Activin B signaling may promote the conversion of normal fibroblasts to scar fibroblasts.

This study is to explore the molecular mechanism of benign bile duct hypertrophic scar formation.Differential proteins between the normal fibroblast (NFB) and scar fibroblast (SCFB) were screened by protein chip assay, and analyzed by pathway-enrichment analysis and function-enrichment analysis. The differential proteins were further tested by ELISA. SiRNA-Act B was transfected to SCFB to down-regulate the expression of Act B. NFB was incubated with rh-Act B. The cell apoptosis and cell cycle were determined by flow cytometry. The expression of Act B, Smad2/3, transforming growth factor-β1 (TGF-β1), endothelin-1 (ET-1), thrombospondin-1 (Tsp-1), and Oncostatin M (OSM) were detected by Western blot.A total of 37 differential proteins were identified in SCFBs by microarray (P < .05), including 27 up-regulated proteins and 10 down-regulated proteins (P < .05). Their function were associated with Activin signaling, synthesis and degradation of extracellular matrix, formation and activation of cytokine, inflammatory reaction, immunoreaction, tissue damage reaction, cell cycle, migration, apoptosis, and secretion, etc. ELISA results showed that the expression of Act B, TGF-β1, ET-1 were higher in SCFBs, while the expression of Tsp-1 and OSM were lower in SCFBs (P < .05). After interfered by siRNA-Act B, the expression of Act B mRNA decreased (P < .05). The percentage of early apoptosis increased (P < .05). The expression of Act B, Smad2/3, TGF-β1 were decreased and Tsp-1, OSM were increased (P < .05). After treatment with rh-Act B, the percentage of G0/G1 phase of NFBs was decreased and that of S phase was increased without significance (P > .05). The expression of Act B, Smad2/3, TGF-β1 were increased (P < .05) and Tsp-1, OSM were decreased (P < .01).There are differentially expressed proteins between SCFBs and NFBs. Activin B signal plays an important role in the process of NFB transforming to SCFB, and TGF-β1, Smad2/3, Tsp-1, and OSM are important participants.

Introduction

Benign bile duct scar (BBS) is very common in biliary surgery. Due to the deep location of the bile duct, BBS is very difficult to be treated with either topical drug administration or surgery.[ Although in some patients, obstruction caused by BBS can be remove by surgery, there is still scar recurrence after surgery in some patients, not only causing pain to patients, but also great challenges to the biliary surgeons.[ Therefore, BBSs have attracted increasing attention by biliary surgeons.

The main pathological changes of BBSs are bile duct scar contracture and scar stenosis. Bile duct healing is accompanied by inflammation and macrophage accumulation, which continuously synthesize and secrete transforming growth factor-β1 (TGF-β1).[ TGF-β1 is a potent mitogen and scarring factor. It can stimulate fibroblasts to proliferate and to express α-actin (α-SMA), thus transforming into myofibroblast (MFB),[ which can secrete extensive extracellular matrix such as collagen, fibronectin, and proteoglycans.[ Inhibition of stromal collagen degradation and deposition of collagen will result in bile duct scar formation and contracture.[ Hyperplasia of scar tissue may cause the bile duct stenosis, unsmooth bile drainage, and cholestasis in liver.[ And, the toxic bile acids may damage liver cells, causing liver atrophy and hyperplasia, followed by cholestatic cirrhosis, portal hypertension, liver failure, and other severe pathological changes.[

Activin (Act) is a glycoprotein hormone firstly isolated from ovary, and belongs to the TGF-β superfamily.[ Act has three molecular subtypes of Act A, Act B, and Act AB, and their receptors are ActRIA, ActRIB, ActRII A, and ActRIIB.[ Acts exert their biological functions through the serine/threonine protein kinase (MAPKs) signaling pathways.[ Studies have found that the Act expression levels are abnormally elevated in mouse skin lesions, especially during granulation tissue formation and in hyperproliferative tissues. The transient overexpression of Act may play a positive role in the healing process of tissue and organ wounds.[ However, if the expression of Act is consistently high, it will lead to the occurrence of fibrosis of tissues and organs.[ Some researchers[ have found that overexpression of Act in the transgenic mice could promote skin damage repair, but the scar tissues were quickly developed at the lesions, forming skin scars.

Herein, we screened the differential proteins between the normal tissue-derived and scar tissue-derived fibroblasts by protein chip assay. The effects of Act B on the other differential proteins in fibroblasts were also investigated. The study will provide experimental evidence for the clinical research and treatment of bile duct hypertrophic scars.

