) and 6.9× (p < 0.01), respectively (Fig. 1B, D and E). Similar results were also observed in the mia-paca2 cells. These results showed that DEX promotes pancreatic cancer cell adhesion, migration and invasion in vitro.
Effects of DEX on cell proliferation, adhesion, migration and invasion in pancreatic cancer cells. Murine pancreatic cancer panc02 cells and human pancreatic cancer mia-paca2 cells were treated with 100 nM DEX or vehicle for the indicated intervals, and (A) cell proliferation, (B) cell adhesion, (C) wound healing, (D) Transwell® cell migration and (E) Transwell® cell invasion were assessed. Representative images and quantification of the results of each assay are shown. Magnification, 100×. Data are expressed as the fold change of control and are presented as the mean ± standard deviation of three independent experiments. ** p < 0.01 vs. vehicle. DEX, dexamethasone.
The effect of CORT (the physiological GC in mice) on pancreatic cancer metastasis in vivo was assessed. As shown in the experimental workflow in Fig. 2A, the model was established by tail vein injection of 1 × 106 mouse pancreatic cancer panc02 cells (Day 0) in BABL/C female nude mice. Subsequently, the mice were randomly divided into two groups and received CORT (50 μg/mL) or vehicle in the drinking water from days 1 to 28, respectively. At the end of the experiment, plasma CORT levels and lung metastasis of pancreatic cancer were assessed. Results showed that the plasma CORT levels (Fig. 2B) and the number of macroscopic lung metastases (1.83 ± 1.22 in vehicle group and 13.71 ± 3.32 in CORT group, p < 0.01) (Fig. 2C) in the CORT group was significantly increased compared with those in the vehicle group. H&E staining showed the gross morphology of pancreatic cancer metastases in the lung (Fig. 2D). No tumor metastasis was observed in the abdominal cavity and other organs, such as the liver. These results further indicate that GCs also promote lung metastasis of pancreatic cancer cells in vivo.
Effects of CORT on lung metastasis of panc02 pancreatic cancer cells in BALB/C nude mice. (A) Diagram of the workflow of animal experiment. (B) On day 28, serum CORT levels in mice were measured. (C) Representative images of lungs with nodules. Number of macroscopic nodules (black arrowhead) on the surface of the lungs was counted. (D) Microphotographs of representative lung sections of metastatic nodules (black arrowhead) stained with hematoxylin and eosin (original magnification ×40). Data are presented as the mean ± the standard error of the mean. n = 7 in vehicle group, and n = 8 in CORT group. ** p < 0.01 vs. vehicle. CORT, corticosterone.
The transmembrane glycoprotein mucin 1 (MUC1) is aberrantly glycosylated and overexpressed in a variety of epithelial cancers, including PDAC, and plays a crucial role in progression and metastasis of tumors [31-33]. However, whether MUC1 was involved in the effect of GCs on metastasis of pancreatic cancer cells remains unknown. Here we demonstrated that DEX significantly increased both the mRNA and protein expression levels of MUC1 in panc02 and mia-paca2 cells in a time-dependent manner (Fig. 3A and B). Treatment with 100 nM DEX for 36 h increased the protein expression levels of MUC1 in panc02 and mia-paca2 cells 3.7-fold (p < 0.01) and 4.3-fold (p < 0.05), respectively, compared with the control cells. The increased MUC1 mRNA and protein were almost completely reversed by RU486, a glucocorticoid receptor (GR) antagonist (Fig. 3C). Knocking down the expression of MUC1 using siMUC1, as confirmed in Fig. 4B, almost completely abrogated the effects of DEX on cell adhesion (Fig. 3D), migration (Fig. 3E) and invasion (Fig. 3F). These results indicated that GC-increased MUC1 is mediated through GR and contributes to the effects of GC on cell adhesion, migration and invasion in pancreatic cancer cells.
