Author Correction: ENMD-1068 inhibits liver fibrosis through attenuation of TGF-β1/Smad2/3 signaling in mice

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

SCientifiC REPORts | 7:5498 | DOI: 10.1038/s41598-017-05190-7 There is an overwhelming amount of data proving that PAR-2 plays an important role in liver fibrosis. PAR-2 antagonists have recently been developed and may represent a novel therapeutic approach in preventing fibrosis in patients with chronic liver disease. N1-3-methylbutyryl-N4-6-aminohexanoyl-piperazine (ENMD-1068) is a novel selective antagonist of PAR-2 and can prevent PAR-2 activation from genomic intervention 10,11 . In the present study, a carbon tetrachloride (CCl 4 ) induced mouse model of liver fibrosis was used to assess the effect of ENMD-1068 on the development of liver fibrosis. The effect of ENMD-1068 treatment in the mouse model was evaluated by collagen production, α-SMA expression, and alanine aminotransferase/aspartate aminotransferase (ALT/AST) level. In addition, HSCs isolated from mice livers were used to analyze the effect of ENMD-1068 treatment in the activation and collagen release of HSCs.

Mouse models. The study protocol was approved by the Animal Experiments and Experimental Animal
Welfare Committee of Capital Medical University (Permit Number: AEEI-2016-150). ICR mice used in this study were purchased from Beijing Vital Laboratory Animal Technology Co. (Beijing, China).
Mouse model of liver fibrosis was induced by injection of CCl 4 . Eight-week-old adult ICR mice received intraperitoneal injections of 1 μL/g body weight of a CCl 4 /olive oil (OO) mixture, 1:9 v/v, twice per week. Control mice were treated with 1 μL/g body weight of OO. Mice were sacrificed at 4 weeks after CCl 4 treatment, on the day after the last injection. 25 mg/kg or 50 mg/kg ENMD-1068 (Abcam, Shanghai, China) with 200 μL of the vehicle were administered intraperitoneally 15 mins before CCl 4 treatment twice per week for 4 weeks (N = 6 per group). As shown in Figs 1, 4 weeks after CCl 4 treatment, the animals were sacrificed by CO 2 exposure and liver tissues were collected for Sirius Red staining, immunohistochemistry staining, RT-PCT and Western blotting. Serum was isolated for analyzing the serum levels of ALT and AST. All animal work was performed under the guidelines of the Ethics Committee of Capital Medical University.
Isolation and culture of primary mouse HSCs. Primary mouse HSCs were isolated from adult male ICR mice by collagenase perfusion and purified by density gradient in Nicodenz (AXIS-SHIELD PoC, Scotland). All cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS). To evaluate the role of PAR-2 antagonist on HSCs activation and collagen induction, HSCs were stimulated with TGF-β1 (5 ng/mL, Abcam, Shanghai, China) for 24 h in the presence or absence of the selective inhibitor of PAR-2 (ENMD-1068, 10 mM). HSCs were serum starved for 24 h before inhibitor incubation. Serum biochemical parameters. On day 28 after CCl 4 injection, serum was collected by orbital venous plexus blood collection. The blood serum was separated, and the levels of AST and ALT were measured by standard enzymatic assay kits. Each assay is a colorimetric assay with detection of a highly-colored end product measured at 490-520 nm by a spectrophotometer (Hitachi 736-10, Beijing, China). The absorbance of each end product is proportional to the enzyme's activity.
Quantitative analysis of liver fibrosis. Liver sections were stained with Sirius Red for collagen visualization. In brief, liver tissues were fixed in PBS containing 4% paraformaldehyde for 24 h and embedded in paraffin. Four micron-thick sections from paraffin-embedded liver tissue were deparaffinized and stained with Sirius Red solution for 1 h. The fibrotic area was assessed by computer-assisted image analysis with Leica Qwin V3 software (Lecia Microsystems, Heidelberg, Germany). The mean value of 15 randomly selected areas per sample was used to estimate the percentage of fibrotic area.  Measurement of calcium mobilization. Primary mouse HSCs were washed and incubated in the dark with 3 mM Oregon green BAPTA-1 AM(Thermo Fisher, Beijing, China) in the presence of 1% pluronic acid for 45 min at 37 °C. The excess of fluorochrome was removed by two washing steps. HSCs were then resuspended at densities of 10 4 cells per well in DMEM medium and incubated with 10 mM ENMD-1068 or with an equivalent volume of dimethylsulfoxide (DMSO) before inducing internal calcium mobilization with 0.2 U/mL trypsin, 10 μΜ SLIGRL-NH 2 or 10 μΜ TFLLR-NH 2 (PAR1 agonist). Changes in fluorescence were recorded as previously described 12 . Transient transfections and reporter gene assays. Primary mouse HSCs received Lipofectamine 2000 (Life Technologies) with TGF-β1/Smad responsive reporter genes REPO TM SMAD (Genomeditech Co., shanghai, China) as previous described 13 . Transfection efficiency was normalized by cotransfection of Renilla luciferase reporter plasmid pRL-TK (Promega, Madison, WI). 24 h after the transfection, cells were treated with TGF-β1(5 ng/mL) or ENMD-1068 (10 mM) for 24 h. The data were derived from three wells processed in parallel and normalised with Renilla luciferase activity.

