Loureirin B Inhibits Hypertrophic Scar Formation via Inhibition of the TGF-β1-ERK/JNK Pathway Open Access

A hypertrophic scar is the pathological outcome of wound healing due to dermal injury [1]. Fibroblasts are an important type of effector cell activated during this process. During the wound-healing process, fibroblasts from the border of the wound migrate to the centre and transdifferentiate into myofibroblasts that abundantly synthesize extracellular matrix (ECM), leading to hypertrophic scar formation [1,2,3]. The mechanism of hypertrophic scar formation is still unclear. Current studies suggest that the abnormal expression of several cytokines is associated with hypertrophic scar formation. One of the most important cytokines associated with fibrotic disease and hypertrophic scarring is transforming growth factor β1 (TGF-β1) [1,4,5]. Through multiple cellular processes, TGF-β1 regulates tissue homeostasis, including cell proliferation, migration, apoptosis, and ECM remodeling [6]. During wound healing, increased TGF-β1 improves tissue regeneration, while a persistent increase in TGF-β1 activates several intracellular signals, such as the Sma- and Mad-related proteins (Smads) [7] as well as those of the mitogen-activated protein kinase (MAPK) pathway. The activation of these pathways promotes the transcription of fibrosis-related molecules [8] and stimulates autocrine release of TGF-β1, leading to a persistent autocrine-positive feedback loop that may result in the overproduction of matrix proteins and subsequent fibrosis [9,10]. Our previous study showed that Loureirin B (LB) can suppress hypertrophic scar formation, and we validated this effect in a rabbit ear scar model [11]. In addition, we found that the phosphorylation of Smad2 and Smad3 induced by TGF-β1 was suppressed by LB. Given that MAPKs are one group of important intracellular proteins that transduce extracellular signals from TGF-β1 [12,13], we investigated whether the inhibitory effect of LB on hypertrophic scarring is associated with the MAPK pathway. In the present study, TGF-β1-stimulated fibroblasts were used to study the response of the MAPK pathway to LB. We found that in TGF-β1-stimulated fibroblasts, the phosphorylation of extracellular signal-regulated protein kinase (p-ERK) and the phosphorylation of c-Jun N-terminal kinase (p-JNK) were suppressed by LB, while the phosphorylation of p38 MAPK kinase (p-p38) was not affected. The contractile capacity of fibroblasts, as well as ECM synthesis, was attenuated through the down regulation of p-ERK and p-JNK. These results suggest that the anti-fibrotic effect of LB is closely associated with inhibition of the ERK/JNK pathway.

All of the experimental procedures were conducted under protocol No: XJYYLL-2013190, which was reviewed and approved by the Institutional Ethical Committee of the Fourth Military Medical University.

Paired normal skin and hypertrophic scar tissue, which were used in our previous experiments [11], were collected from four patients who received no treatment before surgery. The age of the four patients ranges from 18 to 44 years old. Written consent was obtained from patients or their legal guardians. Dermal fibroblasts were isolated and cultured as described previously. Briefly, tissues were trimmed to remove excessive adipose and then rinsed with phosphate buffer solution (PBS) three times. Then, tissues were minced into small pieces and incubated in Dulbecco's Modified Eagle Medium (DMEM, Gibco, Grand Island, New York, USA) containing 0.1% collagenase type I (Sigma, St. Louise, Missouri, USA) at 37 °C for 3 hours. The isolated fibroblasts were then cultured in DMEM containing 10% foetal calf serum (Gibco), 1% penicillin and 1% streptomycin at 37 °C in a humidified atmosphere of 5% CO₂. Fibroblasts from the 3^rd to the 5^th passages were used in all experiments. Before any treatment, fibroblasts reaching 70∼80% confluence were incubated in serum-depleted medium for another 12 hours.

Several 60-mm dishes of normal skin fibroblasts were randomly arranged into different groups (n = 4). LB was obtained from the National Institute for the Control of Pharmaceutical and Biological Products of China the same as we have reported before, and reconstituted in DMSO at a final stock concentration of 20 mg/mL. Recombinant human TGF-β1 was purchased from PeproTech (London, UK) and dissolved in 10 mM citric acid (pH 3.0), yielding a final stock concentration of 10 ng/mL. Previous reports suggest that 5 ng/mL of TGF-β1 can significantly induce transdifferentiation of fibroblasts to myofibroblasts, and our previous experiments suggest that 25 μg/mL of LB can effectively improve scar formation in vivo. Thus, in the present experiment, TGF-β1 was diluted to 5 ng/mL, while LB was diluted to 25 μg/mL.