Up-regulation of MUC1 mediates the effects of DEX on adhesion, migration and invasion of pancreatic cancer cells. Panc02 and mia-paca2 cells were incubated with DEX (100 nM) for the indicated intervals and the (A) mRNA and (B) protein levels of MUC1 were analyzed. Protein expression levels were quantified using densitometry analysis. Panc02 cells were incubated with DEX (100 nM) or/and RU486 (1 μM) for 24 h and the (C) mRNA and protein levels of MUC1 were analyzed. After transfection of cells with either scrambled siRNA (Con siRNA) or MUC1 siRNA, transfected cells were treated with DEX (100 nM) or vehicle control, and (D) cell adhesion, (E) migration and (F) invasion assays were performed. β-actin was used as the loading control. Data are expressed as fold change of control and are presented as mean ± standard deviation of three independent experiments. * p < 0.05, ** p < 0.01 vs. 0 h, vehicle or Con siRNA + vehicle; ## p < 0.01 vs. DEX or Con siRNA + DEX. DEX, dexamethasone; siRNA, small interfering RNA.
MUC1-AKT pathway mediates the effects of DEX on migration and invasion of pancreatic cancer cells. (A) Panc02 and mia-paca2 cells were treated with DEX (100 nM) for the indicated time points, and the protein expression levels of p-AKT (S473) and total AKT were detected using western blotting. Densitometry analysis was used to determine the expression levels of p-AKT (S473). (B) Panc02 and mia-paca2 cells were transiently transfected with either scrambled siRNA (Con siRNA) or MUC1 siRNA and treated with DEX (100 nM) or vehicle for 24 h. The protein expression levels of MUC1, p-AKT (S473) and total AKT were detected by western blotting. β-actin was used as the loading control. (C) Panc02 and (D) mia-paca2 cells were pre-treated with DEX (100 nM) and/or wortmannin (100 nM) for 24 h, and then cell migration and invasion assays were performed. Data are expressed as the fold change of control and are presented as the mean ± standard deviation of three independent experiments. * p < 0.05, ** p < 0.01 vs. 0 h or Con siRNA + vehicle; ## p < 0.01 vs. Con siRNA + DEX. DEX, dexamethasone; siRNA, small interfering RNA; p-, phospho.
MUC1 has been reported to interact with extracellular matrix (ECM) components, initiating downstream signaling pathways, including the PI3-K/AKT pathway, which is important for tumor metastasis [12, 34, 35]. Therefore, we investigated whether the effect of DEX is associated with MUC1-AKT signaling pathway. First, we examined the activity of AKT by detecting the level of phospho (p)-AKT at S473, and found that DEX significantly activated AKT at 12 h and the effect of DEX lasted at least 36 h in the two pancreatic cancer cell lines (Fig. 4A). Knocking down the expression of MUC1 reversed the DEX-increased activity of AKT in pancreatic cancer cells (Fig. 4B), indicating that DEX can activate PI-3K-AKT pathway through up-regulation of MUC1. Furthermore, inhibition of the PI3-K/AKT pathway by wortmannin, a specific inhibitor of PI3K, almost completely abrogated the pro-migratory and pro-invasive effects of DEX in panc02 cells (Fig. 4C). Similar results were also observed in mia-paca2 cells (Fig. 4D). These results indicated that activation of the MUC1/AKT pathway by DEX plays an important role in the pro-migratory and pro-invasive effects of GC.
MUC1/AKT pathway also contributes to the promoting effects of GC on lung metastasis of pancreatic cancer cells in vivoWhether the MUC1-AKT pathway was involved in the pro-metastatic effects of CORT in the xenograft model of pancreatic cancer was next assessed. Stable panc02 cells (panc02-MUC1-shRNA and panc02-control-shRNA) were established using MUC1 shRNA or control shRNA lentivirus, respectively and selected with puromycin. The knockdown of MUC1 was confirmed by western blotting (Fig. 5A). The results showed that compared with the CORT treated control shRNA group, inhibiting the expression of MUC1 with MUC1 shRNA did not alter the serum levels of CORT (Fig. 5B), but significantly reduced the CORT-induced increase in the number of macroscopic lung metastatic foci (0.50 ± 0.34 in vehicle + Con shRNA group, 5.00 ± 1.03 in CORT + Con shRNA group, 0.67 ± 0.49 in vehicle + Muc1shRNA group, 0.83 ± 0.48 in CORT + Muc1shRNA group) (Fig. 5C and D). These data are consistent with the results of panc02 cells in vitro, suggesting that upregulation of MUC1 mediated the pro-metastatic effect of GC on pancreatic cancer cells in vivo.
