Vactosertib

EW-7197, an oral transforming growth factor β type I receptor kinase inhibitor, for preventing peritoneal adhesion formation in a rat model ✩

Abstract

Background: EW-7197 is an oral transforming growth factor β type I receptor kinase inhibitor currently undergoing phase I clinical trials for cancer treatment in the United States. This study evaluates whether EW-7197 prevents peritoneal adhesion formation in a rat model.
Methods: Forty-eight female Wistar rats underwent peritoneal adhesion induction by the creation of peritoneal ischemic buttons and were randomly divided into 4 groups of 12 each. The control group received 0.3 mL vehicle by oral gavage once daily for 7 days after adhesion induction. The 10 mg and 20 mg groups received 10 or 20 mg/kg EW-7197 phosphate dissolved in 0.3 mL vehicle by oral gavage once daily for 7 days after adhesion induction. The rebound group received 20 mg/kg EW-7197 phosphate dissolved in 0.3 mL vehicle by oral gavage once daily for 7 days after adhesion induction followed by 0.3 mL vehicle only by gavage once daily for an additional 21 days. After the respective treatments were completed, the animals were euthanized.

Results: All rats survived until the end of the study without complications. EW-7197 reduced the inci- dence, quality, and tenacity of peritoneal adhesions in a dose-dependent manner. Fibrosis and collagen production were reduced in EW-7197–treated peritoneal ischemic buttons. Transforming growth factor β/Smad2/3 signaling and mesothelial-to-mesenchymal transition were inhibited in EW-7197–treated peri-
toneal ischemic buttons. Discontinuation of EW-7197 was not associated with rebound effects. Conclusion: EW-7197 prevented peritoneal adhesion formation potentially via inhibition of transforming growth factor β1/Smad2/3–induced mesothelial-to-mesenchymal transition in a rat model.

Introduction

Peritoneal adhesions are pathologic bonds (ie, a thin film of connective tissue or a thick fibrous band containing blood vessels and nerves) between abdominal organs and the peritoneum.1 Ad- hesion formation is a normal response to peritoneal injury and oc- curs in approximately 95% of patients after laparotomy.2 Although this process is integral to the healing of the peritoneum, adhesions can occasionally cause significant morbidity, including small-bowel obstruction and female infertility, as well as chronic abdominal or pelvic pain, or both.3

Transforming growth factor β (TGF-β) is one of the most important cytokines in peritoneal adhesion formation.4 Overexpression of TGF-β in the peritoneum, as well as increased peritoneal fluid concentrations of TGF-β, has been associated with an in- creased incidence of adhesions.5,6 In an animal model, exoge- nous TGF-β was found to increase the incidence of adhesions, whereas neutralizing TGF-β antibody has been found to reduce the incidence of adhesions.7,8 TGF-β induces peritoneal fibrosis by stimulating extracellular matrix (ECM) production during adhe- sion formation by promoting mesothelial-to-mesenchymal transi- tion (MMT).9 It also inhibits ECM degradation by increasing tissue inhibitors of metalloproteinase-1.

Several studies have found that blocking the TGF-β signaling pathway with a number of small-molecule adenosine triphosphat– competitive TGF-β type I receptor kinase inhibitors (eg, SB-505124, LY-2157299, and EW-7197) inhibited autophosphorylation of ac- tivin receptor–like kinase (ALK) 5 and TGF-β–induced transcrip- tion of its downstream genes in reporter gene assays.11-13 Among them, EW-7197 has recently progressed to phase I clinical trials for cancer treatment in the United States (ClinicalTrials.gov identifier: NCT02160106). Therefore this study aimed to evaluate whether EW-7197 prevents peritoneal adhesion formation in a rat model.

