Abstract
Background/Aim: Oral squamous cell carcinoma (OSCC) has a nearly 50% global mortality. Physalin A (PA) shows anti-cancer activities, but its role in metastasis remains unclear in OSCC cells. This study intended to determine whether PA inhibits OSCC cell migration and invasion and to clarify the underlying mechanisms.
Materials and Methods: HSC-3 OSCC cells were analyzed using wound-healing, migration, and invasion assays. Atomic force microscopy (AFM) was used to assess morphological changes. Western blotting examined E-cadherin (E-cad), matrix metalloproteinases (MMPs), and urokinase plasminogen activator (uPA). A RasV12/scrib−/− Drosophila model evaluated in vivo tumor suppression.
Results: PA significantly reduced wound closure, migration, and invasion in HSC-3 cells. AFM showed decreased cancer-related morphological alterations. PA increased E-cadherin and reduced MMPs and uPA. PA also inhibited growth factor receptor-bound protein 2 (Grb2)/Ras and phosphatidylinositol 3-kinase (PI3K)/protein kinase B (Akt)/nuclear factor-kappa B (NF-kB) signaling. In vivo, PA suppressed tumor formation and metastasis in RasV12/scrib−/− genotype Drosophila.
Conclusion: PA attenuates HSC-3 OSCC cell migration and invasion by regulating Grb2/Ras, PI3K/Akt/NF-kB, and MMP/uPA pathways, suggesting its potential as an anti-metastatic agent for OSCC.
Introduction
Oral squamous cell carcinoma (OSCC) is one of the most common human tumors worldwide. Both earlier diagnosis and therapeutic techniques have improved the 5-year survival rate for OSCC patients. Still, its mortality rate is nearly 50% and increasing in recent decades (1, 2). According to reports, both regional and distant metastasis are the main reasons for the reduced 5-year survival rate in OSCC patients. Metastasis is associated with lymph node invasion and regional migration, and it may complicate prediction and pose a severe problem in OSCC patients (3-5).
In cancer metastasis progression, many factors, such as cytoskeletal proteins, growth factors, chemokines, and cytokines, can mediate MMP-2 and MMP-9 to participate in extracellular matrix degradation and cell adhesion (6, 7). The decrease of E-cadherin (E-cad) is one of the hallmarks of metastasis, like the process of epithelial-mesenchymal transition (EMT). Both attenuations of growth factor receptor-bound protein 2/Ras (Grb2/Ras) and phosphatidylinositol 3-kinase/protein kinase B/nuclear factor-kappa B (PI3K/Akt/NF-kB) pathways can prevent cell migration and invasion and increase the overall the 5-year survival rate in OSCC patients (8, 9). KRAS mutant genes suppressed the Ras/ERK and PI3K/AKT signaling pathways by repressing SHP2 phosphorylation and forming the SHP2/Grb2/Gab1/SOS1 complex (10).
Physalin A (PA) was isolated from Physalis angulata L (11). The yellow-orange, pearl-like fruit of Physalis angulata grows inside a lantern-shaped calyx. PA is reported to inhibit c-Jun N-terminal kinase (JNK)/activator protein 1 (AP-1) and I kappa B (IκB)/NF-κB signaling pathways and also upregulate antioxidant activity in mouse RAW 264.7 cells (11). PA inhibits tumor growth by downregulating Janus kinase (JAK)/signal transducer and activator of transcription 3 (STAT3) signaling in human non-small cell lung cancer cells (NSCLC) (H292, H358, H1975 cells) (12, 13). It also promotes cell apoptosis through p53-Noxa-reactive oxygen species (ROS) signaling and induces autophagy through the p38-NF-κB pathway in prostate-specific antigens in A375-S2 human melanoma cells (14-16). Recently, it was reported that the PA-induced G2/M phase cell cycle arrest is related to the p38 MAPK/ROS signaling pathways in A549 human non-small-cell lung cancer cells (17). PA inhibits MAPK/NF-κB signaling via the integrin αvβ3 in primary cartilage cells from C57BL/6 mice (18). In our earlier studies, PA triggers apoptosis in human HSC-3 cells by activating caspase and mitochondria dependent signaling pathways (16). PA prevents tumor cell metastasis, but the underlying mechanism has not been elucidated. Thus, this study was the first investigation showing that PA interferes with Grb2-related cellular migration and invasion signaling effects in HSC-3 human OSCC cells in vitro.
Materials and Methods
Materials. PA was isolated from Physalis angulata, which was collected in Nantou County, Taiwan. The 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay kit, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and 4′, 6-diamidino-2-phenylindole (DAPI) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco’s Modified Eagle Medium (DMEM), fetal bovine serum (FBS), penicillin, streptomycin, and trypsin-EDTA were purchased from Invitrogen (Carlsbad, CA, USA). All primary antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA), Santa Cruz Biotechnology (Dallas, TX, USA). Enhanced chemiluminescence (ECL) was obtained from Perkin Elmer (Waltham, MA, USA). IRDye 700DX NHS Ester was purchased from LI-COR Biosciences (Lincoln, NE, USA).
Cell and Drosophila line culture. The HSC-3 human OSCC cells were obtained and described before (17). HSC-3 cells were cultured in flasks containing Dulbecco’s modified Eagle medium (DMEM) with 10% FBS (Gibco, Life Technologies, Inc., Grand Island, NY, USA), 2 mM L-glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin at 37°C in a humidified state of 5% CO2. The Drosophila line of upstream activating sequence (UAS)-RasV12 (RasV12/scrib−/− gene) was kindly given by Prof. Wei-Yong Lin (Graduate Institute of Integrated Medicine, College of Chinese Medicine, China Medical University). Drosophila was cultured at 25°C on a standard medium (19).