Statistical analysis.
Data are presented as mean ± standard deviation (SD). The data between groups were analyzed for statistical differences using SPSS 17.0 statistical software (SPSS Institute, USA) and one-way analysis of variance tests plus subsequent Bonferroni post hoc test. The p-value was two-tailed and considered as statistically significant or highly significant if it was less than 0.05 or 0.01, respectively.

Results
Biochemical parameters. After CCl 4 injection for 4 weeks, there was no mortality in any of the experimental groups. However, all the CCl 4 treated mice showed progressive listlessness, reduced movement and increasing abdominal girth. Liver injury was assessed by determining the serum levels of liver enzymes including ALT and AST. As shown in Fig. 2, the serum level of ALT and AST were markedly increased in the CCl 4treated mice compared with the control mice (270.80 ± 35.6 vs 65.40 ± 9.76 P < 0.01 for ALT; 320.80 ± 45.60 vs 75.30 ± 12.27 P < 0.01 for AST). The increases were suppressed in both ENMD-1068 25 mg/kg (160.34. ± 26.36 vs 270.80 ± 35.6 P < 0.05 for ALT; 165.34 ± 50.30 vs 320.80 ± 45.60 P < 0.05 for AST) and ENMD-1068 50 mg/ kg (120.90 ± 20.16 vs 270.80 ± 35.6 P < 0.05 for ALT; 143.25 ± 25.20 vs 320.80 ± 45.60 P < 0.05 for AST) treatment groups. There was no significant difference between ENMD-1068 50 mg/kg treatment and ENMD-1068 25 mg/kg treatment (P > 0.05). These results showed that ENMD-1068 treatment can alleviate liver injury induced by CCl 4 .

Inhibition of liver fibrosis by ENMD-1068.
To evaluate the effect of ENMD-1068 on liver fibrosis, animals with intraperitoneal injection of CCl 4 that had been allowed to establish liver fibrosis were treated with one of two doses of ENMD-1068 or with saline solution as a negative control. Four weeks later, liver tissues were collected for paraffin sectioning. As shown in Fig. 3, morphometric analysis of Sirius Red staining revealed that collagen was increased significantly in CCl 4 group compared with the control group (4.13% ± 0.51% vs 0, P < 0.05). Collagen deposition was markedly attenuated in the high dose (50 mg/kg) ENMD-1068 group compared with the saline group (0.77% ± 0.16% vs 4.13% ± 0.51%, P < 0.01). Similar results were obtained after administration of the low dose (25 mg/kg) ENMD-1068 in CCl 4 -treated mice (0.92% ± 0.15% vs 4.13% ± 0.51%, P < 0.01). There was no significant difference between ENMD-1068 50 mg/kg treatment and ENMD-1068 25 mg/kg treatment (P > 0.05). These results demonstrated that liver fibrosis was attenuated by ENMD-1068 treatment.
To evaluate the effect of ENMD-1068 on HSC activation, we analyzed the mRNA levels of fibrotic marker in liver tissues, including α-SMA, Col α1(І) and Col α1(III), by real-time qPCR after ENMD-1068 administration for 4 weeks. We found that CCl 4 treatment increased significantly the mRNA levels of α-SMA, Col α1(І) and Col α1(III) compared to the control mice, while ENMD-1068 administration resulted in the reverse effect (Fig. 4).