Methods

Study design

A total of 48 nonpregnant female Wistar rats (250–300 g; Ori- ent Bio, Seongnam, Korea) underwent peritoneal adhesion induc- tion by the creation of peritoneal ischemic buttons and were ran- domly divided into 4 groups of 12 each using computer-generated random numbers.14 The groups were treated as follows by oral gavage once daily for 7 days after adhesion induction. The control group received 0.3 mL of the vehicle (artificial gastric fluid formu- lation; ddH2O 900 mL, conc. HCl 7 mL, NaCl 2.0 g, and pepsin 3.2 g). The 10 mg and 20 mg groups received 10 or 20 mg/kg EW-7197 phosphate dissolved in 0.3 mL vehicle.11,15 The rebound group re- ceived 20 mg/kg EW-7197 phosphate dissolved in 0.3 mL vehicle by oral gavage, followed by 0.3 mL vehicle only by gavage once daily for an additional 21 days. EW-7197 phosphate was provided by the Laboratory of Medicinal Chemistry, College of Pharmacy, Ewha Woman’s University (Seoul, Korea). The dosage of EW-7197 phosphate was determined based on prior studies.11,15 All animals were housed 1 per cage under a 12-hour light/dark cycle in a room at environmental temperature (24 ± 1°C) and moisture (55 ± 10%) and were supplied with food and water ad libitum. The animals were euthanized by carbon dioxide asphyxiation on postoperative day 7 for the control, 10 mg, and 20 mg EW-7197 groups and on postoperative day 28 for the rebound group. This study was ap- proved by the committee for animal research at our institution and conformed to the US National Institutes of Health (NIH, Bethesda, MD) guidelines on the care and use of laboratory animals.

Peritoneal adhesion induction

The rat peritoneal ischemic button model was chosen for this study because it is one of the most consistent and reproducible animal models for postoperative peritoneal adhesions.14 Anesthe- sia was induced by intramuscular injection of 50 mg/kg zolazepam and tiletamine and 10 mg/kg xylazine. The operations were limited to less than 20 minutes for each animal to minimize the drying effect of room air on tissues and risk of infection and were per- formed by the same investigator using powder-free gloves. Four is- chemic buttons spaced 1 cm apart were created on each side of the parietal peritoneum along the paracolic gutter with 4-0 silk sutures as previously described.16 The musculoperitoneal layer was then closed with 4-0 Vicryl sutures, and the skin was closed with 4-0 silk sutures. Postoperatively, all animals received an intramuscular injection of 0.05 mg/kg buprenorphine every 6 hours for 48 hours for pain control. No antibiotics were administered before, during, or after the operation.

Peritoneal adhesion evaluation

After the animals were euthanized, the abdomens were opened via a right lateral U-shaped laparotomy, and the incidence, qual- ity, and tenacity of the peritoneal adhesions at the ischemic button sites were assessed based on the consensus of 2 observers blinded to the group assignment. The incidence was assessed according to the number of peritoneal ischemic buttons with adhesions and was calculated as the fraction of adhesions (expressed as a percentage of the total number of peritoneal ischemic buttons).17 The quality and tenacity were determined according to the adhesion score,18 which is a 4-point scale used for qualitative assessment (0 = none; 1 = filmy, transparent, avascular; 2 = opaque, translucent, avascu- lar; 3 = opaque, capillaries present; and 4 = opaque, larger vessels present), and a 3-point scale was used for tenacity assessment (0 = none; 1 = adhesions fell apart; 2 = adhesions lysed with trac- tion; 3 = adhesions sharply dissected).

Histologic examination

The harvested peritoneal ischemic buttons were fixed in 10% neutral-buffered formalin for 24 hours, embedded in paraffin, and then sectioned (3 μm thick). The tissue samples were stained with hematoxylin and eosin and Masson’s trichrome. The is- chemic buttons were graded for levels of inflammation and fi- brosis using a previously published grading scale.19 The levels of inflammation were scored as follows: 0 = no inflammation; 1 = presence of giant cells, occasional lymphocytes, and plasma cells; 2 = presence of giant cells, plasma cells, eosinophils, and neutrophils; and 3 = presence of numerous inflammatory cells and microabscesses. The level of fibrosis was scored as follows: 0 = no fibrosis; 1 = minimal, loose; 2 = moderate; and 3 = florid, dense. The amount of collagen content was graded using a 5-point scoring scale (1 = mild; 2 = mild to moderate; 3 = moderate; 4 = moderate to severe; and 5 = severe).15 Four sections and 10 fields from each animal were randomly selected and evaluated by 2 observers blinded to the group assignment.