Cell morphological changes and cell viability assays. HSC-3 cells (1 × 105 cells) were placed in a 12-well plate overnight and were incubated with pA at the final concentrations of 0, 2, 4, and 16 μM, for 12, 24, 48, and 72 h. After incubation, cells were examined for morphological changes and photographed under phase contrast microscopy. Cells were then harvested for measuring total cell viability as described previously (16).
Wound healing assay. Cell mobility (wound healing ability) was performed by calculating the percentage of the cells migrated in scratched areas of cell culture plates, as described previously (19). HSC-3 cells (1 × 105 cells/well) were placed overnight in a culture medium well on 12-well plates. Then, a wound was scratched into a line with a sterile pipette tip (200 μl). The wells were then cleaned with PBS (washed twice) to clear the subtracted cells, and then the cells were cultured with 10% FBS medium (DMEM). Cells were treated with PA (0, 0.8, 1.6, 3.1, and 6.2 μM) for various periods. The reason for selecting these concentrations of PA were based on the cell viability assay. Cell mobility expression was detected at 0, 8, 16, and 24 h and photographed with an inverted light microscope (OLYMPUS IX71, Olympus, Tokyo, Japan).
Cell migration and invasion assays. Cells that penetrated the transwell filter were photographed and counted in the inverted light microscope for cell migration and invasion as described previously (20). HSC-3 cells (5 × 104 cells) were starved in a serum-free medium for 4 h before using collagen and 10% Matrigel coated transwell cell culture chambers (8 μm pore size) for measuring the percentage of cell migration and invasion, respectively. In the cell migration assay, the upper chambers of culture inserts were loaded with cells containing different concentrations of PA (0, 0.4, 0.8, 1.6, and 3.1 μM), and the lower chambers’ medium contained 10% FBS. Migrated cells in the collagen-coated transwell basement filter were fixed with 4% cold formaldehyde. Migrated cells were quantified with crystal violet (0.2%) staining solution and photographed in an inverted light microscope (IX71, Olympus) at 200×. The cell invasion assay was performed under the same conditions as the cell migration assay except for matrigel-coated membranes. Matrigel matrix was added to the transwell chamber at 100 μl/well in an incubator for 4 h at 37°C. Cells penetrating the transwell filter were photographed and counted using an inverted light microscope at 200× as described previously (21).
Atomic force microscopy (AFM) assay. The AFM assay explored three-dimensional morphological information of cells as described previously (22). HSC-3 cells (5 × 104 cells/ml) were cultured onto the coverslides overnight and were treated with PA (1.6 and 3.1 μM) or epidermal growth factor receptor (EGFR) inhibitor Gefitinib (Gef) (1.0 μM) for 24 h, and without both PA and Gef as a control. Cells were fixed with 4% paraformaldehyde. In the AFM assay, some core constants included: Spring constant (k, N/m), 2.7 N/m; tip radius, <10 nm; scan size, max 70-80 μm; scan rate, 0.4-0.6 Hz; number of lines, 512′256; force set-point, 1-3 nt. The AFM probe fits into the APP-Nano ACTA series whose tip and cantilever spring have a constant radius. Finally, the cell-binding force of AFM was measured and analyzed with NanoScope software (Bruker Co., Santa Barbara, CA, USA).
Western blotting analysis. HSC-3 cells (1 × 106 cells) were maintained in 6-well plates overnight and incubated with various concentrations of PA (0, 0.8, 1.6, 3.1 and 6.2 μM) for 24 h. The proteins were extracted from the treated cells with cell lysis buffer followed by centrifugation, and total proteins were quantified from each treatment. The quantified proteins were electrophoretically separated by 8-12% SDS-PAGE gel (based on different KDa), and were transferred to polyvinylidene difluoride (PVDF; Cat. No. IPVH00010; Merck Millipore, Burlington, MA, USA) membranes, which were blocked with 5% albumin. The membranes were then stained using primary antibodies [anti- plasminogen activator inhibitor 1 (PAI-1; sc-5297; Santa Cruz Biotechnology), -uPA (sc-59727; Santa Cruz Biotechnology), -E-cad (sc-374067; Santa Cruz Biotechnology), -α-tubulin (α-tub; sc-8035; Santa Cruz Biotechnology), -MMP-1 (sc-21731; Santa Cruz Biotechnology), -MMP-2 (sc-13595; Santa Cruz Biotechnology), -MMP-3 (sc-21732; Santa Cruz Biotechnology), -MMP-9 (sc-393859; Santa Cruz Biotechnology), -tissue inhibitor of metalloprotease (TIMP-1; sc-365905; Santa Cruz Biotechnology), -TIMP-2 (sc-21735; Santa Cruz Biotechnology), -p-IKK (p-IKK; sc-293135; Santa Cruz Biotechnology), -IKK (sc-8032; Santa Cruz Biotechnology), -p-IκB (sc-8404; Santa Cruz Biotechnology), -IκB (sc-1643; Santa Cruz Biotechnology), -NF-κB p65 (sc-8008; Santa Cruz Biotechnology), -NF-κB p50 (sc-8414; Santa Cruz Biotechnology), -p-ERK (9911; Cell Signaling Technology), -ERK (9911; Cell Signaling Technology), -p-c-Jun N-terminal kinase (-p-JNK; 4668; Cell Signaling Technology), -JNK (9252; Cell Signaling Technology), -p-p38 (9211; Cell Signaling Technology), -p38 (9212; Cell Signaling Technology), -p-c-Jun (3270; Cell Signaling Technology), -c-Jun (9165; Cell Signaling Technology), -Grb2 (3972; Cell Signaling Technology), -Ras (3965; Cell Signaling Technology), -p-PI3K (4228; Cell Signaling Technology), -PI3K (4292; Cell Signaling Technology), -p-Akt (9271; Cell Signaling Technology), -Akt (9272; Cell Signaling Technology), -PCNA (sc-56; Santa Cruz Biotechnology) and -β-actin (sc-47778; Santa Cruz Biotechnology)] in 4°C overnight. ECL detected immunoreactive proteins after treatment with horseradish peroxidase-conjugated secondary antibody. After washing and drying, membranes were quantified for each protein using the ImageQuant™ LAS4000 Mimi Molecular Imaging System (GE Healthcare Life Sciences, Pittsburgh, PA, USA).