Effect of ENMD-1068 on α-SMA, p-Smad2 and p-Smad3 expression. To assess whether ENMD-
1068 is effective in inhibiting activation of HSCs in vivo, α-SMA, p-Smad2 and p-Smad3 immunohistochemistry were performed with all liver sections. There were minimal α-SMA, p-Smad2 and p-Smad3 immunostaining detected in the control mice (Fig. 5A1-3). By contrast, immunohistochemical staining for α-SMA, p-Smad2 and p-Smad3 were positive in the CCl 4 group. Very strong α-SMA, p-Smad2 and p-Smad3 staining were observed mainly in the area of collagen production (Fig. 5B1-3), which indicates that the increased expression of α-SMA, p-Smad2 and p-Smad3 relate to collagen production. As shown in Fig. 5E, the content of α-SMA, p-Smad2 and p-Smad3 positives in the CCl 4 group were significantly higher than that in the control and ENMD-1068 group (P < 0.01). Compared with the control vehicle treatment, treatment with ENMD-1068 25 mg/kg and ENMD-1068 week. In ENMD-treated group, mice were administered intraperitoneally 25 mg/kg or 50 mg/kg ENMD-1068 15 mins before CCl 4 treatment twice per week for 4 weeks. All animals were sacrificed at 4 weeks after CCl 4 treatment and liver tissues were co llected to investigate the effect of ENMD-1068 on the mRNA expression of fibrotic marker. The mRNA expression of fibrotic markers was quantified using real-time RT-PCR. 18 S rRNA was used for normalization of PCR data for the genes of α-SMA, Colα1(I) and Colα1(III). N = 6, bars represent SD, * P < 0.05 vs Control, # P < 0.05 vs CCl 4 . 50 mg/kg significantly decreased the expression of α-SMA, p-Smad2 and p-Smad3 (Fig. 5C1-3 and Fig. 5D1-3, P < 0.05). There was no significant difference between ENMD-1068 50 mg/kg treatment and ENMD-1068 25 mg/ kg treatment (P > 0.05).

ENMD-1068 blocks TGF-β1/Smad signaling in primary mouse
HSCs. TGF-β1/Smad signal pathway plays a crucial role in HSCs activation and collagen production. To explore the role of PAR-2 in regulating TGF-β1/Smad signal, HSCs were stimulated with TGF-β1 (5 ng/mL) for 24 h in the presence or absence of the selective inhibitor of PAR-2 (ENMD-1068, 10 mM). As shown in Fig. 7A and B, α-SMA, Col α1(І),Col α1(III), and Smad2/3 C-terminal phosphorylation expression levels were increased significantly following stimulation with TGF-β1 compared with the no treatment HSCs (P < 0.05). The increases were suppressed in ENMD-1068 treated group compared with the TGF-β1 treated only group (P < 0.05). Luciferase assays showed that ENMD-1068 caused significant decrease in luciferase activity compared with TGF-β1 treated HSCs indicating that ENMD-1068 inhibited the Smad transcriptional activity in HSCs (P < 0.05, Fig. 7C). The colocalization of α-SMA and p-Smad2 in HSCs was examined by immunofluorescence imaging with confocal microscopy. As shown in Fig. 7D, α-SMA and p-Smad2 were observed predominantly in the cytoplasm. Furthermore, α-SMA exhibited excellent colocalization with p-Smad2. Colocalization of α-SMA and p-Smad2 were markedly increased after TGF-β1 stimulation for 24 h, which were reduced in ENMD-1068 treated HSCs. These results showed that ENMD-1068 reduced HSCs activation and Col α1(І) and Col α1(III) production through the inhibition of TGF-β1/Smad signal transduction.

