Combination therapy of PKCζ and COX-2 inhibitors synergistically suppress melanoma metastasis

J-4 was acquired from Maybridge Chemical (Cambrige, CBS, UK). Celecoxib was purchased from Meilun Biological Technology (Dalian, China). Antibodies against Vimentin (AF7013), COX-2 (AF7003) and β-actin (T0022) were purchased from Affinity Biosciences (Shanghai, China). Antibodies against E-Cadherin (#14472), phospho-Cofilin (#3311), Cofilin (#3312), phospho-PKCζ (#9378) and PKCζ (#9372) were obtained from Cell Signaling Technology (Cambridge, MA, USA). Antibodies against MMP-2 (sc-53,630) and MMP-9 (sc-21,733) were purchased from Santa Cruz Biotechnology (Dallas, TX, USA.). Phosphatase inhibitor Cocktail tablets were purchased from Roche Molecular Biochemicals (Indianapolis, IN, USA). Z’-LYTE™ KINASE ASSAY KIT-SER/THR 7 PEPTIDE kit (Cat. No. PV3180) and PKCζ (Cat. No.2273) were purchased from Invitrogen (Carlsbad, CA, USA). Methylthiazolyldiphenyl-tetrazolium bromide (MTT) was purchased from Sigma-Aldrich (USA). DMEM medium, fetal bovine serum (FBS) and penicillin/streptomycin were all obtained from Gibco (Thermo Fisher Scientific Inc., USA).

Mouse melanoma cell line B16-F10 was purchased from the Cell Culture Center of Chinese Academy of Medical Sciences (Beijing, China) and cultured according to the instructions. Human melanoma cell line A375 was characterized by Genetic Testing Biotechnology Corporation (Suzhou, China) using short tandem repeat (STR) markers. The cells were cultured in DMEM medium containing 10% FBS and penicillin/streptomycin at 37 °C in an atmosphere containing 5% CO₂.

The Z’-LYTE™ assay was carried out according to the manufacturer’s instruction. Briefly, 20 μL/well reactions were set up in 384-well plates containing kinase buffer, 5 μM ATP, 4 μM ZMTP, 4 Ser/Thr peptide substrate, 50 ng/μL PKCζ and J4 with different concentrations (0, 5, 10, 25, 50, 100 μM). After 1-h incubation, the development buffer was added to each well and further reacted for 1 h, and followed by reaction stopping. The fluorescence signal ratio of coumarin at 445 nm and fluorescin at 520 nm was then calculated to evaluate the kinase inhibitory activity of J4 in the reaction.

MTT assay was used to evaluate the effect of J-4 and Celecoxib on cell proliferation. B16-F10 or A375cells were seeded into 96-well plates at 4000/well, incubated at 37 °C in 5% CO₂. Then cells were treated with J-4 at various doses, Celecoxib (25 μM) or their combination, respectively, for 24 h. MTT reagent was added to each well for further 4 h incubation. The medium was then discarded, and 150 μl of DMSO was added to each well. Subsequently, the plates were shaken for 30 s and the absorbance of each well was measured at 490 nm using a microplate reader (BioTek Epoch, Winooski, VT, USA).

Cell motility was measured using the Wound-healing assay according to protocol described previously [35]. Typically, B16-F10 or A375 cells were seeded into 60 mm dishes at a density of 8 × 10⁵cells/well and incubated for 12 h to grow a monolayer. After that, the culture media were replaced with the fresh culture media containing J4 (25 μM) and/or Celecoxib (25 μM), and the cells were further incubated for 24 h. Next, a linear scratch wound was created across the middle of the well surface using a pipette tip. The cells were then incubated in serum-free medium at 37 °C in 5% CO₂. At predetermined time points (0, 3, 6, 9, 12 and 24 h), the wound widths were quantified and photomicrographs were taken with an IX50 inverted microscope (Olympus, Tokyo, Japan). The experiment was carried out in double blind to eliminate the deviation induced by subjective factors.

Cell invasion in vitro were evaluated by Transwell assays. B16-F10 or A375 cells were pretreated with J-4 and/or Celecoxib at various doses and then seeded into the upper chamber coated with Matrigel matrix (BD Biosciences, MA, USA) at a density of 3.5 × 10⁴ cells/well in serum-free media containing or not containing various doses of J-4 and/or Celecoxib. The lower chambers were filled with media containing 10% FBS. The cells were allowed to migrate for 24 h incubated at 37 °C in 5% CO₂. The cells that migrated through the polycarbonate membrane were stained and counted visually in 5 random fields using the computer-based microcopy imaging system. The dose-effect curve and combination index (CI) was calculated by the CalcuSyn software 2.1. The experiment was carried out in double blind to eliminate the deviation induced by subjective factors.