Electrophoretic mobility shift assay (EMSA). HSC-3 cells (1 × 106 cells) were maintained on 6-well plates overnight, cells were treated with various concentrations of PA (0, 0.8, 1.6, 3.1, and 6.2 μM) for 24 h, harvested for nuclear extracts following the guidelines from the NE-PER Nuclear and Cytoplasmic Extraction kit (Pierce, Rockford, IL, USA). After quantifying the total protein from each sample, 5 μg nuclear extract proteins were further used and followed the LightShift Chemiluminescent EMSA Kit (Pierce). DNA sequences for EMSA analysis with sense (AGTTGAG GGGACTTTCCCAGGC) and antisense (GCCTGGGAA-AGTCCCCTCAACT). Nuclear proteins were incubated in the dark with a biotin-labeled DNA probe (NF-κB oligonucleotide) at room temperature for 20 min. The complementary oligonucleotides on a 5% polyacrylamide gel separated the DNA-protein complex. The gel was transferred to a nylon membrane, which was incubated with streptavidin horseradish peroxidase, followed by enhanced chemiluminescence (ECL; Thermo Scientific Pierce Protein Biology Products) for detecting NF-κB binding.
Confocal laser microscopy assay. HSC-3 cells (1 × 104 cells/well) were incubated on a confocal laser slide and treated with 0 or 3.1 μM of PA for 24 h. Cells were fixed with 4% paraformaldehyde and permeabilized by 0.25% Triton X-100 in 5% PBS-BSA. Cells were labeled with primary antibodies [anti-E-cad (sc-374067; Santa Cruz Biotechnology), and -α-tub (sc-8035; Santa Cruz Biotechnology)] at 4°C overnight. After washing, samples were labeled with a secondary antibody with an IgG Alexa Fluor 488 and Alexa Fluor 594 reagent. Nuclei were stained with DAPI (1.0 μ/ml) in the dark and were visualized and photographed with an SP8X Confocal Spectral Microscope (Leica, Wetzlar, Germany) as described previously (20, 21).
Gelatin zymography assay. HSC-3 cells (5 × 105 cells/well) were cultured in 12-well plates overnight, fresh DMEM containing PA (0, 0.4, 0.8, 1.6, 3.1, 6.2 μM) was added to cells for 24 h, and the conditioned medium was separately collected from each treatment. A 10 μl sample was resolved by 0.2% gelatin-10% SDS-PAGE gel. After electrophoresis, gels were washed with PBS, placed in substrate buffer (50 mM Tris HCl, 1% Triton X-100, 0.02% NaN3, and 5 mM CaCl2, pH 8.0), and then at 37°C shaking for 18 h to digest gelatin by MMPs. Finally, the gel was incubated for 2 h for staining (0.2% Coomassie blue) and washed with a destaining solution (10% acetic acid and 30% methanol) for 20 min. The MMP-2/-9 gelatinolytic activities were measured with clear bands after staining (23).
Fluorescence microscopy in RasV12/scrib−/− Drosophila. In a mosaic analysis with a repressible cell marker, a genomic technique labels a progenitor in single- or multiple-cell distribution at Drosophila (25). The green fluorescent protein (GFP) labeled eye discs were mutant for the tumor suppressor gene scrib and overexpressing oncogenic RasV12, an established genetic cooperation system that induces malignant Drosophila tumors (19, 24). In the quantification of the invasion ratio described, GFP-tagged fluorescent expression in the brain of Drosophila was defined as tumor formation. The fluorescence expression of GFP-tagged outside of the Drosophila brain were recognized as tumor metastasis (25). The fly newborn embryos were laid on the culture medium, which was also administered with various concentrations of PA, and the larvae grew in the medium. The third-instar Drosophila larvae were observed under a fluorescent microscope (Nikon BX51, Tokyo, Japan). The effect of PA on tumors of the third-instar Drosophila larvae with the fluorescent expression of GFP-tagged RasV12/scrib−/− gene, GFP-tagged fluorescent expression was detected using a microscope having GFP fluorescence channel and CCD camera (SPOT; Diagnostic Instruments, Sterling Heights, MI, USA).