Discussion
Liver fibrosis is a common response to chronic liver injury, ultimately leading to cirrhosis and its complications including portal hypertension, liver failure, and hepatocellular carcinoma. Despite the relatively large population of patients suffering from hepatic fibrosis and cirrhosis, no efficient and well-tolerated drugs are available for the treatment of this disorder 14 . Increasing evidence suggests that PAR-2 plays an important role in HSCs activation and liver fibrosis, and PAR-2 antagonists may represent a novel therapeutic approach in preventing fibrosis in patients with chronic liver disease 4, 7-9, 15 . We aimed to evaluate the efficacy of PAR-2 antagonists in suppressing HSCs activation and progression of liver fibrosis. Our results confirmed that ENMD-1068, a protease-activated receptor 2 antagonist, contributes to a reduction in hepatic collagen content and histological fibrosis accompanied by decreased HSCs activation, as demonstrated by the reduced expression of α-SMA. Furthermore, we provide evidence in vitro that inhibition of PAR-2 reduced TGF-β1 induced phosphorylate Smad2 and expression of α-SMA and collagen.
Testing the efficacy of novel therapeutics in an in vivo animal model is a crucial step in a preclinical efficacy evaluation. According to the most widely accepted mechanism for liver fibrosis, HSCs are the primary cell type in the liver responsible for excess collagen synthesis during hepatic fibrosis [16][17][18] . Following liver injury, HSCs undergo a complex transformation or activation process, changing from quiescent, vitamin A-storing cells to activated, myofibroblast-like cells. A large amount of extracellular matrix is then produced by fibroblasts and leads to liver fibrosis. The appearance of the cytoskeletal protein α-SMA is a marker of HSC activity 19 . In the present study, we found that ENMD-1068 inhibits the development of liver fibrosis in a fibrotic mouse model by diminishing α-SMA expression and collagen production. Although increased serum levels of ALT and AST were observed in mice treated with CCl 4 alone, these elevations were reduced in the ENMD-1068-treated CCl 4 groups. In agreement with our results, Knight et al. observed that the progression of liver fibrosis, hepatic collagen gene expression, and hydroxyproline content were reduced in PAR-2 deficient mice with liver fibrosis 8 . Thus, a reduction in α-SMA and collagen expression and ALT/AST levels observed in fibrotic liver tissues in ENMD-1068-treated mice may partially explain the decreased development of liver fibrosis in these mice. Furthermore, the safety and selectivity of ENMD-1068 has been demonstrated 10,11,20 as no mice from the ENMD-1068 50 mg/ kg treatment group were observed to have any obvious pathological changes in the abdomen. However, further investigation of the toxicity of ENMD-1068 is warranted.
Collagen gene expression is controlled by TGF-β1 and Smad family activation in tissues. TGF-β1 is a multifunctional cytokine that plays a crucial role in the regulation of cell growth, differentiation, and biosynthesis of extracellular connective tissue 21,22 . TGF-β1 transduces signal from the membrane to the nucleus through transmembrane type I and II receptors by inducing phosphorylation of specific serine residues of the receptors. Among all Smad family members, Smad2/3 is closely associated with collagen gene expression. To evaluate whether PAR-2 is involved in collagen production, we investigated the expression of type I and III collagen and C-terminal phosphorylation of Smad2 influenced by TGF-β1 with or without ENMD-1068 treatment of isolated primary mouse HSCs. Interestingly, we found that the expression levels of type I and III collagen and TGF-β1 signaling-related phosphorylation of Smad2/3 C-terminal were significantly decreased in the ENMD-1068 treated HSCs. In agreement with our results, Chung et al. revealed PAR-2 synergizes with the TGF-β1 signaling pathway to contribute to renal injury and fibrosis 23   The signals for α-SMA, Collagen I, Collagen III, p-Smad2C and p-Smad3C were normalized to respective bands for β-actin and/or Smad2/3. (C) Primary mouse HSCs were cotransfected with SMAD luciferase reporter plasmid and pRL-TK vector. Twenty-four hours after transfection, cells were treated with TGF-β1(5 ng/mL) in the presence or absence of ENMD-1068 (10 mM) for another 24 h. Luciferase activity was determined with the commercial luciferase reporter assay system. (D) Confocal microscopy images displayed the subcellular localization of α-SMA (red) and p-Smad2 (green) in merged image panels. The nuclei were stained with 4′,6-diamidino-2phenylindole (DAPI; blue)(×200). (mean ± SD; * P < 0.05 vs Blank; # P < 0.05 vs TGF-β1 treatment, N = 3).
(PDAC) and non-PDAC cells of PAR-2 by RNA interference strongly decreased TGF-β1-induced activation of Smad2/3 and Smad dependent transcriptional activity in vitro 24 . Mussbach et al. also revealed that hepatic stellate cell line LX2 was unable to respond to TGF-β1 stimulation with phosphorylation of Smad3 after transfected with a PAR-2 shRNA 25 . These results strongly suggest that a functional cooperation between the TGF-β receptor(s) and PAR-2 is required for activation of Smad and non-Smad signaling. Since Smad2/3 are direct substrates of ALK5 which is in turn phosphorylated by TβRII [26][27][28] , further research is needed to prove that PAR-2 could affect the expression or activity of either ALK5 or TβRII in HSCs.
In conclusion, we have demonstrated that inhibition of PAR-2 activation in mice chronically exposed to CCl 4 leads to a significant reduction in hepatic fibrosis. The mechanism of this effect is likely to be through a reduction in signal crosstalk between PAR-2 and the TGF-β1 signaling pathway. These novel findings suggest that PAR-2 represents a potential therapeutic target for patients with chronic progressive liver fibrosis.