Forty micrograms of total protein was subjected to sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred onto a PVDF membrane (Millipore, Bedford, MA). After blocking with 5% non-fat milk, membranes were incubated at 4 °C overnight with a specific primary antibody, such as rabbit anti-human FN (1:1000, GeneTex, TX, USA), rabbit anti-human Col1α2 (1:1000, Abcam, Cambridge, UK), rabbit anti-human JNK (1:1000, Cell Signaling Technology, Beverly, MA), rabbit anti-human ERK1/2 (1:1000, Cell Signaling Technology), rabbit anti-human p38 (1:1000, Cell Signaling Technology), rabbit anti-human phospho-JNK (1:1000, Cell Signaling Technology), rabbit anti-human phospho-ERK1/2 (1:1000, Cell Signaling Technology), or rabbit anti-human phospho-p38 (1:1000, Cell Signaling Technology). The next day, membranes were washed three times with TBST (Tris-buffered saline with 0.1% Tween-20) and then incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG antibodies (1:3000, Santa Cruz, CA, USA) at 37 °C for 1 hour at room temperature. Antibodies against β-actin (1:1000, Abcam) were used as loading controls. The proteins were visualized with an ECL Kit (Millipore, USA) and Fluor ChemFC (Alpha Innotech, USA).

Scar tissues and autologous skin tissues were embedded in paraffin and cut into 4-μm-thick sections for immunohistochemical staining. Sections were subject to deparaffinization, dehydration and antigen retrieval. Sections were then separately incubated with primary antibodies such as rabbit anti-human FN antibody (1:200, GeneTex), rabbit anti-human Col1α2 antibody (1:200, Abcam) and rabbit anti-human TGF-β1 antibody (1:200, Gene Tex) at room temperature overnight. Next, the sections were incubated with biotinylated secondary antibody; then, streptavidin-biotin-horseradish peroxidase was used for signal amplification and diaminobenzidine (DAB) for staining, following the instructions of the SP-9001 Histostain TM Kit (ZSJQ, Beijing, China). The images were obtained and analysed using the Image-Pro Plus system 6.0.

The contractile capacity of fibroblasts was measured using a gel contraction assay. Fibroblast-embedded collagen gels were prepared, as described previously in the literature [14]. Briefly, four 24-well plates were pre-treated with 0.2% BSA for 1 hour. A 0.5-mL suspension containing 1 × 10⁶ fibroblasts and 1.4 mg/mL collagen were added into the wells. Then, fibroblasts in the wells were treated with TGF-β1 (5 ng/mL), LB (25 μg/mL), TGF-β1 (5 ng/mL) + LB (25 μg/mL), TGF-β1 (5 ng/mL) + SP600125/PD98059 (30 μmol/L), TGF-β1 (5 ng/mL) + LB (25 μg/mL) + SP600125/PD98059 (30 μmol/L) or dimethyl sulfoxide (DMSO, control). The gel solutions were added to 24-well plates (600 μL per well) and incubated at 37 °C for 24 hours for polymerization, followed by mechanical detachment from the sides of the wells. The images of gels were captured at 0, 24 and 48 hours after the gels were released, and the images were analysed using Image Pro Plus 6.0 software.

Results were presented as the mean ± SEM. Data were analysed for significance by analysis of variance (ANOVA) using SPSS 17.0 software (Chicago, USA). p<0.05 was considered statistically significant.

Expression of ECM is higher in hypertrophic scar tissue than in normal skin [15]. During this experiment, we detected the expression of Col1, FN and TGF-β1 by immunohistochemistry. The results showed that expression of all three proteins increased in hypertrophic scar tissue compared with autologous skin (n = 4), which suggested that the samples we selected were consistent with the pathological standard of hypertrophic scar tissue (p<0.05).