Materials and methods

Sample collection

The bile duct scar tissues and the surrounding normal bile duct tissues were from 22 patients with intrahepatic bile duct stones (8 men and 14 women; age 28–45 years old). Informed consent was obtained and the study was approved by the ethics review board of Kunming Medical University.

Isolation and culture of primary fibroblasts

The tissues were placed in a petri dish, rinsed 3 times with PBS to remove the blood, bile, and stones, and soaked in the medium containing penicillin and streptomycin for 30 minutes. The tissues were cut into pieces of about 1 mm × 1 mm in size. Then the tissue pieces were attached to a culture flask. The moisture on the tissues was dried to make the tissue pieces better adhere to the tissue flask. When the water on the tissue pieces was dried, complete medium was carefully added to just overwhelm the tissue pieces. The culture flask was placed in an atmosphere containing 5% CO2 at 37 °C, and the medium was changed every 3 days. The isolated primary normal fibroblast (NFB) and scar fibroblast (SCFB) were cultured in complete medium at 37 and 5% CO2.

Protein chip assay

The NFB and SCFB cells were collected when the confluency was >80% and lysed in a lysing buffer containing protease inhibitors on ice for 30 minutes. The cell lysates were centrifuged at 14,000 rpm for 15 minutes at 4 °C. The supernatants were dialyzed to PBS (pH = 8.0) at 4 °C. The dialysis buffer was changed every 3 hours for three times. The protein concentration was determined by a bicinchoninic acid (BCA) method, and then the proteins were labeled with biotin.

RayBio AAH-BLG-1 Kit (RayBiotech, Norcross, GA) was used. Briefly, glass chips were equilibrated for 20 to 30 minutes and dried for 1 to 2 hours at room temperature. To each well of the chip, 400 μL blocking solution was added and incubated for 30 minutes. After removing the blocking solution, 400 μL sample solution was added into each well and incubated overnight at 4 °C. After that, the sample solution was discarded and washed with washing solution I (RayBiotech, Norcross, GA) 4 times (5 minutes each time) and washing solution II (RayBiotech, Norcross, GA) 3 times (5 minutes each time). The samples were stained with Cy3 (RayBiotech, Norcross, GA) and incubated for 2 hours at room temperature in the dark, and then washed again. The fluorescence of samples was detected by an Axon GenePix laser scanner system (RayBiotech, Norcross, GA). The images were quantitatively analyzed by the AAH-BLG-1 data analysis software (RayBiotech, Norcross, GA). The differentially expressed proteins were screened based on signal values more than 100 and Fold change more than 1.5 or less than 0.66, after removing the background by selecting Normalization 2.

Differential protein analysis

The GO annotations of the differential proteins were performed using online tool (http://www.geneontology.org). Using the DAVID Bioinformatics Resources (National Institutes of Health, MD, USA), the GO functional annotations and biological pathway analysis of the differential proteins were carried out, and the enrichment classes of the related functions of the differential proteins were obtained.

ELISA

The bile duct scar tissues and normal bile duct tissues were lysed with RIPA (containing protease inhibitors) on ice bath for 30 minutes, and then centrifuged at 14,000 rpm for 15 minutes at 4 °C. The supernatant was collected. The samples were added to a 96-well plate and incubated for 2.5 hours at room temperature. After that, biotin-labeled antibodies were added and incubated for 1 hour. Then, streptavidin was added and incubated for 45 minutes. Color development was conducted with tetramethylbenzidine (TMB). The absorbance at 450 nm was measured with a plate reader (Bio-Rad Laboratories Inc., Hercules, CA).

SiRNA transfection

SNFB cells were seeded into a 24-well plate at a concentration of 1 × 105/well, and incubated for 24 hours. Then, 1.25 μL of the siRNA of Act B (20 μM) (siRNA-Activin B kit, Guangzhou RiboBio Co., Ltd., Guangzhou, China) was diluted with 50 μL 1× riboFECTTMCP buffer (riboFECTTMCP transfection reagent, Guangzhou RiboBio Co., Ltd., Guangzhou, China) and incubated at room temperature for 5 minutes. Then, 5 μL riboFECTTMCP reagent (riboFECTTMCP transfection reagent, Guangzhou RiboBio Co., Ltd., Guangzhou, China) was added and mixed. The mixture was mixed with 443.75 μL cell culture media, added to the cells and incubated for 48 hours. The transfection efficiency was observed under an inverted fluorescence microscope (Olympus Co., Tokyo, Japan).