Statistical analysis. All data were presented as the mean ± standard deviation (SD) obtained from three independent experiments (n=3). IBM SPSS Statistics for Windows Version 25.0 (IBM Corp., Armonk, NY, USA) was used to perform statistical analysis. We have analyzed the data using the Kruskal–Wallis test for comparisons involving more than two groups. Following a significant result, we applied Dunn’s post hoc test to identify specific differences between groups. The α level was set at 0.05, and p-values less than 0.05 were considered significant.
Results
PA induced cell morphological changes and decreased cell viability in HSC-3 cells. After cells were treated with PA at various concentrations for various time periods, they were examined for morphological changes, as shown in Figure 1A. Results indicated PA induced cell morphological changes in a dose-dependent manner (Figure 1A). The cells were then separately collected and assayed for cell viability (total viable cell number), and the results are presented in Figure 1B. PA decreased cell viability in a dose-and time-dependent manner (Figure 1B).
Physalin A (PA) induced cell morphological changes and decreased cell viability in HSC-3 cells. Cells (1×105 cells/well) were plated in 12-well plates and incubated in DMEM medium containing with 10% FBS overnight and were treated with PA (0, 2, 4, 8, and 16 μM) for 12, 24, 48, and 72 h. (A) Cell morphological changes were examined and photographed under contrast phase microscopy at 200×. (B) Statistical analysis of the total viable cell number (cell viability). Data were analyzed using the Kruskal–Wallis test for comparisons involving more than two groups, followed by Dunn’s post hoc test. All data are presented as the mean±standard deviation (SD). *p<0.05, **p<0.01, and ***p<0.001 as compared with the control group.
PA inhibited wound healing in HSC-3 cells. Wound healing assays were used in HSC-3 cells after various doses of PA were applied to measure cell mobility. Observations were completed at 0, 8, 16, and 24 h after PA treatment, and the results are shown in Figure 2A and B. The control group did not significantly inhibit healing after 24 h, and wounds from the treated area in HSC-3 cells were fully closed to healing (Figure 2A). Cells treated with 0.8 μM PA for 24 h showed minimal inhibitory effects approaching complete wound closure. However, after 1.6 μM of PA was applied, it showed variable inhibitions for 16 and 24 h. PA at 1.6, 3.1, and 6.2 μM treatment groups showed extremely variable inhibitions of slow cell movement for 8, 16, and 24 h compared to control groups in HSC-3 cells in vitro (Figure 2B).
Physalin A (PA) attenuated the tumor cell wound healing in HSC-3 cells. Cells (1×105 cells/well) were incubated in DMEM medium with 10% FBS and the cell wound healing assay was performed using silicone culture cannulas on 35 mm dishes overnight. Then, they were treated with PA (0, 0.8, 1.6, 3.1, and 6.2 μM) for 0, 8, 16, and 24 h. (A) Cell mobility was observed at 0, 8, 16, and 24 h using an inverted light microscope at 200×. (B) Statistical analysis of the wound healing assays. Data was analyzed using the Kruskal–Wallis test for comparisons involving more than two groups, followed by Dunn’s post hoc test. All data are presented as the mean±standard deviation (SD). *p<0.05, **p<0.01, and ***p<0.001 as compared with the control group.
PA decreased cell migration and invasion in HSC-3 cells. HSC-3 cells were treated with various doses of PA and were measured for cell migration and invasion. The results of the cell migration assay were compared to the control group (Figure 3A and B). PA treated at 0.4, 0.8, 1.6, and 3.1 μM showed a significantly decreased cell migration at 63.0±9.7%, 37.2±5.6%, 30.0±11.8%, and 10.7±7.7%, respectively (Figure 3B). In the invasion assay, PA treatment significantly decreased cell invasion compared to the control group (Figure 3C and D). Specifically, cells treated with PA at 0.4, 0.8, 1.6, and 3.1 μM showed invasion rates of 84.0±11.5%, 59.7±8.6%, 30.9±5.1%, and 18.7±1.7%, respectively (Figure 3D). Both of the effects of PA were dose-dependent.
Physalin A (PA) attenuated the tumor cell migration and invasion of HSC-3 cells. (A) HSC-3 (3×104 cells) were starved in serum-free DMEM for 4 h and then treated with PA (0, 0.4, 0.8, 1.6, and 3.1 μM) for 24 h in the upper chambers. The lower chambers were in a 10% FBS DMEM. The cell migration assay used 4% formaldehyde for fixation and crystal violet (0.2%) for staining. The migration cell numbers were quantified using an inverted light microscope at 200× (A) and calculated by ImageJ (B). A cell invasion assay was performed under the same conditions as the cell migration assay except for 10% matrigel-coated membranes. The diluted mixture was added to the transwell chamber at 100 μl/well for 4 h. Cells that had penetrated through the transwell filter were imagined using an inverted light microscope (C) and calculated by ImageJ (D). Data were analyzed using the Kruskal–Wallis test for comparisons involving more than two groups. Data were represented as the mean±standard deviation (SD). *p<0.05, **p<0.01, and ***p<0.001 compared to the control group.
PA attenuated cell morphological size and shape change in HSC-3 cells. Thus, Gef was used as a positive control in the AFM experiment. Results from the AFM assay are presented in Figure 4A and B. Figure 4A indicates that the height sensor shows a long shape in the control on HSC-3 cells. However, after PA treatment at 1.6 and 3.1 μM, the cells were more oval, and Gef (1.0 μM) treatment showed similar trends. That is consistent with the 4D image results presented in Figure 4B. After calculating the horizontal distance (μm) of PA and Gef treatment, both decreased significantly compared to the control (Figure 4C). In the control group, the horizontal distance was 69.0±3.2 μm; after PA treatment at 1.6 μM, it significantly decreased to 46.6±4.9 μm, and at 3.1 μM it decreased to 31.0±5.5 μm. Treatment with Gef resulted in a horizontal distance of 35.00±2.1 μm (Figure 4C).
