Rg3 (purity, 98.6%) was purchased from Dalian Fusheng Pharmaceutical, Ltd. (Dalian, China). Rg3 was dissolved in dimethyl sulfoxide (dMSo) and filtered through a 0.22 μm bacterial filter. The mixture was then diluted with dulbecco's modified eagle's medium (dMeM; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) containing 10% fetal bovine serum (FBS; HyClone; GE Healthcare life Sciences, logan, UT, USA) to form the final concentrations (50 or 100 μg/ml). The final concentrations of Rg3 were selected according to previous studies and based on existing data regarding the effective dose (15,16). The final concentrations of DMSO in the culture medium were <0.1%.

A total of 15 Asian patients with keloids were recruited to the present study. Keloid scar specimens were obtained from these patients, which were aged between 22 and 35 years old, without systemic diseases. All patient information is provided in Table I. The lesions were diagnosed as keloids according to clinical appearance, symptoms, persistence for >1 year and extension beyond the original margins. All of the patients' keloids were in the active stage and none had undergone prior treatment. Prior to surgery, all patients were informed of the purpose and procedure of the present study and agreed to provide their resected lesion masses. The whole keloid tissues were completely removed, following administration of local anesthesia, from the skin of the neck, chest, abdomen and upper limb, according to the standard surgical procedures. Prior written informed consent was obtained from all participants and the present study was approved by the Ethics Committee of Shanghai Ninth People's Hospital affiliated Shanghai Jiao Tong University School of Medicine (Shanghai, China).

The adipose tissues and epidermis were removed from the samples using sterilized scissors; the remaining dermis was cut into 1×1 mm sections and digested in 0.25% collagenase for ~4 h at 37°C. Following centrifugation at 300 × g at room temperature for 5 min, the supernatant was discarded and the remaining precipitate, including abundant KFs and several pieces of dermis tissue was cultured in high glucose DMEM (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS, penicillin (100 U/ml) and streptomycin (100 μg/ml) (Sigma-Aldrich; Merck KGaA, darmstadt, Germany) in 100 mm dishes at 37°C in a humidified incubator containing 5% Co₂. The culture media were replaced every 3 days. The fibroblasts gradually attached to the dish or migrated out of the small pieces of dermis tissue within 7–10 days. KFs between passages 2 and 4 were used in the present study.

The present study analyzed Rg3-induced cell death by observing morphological alterations. KFs were seeded in 6-well plates at a density of 8×10⁴ cells/well and were cultured at 37°C in an atmosphere containing 5% Co₂. Medium with DMSO (<0.1%) and with various concentrations of Rg3 (50 or 100 μg/ml) was added to the grouped wells. Following a 72 h incubation at 37°C, KFs in the various groups were observed and images were captured under an inverted microscope (Nikon IX70; Nikon Corporation, Tokyo, Japan).

The viability of KFs treated with or without Rg3 was determined using the Cell Counting kit-8 (CCK-8) assay (dojindo Molecular Technologies, Inc., Kumamoto, Japan). The KFs were cultured in serum-free medium for 24 h for synchronization. Subsequently, the cells were incubated with DMEM containing 10% FBS for the following experiments. KFs were seeded in 96-well plates at a density of 1×10⁴ cells/ml and were incubated overnight for 24 h at 37°C in an atmosphere containing 5% CO₂. Medium with various Rg3 concentrations (50 or 100 μg/ml) was then added to the wells. After treatment for 1, 2, 3, 4 and 5 days, 10 μl CCK-8 was added to each well, and the cells were incubated at 37°C for 2.5 h according to the manufacturer's protocol. Aliquots (100 μl) of incubated medium were pipetted into a 96-well plate and colorimetric absorbance was recorded at 450 nm using a microplate reader (Thermo Labsystems, Helsinki, Finland).

The present study applied the FCM method to measure and analyze the rate of apoptosis according to the instructions provided by the Annexin V-FITC kit (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). The KFs were incubated in 6-well plates with medium containing various Rg3 concentrations (50 or 100 μg/ml). Following a 72 h incubation at 37°C, the KFs in each well were collected by centrifugation at 300 × g at room temperature for 5 min and washed twice with cold PBS. These cells were then resuspended in 500 μl binding buffer, and incubated with 10 μl Annexin V-FITC for 10 min at room temperature in the dark. Subsequently, 10 μl propidium iodide was added to the cells for 10 min at room temperature in the dark; the reaction was terminated by chilling in an ice-bath. Analysis of apoptotic rate was conducted using a flow cytometer (BD Biosciences, San Jose, CA, USA); >10,000 cells from each well were counted and the apoptotic percentage, as well as the percentage of necrotic cell death, was quantitatively analyzed using CellQuest software 5.1 (BD Biosciences).