Reverse transcription PCR

The total RNA of cells was extracted with Trizol, and the concentration was determined by absorbance at 260 nm. The RNA was reverse transcribed into cDNA, and the target genes were amplified with a reverse transcription Polymerase Chain Reaction (RT-PCR) Reverse Transcription Kit (Thermo Fisher Scientific Inc., Waltham, MA) using the following condition: pre-denaturation at 94 °C for 5 minutes, followed by 35 cycles of denaturation at 94 °C for 1 minute, annealing at 53 °C for 1 minute and extension at 72 °C for 1 minute, and final extension at 72 °C for 10 minutes. The Polymerase Chain Reaction (PCR) product was then detected by agarose gel electrophoresis and visualized by a Gel Imaging System (Bio-Rad Laboratories, CA). The optical density of each band was measured by Image J software (National Institutes of Health, MD, USA).

Recombinant human Act B (rh-ActB) protein incubation with the normal fibroblasts

The NFB were seeded into a 6-well plate at 1 × 105 cells/well. After cell adhesion overnight, the medium was changed with fresh medium containing 10 ng/mL rh-Act B (PeproTech, Inc., Rocky Hill, NJ), and cultured for another 48 hours. The cells were collected for the detection of cell cycle and expression of differentially expressed proteins.

Flow cytometry

After 72 hours of siRNA Act B transfection, the cells were collected, re-suspended with pre-cooled PBS and fixed with pre-cooled 70% ethanol at 4 °C. After 24 hours, the cells were centrifuged at 1000 rpm for 5 minutes, and then re-suspended with pre-cooled PBS. The cells were stained with popidium iodide (PI) for 30 minutes at 37 °C in the dark. Flow cytometry was used to detect the cell cycle. The results were analyzed by ModFit software (Verity Software House, ME, USA) (Version 3.1).

For the detection of cell apoptosis, the cells were collected as described above and stained with Annexin V-FITC and PI before flow cytometry analysis.

Western blot

After 48 hours of transfection/overexpression, cells were collected and lysed with cell lysing buffer containing protease inhibitors on ice. The lysate was centrifuged at 14,000 rpm for 15 minutes at 4 °C, and the supernatant was collected. After determining the protein concentration, the sample was subjected to polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a Poly vinylidene fluoride (PVDF) membrane. The membrane blocked with 5% skimmed milk for 1 hour. Then, the membrane was incubated with primary antibodies (mouse anti-human) of Act B, endothelin-1 (ET-1), thrombospondin-1 (Tsp-1), TGF-β1, Oncostatin M (OSM), and Smad2/3 (Santa Cruz, CA) overnight at 4 °C. After washing, the membrane was incubated with the secondary antibody at room temperature for 1 hour. The membrane was finally colored with chemiluminescence reagent. The protein bands were visualized in a gel imager (Bio-Rad Laboratories Inc., Hercules, CA).

Statistical analysis

The statistical analysis was performed with the statistical software SPSS 17.0 (SPSS Inc., Chicago, IL). Measurement data are expressed as mean ± standard deviation (SD). For normally distributed data, t test was used for the comparison between 2 independent samples, and one-way analysis of variance (ANOVA) was used for multi-sample comparisons. For the non-normally distributed data, Wilcoxon rank sum test was used for the comparison of 2 samples, while Kruskal-Wallis rank sum test was used for the comparison of multiple samples. A P value <.05 was considered as statistically significant.

Results

Identification and analysis of differential proteins

Protein chip analysis of 507 proteins in NFB and SCFB, was performed. According to the criteria of signal value >100, fold change >1.5 or <0.66, a total of 37 differential proteins were identified. Compared with NFBs, there were 27 up-regulated proteins and 10 down-regulated proteins in SCFBs. Among them, levels of TGF-β1 and endothelin were increased in SCFB cells (Table 1), compared with NFB cells.

Table 1

Representative differential proteins by protein chip analysis.

In order to demonstrate the pathways and functions of the differential proteins, DAVID software was used to analyze these proteins. A total of 18 pathway enrichment (corrected P-value <.05) (Table 2) and 195 functional enrichment (correlated P-value <.05) (Table 3   ) were found.

Table 2

Enrichment analysis of different protein pathways.

Table 3

Enrichment analysis of different protein functions.

Pathway enrichment analysis (Table 2) revealed that the differentially expressed proteins were mainly involved in cytokine–cytokine receptor binding, JAK-STAT signaling pathway, chemokine and chemokine receptor binding, activin signaling pathway, collagen degradation signaling pathway, PI3K-Akt signaling pathway, TNF signaling pathways, TGF-β signaling pathways, insulin/insulin-like growth factor signaling pathways, and extracellular matrix degradation signaling pathway.

