Myristoylated alanine‐rich C kinase substrate phosphorylation promotes cholangiocarcinoma cell migration and metastasis via the protein kinase C‐dependent pathway

Human CCA specimens VSports.   The 60 paraffin‐embedded liver tissues collected from primary tumors of CCA patients were obtained from the specimen bank of the Liver Fluke and Cholangiocarcinoma Research Center, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. The research protocols were approved by the Human Research Ethics Committee, Khon Kaen University (no. HE471214), and informed consent was obtained from each patient before surgery.

Cell culture and chemicals.  The CCA cells were cultured in Ham’s F‐12 medium supplemented with 44 mm NaHCO₃ and 10% fetal bovine serum in a humidified atmosphere containing 5% CO₂. The medium was supplemented with penicillin (100 units/mL) and streptomycin (100 mg/mL). 12‐O‐tetradecanoylphorbol‐13‐acetate (TPA) was purchased from Sigma–Aldrich (St Louis, MO, USA) and GF109203x was from Invitrogen (Carlsbad, CA, USA). All other chemicals used were of analytical grade.

Isolation of total RNA and quantitative real‐time‐polymerase chain reaction.  The isolation of total RNA and quantitative real‐time–polymerase chain reaction (qRT–PCR) was as described previously^{( 11 )} with MARCKS‐specific primers.

mRNA in situ hybridization.  A non‐radioactive in situ hybridization (ISH) method for the detection of hamster MARCKS mRNA in paraffin sections was performed according to Yamada et al. ^{( 32 )} The immunoreactive products were visualized using diaminobenzidine (DAB) as the chromogen, with hematoxylin counterstaining. The stained sections were viewed under a standard light microscope.

Immunohistochemistry.  MARCKS or phosphorylated MARCKS (pMARCKS) was detected on the paraffin‐embedded sections using standard immunohistochemistry protocols. The sections were incubated with the 1:1000 anti‐MARCKS or 1:200 anti‐pMARCKS antibody (Santa Cruz, Santa Cruz, CA, USA) at 4°C overnight. After that, the sections were incubated with peroxidase‐conjugated Envision secondary antibody (DAKO, Glostrup, Denmark). Peroxidase activity was observed using the DAB solution as the substrate, with hematoxylin counterstaining. The staining frequency of MARCKS was semiquantitatively scored based on the percentage of positive cells as: 0% = negative; 1%–25% = +1; 26%–50% = +2; and >50% = +3. The intensity of the MARCKS expression was scored as weak = 1, moderate = 2, and strong = 3.

Constructing the retroviral vector encoding MARCKS and the retroviral transduction of MARCKS in CCA cells.  MARCKS cDNA (a kind gift from Professor Naoaki Saito, Kobe University, Kobe, Japan) with the EcoRI site at both the 5′ and 3′ termini was produced by polymerase chain reaction (PCR). After that, the PCR products were subcloned into a pMXs‐IG retroviral vector (kindly provided by Professor Toshio Kitamura, University of Tokyo, Tokyo, Japan). pMXs‐MARCKS was transfected into Plat‐A packaging cells using FuGene6 (Roche, Mannheim, Germany). KKU‐M156 CCA cells that express endogenous MARCKS at a low level were incubated in the virus‐containing supernatants with 8 μg/mL polybrene (Sigma–Aldrich). The green fluorescent protein‐expressing infected cells were sorted by Cell Sorter (FACS Vantage SE; Beckton Dickinson, San Jose, CA, USA).

PKC activity.  PKC activity was detected by using a non‐radioactive PKC activity assay kit according to the manufacturer’s protocol (Promega, Madison, WI, USA). In brief, the protein extract was incubated with the PKC reaction mixture, which contains a PKC‐specific peptide substrate (PepTag C1). The migration of the phosphorylated peptide toward the cathode (+) was then detected using agarose gel electrophoresis, and the gel image was analyzed and quantified using an ImageQuant Imager (GE Healthcare, Buckinghamshire, UK).

Transient knocking down of MARCKS using siRNA.  KKU‐M214 CCA cells (with a high endogenous MARCKS expression; 5 × 10⁴ cells) were seeded into a 12‐well plate for 24 h before transfection. The siRNA specific for human MARCKS mRNA (5′‐CTACACTTGGGCTCCTTTT‐3′) was obtained from JbioS (Saitama, Japan). The cells were transfected either with 20 pM siMARCKS or with a scrambled siRNA or with no siRNA sequence at all. Transfections were carried out using the oligofectamine system (Invitrogen) according to the manufacturer’s instructions. After siRNA transfection, the plates were incubated at 37°C for 24–72 h for further analysis.