Physalin A (PA) inhibited the cell length and morphological changes of HSC-3 cells. HSC-3 cells were treated with 1.6 and 3.1 μM PA for 24 h and Gef 1.0 μM for the positive control. (A) PA inhibited length (horizontal distance) and anterior morphological changes of the high sensor of three-dimensional (3D) images and side vision of the 3D displays and stereo vision in the AFM assay. (B) PA inhibited the cells’ horizontal distance; distance (μm) was calculated with NanoScope software. All Atom Force Microscopy (AFM) analysis data were replicated in three independent experiments (n=3). We applied Dunn’s post hoc test to identify specific differences between groups. Data are represented as the mean±standard deviation (SD). *p<0.05, **p<0.01, and ***p<0.001 compared to the control group.
PA influenced the tight and adherent junctions related protein levels. To further understand tight and adherent junction factors involved in cell migration, western blotting and confocal laser microscopy were used on HSC-3 cells after incubation with various doses of PA. The results are presented in Figure 5A. PA (6.2 μM) increased the expression of PAI-1 at 2.13±0.1 fold (Figure 5B) and E-cad at 2.49±0.3 fold (Figure 5C). However, PA at 6.2 μM decreased the expression of uPA at 0.51±0.02 fold (Figure 5B) and α-tub at 0.51±0.3 fold (Figure 5C). Confocal laser microscopy also showed the increased expression of E-cad but decreased expression of α-tub (Figure 5D) at cell-cell junctions with PA at 3.1 μM treatment for 24 h in HSC-3 cells.
Physalin A (PA) affected E-cad and α-tubulin expression in HSC-3 cells. HSC-3 cells (1×106 cells/well) were pretreated with PA (0, 0.8, 1.6, 3.1, and 6.2 μM) for 24 h. (A) The protein expression of PAI-1, uPA, E-cadherin, α-tubulin, and β-actin were detected by western blotting. (B) and (C) show statistical analyses of the cytoskeleton-related protein assays. HSC-3 cells were pretreated with PA (0 and 3.1 μM) for 24 h. Confocal laser microscopy assay detected E-cadherin (D) and α-tubulin (E) expression in HSC-3 cells. We applied Dunn’s post hoc test to identify specific differences between groups. All data were presented as the mean±standard deviation (SD) (n=3). *p<0.05, **p<0.01, ***p<0.001 compared to the control group.
PA affects the migration-related protein expression of MMPs and TIMPs. To recognize whether PA modulates the protein expression involved in cell migration and invasion, we analyzed HSC-3 cells with western blotting and gelatin zymography, respectively. Results from the protein assay are present in Figure 6A-E. Figure 6A shows that PA inhibited the protein levels of MMP-1, MMP-2, MMP-3, and MMP-9 but increased the protein levels of TIMP-1 and TIMP-2, based on the bands compared to controls. PA at 6.2 μM decreased the protein expression of MMP-1 at 0.71±0.04 fold, MMP-2 at 0.33±0.02 fold, MMP-3 at 0.54±0.07 fold, and MMP-9 at 0.45±0.02 fold (Figure 6B). However, PA at 6.2 μM increased the protein expression of TIPM-1 at 10.71±0.6 fold and TIMP-2 at 7.00±0.4 fold (Figure 6C). Furthermore, the gelatin zymography assay also showed that PA inhibited MMP-2/-9 activities in a dose-dependent manner. PA at 6.2 μM reduced the release of MMP-2 at 0.44±0.12 fold (Figure 6D) and MMP-9 at 0.3±0.03 fold (Figure 6E).
Physalin A (PA) decreased MMPs/uPA but increased TIMPs/PAI-1 signaling in HSC-3 cells. HSC-3 cells (1×106 cells/well) were pretreated with PA (0, 0.8, 1.6, 3.1, and 6.2 μM) for 24 h. (A) Matrix metalloproteinase (MMP)-1, -2, -3, -9, Tissue inhibitor of metalloproteinase (TIMP)-1, and TIMP-2 protein expression was analyzed with Western blot. (B) Statistical analysis used Dunn’s post hoc test for the MMP-related intensity of the PA group and was compared with controls. (C) Relative intensity for TIMP-1 and TIMP-2 proteins was statistically analyzed by comparing PA and control groups. HSC-3 cells (5×105 cells/well) were starved for 24 h in a serum-free DMEM containing PA (0, 0.8, 1.6, 3.1, and 6.2 μM). Then, culture medium was collected and proteins separated on a 0.2% gelatin-10% SDS-PAGE gel. The gelatin-containing gel was treated twice with Triton X-100 (2.5%) solution, developing buffer, and 0.2% Coomassie blue with the gelatin zymography assay. Statistical analysis of the MMP-2 (D) and MMP-9 (E) by the gelatin zymography assay. Dunn’s post hoc test was used to identify specific differences between groups. Data is represented as the mean±standard deviation (SD) (n=3). *p<0.05, **p<0.01, and ***p<0.001 compared to the control group.