Functional enrichment analysis (Table 3   ) revealed that differential proteins were mainly involved in the synthesis and degradation of extracellular matrix, cytokine production and activation, inflammatory immune responses, tissue injury responses, cell cycle, cell proliferation, cell migration, cell viability, apoptosis, cell secretion, activin binding, and collagen synthesis and degradation.

The above results suggest that the differential-proteins-involved biological processes and molecular functions may be closely related to the occurrence and development of scars. In both the pathway enrichment and functional enrichment analysis, activin was involved.

Verification of the proteins related to the scar formation

To verify the identified proteins related to the scar formation, ELISA was performed. In general, the expressions of Act B, TGF-β1, ET-1, Tsp-1, and OSM were in consistent with the protein chip analysis. The levels of Act B (130.80 ± 58.46 pg/mL vs 88.83 ± 51.01 pg/mL) (Fig. 1A), TGF-β1 (10.31 ± 4.45 ng/mL vs 5.18 ± 2.68 ng/mL) (Fig. 1B) and ET-1 (107.63 ± 18.04 pg/mL vs 59.75 ± 12.49 pg/mL) (Fig. 1C) in the scar tissues were significantly higher than those of the normal tissues (P < .05). However, compared with normal tissues, the levels of Tsp-1 (672.42 ± 193.56 ng/mL vs 1311.47 ± 278.05 ng/mL) (Fig. 1D) and OSM (296.49 ± 72.28 pg/mL vs 485.52 ± 78.91 pg/mL) (Fig. 1E) in the scar tissues were significantly lower (P < .05). These results indicate that the changing trends of the identified proteins are consistent with the results of protein chip assay.

Figure 1

The expression of related proteins tested by ELISA. (A) Act B; (B) TGF-β1; (C) ET-1; (D) Tsp-1; (E) OSM. ∗P < .05, compared with the NFBs. ET-1 = endothelin-1; NFB = normal fibroblast; OSM = Oncostatin M, TGF-β1 = transforming growth factor-β1; Tsp-1 = thrombospondin-1.

The effect of downregulation of Act B on the cell apoptosis

To detect whether the transfection of siRNA-Act B was successful, cell fluorescence was observed. After the siRNA-Act B was transfected into SCFB, the transfected cells showed red fluorescence under a fluorescence microscope (Fig. 2A). Then, the mRNA level of Act B was determined by RT-PCR. As shown in Fig. 2B, Act B mRNA was significantly decreased in cells transfected with siRNA Act B than control (P < .01). This indicates that the siRNA-Act B was successfully transfected into the SCFBs.

Figure 2

Knockdown of Act B by siRNA-Act B. The SCFBs were transfected with siRNA-Act B to down-regulate the expression of Act B. (A) The SCFBs observed under a fluorescence microscope after siRNA-Act B transfection. Magnification: 100×. (B) The expression of Act B mRNA after siRNA-Act B transfection detected by RT-PCR, and GAPDH was used as internal standard. (C) The relative value of Act B mRNA in the siRNA Act B group and the Control group. ∗P < .05, compared with the Control group. GAPDH = glyceraldehyde-3-phosphate dehydrogenase, SCFB = scar fibroblast.

Cell apoptosis of SCFB was detected after the transfection of siRNA-Act B. The percentage of early apoptosis in the siRNA-Act B group was 15.72 ± 1.33%, which was significantly increased compared with that of the control group (0.32 ± 0.02%) (P < .01) (Fig. 3). The percentage of late apoptosis in the siRNA-Act B group was 0.96 ± 0.06%, which was also significantly higher than that of the control group (0.08 ± 0.02%) (P < .01) (Fig. 3). This demonstrates that interference of the expression of Act B in SCFB increases the cell apoptosis.

Figure 3

The effect of Act B interference on the apoptosis of SCFBs. (A) Cell apoptosis was detected by flow cytometry. (B) The percentage of cells at early and late apoptosis phases in the siRNA-Act B interfering group and the Control group. ∗P < .05, compared with the Control group. SCFB = scar fibroblast.