Quantitative real-time polymerase chain reaction analysis

Total RNA was isolated from the harvested peritoneal ischemic buttons using QIAzol lysis reagent (Qiagen, Valencia, CA) follow- ing the manufacturer’s protocol. RNA concentration and quality were assessed using NanoDrop 2000 UV-Vis spectrophotometer (Thermo Scientific, Burlington, ON, Canada) and 1 μg total RNA was used to synthesize complementary DNA using Oligo(dT)15 (Elpis Biotech, Daejeon, Korea) and RevertAid H Minus reverse transcrip- tase (Thermo Scientific) following the manufacturer’s protocol. The synthesized complementary DNA was amplified using the LightCy- cler 480 SYBR Green 1 Master Mix and Light Cycler 480 (Roche, Indianapolis, IN). The primers used, Snai1, epithelial cadherin (E- cadherin), α–smooth muscle actin (α-SMA), and glyceraldehyde 3- phosphate dehydrogenase, have been described in previous stud- ies19-26 and are listed in Supplemental Table 1. The following quantitative real-time polymerase chain reaction (qPCR) conditions were used: 95°C for 15 minutes; 45 cycles of 30 seconds at 95°C, 30 seconds at 60°C, and 30 seconds at 72°C; and 95°C for 15 min- utes. Melting curve analyses were performed to ensure that spe- cific polymerase chain reaction products were generated. Gene ex- pression was analyzed using the comparative CT method and nor- malized to glyceraldehyde 3-phosphate dehydrogenase messenger RNA (mRNA) levels, expressed as the relative fold change.

Western blot analysis

Total protein was extracted from the peritoneal ischemic buttons of each rat using radioimmunoprecipitation assay buffer containing protease and phosphatase inhibitors (Roche), and the concentrations were measured using a Pierce bicinchoninic acid protein assay kit (Thermo Scientific), following the manufacturer’s instructions. Then, 20 μg protein was loaded and separated using 10%, 15%, or 4% to 15% polyacrylamide gels (Bio-Rad, Hercules, CA) and then transferred onto polyvinylidene difluoride membranes. The membranes were cut according to the molecular sizes and blocked with 5% bovine serum albumin or skim milk for 1 hour at room temperature, followed by incubation with the following primary antibodies at 4°C overnight: anti-Smad2/3, anti–phospho- Smad2/3 (p-Smad2/3), and antitubulin antibody (all 1:1000; Cell Signaling Technology, Danvers, MA). Then the membranes were incubated with the horseradish peroxidase-conjugated secondary antibodies (1:1000; Jackson ImmunoResearch, West Grove, PA) at room temperature for 1.5 hours. The membranes were finally de- veloped with SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific), and the protein bands were imaged using EZ-Capture MG (ATTO, Tokyo, Japan). Densitometric values of the bands were quantified and expressed as their ratio to tubulin using the ImageJ software (NIH).

Immunohistochemical analysis

Formalin-fixed, paraffin-embedded sections of the peritoneal is- chemic buttons were immunohistochemically analyzed. Deparaf- finized sections (3 μm thick) were placed in sodium citrate buffer (pH 6.0), heated for antigen retrieval, and mounted on slides. The slides were permeabilized with 0.025% Triton-X 100 (Sigma- Aldrich, St. Louis, MO) in Tris-buffered saline and blocked with 10% fetal bovine serum. The sections were then incubated with anti–p- Smad2/3 antibody (1:100) at 4°C overnight. Sections were treated with 0.3% H2O2 to block the endogenous peroxidase prior to incu- bation with horseradish peroxidase-conjugated secondary antibod- ies (1:100). Then 3.3r-diaminobenzidine (DAB) tetrahydrochloride (Dako, Glostrup, Denmark) was applied for color development, and the sections were counterstained with hematoxylin and eosin.

Immunofluorescence analysis

Peritoneal ischemic buttons were harvested and fixed with paraformaldehyde perfusion fixative (Electron Microscopy Sciences, Hatfield, PA) overnight at 4°C, followed by sequential incubation in 15% and 30% sucrose solution. The fixed samples were then em- bedded in optimal cutting temperature compound, and 6-μm-thick cryostat sections were prepared and mounted on slides for double- immunofluorescence staining. Before staining, the slides were air dried, fixed, and then permeabilized in ice-cold acetone for 10 minutes at –20°C. Then the slides were blocked with 10% fetal bovine serum in 1 × phosphate-buffered saline with 0.1% Tween 20. Tissue sections were first incubated with the anti–α-SMA antibody (1:300; Abcam) overnight at 4°C, washed with phosphate-buffered saline, and incubated with Alexa Fluor 488 (Invitrogen, Burlington, ON, Canada). After washing, the slides were incubated for 1 hour at room temperature with the anti–E-cadherin antibody (1:200; Cell Signaling Technology) followed by Alexa Fluor 594 (Invitrogen). The nuclei were visualized with 4r-6-diamidino-phenylindole (In- vitrogen). The negative controls, in which primary antibodies were substituted with isotypic nonimmune immunoglobulin G, did not have any detectable labeling.

