The edible gintonin fraction was prepared from Panax ginseng by applying a previously described method [7]. Gintonin was dissolved in saline and diluted with medium before use. RPMI 1640 medium, DMEM, fetal bovine serum (FBS), horse serum (HS), penicillin, and streptomycin were obtained from Invitrogen (USA). LPA (1-oleoyl-2-hydroxy-sn-glycero-3-phosphate; 857130P) was obtained from Avanti Polar Lipids (Alabaster, USA). All other reagents used in the experiments, including BAPTA-AM and N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES), were obtained from Sigma-Aldrich (USA).

HCE cells (RCB 2280 HCE-T) were kindly provided by the Riken BRC Cell Bank (Japan) [12] and were grown in RPMI 1640 medium supplemented with 10% heat-inactivated HS, 5% heat-inactivated FBS, 100 units/mL penicillin, and 100 µg/mL streptomycin in a humidified atmosphere with 5% CO₂ at 37℃.

Male rabbits (New Zealand White, 8 weeks) were purchased from Koatech (Korea). The rabbits were housed one per cage, allowed access to water and food ad libitum, and maintained under constant temperature (23℃ ± 1℃) and humidity (55% ± 5%) under a 12 h light/dark cycle (lights on 07:00 to 19:00 h). All experiments in this study were conducted in accordance with the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health. The study protocol was approved by Konkuk University (permit No. 13-187).

Cytosolic calcium concentration was determined by performing dual excitation spectrofluorometric analysis of cell suspensions loaded with Fura-2 AM (Sigma-Aldrich, USA) as previously described [12]. Briefly, HCE cells (either untreated or treated with nerve growth factor) were harvested with trypsin/EDTA solution and resuspended in HEPES-buffered saline solution (HBS; 120 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 1.5 mM CaCl₂, 10 mM glucose, 25 mM HEPES, pH 7.4). The cells were incubated with Fura-2 AM (final concentration 2.5 µM) in HBS at 37℃ for 30 min. Centrifugation was used to remove extracellular Fura-2 AM. Each aliquot of 3 × 10⁶ cells was put into a cuvette, and cytosolic calcium mobilization was measured by using a RF-5301PC spectrofluorophotometer and Supercap software (excitation wavelengths, 340 nm and 380 nm; emission wavelength, 520 nm; Shimadzu, Japan).

HCE cells were washed with cold phosphate-buffered saline (PBS), collected, and lysed with buffer (20 mM Tris-HCl, pH 7.5, 2 mM EDTA, 5 mM dithiothreitol, 0.32 M sucrose, and a protease inhibitor cocktail) to obtain membrane fractions. The HCE cell lysates were centrifuged at 12,000 × g for 30 min at 4℃; the pellets were resuspended in the same lysis buffer (1% Triton X-100, incubated on ice for 45 min) and centrifuged again at 12,000 × g for 30 min at 4℃. The supernatants were collected and used as membrane fractions. Total cell protein was measured by the BCA method (Thermo Scientific, USA). HCE cell membrane fractions were separated by performing sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on 10% gels and then transferred to polyvinylidene difluoride membranes. After that, horseradish peroxidase-conjugated goat anti-mouse secondary antibody (Millipore, USA) and 16B12 monoclonal antibody (Millipore) were added, and the results were visualized by enhanced chemiluminescence assay (GE Healthcare, UK).

HCE cells were first incubated in 150 mm dishes, serum-starved in M199 containing 1% FBS for 6 h, and treated with M199 containing 1% FBS in the presence of gintonin. Subsequently, the HCE cells were washed with cold PBS (pH 7.4), collected, and lysed with modified radioimmunoprecipitation assay buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% NP-40, 0.25% sodium deoxycholate, 1 mM ethylene glycol tetraacetic acid [EGTA], 2 mM sodium orthovanadate, a protease inhibitor cocktail, and a phosphatase inhibitor cocktail). The HCE cell lysates were analyzed by performing SDS-PAGE and immunoblotting with rabbit anti-phospho-ERK polyclonal antibody (1:1,000) (Cell Signaling Technology, USA). The resulting blot was stripped and re-probed with rabbit anti-ERK polyclonal antibody (1:1,000) (Cell Signaling Technology). Data collecting processes were performed with an LAS-3000 luminescent image analyzer and Multi Gauge software (Fujifilm, Japan).