Western blot analysis.  The cells were treated with 10 ng/mL TPA in serum‐free medium for 12 h, or 5 μm GF109203x for 30 min before TPA treatment. The cell lysate was electrophoresed and transferred to a polyvinylidene fluoride membrane (Immobilon‐P; Millipore, Bedford, MA, USA). The membranes were incubated with primary antibodies at 4°C overnight and secondary antibodies at room temperature for 1 h. Peroxidase activity was detected by Western lightning chemiluminescence reagents (Perkin Elmer, Boston, MA, USA). The apparent density of the bands on the membranes was estimated using densitometry.

Confocal microscopic analysis.  The cells were cultured on glass microscope slides, then immunostained using primary antibodies specific to MARCKS or pMARCKS (Santa Cruz). Either anti‐goat immunoglobulin G (IgG) Alexa 488 or anti‐rabbit IgG Alexa 555 (Molecular Probes, Eugene, OR, USA) was used as the secondary antibody. For pMARCKS and F‐actin colocalization, either donkey anti‐rabbit IgG Alexa 488 or pholloidin‐actin staining (Molecular Probes) was used as the secondary antibody. The stained slides were visualized using a fluoview Olympus confocal laser microscope (Tokyo, Japan) with a ×40 objective lens.

Re‐attachment assay.  In total, 4 × 10⁴ CCA cells in serum‐free medium were plated onto fibronectin‐coated 24 well plates (BD Bioscience, San Jose, CA, USA) and allowed to attach overnight. All assays were performed in triplicate in each of two independent experiments. After incubation, the wells were washed three times with phosphate‐buffered saline. The attached cells were then fixed and stained with Diff‐Quick staining solution (Sysmex, Japan) and counted under a microscope.

Migration and invasion assay.  In total, 4 × 10⁴ siMARCKS transfected cells were incubated with 10 ng/mL TPA or 5 μm GF109203x in serum‐free medium for 30 min before TPA treatment. After that, the cells were seeded into the upper chamber of migration or invasion Transwell (8 μm pore size; Becton Dickinson, San Jose, CA, USA) and aliquots of Ham’s F 12 medium supplemented with 10% w/v fetal bovine serum were placed in the lower chamber. After incubation at 37°C for 18 h, the cells in the upper surface of the filter were scraped off. Cells that had migrated to the underside of the filter were fixed and stained with Diff‐Quick staining solution and then counted under a microscope. The mean values of 10 low‐power fields (×100 magnification) were determined. Assays were done in triplicate, and two independent experiments were repeated.

Statistical analysis.  Data were expressed as mean ± SD and statistically analyzed using the Student’s t‐test. The association of MARCKS expression in CCA tissue and patients’ histopathological data was analyzed and patient survival was calculated according to the Kaplan–Meier log–rank test. A P‐value <0.05 was considered statistically significant.

MARCKS mRNA upregulation in liver tissue during CCA development.  QRT–PCR and mRNA ISH were performed to determine the MARCKS mRNA level and localization in the tissues at different time points after treatment with either Ov or NDMA or both. The MARCKS mRNA level was found to be increased after Ov plus NDMA administration from weeks 1 to 3 and was strongly upregulated again at week 26 when the CCA developed (Fig. 1). mRNA ISH showed little or no staining in the normal bile duct at all time points of the control group. When Ov was administered together with NDMA, a strong positive expression was found in cancer cells at 26 weeks, whereas only faint to moderate staining was observed in the bile duct at 4, 12, and 24 weeks. At 1, 2, or 3 weeks of Ov plus NDMA treatment, moderate to strong expression was observed in inflammatory cells (Fig. 2).