Up-regulation of MUC1 mediates the pro-metastatic effects of CORT in nude mice. BALB/c nude mice were randomly divided into two groups and were injected into the tail veins with panc02 cells stably transfected with MUC1 or control shRNA, respectively. Subsequently, mice in each group were randomly divided into two groups and provided water containing CORT or vehicle for 28 days, respectively. (A) MUC1 shRNA interference efficiency was determined by western blotting. (B) On day 28, serum CORT levels in mice were measured. (C) Images of lungs with metastatic foci, and (D) number of lung metastases counted in all groups. Data are presented as mean ± the standard error of the mean. n = 6 mice per group. ** p < 0.01 vs. vehicle + Con shRNA or vehicle + MUC1 shRNA; ## p < 0.01 vs. CORT + Con shRNA. CORT, corticosterone; shRNA, short hairpin RNA.
Rho-associated kinases ROCK1 and ROCK2 are downstream effectors of RhoA and RhoC, and are serine/threonine kinases. Accumulated evidence has shown that activated ROCK contributes to tumor metastasis by promoting migration of tumor cells [36-38]. Therefore, we examined whether ROCK1/2 was involved in the pro-migratory and pro-invasive effects of DEX in pancreatic cancer cells. The results showed that DEX up-regulated the mRNA (Fig. 6A) and protein (Fig. 6B) levels of ROCK1/2 in a time-dependent manner in pancreatic cancer cells. Besides DEX also rapidly activated the RhoC at 30 minutes (but not RhoA) (Fig. 6C) and then activated ROCK2 by increasing its phosphorylation levels in panc02 cells (Fig. 6D). The increased protein levels of ROCK1/2 and the activation of ROCK2 were reversed by RU486 (Fig. 6E and F), indicating the stimulatory effect of DEX was mediated through GR. Moreover, inhibiting ROCK1/2 activity with Y-27632, a ROCK1/2 specific inhibitor, completely abrogated the pro-migratory (Fig. 6G) and pro-invasive effects (Fig. 6H) of DEX in pancreatic cancer cells, indicating that up-regulation and activation of ROCK1/2 are important in the GC-mediated effects in pancreatic cancer cells.
Up-regulation and activation of ROCK1/2 mediates the effects of DEX on migration and invasion of pancreatic cancer cells. Panc02 and mia-paca2 cells were treated with 100 nM DEX for the indicated time periods, the mRNA (A) and protein (B) expression levels of ROCK1/2 were detected. Panc02 cells were incubated with DEX (100 nM) for the time indicated and the GTP-bound RhoA and RhoC were analyzed by Rho-GTP pull-down assay (C), and the p-ROCK2 levels were analyzed by western blotting (D). Panc02 cells were incubated with DEX (100 nM) or/and RU486 (1 or 2 μM) for 24 h and the protein levels of ROCK1/2(E) and p-ROCK2 (F) were analyzed. β-actin was used as the loading control. Densitometry analysis was used to quantify protein expression levels. Cells were pre-incubated with 10 μM Y-27632, a ROCK1/2 inhibitor, and/or 100 nM DEX for 24 h, cell migration (G) and invasion (H) assays were performed. Data are expressed as the fold change of control and are presented as the mean ± standard deviation of three independent experiments. * p < 0.05, ** p < 0.01 vs. 0 h or vehicle + DMSO; # p < 0.05, ## p < 0.01 vs. DEX + DMSO. p-, phospho; ROCK, Rho-associated kinase; DEX, dexamethasone.
A large number of experimental and clinical studies have demonstrated that the MMPs, such as MMP2, 3, 7 and 9 play vital roles in degradation of the matrix proteins and destruction of the basilar membrane, facilitating cancer metastasis [39-43]. However, the involvement of MMPs has not been previously reported to be associated with the effects of GCs on invasion and metastasis of tumor cells. Therefore, the effect of DEX on the protein expression levels of MMP2, 3, 7 and 9 in panc02 and mia-paca2 cells was determined using western blotting. The results showed that DEX treatment resulted in up-regulated protein expression levels of MMP 3 and 7 in a time-dependent manner, but did not affect the expression of MMP2 and 9 (Fig. 7A), and the increased MMP3/7 proteins were reversed by concurrent treatment with RU486 (Fig. 7B). Inhibiting the activity of MMPs in pancreatic cancer cells with BB-94, a pan-MMP inhibitor, reversed the increased invasion induced by DEX in pancreatic cancer cells (Fig. 7C). These results suggest that up-regulation of MMP3/7 by GC is mediated through GR and involved the pro-invasive effects of GC in pancreatic cancer cells.