The effect of downregulation of Act B on the expressions of related proteins of SCFB

After transfection of siRNA-Act B, the expressions of Smad2/3, TGF-β1, ET-1, Tsp-1, and OSM in SCFB were detected by Western blot. For the siRNA-Act B group, the expressions of Act B, Smad2/3, TGF-β1, and ET-1 decreased, among which the reduction of Act B, Smad2/3, and TGF-β1 was significant compared with the normal group (P < .05) (Fig. 4). In addition, the expressions of Tsp-1 and OSM in the siRNA-Act B group increased compared with the normal group (P < .05) (Fig. 4). These results indicate that the interference of the expression of Act B in the scar fibroblasts can reverse the expression of related proteins.

Figure 4

The effect of Act B interference on the protein expression. Western blot was used to detect protein expression. (A) Representative Western blot results. (B) The gray values of the protein bands. ∗P < .05, compared with control group.

The effect of rh-Act B on the cell cycle of NFB

To determine the effect of Act B on cell cycle, flow cytometry was performed. After treatment with rh-Act B, the percentage of G0/G1 phase and G2/M phase of the NFBs decreased and the percentage in S phase increased in the ActB overexpression group, compared with control group, but the differences were not significant (P > .05) (Fig. 5). This indicates that overexpression of Act B in fibroblasts may not affect the proliferation.

Figure 5

The effect of Act B overexpression on the cell cycle. (A) Representative flow cytometry results of the control group and the Act B overexpression group. (B) The percentage of cells at different phases in the Act B over-expression group and the Control group.

The effect of rh-Act B on the related protein expressions of NFB

To determine the effect of rh-Act B expression on the protein levels of Act B, Smad2/3, TGF-β1, ET-1, Tsp-1, and OSM in NFB, Western blot was performed. After treated with rh-Act B, NFBs showed increased expressions of Act B, Smad2/3, TGF-β1, and ET-1 compared with the untreated NFBs, among which the differences in Act B, Smad2/3, and TGF-β1 were significant (P < .05) (Fig. 6). In addition, the levels of Tsp-1 and OSM were significantly lower than those of the control group (P < .01) (Fig. 6). This suggests that rh-Act B treatment of the normal fibroblasts can reverse the expression of related proteins.

Figure 6

The effect of Act B overexpression on the related protein expressions. (A) Protein bands detected by Western blot after the treatment with rh-Act B. (B) The gray values of the protein bands. ∗P < .05, compared with the Control group.

Discussion

Fibroblasts in normal skin and scar tissues had different sub-structures,[ and they are very active in scar tissues.[ Study[ has shown that the fibroblasts of the normal dermis and hypertrophic scars are heterogeneous, and these subpopulations are different in cell morphology, cell proliferation, collagen synthesis, growth factors, cytokine production, and involvement in inflammatory reactions. However, there are no specific protein markers for differentiating normal and scar fibroblasts, so they cannot be identified effectively.

Protein chip is a high-throughput rapid protein detection technology developed in recent years and has been used in basic research, clinical diagnosis, treatment, and prognosis of many diseases.[ This study used this technology to screen the differential proteins between normal bile duct fibroblasts and benign bile duct scar fibroblasts. Totally, 37 differential proteins were identified, including 27 up-regulated proteins and 10 down-regulated proteins. Their function were associated with signaling by Activin, synthesis and degradation of extracellular matrix, formation and activation of cytokine, inflammatory reaction, immunoreaction, tissue damage reaction, cell cycle, migration, apoptosis, and secretion, etc.

TGF-β1 is currently the most recognized and the most important scar factor.[ It can increase vascularization, induce migration of fibroblasts, monocytes, and macrophages to the site of injury and promote fibrosis at the injury site.[ It has also been reported that TGF-β1 plays an important role in the development of benign bile scar.[ In this study, we found that TGF-β1 was differentially expressed in scar fibroblasts. This was further confirmed by ELISA, which found elevated levels of TGF-β1. Our result further verifies that TGF-β1 is an important factor during scar formation.

It is found that fibroblasts can also produce ET-1, which is a potent mitogen for fibroblasts and stimulates the expression of proto-oncogenes such as c-fos and c-myc.[ ET-1 can act synergistically with a variety of cytokines and play an important role in the development of scars. For example, ET-1 cooperates with TGF-β1 to enhance fibrogenic effects.[ ET-1 can also cooperate with vascular endothelial growth factor (VEGF) to significantly increase the proliferation, migration, and invasion ability[ of endothelial cells, and induce neovascularization.[ In addition, ET-1 can cooperate with basic fibroblast growth factor to promote scar formation.[ It has also been reported that, the ET level in hypertrophic scar tissue gradually increases, reaches the peak in the proliferative phase, gradually decreases in the remission period, and returned to normal in the mature period, which is consistent with the neovascularization and collagen fiber distribution in the scar tissue at different stages.[ In this study, the results of the protein chip assay showed that the expression level of ET protein in SCFB cell samples was significantly higher than that of NFB cells, suggesting that the clinical samples obtained in this experiment are in the proliferative phase of bile duct scars.