Statistical analysis

The Student’s t test or one-way analysis of variance with Tukey post hoc test was used to compare continuous variables as appro- priate. All statistical analyses were performed using SPSS software (Version 21.0; IBM Corporation, Armonk, NY).

Results

Peritoneal ischemic button establishment was successful in all rats without procedure-related complications. All animals survived until the end of the study without EW-7197-related complications (eg, delayed wound healing and infection).
EW-7197 reduced incidence, quality, and tenacity of peritoneal adhesions in a dose-dependent manner

To determine whether EW-7197 prevented peritoneal adhesion formation in a rat model, we evaluated the incidence and adhe- sion score of peritoneal adhesions at the ischemic button sites af- ter the animals were euthanized. As shown in Table 1 and Fig. 1, EW-7197 reduced the incidence, quality, and tenacity of peritoneal adhesions. The mean incidence, quality score, and tenacity score were significantly different among the control and 10 mg and 20 mg EW-7197 groups (P < .05). For the mean incidence, both the 10 mg and 20 mg EW-7197 groups had significantly lower means than the control group did (P < .05), and that of the 20 mg EW-7197 group was significantly lower than the 10 mg EW-7197 group (P <.05). Similarly, for the mean quality score, both the 10 mg and 20 mg EW-7197 groups had significantly lower means than the con- trol group did (P < .05), and that of the 20 mg EW-7197 group was significantly lower than the 10 mg EW-7197 group (P < .05). The mean tenacity score was also significantly lower in the 10 mg and 20 mg EW-7197 groups than it was in the control group (P < .05) and was also significantly lower in the 20 mg EW-7197 group than in the 10 mg EW-7197 group (P < .05). These results indicate that oral administration of EW-7197 for 7 days after peritoneal adhe- sion induction reduced adhesion formation in the rat model in a dose-dependent manner. Fibrosis and collagen production were reduced in EW-7197– treated peritoneal ischemic buttons.Subjects with peritoneal adhesion exhibit morphologic changes characterized by inflammation and fibrosis of the peritoneum,1 and therefore we graded the peritoneal ischemic buttons for levels of inflammation and fibrosis in the control and 10 mg and 20 mg EW- 7197 groups. In addition, we graded the collagen content in the ischemic buttons because excessive ECM production is a hallmark of fibrosis.27 As shown in Table 2 and Fig. 2, fibrosis and colla- gen production were reduced in the EW-7197–treated peritoneal ischemic buttons. Hematoxylin and eosin staining indicated that the difference in mean inflammation score was not statistically significant among the 3 groups (P > .05). The difference in the mean fibrosis score among the 3 groups was found to be statis- tically significant (P < .05). Specifically, the 20 mg EW-7197 group had a significantly lower mean fibrosis score than the control, and 10 mg EW-7197 groups did (P < .05), but the mean fibrosis score was not significantly different between the control and 10 mg EW- 7197 groups (P > .05). Masson’s trichrome staining indicated that the difference in the mean collagen production score among the 3 groups was statistically significant (P < .05). Specifically, the mean collagen deposition score was significantly lower in the 10 mg and 20 mg EW-7197 groups than it was in the control group (P < .05), and it was significantly lower in the 20 mg EW-7197 group than in the 10 mg EW-7197 group (P < .05). These results imply that EW-7197 reduced peritoneal fibrosis and collagen production but not inflammation during adhesion formation in the rat model. Figure 1. EW-7197 reduced incidence, quality, and tenacity of peritoneal adhesions. Bar graph of (A) incidence, (B) quality score, and (C) tenacity score of peritoneal ad- hesions. ∗P < .05. Bar, mean; error bar, standard deviation. Images show (D) severe peritoneal adhesion formation (arrows, 3 quality score; arrowheads, 4 quality score) in control group, (E) moderate peritoneal adhesion formation (arrow, 1 quality score; arrowheads, 3 quality score) in the 10 mg EW-7197 group, and (F) mild peritoneal adhesion formation (arrows, 1 quality score) in the 20 mg EW-7197 group. TGF-β/Smad2/3 signaling and MMT were inhibited in EW-7197–treated peritoneal ischemic buttons.The Smad-dependent TGF-β signaling pathway plays an impor- tant role in peritoneal adhesion formation.4 To determine if TGF- β/Smad2/3 signaling was inhibited by EW-7197, western blot anal- yses of the markers of TGF-β/Smad2/3 signaling (ie, Smad2/3 and p-Smad2/3) were performed. As shown in Fig. 3, TGF-β/Smad2/3 signaling was inhibited in EW-7197–treated peritoneal ischemic buttons. Phosphorylation of Smad2/3 is an obligatory step of in- tracellular canonical TGF-β signaling.28 The mean protein expres- sion of