Adhesion assay was performed as described previously [34]. Briefly, B16-F10 or A375 cells were treated with J4 (25 μM) and/or Celecoxib (25 μM) for 6 h, trypsinized, and re-suspended in serum-free media at a density of 3 × 10⁵ cells/mL. After incubation for additional 30 min, 1.5 mL of cell suspension was placed in 35 mm dishes containing glass cover slips that were coated with 10 ng/mL fibronectin. After further incubations for 5, 15 and 30 min, the cells were washed, fixed and counted in five separate fields under a light microscope. The experiment was carried out in double blind to eliminate the deviation induced by subjective factors.

F-actin was quantified by methanol extraction of Oregon Green 568/phalloidin–stained cells as described previously [24]. Briefly, B16-F10 or A375 cells were plated and cultured for 18 h in complete medium followed by further culturing in serum free medium for 3 h. Cells were then treated with the indicated inhibitors or DMSO for 2 h and stimulated by 50 ng/mL EGF at 37 °C. Cells were fixed, permeabilized, and stained in the dark with Oregon Green 568 phalloidin diluted in F-buffer (10 mM HEPES, 20 mM KH₂PO₄, 5 mM EGTA, 2 mM MgCl₂, PBS, pH 6.8) at room temperature for 60 min. After five washes, bound phalloidin was extracted with methanol at 4 °C and subjected to fluorescence analysis at 578 nm excitation and 600 nm emission. At the same time, an aliquot of cells were analyzed by a bicinchoninic acid assay (Pierce, Thermo Fisher Scientific Inc., USA) to determine total protein in the sample. Fluorescence signals were normalized against total protein. Results were expressed as relative F-actin content, where.

For observation of F-actin filaments, the cells were fixed and stained with rhodamine phalloidin (14 μM; Cytoskeleton, Denver, USA) in the dark for 30 min and finally imaged using a laser scanning confocal microscope (LSCM) (FV1000; Olympus, Tokyo, Japan).

Total RNA from cells pretreated with J-4 and/or Celecoxib was extracted by using Trizol. Then, RNA was transcribed by using a FastQuant RT kit (TIANGEN, China). The amplification reaction was carried out for 35 cycles. Each cycle consisted of denaturation for 1 min at 95 °C, annealing for 45 s and an extension for 1 min at 72 °C. A final extension step at 72 °C for 5 min terminated the amplification. The primer sequences as previous reports [23, 36, 37], the predicted amplicon sizes, and the annealing temperatures are depicted in Table 1.

C57BL/6 mice, 5–6 weeks old, were purchased from the Food and Drug Verification Institute (Beijing, China). All animal experiments were approved by the Animal Ethics Committee of Tianjin Medical University and complied with its regulations. A mouse model of melanoma lung metastasis was constructed by injection of B16-F10 cells into mice (5 × 10⁴cells/mouse) via tail vein. Compound treatment started the next day after melanoma cells injection. The mice were intravenously injected with normal saline, J-4 (20 mg/kg), Celecoxib (20 mg/kg), or their combination every three days, respectively. Mice were sacrificed after 3-week treatment, and the lungs were separated to examine the number of lung metastasis nodules. Then the lungs were homogenized and incubated in 1 M NaOH containing 10% DMSO at 80 °C for 2 h to measure the melanin content [38]. Then the homogenate were centrifuged and the absorbance of supernate was read at 490 nm. The relative melanin content was calculated as follows:

Animal activity and body weight were monitored during the entire experiment period to assess acute toxicity. The liver and lung of each mouse were fixed by formalin and examined by hematoxylin-eosin (HE) staining.

J-4 is an effective small-molecule inhibitor of PKCζ screened via a Z’-LYTE™ KINASE ASSAY KIT and its molecular structure is shown in Fig. 1A. J4 inhibited the activity of PKCζ in a dose-dependent manner, and its half maximal inhibitory concentration (IC₅₀) was calculated to be about 10 μM (Fig. 1B), which was consistent with previous reports [35]. MTT assays revealed that J-4 alone or combined with Celecoxib had slightly influence on viability of A375 (Fig. 1C) and B16-F10 (Fig. 1D) cells at indicated doses, which would not interfere with the following assays of motility properties.

Cell invasion is a critical step in cancer metastasis. To investigate the synergistic effects of J-4 combined with Celecoxib on the invasion of melanoma cells, the Transwell assay was performed. The cells were treated with J-4 (0.1, 1, 5, 10, 20 and 25 μM), Celecoxib (0.1, 1, 5, 10, 20 and 25 μM) and their combination (1:1), respectively. The results of J-4 (25 μM) combined with Celecoxib (25 μM) were shown, which significantly enhanced capability for suppressing the invasion of B16-F10 (Fig. 2A) and A375 (Fig. 2B) cells compared with mono-treatments with J4 or Celecoxib. The dose-effect curve and CI in A375 (Fig. 2C) and B16-F10 cells (Fig. 2D) were calculated by CalcuSyn software 2.1 according to previous reports [39]. The CI at various doses was less than 1, indicating a synergistic effect in the combination of J-4 and Celecoxib.