KFs were added to a 10 cm culture dish at a density of 5×10⁴ cells/ml. After 48 h, the medium was changed and the cells were cultured for 72 h in fresh medium with or without various Rg3 concentrations (50 or 100 μg/ml). Subsequently, KFs were harvested and total RNA was extracted using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Reverse transcription was then performed; briefly, cdNA was synthesized from 2 μg total RNA using oligo(dT) and AMV reverse transcriptase (Promega Corporation, Madison, WI, USA). The reverse transcription was conducted according to manufacturer's protocol. RNA integrity and the success of reverse transcription were monitored by qPCR amplification of GAPdH transcripts. qPCR was conducted using the Power SYBR-Green PCR master mix (2X) (Applied Biosystems; Thermo Fisher Scientific, Inc.) on a real-time thermal cycler (Mx3000P™ qPCR system; Agilent Technologies, Inc., Santa Clara, CA, USA). The mixture was incubated at 95°C for 10 min, followed by 40 cycles (30 sec at 95°C, 30 sec at annealing temperature listed in Table II), and was finally incubated at 72°C for 5 min. The amplified products were normalized against the internal reference gene (GAPDH). GAPGH was amplified as an internal control, and relative gene expression analysis was performed using the 2^−ΔΔCq method (17). The primers for qPCR analysis are listed in Table II.

The scratch wound assay was used to evaluate the migration of KFs (18). Briefly, KFs (2×10⁵ cells/well) were plated into 6-well culture plates and were incubated until they reached ~100% confluence. A scratch wound was generated on the cell monolayer using a sterile 200 μl pipette tip, in order to form a cell-free 'wound' ~0.83±0.05 mm in width. The cell cultures were incubated with fresh medium containing various Rg3 concentrations (50 or 100 μg/ml). Digital images of each wound were captured under a Nikon Eclipse E200 microscope (Nikon Corporation) immediately (0 h), and 24 and 48 h after scratch generation. Cell migration was analyzed using the commercial software Image Pro-Plus version 6.0 (Media Cybernetics, Inc., Rockville, MD, USA). Data (means ± standard deviation, n=3) are expressed as the percentage of the scratched cell-free zone filled with KFs. For each sample, images were captured from three random views to obtain the mean value. The final mean percentages and standard deviation were determined from three KF samples.

KFs were grown in 6-well plates at a density of 2×10⁴ cells/ml. Following incubation for 24-36 h at 37°C in an atmosphere containing 5% CO₂, medium with or without various Rg3 concentrations (50 or 100 μg/ml) was added to the wells. After 72 h, KFs were fixed with 4% paraformaldehyde at 4°C overnight and permeabilized using 0.3% Triton X-100 at room temperature for 1 h. Nonspecific binding sites were blocked with normal goat serum (Sigma-Aldrich; Merck KGaA) at 37°C for 30 min. Subsequently, the KFs were incubated overnight at 4°C with primary rabbit anti-human antibodies against type Ⅰ collagen (1:500; ab34710), α-SMA (1:100; ab5694) and Ki-67 (1:1,000; ab15580) (Abcam, Cambridge, MA, USA). Subsequently, the cells were incubated with appropriate fluorescent goat anti-rabbit secondary antibodies (111-035-003; Jackson Immunoresearch Laboratories, Inc., West Grove, PA, USA) at room temperature for 1 h. DAPI was used to stain the nuclei prior to image acquisition. Images of the positive cells (green) and DAPI nuclear staining (blue) were captured using a fluorescence microscope (Olympus Corporation, Tokyo, Japan). The incorporation ratio of Ki-67-positive cells was determined using the following equation: Number of Ki-67-positive cells/number of total cells. The results were counted in five randomly selected fields.