PA inhibited the PI3K/Akt/NF-κB signaling pathway. Results from western blotting in Figure 7A and B showed that PA at 1.6-6.2 μM significantly inhibited p-IKK and p-IκB, respectively. Results from density analysis showed that a higher concentration of PA (6.2 μM) decreased the protein expression of p-IKK at 0.08±0.02 fold and p-IκB at 0.09±0.01 fold (Figure 7B). PA inhibited the nuclear translocation of p65 and p50, respectively (Figure 7C). Besides, PA (6.2 μM) decreased the protein expression of p65 at 0.18±0.02 fold and p50 at 0.22±0.04 fold in HSC-3 cells (Figure 7D) in the nucleus. Therefore, according to the EMSA analysis results, PA at 1.6-6.2 μM significantly inhibited NF-κB binding to DNA (Figure 7E). A higher concentration of PA (6.2 μM) could suppress the expression of NF-κB binding to DNA at 0.22±0.04 fold in HSC-3 cells, which are dose-dependent (Figure 7F).
PA inhibits the PI3K/Akt/NF-κB pathway in HSC-3 cells. HSC-3 cells (1×106 cells/well) were pretreated with PA (0, 0.8, 1.6, 3.1, and 6.2 μM) for 24 h. (A) Protein expression of phospho-Inhibitory-κB Kinase (p-IKK), IKK, phospho-Inhibitor of κB (p-IκB), and IκB in HSC-3 cells. (B) Analysis of the p-IKK, IKK, p-IκB, and IκB protein assay. (C) The nuclear translocation of transcription factors p65 NF-κB (p65) and p50 NF-κB (p50) were detected by western blotting. (D) Statistical analysis of the p65 and p50 protein assay. (E) The protein binding of p65 to DNA was measured with an EMSA assay. (F) Dunn’s post hoc test was applied to identify specific differences between groups of the p65 binding to DNA assay in statistical analysis. All data are presented as the mean±standard deviation (SD) (n=3). *p<0.05, **p<0.01, and ***p<0.001 as compared with the control group.
PA suppressed the Grb2/Ras/MAPK signaling pathway. The results from western blotting are presented in Figure 8A and C, and protein expression based on Figure 8A and D are shown in Figure 8B, C, E, and F. These results showed that PA induces dose-dependent effects to inhibit the phosphorylated protein expression of the MAPK signaling pathways in HSC-3 cells (Figure 8A). PA inhibited the phosphorylated protein level of p-ERK, p-JNK, p-p38, and p-c-Jun (Figure 8A). PA at 6.2 μM decreased protein expression of p-ERK at 0.23 fold, p-JNK at 0.17±0.01 fold, p-p38 at 0.21±0.01 fold (Figure 8B), p-c-Jun at 0.51±0.06 fold, p-c-Fos at 0.32±0.05 fold (Figure 8C). PA inhibited the expression of Grb2 and Ras in HSC-3 cells (Figure 8D). PA (6.2 μM) decreased the protein expression of Grb2 and Ras at 0.48±0.06 and 0.10±0.01 fold, respectively (Figure 8E). Moreover, PA inhibited the phosphorylation of PI3K and Akt; PA at 6.2 μM decreased the protein expression of p-PI3K by 0.24±0.02 fold and p-Akt by 0.21±0.12 fold separately (Figure 8F).
Physalin A (PA) inhibited the Grb2/Ras signaling. HSC-3 cells (1×106 cells/well) were treated with PA (0, 0.8, 1.6, 3.1, and 6.2 μM) for 24 h. (A) Western blots of phospho-Extracellular signal-regulated kinase (p-ERK), phosphor- c-Jun N-terminal kinase (p-JNK), phosphor-p38 (p-p38), and phosphor-c-jun (p-c-Jun). (B) Statistical analysis of the p-ERK, p-JNK, p-p38, and p-c-Jun protein assay. The protein expression of Growth factor receptor bound protein 2 (Grb2) and Ras and the phosphorylation of Phosphoinositide 3-kinase (PI3K) and Protein Kinase B (PKB; Akt) (C) were analyzed with Western blot. The statistical analysis of the Grb2 and Ras (D) p-PI3K and p-Akt (E) proteins. Dunn’s post hoc test was used to identify specific differences between groups. Data are represented as the mean±standard deviation (SD) (n=3). *p<0.05, **p<0.01, and ***p<0.001 compared to the control group.
PA reduced the cancer formation and metastasis via suppressed RasV12/scrib−/− gene expression. The fluorescence expression of tagged GFP in the brain of larvae in Drosophila was defined as tumor formation, and outside the brain of larvae in Drosophila was recognized as tumor metastasis. The quantification of the invasion ratio was described as GFP-tagged fluorescent expression (19, 26). PA could inhibit RasV12/scrib−/− fluorescence expression to 27.5±5.0% in 0.1 μM, and PA (0.2 μM) inhibited RasV12/scrib−/− fluorescence expression to 14.5±0.5% (Figure 9A and B) in third-instar larvae of Drosophila. Concerning tumor metastasis expression, PA at 0.1 μM inhibited RasV12/scrib−/− tumor expression to 33.9±3.9%, and PA 0.2 μM inhibited RasV12/scrib−/− tumor expression to 17.9±2.6% (Figure 9A and B) in the larvae of Drosophila.