Immunohistochemical analysis of MARCKS and pMARCKS in human CCA tissues.  Of the 60 CCA patients studied, 19 were males and 41 were females. The mean age was 54.95 ± 9.1 years (range: 33–74 years). Most of the patients were at an advanced stage, with 61% metastasis. The certain immunoreaction of MARCKS and pMARCKS was observed in tumor cells (Fig. 3a,c), whereas it was only weakly expressed in normal bile duct epithelia and hepatocytes. Positive staining for the MARCKS protein was illustrated in the cell membrane, whereas pMARCKS was localized in the cytoplasm of the tumor biliary cells. Immunohistochemical staining for MARCKS showed positive staining in 53 cases (88%), whereas positive staining for pMARCKS was seen in 49 (81%) CCA patients. A percentage of the 53 CCA patients exhibited a high expression of both forms with statistical significance (P = 0.032). Neither an increased level of MARCKS nor pMARCKS was associated with age, sex, or histological grading. Of the 53 cases with MARCKS positive staining, a high level of this protein was significantly correlated with a positive metastasis status (P = 0.039). Moreover, of the 49 cases of pMARCKS positive staining, a high level of pMARCKS was significantly correlated with a positive metastasis status (P = 0.009). The cumulative survival analysis revealed that a high level of MARCKS, but not pMARCKS, showed a significant correlation with 5‐year survival (P = 0.02) (Fig. 3b,d). To prove whether the correlation between a high expression of MARCKS and poor prognosis of CCA patients was due, at least in part, to metastasis induction, we investigated in depth the mechanisms by which MARCKS influences the adhesion, migration, and invasion of CCA cells.

PKC induces the translocation and phosphorylation of MARCKS in CCA cell lines upon TPA stimulation.  The subcellular localization of MARCKS upon stimulation with TPA was investigated using a PKC activator. In the M156 CCA cell line, MARCKS was overexpressed; in the M214 CCA cell line, high endogenous MARCKS expression was either exposed to 10 ng/mL TPA or to 5 μm GF109203x (PKC inhibitor) for 30 min prior to TPA treatment. Confocal laser microscopy was performed to visualize the intracellular localization of MARCKS and pMARCKS. In non‐stimulated cells, MARCKS was mostly located at the plasma membrane, and pMARCKS was evenly distributed throughout the cytoplasm. After stimulation with TPA, MARCKS was translocated to the cytoplasm and perinuclear area, and pMARCKS became highly concentrated at the perinuclear area. With prior PKC inhibitor treatment, MARCKS and pMARCKS distribution and abundance were similar to those observed in non‐stimulated cells (Fig. 4a). In addition, TPA induced the pMARCKS protein (Fig. 4b,c) and PKC activity up to 1.8‐fold (Fig. 4d) over untreated cells.

MARCKS is associated with the re‐attachment of CCA cell lines.  To address the functional importance of MARCKS, gain and loss of function experiments were carried out using M156 and M214 CCA cell lines. M156 cells were infected with the retroviral vector encoding a human MARCKS cDNA to construct the stable overexpressing MARCKS cell line. The overexpression of MARCKS was clearly observed in infected cells, as determined by western blot analysis (Fig. 5a). As shown in Figure 5(c,d), MARCKS‐overexpressing M156 showed a 2.5‐fold increase in cell re‐attachment over the untransfected M156 cell line. Suppressing MARCKS mRNA function in the M214 CCA cell line by siRNA decreasing cell re‐attachment by up to twofold. A similar effect was observed in the knocking down of MARCKS‐overexpressing M156 cells (data not shown).

TPA‐induced MARCKS phosphorylation promotes cell migration and invasion in CCA cell lines.  To investigate the function of pMARCKS on CCA cell migration and invasion upon TPA stimulation, the M214 cell line was transiently transfected either with MARCKS siRNA to suppress MARCKS protein translation or with the scrambled siRNA as a control. After 48 h of transfection, the western blot analysis revealed that the level of the MARCKS protein was diminished (Fig. 5b). In comparison, modified Boyden chamber migration and invasion assays were performed, as shown in Figure 5(e–h). After TPA stimulation, siRNA against MARCKS decreased cell motility and invasion approximately five‐fold and two‐fold, respectively. After the PKC inhibitor was added, fewer numbers of cell migration and invasion were observed. The same result was observed in the knocking down of MARCKS‐overexpressing M156 cells (data not shown).

Phosphorylation of MARCKS affects F‐actin disassembly upon TPA stimulation.  This study investigated the subcellular localization of MARCKS and F‐actin upon stimulation with TPA. M214 cells were exposed to 10 ng/mL TPA or 5 μm GF109203x. The immunofluorescent staining of MARCKS and F‐actin was performed to visualize the intracellular localization of MARCKS and F‐actin. In non‐stimulated cells, MARCKS was found in cell membranes and cytoplasms, whereas F‐actin formation was seen in the plasma membranes. After stimulation with TPA, MARCKS became highly expressed around the perinuclear area, and F‐actin disassembly was rearranged. Prior to PKC inhibitor treatment, MARCKS and F‐actin distribution and abundance were similar to that observed in the control cells (Fig. 6).