Tsp-1 is one of the endogenous anti-angiogenic factors, which can inhibit the proliferation and induce the apoptosis of endothelial cells, and inhibit angiogenesis.[ VEGF is the most potent pro-angiogenic factor,[ while Tsp-1 is a newly discovered factor that strongly inhibits angiogenesis. The expression of VEGF and Tsp-1 is associated with pathological scar formation. VEGF may promote scar hyperplasia by inducing angiogenesis,[ whereas Tsp-1 inhibits VEGF elevation. The reduction of Tsp-1 will lead to the proliferation of blood vessels in pathological scars, resulting in the formation of pathological scars. It has been reported that[ the positive rate of Tsp-1 mRNA expression in skin hypertrophic scars and keloids was lower than that in normal skin tissues. Consistently, in the present study, Tsp-1 level was decreased in scar cells by protein chip assay and ELISA validation.

OSM belongs to interleukin-6 (IL-6) family,[ and can inhibit a variety of tumor cells, such as melanoma cell A375 and lung cancer cell H298.[ OSM is also involved in the pathophysiological process of liver regeneration.[ OSM inhibits normal fibroblast apoptosis, promotes normal fibroblast growth, and stimulates collagen secretion.[ Levy et al[ reported that OSM level was low in normal liver tissues but increased in liver cirrhosis. In our study, OSM level was also decreased in scar cells than control cells.

Act belongs to TGF-β superfamily,[ and can promote the differentiation of epithelial cells,[ promote the proliferation of fibroblasts, promote the synthesis of extracellular matrix, and promote the healing of skin wounds. Overexpression of Act causes inflammatory reactions and fibrosis of tissues and organs.[ Some scholars have found that[ Act over-expression in transgenic mice can promote skin damage repair, but at the same time it also induces formation of skin scars. Act B stimulates cell proliferation at the site of skin injury and promotes wound healing through the RhoA-Rock-JNK-cJun signaling pathway.[ In this study, protein chip assay showed that Act B was higher in fibroblasts than in normal fibroblasts. ELISA results also confirmed the results of protein chip assay. By transfecting siRNA-Act B into scar fibroblasts to down-regulate the expression of Act B in SCFBs, the apoptosis of SCFBs increased. The levels of TGF-β1, Smad2/3, and ET-1 decreased, while the levels of Tsp-1 and OSM increased. By contrast, after treatment with rh-Act B, the expression levels of TGF-β1, Smad2/3, and ET-1 were elevated, whereas those of Tsp-1 and OSM decreased. Thus, we suppose that Act B may regulate the expression of the above mentioned proteins, thus playing an important role in the conversion of normal bile duct fibroblasts to scar fibroblasts.

In conclusion, 37 differential proteins were identified from scar fibroblasts. Among them, the proteins associated with scar formation such as Activin B, TGF-β1, ET-1, Tsp-1, and OSM were identified and verified. Activin B signaling plays an important role in the conversion of normal fibroblasts to scar fibroblasts, and TGF-β1, Smad2/3, Tsp-1, and OSM are important participants. This study may provide evidence for the investigating the pathogenesis of benign biliary hypertrophic scar.

Author contributions

Conceived and designed the experiments: Shi-Kang Deng, Jian-Zhong Tang, Yan Jin and Ping-Hai Hu. Performed the experiments: Shi-Kang Deng, Jian-Zhong Tang, Yan Jin, Jun-Feng Wang and Ping-Hai Hu. Analyzed the data: Shi-Kang Deng, Jian-Zhong Tang and Ping-Hai Hu. Contributed reagents/materials/analysis tools: Shi-Kang Deng, Jian-Zhong Tang, Ping-Hai Hu, Jun-Feng Wang and Xiao-Wen Zhang. Wrote the paper: Shi-Kang Deng, Jian-Zhong Tang, Yan Jin, Xiao-Wen Zhang and Ping-Hai Hu. Access to full-text articles: Shi-Kang Deng, Jian-Zhong Tang and Ping-Hai Hu.

Table 3 (Continued)

Enrichment analysis of different protein functions.

Table 3 (Continued)

Enrichment analysis of different protein functions.

Table 3 (Continued)

Enrichment analysis of different protein functions.