Western blot analysis was performed as described previously (19), using primary antibodies specific for type I collagen, type III collagen, fibronectin, phosphorylated (p)-Smad2, p-Smad3, total-Smad2/3, Smad7, p-extracellular signal-regulated kinase (eRK)1/2 and total-eRK1/2 (all purchased from Cell Signaling Technology, Inc., Danvers, MA, USA). Briefly, tissues were collected, and total cellular protein was extracted in 100 μl RIPA lysis buffer (Beyotime, Shanghai, China) supplemented with 1 mM phenylmethylsulfonyl fluoride (PMSF) (Beyotime) and 50 μl/ml protease inhibitor cocktail (Sigma-Aldrich; Merck KGaA) at 4°C for 30 min. Protein concentration was measured using the bicinchoninic acid (BCA) method. Samples containing 20 μg protein (2 μg/μl) were boiled, subjected to SDS-PAGE on 10% Tris-Glycine gels, and then electrophoretically transferred to polyvinylidene fluoride membranes. The membranes were blocked with 5% fat-free milk for 1 h at room temperature, and the blots were incubated with appropriate primary antibodies [type I collagen (1:2,000; ab34710; Abcam); type III collagen (1:2,000; ab7778; Abcam); fibronectin (1:2,000; ab32419; Abcam); p-Smad3 (1:1,000; 8828; Cell Signaling Technology, Inc.); Smad2/3 (1:1,000; 8685; Cell Signaling Technology, Inc.); Smad7 (1:2,000; ab190987; Abcam); p-eRK1/2 (1:1,000; 4370; Cell Signaling Technology, Inc.); eRK1/2 (1:1,000; 4695; Cell Signaling Technology, Inc.)] overnight at 4°C. The membranes were then incubated with horseradish peroxidase-linked secondary antibodies (Beyotime) for 1 h at room temperature. The goat anti-rabbit secondary antibodies were purchased from Beyotime (cat. no. A0208) with a dilution of 1:1,000 for 1 h at room temperature. The blots were visualized using a Super-GL enhanced chemiluminescence reagent (Novland, Shanghai, China) and were exposed onto KodAK X-omat BT Film (Kodak, Rochester, NY, USA). The results were analyzed using a digital imaging system equipped with AlphaEaseFC software V (Alpha Imager 2000; ProteinSimple, San Jose, CA, USA).

Transwell invasion chambers (membrane pore size, 8 μm) coated with Matrigel (BD Biosciences) were placed into 24-well plates. Following an overnight culture in serum-free medium, KFs (1×10⁵ cells/well) were added to the upper chambers and were incubated with or without various Rg3 concentrations (50 or 100 μg/ml) for 24 and 48 h. Normal medium containing serum was placed into the lower chambers. After 24 or 48 h, cells that remained on the upper surface of the membrane were completely removed using a cotton swab. Cells that crossed the Matrigel and migrated to the lower side of the Transwell insert were fixed with 4% paraformaldehyde for 5 min at room temperature and stained with DAPI. The number of cells that invaded across the membrane was counted in five random fields under an Olympus CX40 fluorescence microscope (olympus Corporation).

Keloid tissues removed from the patients were cut into 1-mm (1×2×5 mm) tissue explants. Once the tissue sections adhered to the bottom of the dish, medium with or without various Rg3 concentrations (50 or 100 μg/ml) was added to the wells. Explants were added to 3.5-cm culture dishes. After 6 days, the tissue blocks were fixed with 4% paraformaldehyde for 24 h at 4°C and were embedded in paraffin. Immunohistochemical staining was performed using a peroxidase-labeled streptavidin-biotin technique. Briefly, tissues embedded in paraffin were cut into 4-μm sections and placed onto glass slides. After antigen retrieval, the sections were incubated overnight at 4°C with primary antibodies. Subsequently, each section was incubated with an appropriate secondary antibody and then detected by the formation of a streptavidin-biotin-horseradish peroxidase complex (Zhongshan-Jinqiao, Beijing, China). Immunostaining was considered positive when the cells were stained brown after the addition of 3% 3,3′-diaminobenzidine reagent. Sections stained by isotype-matched IgG instead of primary antibody were used as negative controls. Goat anti-rabbit antibody (SPN-9001) was purchased from Zhongshan-Jinqiao. Paraffin-embedded sections were used for histological staining. Slides were then incubated overnight at 4°C with primary antibodies against human type I collagen (1:200; ab34710), type III collagen (1:200; ab7778), CD31 (1:100; ab28364) and CD34 (1:100; ab81289) (Abcam). Subsequently, each section was incubated with the appropriate secondary antibody (Zhongshan-Jinqiao) and with DAB. All the tissue sections were observed under a microscope (Nikon Corporation) at a ×400 magnification and images of five random views were captured for each group. The relative density of collagen was analyzed using Image Pro Plus 6.0 software. The number of microvessels was counted in five random fields under a microscope (olympus Corporation).