Given that the MAPK pathway is one of the most important pathways in fibrosis and TGF-β1 is the most important cytokine involved in hypertrophic scar formation, we examined whether TGF-β1 stimulation could separately increase phosphorylation of ERK, JNK and p38 in normal-skin-derived fibroblasts. As shown in Fig. 2, thirty minutes after stimulation with TGF-β1, the expression of p-JNK was up regulated. Next, we stimulated the fibroblasts with TGF-β1, LB, both or none. The results show that LB suppressed the up regulation of p-JNK, Col1 and FN that was induced by TGF-β1. During this process, the expression of Col1 and FN was also reduced. The effect of SP600125, an inhibitor of JNK, was similar to that of LB (n = 4) (p < 0.05).

As shown in Fig. 3, similar to the results for the JNK pathway, 5 minutes after stimulation with TGF-β1, the expression of p-ERK was up regulated. LB suppressed the up regulation of p-ERK that was induced by TGF-β1. During this process, the expression of Col1 and FN was also reduced. PD98059, an inhibitor of ERK, had a similar effect as that of LB (n = 4) (p < 0.05).

As shown in Fig. 4, similar to the results of the ERK pathway, 5 minutes after stimulation with TGF-β1, the expression of p-p38 was up regulated. However, LB did not suppress the up regulation of p-p38 that was induced by TGF-β1, which indicated that the p38 MAPK pathway participates in TGF-β1-stimulated fibrosis, while there seems to be no relationship between the anti-fibrosis effect of LB and the p38 MAPK pathway (n = 4) (p < 0.05).

Given that the hypertrophic scar tissue shows increased contraction compared to normal skin, it is important to know whether LB affects the contraction capacity of TGF-β1-stimulated fibroblasts as well as to understand the mechanism of such an effect. Fibroblasts were seeded in a gel and stimulated by permutations of TGF-β1, LB and subsequently, an inhibitor of ERK/JNK. Gel size was observed at 0, 24 and 48 hours after stimulation. Fibroblasts contracted the surrounding matrix, leading to the contraction of the gel. The results clearly demonstrate that LB attenuated the contraction of fibroblasts that was induced by TGF-β1 (n = 4). Based on our previous investigation, we hypothesized that the inhibition of p-ERK and p-JNK suppresses contraction in TGF-β1-stimulated fibroblasts. Thus, we added an inhibitor of ERK and JNK to the gel. Similarly to LB, these two inhibitors suppressed contraction in TGF-β1-stimulated fibroblasts (n = 4).

To further investigate the role of ERK and JNK in LB-mediated inhibition of scar formation, we cultured human hypertrophic scar tissue ex vivo [16]. The tissues were incubated in medium containing LB solution. As shown in Fig. 6, the detection of p-ERK and p-JNK protein levels by western blotting showed that both of these two proteins, as well as FN and Col1, were downregulated in hypertrophic scar tissue after LB stimulation (n = 4).

We demonstrated that LB down regulated expression of p-JNK and p-ERK in TGF-β1-stimulated fibroblasts, which can then inhibit ECM synthesis. Recent studies suggest that compared with normal skin tissue, fibroblasts in hypertrophic scar tissue tend to transdifferentiate into myofibroblasts [17], which results in an increase in myofibroblast contractions and synthesis of ECM. The major components of ECM in hypertrophic scar tissue include Col1, collagen 3, FN and proteoglycan. FN participates in cell-to-cell adhesion, cell to ECM adhesion, cell migration and differentiation. FN, which is important for the maintenance of cell structure, also has some affinity to collagen [18]. In the present study, we used FN and Col1 expression to investigate the ECM. Cytokines play an important role in hypertrophic scar formation, and TGF-β1 is the most important of these cytokines. A TGF-β1-mediated signalling pathway is believed to be closely associated with scar formation. TGF-β1 binds to a receptor, which then facilitates cell transdifferentiation and ECM deposition. TGF-β1-stimulated fibroblasts were used here to simulate the process of scar formation.

During this period, the expression of Col1 and FN significantly increased. However, LB inhibited the expression of these two molecules. In a previous study, we confirmed that LB improved scar appearance and reduce collagen synthesis in a rabbit ear scar model. In this study, we collected hypertrophic scar samples from the adults who were included in the previous study. These cases were established using the clinical standards of hypertrophic scar diagnoses. Then, we examined the expression of FN and Col1 in scar tissues using immunohistology. As shown in Fig. 1, the expression of TGF-β1, FN and Col1 increased in hypertrophic scar tissue. The results further confirmed the diagnosis of the samples and laid a foundation for further experiments.