Proliferation of HCE cells was determined by using the XTT (2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide) incorporation assay, which measures DNA synthesis. HCE cells were seeded at 3 × 10³ cells per well into 96-well plates coated with a 0.1% gelatin solution. After 48 h, cells were washed with DMEM/F12 and incubated for 6 h with DMEM/F12 without FBS. The HCE cells were washed with fresh DMEM/F12 again and incubated with gintonin at the indicated concentrations. After 48 h, cell proliferation was assessed with 25 µL/well of XTT solution, according to the supplier's instructions (Sigma-Aldrich). Briefly, XTT reagent in DMEM/F12 was added to the cells and followed by an additional 4 h incubation. After incubation, optical densities were measured by an Opsys microplate reader (Thermo LabSystems, USA) at 450–490 nm, with a 630 nm reference filter.

A cell migration experiment was performed as previously described [16]. The chemotactic motility of HCE cells was measured by using modified Boyden chambers in a 48-well plate (Neuro Probe, USA). The polycarbonate membrane with 8 µm pores was coated with 0.1 mg/mL collagen for 30 min, followed by drying for more than 1 h. Gintonin in M199 (0.1% bovine serum albumin) was added to the lower wells of the chambers. The chambers were assembled with the membrane coated with collagen type I (BD Biosciences, USA). Cells (5 × 10⁴ cells/well) were added to the upper wells of the chambers and incubated for 70 to 80 min at 37℃. The HCE cells on the membrane were fixed and stained with Diff Quik (Sysmex, Japan) and mounted on slide glass. Non-migrated cells were removed with Kimwipes. The migrated HCE cells were counted in four fields under a light microscope at a magnification of 200×.

The rabbits were divided into three groups (four rabbits in each group) to evaluate corneal wound healing: ethanol exposure + control (saline 1 drop/eye), ethanol exposure + gintonin (200 µg/eye), and ethanol exposure + Solcorin (1 drop/eye) (Solcoseryl 120 concentrate; Hanlim Pharmaceuticals, Korea) treatment groups. To remove the corneal epithelium from each group, a method using ethanol and an Amoils epithelial scrubber (Innovative Excimer Solutions, Canada) was adopted [15]. Before the removal of the corneal epithelium, topical anesthetics, proparacaine (Alcaine; Alcon, Belgium) and antibiotic, ofloxacin (Effexin; Ildong Pharmaceutical, Korea) eye drops were applied. The eyes were prepared for surgery with 0.2% povidone-iodine solution to sterilize the ocular surface. Absolute ethanol (500–1000 µL/eye) was dropped into the cornea with a disposable syringe; thus, all corneal epithelia were de-epithelialized. The de-epithelialized epithelia were removed with an Amoils epithelial scrubber. Fluorescein (Fluorets; Chauvin, France) was used to visualize the area of epithelial denudement. An impregnated strip with fluorescein was applied to the conjunctiva overlying the dorsal sclera via the application of a drop of saline. Next, the corneal surface was flushed carefully with 10 mL of sterile saline to remove the remnant fluorescein stain. The stained area was photographically recorded by 8× portable slit-lamp biomicroscopy (Hawk Eye; Dioptrix, France) with manual retaining of the rabbit's eye under cobalt blue light.

For 48 h after operation, topical antibiotic eye drops (Effexin, Ildong Pharmaceutical) were applied to rabbits in all groups every 6 h. In addition to antibiotic eye drops, gintonin or Solcorin was also applied in the ethanol exposure + gintonin (200 µg/eye) or ethanol exposure + Solcorin (1 drop/eye) treatment groups, respectively, every 6 h. The ethanol exposure + control groups were only treated with saline (1 drop/eye) every 6 h. After 48 h, corneal re-epithelialization was evaluated by applying the fluorescein staining test after using local anesthetic eye drops (Alcaine; Alcon). After staining the cornea with fluorescein, the ocular surface was photographed by using 8× slit-lamp biomicroscopy under cobalt blue light. In the obtained images, the Image-Pro Plus program (ver. 4.5.0.29; Media Cybernetics, USA) was used to measure the area of fluorescein uptake in the cornea. The area of the healed corneal epithelium was determined by subtracting the area of postoperative fluorescein staining from the area of preoperative fluorescein staining.