Smad2/3 was not significantly different among the 3 groups (P > .05), whereas the mean protein expression of p-Smad2/3 was significantly lower in the 10 mg and 20 mg EW-7197 groups than it was in the control group (P < .05). These results suggest that EW- 7197 does not alter the total level of Smad2/3 but merely atten- uates the phosphorylation of Smad2/3 in the peritoneal ischemic buttons. Figure 2. Fibrosis and collagen production were reduced in EW-7197–treated peritoneal ischemic buttons. Bar graph of (A) inflammation, (B) fibrosis, and (C) collagen production scores of peritoneal ischemic buttons. ∗P < .05. Bar, mean; error bar, standard deviation; NS, not significant. Hematoxylin and eosin–stained sections of peritoneal ischemic buttons showing (D) dense fibrosis with increased admixed inflammatory cells in the control group, (E) moderate fibrosis with increased admixed inflammatory cells in 10 mg EW-7197 group, and (F) loose fibrosis with increased numbers of admixed inflammatory cells in 20 mg EW-7197 group (original magnification 200 ×). Masson’s trichrome-stained sections of peritoneal ischemic buttons showing (G) moderate to severe collagen production in the control group, (H) moderate collagen production in the 10 mg EW-7197 group, and (I) mild collagen production in the 20 mg EW-7197 group (original magnification 200 ×). TGF-β1 plays an important role in MMT and, therefore real-time qPCR analyses of the markers of MMT (ie, Snai1, E-cadherin, and α-SMA) were performed on the peritoneal ischemic buttons.As shown in Fig. 4, MMT was inhibited in EW-7197–treated peri- toneal ischemic buttons. Snai1 is a transcription factor that re- presses E-cadherin and thereby induces MMT.9 The 10 mg and 20 mg EW-7197 groups had significantly lower mean mRNA expres- sion of Snai1 than that of the control group (P < .05), and the mRNA expression of E-cadherin was significantly higher in the 20 mg EW-7197 group than it was in the control group (P < .05). Mesothelial cells undergoing MMT acquire the expression of the smooth muscle cell marker α-SMA.9 The mRNA expression of α- SMA was significantly lower in the 10 mg and 20 mg EW-7197 groups than it was in the control group (P < .05). These results suggest that EW-7197 inhibited MMT in the peritoneal ischemic buttons. To determine if TGF-β/Smad2/3 signaling and MMT were inhibited by EW-7197, the peritoneal ischemic buttons were im- munohistochemically stained for p-Smad2/3 and immunofluores- cently stained for E-cadherin and α-SMA. As shown in Fig. 5, TGF-β/Smad2/3 signaling and MMT were inhibited in EW-7197–treated peritoneal ischemic buttons. The p-Smad2/3–positive mesothelial monolayer was most and least prominent in the 20 mg EW-7197 and control groups, respectively, indicating that phosphorylation of Smad2/3 was markedly obliterated in the mesothelial monolayer. The E-cadherin–positive mesothelial monolayer was most and least prominent in the 20 mg EW-7197 and control groups, respectively,whereas the α-SMA–positive mesothelial monolayer was most and least prominent in the control and 20 mg EW-7197 groups, respec- tively, indicating that MMT was inhibited by administration of EW- 7197. Figure 3. Transforming growth factor β (TGF-β)/Smad2/3 signaling, and mesothelial-to-mesenchymal transit were inhibited in EW-7197–treated peritoneal ischemic buttons. (A) Bar graph of mean protein expression of Smad2/3 with representative bands. (B) Bar graph of mean protein expression of p-Smad2/3 with representative bands. Molecular weight for Smad2/3, p-Smad2/3, and tubulin are 52/60 kDa, 60 kDa, and 52 kDa, respectively. ∗P < .05. Bar, mean; error bar, standard deviation; NS, not significant. Figure 4. Mesothelial-to-mesenchymal transition was inhibited in EW-7197–treated peritoneal ischemic buttons. Bar graph of mean messenger RNA (mRNA) expression of (A) Snai1, (B) E-cadherin, and (C) α–smooth muscle actin (α-SMA). ∗P < .05. Bar, mean; error bar, standard deviation; NS, not significant. Discontinuation of EW-7197 was not associated with rebound effects We evaluated whether discontinuation of EW-7197 administra- tion was associated with rebound effects. In the rebound group the animals were dosed with 10 or 20 mg/kg EW-7197 phosphate dis- solved in 0.3 mL vehicle by oral gavage once daily for 7 days after peritoneal adhesion induction followed by 0.3 mL vehicle only by gavage once daily for an additional 21 days. The 20 mg EW-7197 group was used as a control. As shown in Table 1 and Fig. 6, dis- continuation of EW-7197 was not associated with rebound effects. No significant differences were identified between the rebound and the 20 mg EW-7197 groups in mean incidence, quality score, and tenacity score (P > .05). This observation indicated that discontin- uation of EW-7197 treatment did not exacerbate the development of peritoneal adhesion after the 7-day treatment in the rat model.