Effects of DEX on the expression of MMPs in pancreatic cancer cells. (A) Panc02 and mia-paca2 cells were treated with 100 nM DEX for the indicated time periods, and the protein expression levels of MMP2, 3, 7 and 9 were analyzed. (B) Panc02 cells were incubated with DEX (100 nM) or/and RU486 (1 μM) for 24 h and the protein levels of MMP3/7 were analyzed. β-actin was used as the loading control. Densitometry analysis was used to quantify protein expression levels. (C) Cells were pre-incubated with a pan-MMP inhibitor BB-94 (10 μM) and/or DEX (100 nM) for 24 h, and cell invasion was assessed. Data are expressed as fold change of control and are presented as mean ± standard deviation of three independent experiments. * p < 0.05, ** p < 0.01 vs. 0 h or vehicle + DMSO; # p < 0.05, ## p < 0.01 vs. DEX + DMSO. MMP, matrix metalloproteinase; DEX, dexamethasone.
In the present study, it was shown that treatment pancreatic cancer cells with 100 nM of DEX significantly enhanced cell adhesion, migration and invasion in vitro. Administration of CORT in drinking water (50 μg/mL) also resulted in significant increases in the concentration of plasma CORT levels and in the number of lung metastatic foci in experimental models of pancreatic cancer metastasis in immunodeficient BALB/c nude mice. These results indicated that GCs promote the metastasis of pancreatic cancer cells through their direct effects on cell migration and invasion, rather than through their immunosuppressive effect in vivo. Since GCs/GR pathway exerts its biological effects through regulating gene expression, we examined the influence of GCs on the expression and activity of known metastasis-related molecules and downstream signaling pathways to elucidate the mechanisms by which the GCs exerted their effects in pancreatic cancer cells.
MUC1, a large transmembrane epithelial mucin glycoprotein, has been reported to interact with extracellular matrix (ECM) components and initiate downstream signaling pathways, including the PI3-K/AKT pathway, which plays a key role in tumor progression and metastasis including pancreatic cancer [31, 44-46]. Our previous study showed that up-regulation of MUC1 is involved in DEX-induced pro-adhesion, AKT activation, and enhancement of chemotherapeutic resistance in ovarian cancer cells [12]. However, whether MUC1 was involved in the effect of GCs on metastasis of pancreatic cancer cells remains unknown. Here, we show that DEX also induced expression of MUC1 in pancreatic cancer cells. Knocking down the expression of MUC1 not only reversed the enhancing effects of DEX on cell adhesion and AKT activation, but also obviously abrogated the promoting effects of DEX on cell migration, invasion and lung metastasis of pancreatic cancer cells. These data indicated that enhanced MUC1/AKT pathway mediates the promoting effects of GCs on metastasis of pancreatic cancer cells in vitro and in vivo.
Rho/ROCK signaling plays an important role in the migration and metastasis of various types of cancer, including pancreatic cancer cells [47-49]. Our previous study demonstrated that the increased ROCK1/2 proteins by DEX through enhancing protein stability mediated the promoting effects of GCs on migration, invasion and metastasis of melanoma cells [11]. In the present study, we showed that ROCK1/2 also contributes to the pro-migratory and pro-invasive effects of DEX in pancreatic cancer cells. However, the mechanism underlying up-regulation of ROCK1/2 by DEX in pancreatic cancer cells differs from that observed in melanoma cells. In pancreatic cancer cells besides increasing both mRNA and protein levels of ROCK1/2, DEX also rapidly increasing the phosphorylation of ROCK2 in panc02 cells. Since increased ROCK2 protein by DEX occurs at 12 h, whereas DEX activated ROCK2 at 2 h, suggesting that early activation of ROCK2 is independent of the DEX-induced expression of ROCK2 in panc02 cells. Previous studies have identified ROCKs as downstream effectors of Rho subfamily of small GTPases (RhoA and RhoC) [50, 51], therefore we further examined the effect of DEX on activation of RhoA and RhoC, the result showed that DEX could rapidly activate RhoC at 30 minutes, but had no significant effect on RhoA activation within 4 h, suggesting that the early activation of ROCK2 is due to the rapid activation of RhoC. This may be a non-genomic effect of GC, as the classical genomic mechanism of steroid action usually occurs between >30 minutes to hours or days due to the need for de novo protein synthesis [52], and the precise mechanism will need to be further studied.