The rabbits were divided into five groups (four rabbits in each group) for the evaluation of corneal wound healing: normal control, ethanol exposure + control (saline 1 drop/eye), ethanol exposure + gintonin (200 µg/eye), ethanol exposure + Solcorin (drop/eye), and gintonin (200 µg/eye) alone groups. Topical antibiotic eye drops (Effexin, Ildong Pharmaceutical) were applied to rabbits in all groups every 6 h for 48 h. In addition to antibiotic eye drops, gintonin or Solcorin was also applied to the ethanol exposure + gintonin (200 µg/eye), ethanol exposure + Solcorin (1 drop/eye) or gintonin (200 µg/eye) alone treatment groups, respectively, every 6 h. The normal control and ethanol exposure + control groups were only treated with saline (1 drop/eye) every 6 h. After 48 h, euthanasia was achieved via intravenous injection of KCl (Potassium Chloride 40 inj; DAI HAN Pharm, Korea) under propofol anesthesia (6 mg/kg intravenously; Provive 1%; Myungmoon Pharm, Korea). Corneas from individuals of each group were then removed, fixed with 4% paraformaldehyde solution, and embedded in paraffin. Five-micrometer-thick paraffin sections were prepared and stained by hematoxylin and eosin (H&E) as previously described [19]. To detect DNA fragmentation, an ApopTag Peroxidase In Situ Apoptosis Detection Kit (S7100; Millipore) was used to analyze the paraffin sections [14]. To assess the degree of cell proliferation in corneal wound healing, immunohistochemical analysis for proliferating cell nuclear antigen (PCNA) was performed on the paraffin sections by using mouse anti-PCNA antibody (1:2,000; Santa Cruz Biotechnology, USA) as previously described [3]. The H&E stained sections, DNA fragments, and immunohistochemistry results were analyzed under a light microscope.

The peak increase of cytosolic free Ca²⁺ at different concentrations of gintonin were plotted to obtain a concentrationresponse curve of the effects of gintonin. Origin software (OriginLab, USA) was used to fit the data to the Hill equation: y/y_max = [A]^nH/([A]^nH + [EC₅₀]^nH), where y is the peak at a given concentration of gintonin, y_max is the maximal peak in the absence of gintonin, EC₅₀ is the concentration of gintonin producing a half-maximal effect, [A] is the concentration of gintonin, and nH is the Hill coefficient. All values are presented as the mean ± SEM. Differences between control vehicle and treatment values were analyzed by using Student's t-test.

For the corneal wound healing study in rabbits, all results obtained are expressed as mean ± SD. Statistical analysis was undertaken by using ANOVA with Tukey's honestly significant difference test to evaluate differences in the area of healing between each groups. In each test, p < 0.05 was considered significant. Statistical tests were performed with SPSS (ver. 22; IBM, USA).

Initially, we examined the effects of gintonin on cytosolic free Ca²⁺ in HCE cells. Gintonin treatment induced [Ca²⁺]_i mobilization in HCE cells and the effects of gintonin occurred in reversible and concentration-dependent manners (panels A and B in Fig. 1). The EC₅₀ was 0.04 ± 0.01 µg/mL. Gintonin-mediated [Ca²⁺]_i mobilization began without a detectable lag and reached peak values within a few seconds; then, the extent of [Ca²⁺]_i mobilization decreased and approximately returned to its basal level. We observed that treatment of LPA C_18:1 to HCE cells also induced [Ca²⁺]_i mobilization, similar to that observed with gintonin (data not shown). The repeated treatment of HCE cells with gintonin did not induce desensitization to [Ca²⁺]_i mobilization (panel A in Fig. 1). These results show that both gintonin and LPA C_18:1 can induce [Ca²⁺]_i mobilization in HCE cells.

Because HCE cells express several subtypes of endogenous LPA receptors [2730], we examined the effects of gintonin on [Ca²⁺]_i mobilization in the absence or presence of the LPA1/3 receptor antagonist Ki16425 (10 µM). Treatment of Ki16425 with gintonin clearly blocked gintonin-mediated [Ca²⁺]_i mobilization (panel C in Fig. 1). The active phospholipase C inhibitor U-73122 (10 µM), inositol 1,4,5-trisphosphate receptor antagonist 2-APB (100 µM), and intracellular Ca²⁺ chelator (BAPTA-AM, 100 µM) also attenuated gintonin-mediated [Ca²⁺]_i mobilization in HCE cells (panel D in Fig. 1). These results show that gintonin, via activation of the LPA receptor-phospholipase C-intracellular IP₃ receptor-Ca²⁺ signaling transduction pathway, induces the mobilization of intracellular Ca²⁺ stores to increase [Ca²⁺]_i in HCE cells.