All assays were performed in triplicate and all results are presented as the means ± standard deviation. The differences among the three groups were measured using one-way analysis of variance and differences between two groups were determined using Turkey's post-hoc statistical method. SPSS 21.0 software (IBM Corp., Armonk, NY, USA) was used for statistical analysis. P<0.05 was considered to indicate a statistically significant difference.

Since excessive and abnormal proliferation of fibroblasts has been reported in keloids, the present study investigated the regulatory function of Rg3 on the proliferation of KFs using a cell viability assay (CCK-8 assay). In the Rg3-treated groups, cell proliferation was inhibited from the second day (after 48 h of drug incubation) compared with in the control group (Fig. 1A). There were significant differences between the Rg3-treated groups and the 0 μg/ml-Rg3-treated group (P<0.05). In addition, an apparent difference in cell proliferation between the 50 and 100 μg/ml Rg3-treated groups was detected from the third day (after 72 h of drug incubation, P<0.05). There was also a marked difference detected between the DMSO and Rg3-treated groups (P<0.05); however, no obvious difference was observed between the 0 μg/ml-Rg3-treated and DMSO groups. No statistical differences were detected in cell proliferation rate among the groups on day 1. The present study also investigated the suppressive effects of Rg3 towards fibroblast growth using Ki-67 immu-nofluorescence in combination with dAPI nuclear staining (Fig. 1B and C). The expression of the proliferation marker Ki-67 was markedly decreased in Rg3-treated groups, as fewer Ki-67-positive KFs were detected compared with in the control group (Fig. 1C). The ratio of Ki-67-positive cells was significantly different between the untreated and Rg3-treated groups (P<0.05; Fig. 1B). The effects of various concentrations of Rg3 (50 or 100 μg/ml) on KF morphology were investigated by microscopy after 72 h. KF density was gradually decreased as the concentrations of Rg3 increased (P<0.01; Fig. 1D and E). These results indicated that KF proliferation was effectively inhibited by certain concentrations of Rg3.

The apoptotic rate was gradually increased as the concentration of Rg3 increased (Fig. 2A). However, statistical analysis demonstrated that a significant difference only existed between the control and 100 μg/ml-Rg3-treated groups, with regards to early apoptosis (P<0.05; Fig. 2B). With regards to late and total apoptosis, significant differences existed not only between the control and 100 μg/ml-Rg3-treated groups (P<0.01), but also between the two Rg3-treated groups. Notably, there was no significant difference between the control and 50 μg/ml-Rg3-treated groups in each phase of apoptosis. The majority of apoptotic KFs in each group were revealed to be undergoing late apoptosis.

The expression levels of types I and III collagen, fibronectin, α-SMA, connective tissue growth factor (CTGF), interferon (IFN)-γ and transforming growth factor (TGF)-β3 were detected in KFs using qPCR, western blot analysis, immunofluorescence and immunohistochemical analysis (Fig. 3). The mRNA expression levels of types I and III collagen, fibronectin, α-SMA and CTGF were significantly decreased in the Rg3-treated groups compared with in the control group (P<0.05), thus indicating that Rg3 may exert antifibrogenic effects in KFs (Fig. 3A). In addition, there were marked differences between the 50 μg/ml and 100 μg/ml-Rg3-treated groups (P<0.05) with regards to the mRNA expression levels of type I collagen, fibronectin and α-SMA; however, no significant difference was detected between these groups with regards to type III collagen and CTGF. Furthermore, the mRNA expression levels of IFN-γ and TGF-β3 were significantly increased in the Rg3-treated groups compared with in the control group (P<0.05), thus indicating that Rg3 may strengthen antifibrotic effects in KFs (Fig. 3A). Intracellular localization of type I collagen and α-SMA was examined by immunofluorescence microscopy using the corresponding antibodies. The expression of type I collagen and α-SMA was visibly weak in the Rg3-treated groups; the most obvious effects were detected in the 100 μg/ml-Rg3-treated group (Fig. 3B). Western blot analysis indicated that types I and III collagen, and fibronectin protein expression levels were decreased in KFs as the concentration of Rg3 increased (Fig. 3C). This tendency was particularly obvious with regards to type I collagen and fibronectin. The immunohistochemical analysis of types I and III collagen in keloid explant cultures indicated that Rg3 suppressed types I and III collagen synthesis compared with in the control group (Fig. 3D). In addition, a marked difference was detected among the three groups with regards to the relative density of collagen I (P<0.05; Fig. 3E). With regards to type III collagen, a statistical difference existed between the untreated group and the treated groups, but not between the two treated groups (Fig. 3E). These results suggested that Rg3 may downregulate the expression of profibrotic genes and proteins (types I and III collagen, fibronectin, α-SMA and CTGF) and upregulate the expression of antifibrotic genes (IFN-γ and TGF-β3) in KFs.