Physalin A (PA) inhibited tumor formation and metastasis in RasV12/scrib−/− Drosophila. (A) The Drosophila with RasV12/scrib−/− genes were transferred on a culture medium in the culture tube. The embryos were hatched and grown to the early larvae stage with various concentrations of PA (0, 0.1, and 0.2 μM) per oral in the medium. (B) Tumor formation in the brain of Drosophila was quantified by assessing the fluorescent signal from GFP markers. The arrow bar points to the fluorescent RasV12/scrib−/− gene expression. The fluorescent appearance rate of GFP markers outside of the brain in the Drosophila was considered tumor metastasis. We applied Dunn’s post hoc test to identify specific differences between groups. All data are presented as the mean±standard deviation (SD). *p<0.05, **p<0.01, and ***p<0.001 compared to the RasV12/scrib−/− genotype only group.
Discussion
Gab2 and its homologs play a vital role in EGFR signaling and promote cell proliferation, motility, and lymph node metastasis, thereby promoting the growth of OSCC cells (26). The Grb2/Ras signaling pathway may drive tumor cell metastasis, promoting angiogenesis, adhesion, migration, and invasion in OSCC patients (27). Ras-activated PI3K/Akt/NF-κB pathways are related to the MMP-related metastatic effects (28, 29). The p38-MAPK and MMP-2/-9 signaling pathways are associated with cell proliferation and migration in oral cancer cells (30). Overexpressed MMPs are often highly correlated with the metastatic potential of human cancers (31, 32). Other studies indicated that PA exerts a toxicological effect on the tumor cells by inducing G2/M phase cell cycle arrest and inducing cell apoptosis through activated p53/p38/ROS signaling and up-regulating up regulated the activities of caspase-8 and 3 (33, 34). It has also been reported that PA influences cell colony formation and migration in breast cancer cells (35). However, the mechanism of PA metastasis inhibition is still unclear. At first, HSC-3 cells were treated with various concentrations of PA for different time periods and the cell morphological changes were further examined; results indicated that PA induced cell morphological changes and decreased total viable cell number (cell viability) in a concentration and time-dependent manner in HSC-3 cells (Figure 1A and B). Therefore, we selected a low concentration for further cell migration and invasion experiments.
A wound-healing assay usually measures the cancer cell mobility to investigate reversing cell migration and the coordination of the cell population (36). The comparison of Figure 2A and B indicates that PA inhibited cell mobility in a dose- and time-dependent manner in HSC-3 cells. The reason for selecting these doses of PA in this study is based on our earlier study. Thus, we selected the low doses (without inducing cell death) of PA for this study. The transwell migration assay is respected and vital for various biomedical investigations, such as tumor biology, immunology, and oncogenic pathology (37, 38). Also, the transwell invasion assay is used to quantify cell movement. It is necessary to recognize inhibitors of cancer invasion, which have constrained the progress of invasion assays (40). Herein, the results presented that PA suppressed cell migration (Figure 3A and B) and invasion (Figure 3C and D) in HSC-3 cells in vitro.
Gef is an oral first-line selective inhibitor of epidermal growth factor receptor (EGFR) for patients with EGFR-mutant advanced NSCLC (39). With analysis of variance from the AFM examination, PA and Gef both affect cell morphological changes of HSC-3 cells after being compared with control cells (Figure 4A and B). PA prevented cell morphological changes, which are associated with cell migration and invasion in HSC-3 cells.
Both uPA and uPA receptor (uPAR) enhance tumor cell migration and, thus, play a crucial role in cell migration development, resulting in poor prognosis of cancer patients (39, 41). However, PAI-1 interacts with cell surface receptors, which induces intracellular signaling and inhibits cell migration activity (42). PA suppressed the release and the protein expression of uPA (Figure 5A and B) but stimulated the protein expression of PAI-1 in HSC-3 cells (Figure 5A and B). Herein, PA improved the expression of PAI-1 but decreased the expression of uPA to attenuate cell migration in HSC-3. Also, E-cad is a cytoskeleton-associated protein mediating epidermal growth factors and establishing tight and adherent junctions to maintain epithelial cell polarity that may prevent cell migration and proliferation (43, 44). Still, α-tub is the predictive marker affecting adhesion kinetics to influence cell migration and expansion through tubulin acetylation (45). PA increased E-cad expression (Figure 5A, C, and D) and decreased α-tub expression (Figure 5A, C, and D) in HSC-3 cells.
Previous reports indicated that MMP-2 and -9 decreased E-cad protein levels and increased tubulin expression, which caused tumor cell migration and invasion (46-48). However, TIMPs and PAI-1 can inhibit MMPs and uPAR activation (49). Herein, PA suppressed the release and the protein expression of MMP-1, -2, -3, and -9 (Figure 6A and B) but stimulated both TIMP-1 and TIMP-2 in the HSC-3 cells (Figure 6A and C). Therefore, in the present study, PA could influence the MMP/TIMP signaling pathways to inhibit cell invasion and migration in human oral cancer HSC-3 cells.
In OSCC cells, the PI3K/Akt/NF-κB pathways regulate cell metabolism, invasion, and metastatic proliferation (50, 51). Furthermore, NF-kB activates MMP-2/-9 transcript expression to induce tumor cell metastatic and lymph node metastatic capacity (52). The present studies indicated that PA inhibited the phosphorylation of IKK and IkB (Figure 7A and B). PA decreased the intracellular translocation of p65 NF-κB and p50 NF-κB transcription factors in HSC-3 cells (Figure 7C and D). Based on these observations, PA suppressed cancer cell migration and invasion, possibly inhibiting the IKK/IkB/NF-κB signaling pathways in HSC-3 cells.