In addition, we examined whether gintonin treatment in HCE cells affects ERK1/2 phosphorylation. Gintonin treatment of HCE cells stimulated ERK1/2 phosphorylation in concentration- and time-dependent manners (panels A and C in Fig. 2). Gintonin-mediated ERK1/2 phosphorylation was saturated at 1 µg/mL gintonin (panel C in Fig. 2), and maximal ERK1/2 phosphorylation was observed after 5 min of gintonin treatment (panel B in Fig. 2). The effects of gintonin declined to approximately the basal level after 60 min (panel D in Fig. 2). These results indicate that gintonin-mediated mobilization of [Ca²⁺]_i is linked to rapid ERK1/2 phosphorylation in HCE cells.

Next, we examined the effects of gintonin on the proliferation and migration of HCE cells. Gintonin treatment of HCE cells for 48 h stimulated cell proliferation in a concentration-dependent manner (panel A in Fig. 3A) and induced cell proliferation in time-dependent manner with a treatment of 0.3 µg/mL gintonin (panel B in Fig. 3). The EC₅₀ of cell proliferation was 0.03 ± 0.01 µg/mL (panel A in Fig. 3). In addition, gintonin treatment of HCE cells induced cell migration in a concentration-dependent manner, but the presence of Ki16425 blocked gintonin-mediated cell migration (panel B in Fig. 4). These results indicate that gintonin stimulates HCE cell proliferation and migration through LPA receptors.

We then investigated the therapeutic effects of gintonin on corneal damage induced by ethanol in a rabbit model. Representative results of the fluorescein dye test at 48 h after operation are shown in Fig. 5 and Table 1 shows a summary of the eroded epithelium size. Compared to the ethanol-exposed control saline treatment (panel A in Fig. 5), gintonin treatment significantly increased the healing area by 50.14% ± 14.35% (panel B in Fig. 5). In addition, compared to the control saline treatment, Solcorin treatment also increased the healing area (panel C in Fig. 5 and Table 1). These results indicate that treatments with gintonin or Solcorin have therapeutic effects on corneal damage.

We investigated the histopathological effects of gintonin on ethanol-induced corneal damage in a rabbit model. Normal corneas are composed of corneal epithelium with stratified squamous epithelial cells (5–7 cell layers thick), the anterior limiting membrane (Bowman's membrane), stroma (approximately 90% of the cornea) with resident keratocytes, the posterior limiting membrane (Descemet's membrane), and the endothelium of the inner surface (Figs. 5 and 6). Two days after ethanol exposure, the epithelial cell nuclei disappeared or were markedly decreased in the damaged area of the cornea (panel B in Fig. 6); however, epithelial cells in regenerating, proliferating, and migrating stages were observed at the margin of the damaged area (arrows in panel B in Fig. 6). Additionally, infiltration of inflammatory cells was increased in the superficial stroma (panel B in Fig. 6). Interestingly, compared to the ethanol-exposed group, gintonin treatment in the form of eye drops (200 µg/eye) increased the number of epithelial cells in the superficial layer of the epithelium, as well as the epithelial thickness, but decreased the infiltration of inflammatory cells in the stroma.

Solcorin treatment showed effects similar to those of the gintonin treatment (panel D in Fig. 6). Three days after ethanol exposure, compared to that observed in the control group, the number of PCNA-immunoreactive cells decreased in the epithelium and stroma (panels F and G in Fig. 6). Moreover, compared to that observed in the ethanol-exposed group, gintonin treatment increased PCNA immunoreactivity; these results were similar to those observed in the Solcorin-treated group (panels H and I in Fig. 6). In addition, compared to the normal group, after 3 days, ethanol exposure increased the number of apoptotic cells in the epithelium and stroma (panels K and L in Fig. 6). Compared with that observed in the ethanol-exposed group, gintonin or Solcorin treatment decreased the number of apoptotic cells in the epithelium and stroma (panels M and N in Fig. 6). Treatment with gintonin alone (200 µg/eye) did not affect the histological structure of the cornea (panels E, J and O in Fig. 5). These results confirm that gintonin has therapeutic histochemical effects in a rabbit model of corneal damage.