The results of a scratch wound assay indicated that Rg3 was able to markedly inhibit KF migration (Fig. 4A). After 24 h, KFs in the control group had migrated 55.43±6.03% (mean ± standard deviation) of the scratched area, whereas 50 μg/ml-treated KFs had migrated 44.08±7.66% and 100 μg/ml-treated KFs had migrated only 32.60±6.13%. There were significant differences between the 100 μg/ml-Rg3-treated and control groups (P<0.01), but not between the 50 μg/ml-Rg3-treated and control groups (Fig. 4B). After 48 h, KFs in the control group had migrated 89.46±4.12% of the scratched area, whereas 50 μg/ml-treated KFs had migrated 60.96±2.85% and 100 μg/ml-treated KFs had migrated only 49.15±5.79%. Significant differences were detected among the three groups (P<0.05; Fig. 4B). These results indicated that Rg3 may markedly suppress the migration of KFs.

The invasive capability of KFs was investigated using a Transwell invasion assay. The number of KFs that migrated across the Matrigel-coated polycarbonate membrane to the lower chambers was markedly increased in the untreated group compared with in the Rg3-treated groups at 24 and 48 h (Fig. 5A). Rg3 diminished the invasive capability of KFs by reducing the number of cells that migrated across the Matrigel-coated membrane in a concentration-dependent manner; a significant difference was detected among the three groups at both time-points (P<0.01; Fig. 5B). Furthermore, to validate the inhibitory effects of Rg3 towards the invasive capability of KFs, the mRNA expression levels of matrix metal-loproteinase (MMP)−1 and MMP-3, which serve a pivotal role in cell invasion during wound healing and cancer metastasis, were detected (20). The results indicated that the mRNA expression levels of MMP-1 and MMP-3 were significantly downregulated in Rg3-treated groups (P<0.01; Fig. 5C). These results revealed that Rg3 may suppress the invasive capability of KFs.

The number of CD31 and CD34 positively stained microvessels was decreased in Rg3-treated groups (Fig. 6A). The results indicated that microvessel density was reduced by ~1/2 in the 50 μg/ml-treated group and by ~2/3 in the 100 μg/ml-treated group compared with in the control group (Fig. 6B). Significant differences existed among the three groups (P<0.01) with regards to CD31⁺ and CD34⁺ microvessels.

The mRNA expression levels of TGF-β1, which has been reported to be highly expressed in KFs (21), VEGF, which is associated with malignant diseases (22), and plasminogen activator inhibitor-1 (PAI-1), which is strongly increased by TGF-β1, were significantly decreased in the Rg3-treated groups compared with in the control group (P<0.05; Fig. 7A). Whereas, the mRNA expression levels of Smad7, which is a negative feedback regulator in the TGF-β1/Smad pathway, was markedly increased in response to Rg3 compared with in the control group (P<0.01), thus indicating that TGF-β1-induced decreases in Smad7 expression were reversed by Rg3 in a concentration-dependent manner (Fig. 7A). Statistical analysis indicated that there were significant differences in the expression levels of all genes among the three groups (P<0.05). The protein expression levels of p-Smad2 and p-Smad3, which are enhanced by TGF-β1, were markedly decreased in Rg3-treated KFs (Fig. 7B). In addition, p-eRK1/2 expression was suppressed by Rg3 treatment in KFs (Fig. 7B). However, the protein expression levels of Smad7 were increased in the Rg3-treated groups compared with in the control group, which was similar to the findings of the qPCR analysis (Fig. 7A). The protein expression levels of total Smad2/3 and total eRK1/2 remained almost unchanged in the three groups.