The signaling pathway of MAPKs is crucial in oral cancer progression because it involves tumor cell proliferation, differentiation, apoptosis, and angiogenesis; they also involve cell invasion and metastasis. MAPKs, including JNK, p38, and ERK, play critical roles in cell migration and invasion (53, 54). Results from Figure 8A and B indicated that PA inhibited p-ERK, p-JNK, p-p38, and p-c-Jun; therefore, we suggest that PA decreased the related protein expression of MAPK pathways in HSC-3 cells. Grb2 is a member of the Gab/Daughter of the sevenless family of proteins, which activates receptor tyrosine kinases (RTKs) (55). To elucidate this role, a set of studies showed that Grb2 activated the PI3K/Akt/NF-kB pathways to activate cell migration and tumor progression in OSCC cells (56). Conversely, inhibited Grb2 expression suppressed the Ras-ERK MAPKs and PI3K/AKT/NF-κB pathways (57). Effects of Gef inhibit EGFR for EGFR-mutant advanced NSCLC (36). The mechanism by which of Gef acts involves the reduction of metastasis-related proteins levels, including basic fibroblast growth factor (bFGF), MMP-2, and MMP-9. Gef decreased the expression of MMPs and increased cleaved caspase-3 activities to mediate apoptosis and autophagy of human oral cancer (58). Here, we describe that PA inhibited the protein expression of the Grb2/Ras signaling pathway (Figure 8C and D) and PI3K/Akt/NF-κB pathways in HSC-3 cells (Figure 8C and E). Our results indicate that PA inhibits the Grb2/Ras and PI3K/Akt/NF-κB signaling pathways in the HSC-3 cells in vitro.
Ras protein expression levels were quantified by measuring the fluorescence of tagged GFP out of the brain as the tumor metastasis in the RasV12/scrib−/− expression in the third-instar larvae of Drosophila (59). The fluorescence expression of GFP-tagged RasV12/scrib−/− was observed in the brain of Drosophila during tumor formation (59). The fluorescence expression of tagged GFP outside of the brain of Drosophila corresponded to tumor metastasis. The results demonstrate that PA effectively inhibits the RasV12/scrib−/− gene synergistically with the fluorescence expression of cancer formation and metastasis (Figure 9A and B). It indicated that PA suppresses Ras protein expression in the GFP-tagged RasV12/scrib−/− fluorescence to explore the Ras gene as a critical role in cancer formation and migration in Drosophila.
Conclusion
PA interference with Grb2/Ras signaling inhibited PI3K/Akt/NF-κB signaling pathways and decreased the tumor cell migration-related factors MMPs and uPA. Consequently, PA increased the E-cadherin tight junction protein levels to attenuate the tumor cell migration effect in the HSC-3 cells. We confirm that PA inhibits Ras expression in Drosophila. Overall, based on these observations, we investigated the possible signaling pathways by which PA inhibits HSC-3 cell migration and invasion in vitro, as shown in Figure 10. PA might be developed as a novel agent for preventing cell migration and invasion in HSC-3 cells in vitro in the future.
The proposed model of PA attenuates cell migration and invasion effects in HSC-3 cells. Together, the results indicated that PA inhibited the Grb2/Ras, the MAPK, the PI3K/Akt/NF-κB, and the MMP/uPA signaling pathways that may lead to PA attenuating the tumor cell mobility, migration and invasion in HSC-3 cells.
Acknowledgements
The identification of Physalis angulata was performed by Professor Chao-Lin Kuo, China Medical University, Taichung, Taiwan. The isolation and molecular structure identification of PA was done by Professor Yueh-Hsiung Kuo, China Medical University, Taichung, Taiwan. The experiments and resulting data analysis were performed through the Medical Research Core Facilities, Office of Research & Development at China Medical University, Taichung, Taiwan.
Footnotes
Authors’ Contributions
Yi-Shih Ma: Investigation, Writing – original draft, Data curation; Fu-Shin Chueh: Data curation, Investigation, Writing – original draft; Yueh-Hsiung Kuo: Data curation, Investigation, Methodology, Resources; Yu-Seng Hsieh: Data curation, Investigation, Methodology; Sung-Nien Yu: Conceptualization, Investigation, Methodology; Jiann-Yeu Chen: Data curation, Methodology; Wei-Yong Lin: Conceptualization, Formal analysis, Methodology; Jaw-Chyun Chen: Data curation, Investigation, Methodology; Chiu-Ying Chen: Data curation, Investigation, Methodology; Wen-Tsong Hsieh: Conceptualization, Resources, Project administration, Supervision, Writing – review & editing; Yi-Ping Huang: Conceptualization, Supervision, Project administration, Writing – review & editing.
Conflicts of Interest
The Authors declare that they have no known competing financial interests or personal relationships that could influence the work reported in this paper.
Funding
This work was supported by the National Science and Technology Council, Taiwan (grant no. NSTC 113-2320-B-039-029; NSTC 111-2813-C-039-161-B). This work is also supported by the Chinese Medicine Research Center, China Medical University, under the Higher Education Sprout Project, Ministry of Education, Taiwan (grant no. CMRC-CENTER-7). It was also supported by China Medical University and Asia University (CMU107-ASIA-24 and CMU103-ASIA-21).
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received November 4, 2025.
- Revision received November 24, 2025.
- Accepted December 2, 2025.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
Jump to section
Related Articles
Cited By...
- No citing articles found.