To clarify the possible mechanism of LB-mediated inhibition of hypertrophic scar formation, we examined the phosphorylation of Smad2 and Smad3 in TGF-β1-stimulated fibroblasts before and after LB stimulation. We found that the phosphorylation of Smad2 and Smad3 was inhibited by LB. In addition to the Smad pathway, the MAPK pathway also plays an important role in hypertrophic scar formation and fibrosis [19,20,21,22]. MAPK is a serine/threonine kinase composed of extracellular signal regulated kinases (ERK), c-Jun N-terminal kinase (JNK) and p38 MAP kinase (p38 MAPK), which can be activated by TGF-β1 and participates in cell proliferation, differentiation and apoptosis. MAPKs can be activated in liver fibrosis, increasing Col1α2 transcription [23,24]. The MAPK pathway also plays a key role in myocardial fibrosis, lung fibrosis and renal fibrosis [25,26,27,28]. In normal skin, exogenous TGF-β1 can activate p38, which can combine with the Col1α2 promotor and increase the transcription of Col1α2. JNK is a key factor in TGF-β1-stimulated lung fibrosis [29]. Inhibition of JNK, ERK and p38 can partially reverse TGF-β1-induced epithelial-to-mesenchymal transition (EMT) and fibrosis [5,30]. We are interested in knowing whether MAPK plays a role in the inhibitory effects of LB on scar formation. However, we should first clarify whether the phosphorylation of MAPKs is increased in TGF-β1-stimulated fibroblasts. As shown in the results, TGF-β1 stimulation elevated the expression levels of p-ERK, p-JNK and p-p38. Next, we investigated whether MAPK participates in the LB-mediated inhibition of scar formation. The experiments showed that LB significantly inhibited the TGF-β1-mediated induction of Col1 and FN expression. During this period, the expression of p-ERK and p-JNK decreased while p-p38 did not change, which indicates that ERK and JNK responded to LB stimulation.

To further verify our results, TGF-β1-stimulated fibroblasts were separately co-cultured with an inhibitor of ERK and JNK. As a result, the expression of both Col1 and FN was significantly reduced. It is clear that ERK and JNK are associated with TGF-β1-stimulated ECM expression in fibroblasts. Overall, our cellular experiments demonstrate that LB inhibited the expression of ECM in fibroblasts through the ERK and JNK pathway.

Previous research suggests that enhanced contractile capacity is related to a high level of α-SMA expression in myofibroblasts [8,30]. The contractile capacity of TGF-β1-stimulated fibroblasts was significantly increased compared to that of normal fibroblasts [31], which is one of the most important characteristics of myofibroblasts in hypertrophic scar tissue [32,33]. We used a gel contraction assay to examine the contractile capacity of fibroblasts after exposure to LB. As shown in Fig. 5, LB, as well as an inhibitor of JNK and ERK, significantly inhibited the fibroblast contraction that was induced by TGF-β1. All of these results suggest that LB inhibits hypertrophic scar formation through the inhibition of ERK and JNK phosphorylation.

To further clarify the response of ERK and JNK to LB stimulation in hypertrophic scar formation, we designed experiments at the tissue level. There are several animal models used for scar research, including a heterologous scar transplantation model, a chemically induced scar model and animal models with scarring in specific locations [34,35,36,37]. All of these models do not completely simulate hypertrophic scar formation in humans. Yasuoka et al reported that tissue culture better mimics in vivo environments [16]. Scar tissues were cultured in medium that included LB. The protein expression levels of p-ERK and p-JNK were observed using western blotting. The results demonstrated that compared with the control group, LB significantly down regulates the expression of p-ERK and p-JNK as well as the expression of Col1 and FN.

Overall, we have shown that in both TGF-β1-stimulated fibroblasts and scar tissue, the stimulation of LB suppressed phosphorylation of ERK and JNK (Fig. 7). Similarly, the inhibitors of either ERK or JNK down regulated the expression of ECM and inhibited cell contraction. These results suggest that LB can inhibit scar formation though the ERK and JNK pathway.

The National Natural Science Foundation of China supported this study (Grant Number: 81171811, 81372069).

The authors state no conflict of interest.

T. He, X. Bai and L. Yang contributed equally to this work.