The migration of B16-F10 and A375 cells were evaluated using the Wound-healing assay. Compared with control or mono-treatment with J-4 (25 μM) or Celecoxib (25 μM), co-treatment exhibited more potent inhibitory effect on cell migration in B16-F10 (Fig. 3A, B) and A375 cells (Fig. 3C, D). Little mobile was observed with combined treatment after the scratch wound had been healed in control group. The striking differences in the migration distances indicated that the combination of J-4 and Celecoxib severely inhibited the migration of melanoma cells.

Cell chemotaxis depends on cell adhesion and actin polymerization. Adhesion assays were performed to assess the effect of J-4 combined with Celecoxib on melanoma cells adhesion. Although treatment with J-4 and/or Celecoxib resulted in a marked reduction in numbers of adherent cells after EGF stimulated for 5, 15 and 30 min, J-4 combined with Celecoxib exhibited more significant inhibition than mono-treatment with J-4 or Celecoxib (Fig. 4A, B). EGF induced actin polymerization was determined by F-actin content and LSCM based immunofluorescence. As shown in Fig. 4C, D, mono-treatment with Celecoxib had slightly influence on EGF induced F-actin formation. When Celecoxib combined with J-4, the two phase peaks of actin polymerization at 15 s and 60s, depending on Cofilin and PI3K [26], respectively, were eliminated. As observed by LSCM (Fig. 4E), F-actin accumulated at the cell leading edges, which further caused the deformations of B16-F10 and A375 cells under the stimulation of EGF. However, the phenomena of F-actin accumulation and cell deformation almost disappeared both in B16-F10 and A375 cells after exposure to combined treatment for 24 h. Taken together, J-4 combined with Celecoxib severely impaired cell adhesion and actin polymerization during melanoma cells motility.

To confirm J-4 combined with Celecoxib suppress melanoma cells chemotaxis in a PKCζ and COX-2 dependent manner, Western blotting assays were performed to analyze the expression of p-PKCζ, p-cofilin and COX-2 under EGF stimulation. Cofilin, an actin binding protein, plays an important role in actin polymerization and serves as an indicator of PKCζ related signal pathway. As shown in Fig. 5A, B, mono-treatment of J-4 decreased the phosphorylation of PKCζ and Cofilin induced by EGF, while Celecoxib reduced the over-expression of COX-2. However, co-treatment simultaneously reduced their expressions more significant than J-4 or Celecoxib, suggesting a synergistic but not additive effect existed in the combination of J-4 and Celecoxib. In RT-PCR results (Fig. 5C), total mRNA of PKCζ had no significant variation, indicating combined treatment affected the activity rather than expression of PKCζ. Total mRNA of COX-2 with co-treatment declined more than mono-treatment with Celecoxib both in B16-F10 and A375 cells, suggesting J-4 enhanced the inhibitory effect of Celecoxib on COX-2. Taken together, J-4 combined with Celecoxib synergistically suppressed the activity of PKCζ and expression of COX-2. In addition, after treatment with the combination of J-4 (25 μM) and Celecoxib (25 μM), the expression of E-Cadherin increased more than 2- and 3-fold in B16-F10 and A375 cells, respectively, while the expression of Vimentin both decreased about 50% in the two cell lines (Fig. 5D). As reported previously [40, 41], PKCζ plays an important role in the secretion of MMP-2/MMP-9 and the results of mono-treatment of J-4 further support it. However, co-treatment with J-4 and Celecoxib decreased the expression of MMP-2/MMP-9 more than each mono-treatment (Fig. 5D). The results indicate the combination of two inhibitors could induce MET in melanoma cells and decrease the expression of MMP-2/MMP-9.

B16-F10/C57BL mouse melanoma lung metastasis model is a classic method to evaluate cell metastasis capability in vivo [42, 43]. In order to test the efficacy of J-4 combined with Celecoxib in preventing tumor lung metastasis, B16-F10 cells were intravenously injected into C57BL/6 mice, which were then treated with J-4 (20 mg/kg), Celecoxib (20 mg/kg) and their combination for 3 weeks, respectively. Compared with control and mono-treatments, co-treatment with J-4 and Celecoxib exhibited more efficient in reducing pulmonary metastatic nodules and almost blocked melanoma lung metastasis (Fig. 6A, B). The result was further confirmed by the melanin content determination (Fig. 6C). No notable variation of body weight was observed during the entire experiment period (Fig. 6D), suggesting that co-treatment with J-4 and Celecoxib was low toxic in mammals. To further confirm the safety of the combination therapy, the liver and kidney of each mouse was analyzed by HE staining. No necrosis was observed in J-4, Celecoxib or their combination group (Fig. 6E). All above results signified that J-4 combined with Celecoxib possessed potent inhibitory effects on cancer metastasis but showed no significant toxicity.