Matrix metalloproteinases (MMPs) are a family of proteolytic enzymes that degrade multiple components of the ECM [53]. Among these MMPs, enhanced expression and activity of MMP2, 3, 7 and 9 have been reported to play vital roles in degrading the ECM and the basement membrane, in-turn promoting cancer metastasis [39-43, 54]. In the present study, it was shown that DEX increased the protein expression levels of MMP3 and MMP7. Although DEX did not affect the expression of MMP2 and MMP9, MMP-7 was reported to activate MMP2 and MMP-9 in vitro [55]. Inhibiting the activity of MMPs reversed DEX-enhanced invasion in pancreatic cancer cells, indicating that up-regulation of MMP3/7 is involved in the pro-invasive effects of DEX in pancreatic cancer cells. Our previous study demonstrated that GCs promote the invasion of melanoma cells not by up-regulating the expression of MMP3 and MMP7, but by down-regulating the expression of tissue inhibitor of MMP (TIMP)-2 [11], suggesting that the pro-invasion effect of DEX in different types of tumor cells is mediated by different members of MMPs and TIMPs families.
It is well known that mechanism of GC/GR regulating gene expression is very complicated, which occurs through direct or indirect regulation of transcription or post-transcription. In this study we found DEX up-regulated the expression of MUC1, ROCK1/2 and MMP3/7, the effect of DEX was blocked by RU486. Although RU486 is antagonists of GR and progestogen receptor (PR), the previous study showed that the normal exocrine epithelial cells of the pancreas and pancreatic adenocarcinomas did not exhibit PR [56]. In addition, an important point is that a higher dose of RU486 is required to obtain an anti-GC effect than an anti-progestin effect, so we used a high concentration of RU486 (1 or 2 μM) in our experiment [57, 58]. Therefore, the above effects of DEX are mediated by GR.
Classically, GR regulates gene transcription usually through direct binding to glucocorticoid response elements (GREs) in the promoter region of target genes. Since GC increase mRNA levels of MUC1, ROCK1/2 and MMP3/7 [58, 59], we analyzed a 2 kb fragment of the promoter sequence upstream of the transcriptional start site of the above genes using JASPAR [60], an open-access database, and found no consensus GRE except several potential GRE half sites and poor full sites. In addition, direct transcription regulation dependent on GR-DNA binding at the classical GRE sites usually occurs within 4 h [61]. For example, DEX significantly induced glucocorticoid-induced leucine zips (GILZ) mRNA expression at 2 h [62]. However, in our study, the DEX-induced mRNA expression usually occurred at 24 h, suggesting that DEX/GR up-regulated the expression of these genes not directly, but indirectly. Previous studies have shown that DEX/GR can regulate gene expression by activating other transmembrane signaling pathways and their downstream transcription factors [63]. For example, cortisol induced the mRNA expression of MMP7 through activating transcription factor AP-1 in amnion fribroblast [58]. How DEX/GR indirectly regulates the expression of these genes is very interesting and needs further study.
In the present study it was shown that GCs significantly promoted cell adhesion, migration, and invasion of pancreatic cancer cells in vitro and their lung metastasis in vivo. The action of GCs was mainly mediated by the MUC1-PI3K/AKT pathway, ROCK1/2 pathway and MMP3/7. These findings suggest the importance of reducing stress and GC administration in patients with pancreatic cancer to avoid an increased risk of cancer metastasis.
Glucocorticoid, GC; glucocorticoid receptor, GR; dexamethasone, DEX; corticosterone, CORT; mucin 1, MUC1; Rho-associated kinase, ROCK; matrix metalloproteinase, MMP; pancreatic ductal adenocarcinoma, PDAC; extracellular matrix, ECM; glucocorticoid response element, GRE
The present work was funded by the National Natural Science Foundation of China (No. 81472690).
The authors have nothing to disclose.