Discussion

In this study the incidence and severity of peritoneal adhesions were significantly lower in the EW-7197–treated groups than they were in the control group. In addition, the histologic analysis indi- cated that fibrosis and collagen deposition were significantly attenuated in EW-7197–treated peritoneal ischemic buttons. These results indicate that EW-7197 suppressed peritoneal adhesion for- mation in the rat model. TGF-β1/Smad2/3–induced MMT has been implicated in peritoneal adhesion formation.9 As shown by the real-time qPCR, western blot, immunofluorescence, and immuno- histochemical analyses, TGF-β1/Smad2/3 signaling and MMT were significantly inhibited in EW-7197–treated peritoneal ischemic but- tons. Collectively, these results suggest that EW-7197–induced sup- pression of peritoneal adhesion formation is most likely caused by inhibition of TGF-β1/Smad2/3–induced MMT (Fig. 7).

Although EW-7197 significantly decreased the mean fibrosis and collagen production scores in this study, regulation of the accumulation of inflammatory cells was not evident as revealed by the histologic analysis. The similar inflammation scores of the groups might have resulted from the decreased inflammatory re- sponse on postoperative day 7. In response to surgical injury, in- flammatory cells increasingly accumulate in the injured tissues to remove necrotic cells and ECM components. To promote tissue re- generation, the accumulation of inflammatory cells is gradually at- tenuated, and these cells in the injured tissue are also gradually replaced by fibroblasts 5 to 6 days later.29–32 In this study, inflam- mation was scored on postoperative day 7, and therefore it is likely that there was no difference in the score among the groups. In ad- dition, other reasons for this finding might be the pharmacologic mechanism of action of EW-7197 and the unclear inflammation-related role of TGF-β in peritoneal adhesion formation. EW-7197 is a novel inhibitor of the TGF-β type I receptor kinase (ALK-5),which can suppress the biological effect of TGF-β by specifically inhibiting the ALK-induced phosphorylation of Smad 2/3, the key signal transducers of TGF-β. TGF-β also plays a pleiotropic role in inflammation.11,12,33 It exerts either anti-inflammatory effect or proinflammatory effects, depending on the amount of TGF-β and local extracellular factors that influence the activity of TGF-β.34 Al- though TGF-β is well known to exert profibrotic effects on peri- toneal adhesion formation, whether and how TGF-β regulates the inflammatory response in this process is unclear.9,35 It is possible that TGF-β might have no or only a weak effect on inflammation in peritoneal adhesion formation. Therefore there was no differ- ence in the inflammation score among the control and 10 mg and 20 mg EW-7197 groups.

Figure 5. Immunohistochemical (p-Smad2/3) and immunofluorescence (4r ,6-diamidino-2-phenylindole [DAPI], blue; E-cadherin, green; and α–smooth muscle actin [α-SMA], red) stained sections of peritoneal ischemic buttons showed most prominent p-Smad2/3–positive and E-cadherin–positive mesothelial monolayer in the 20 mg EW-7197 group and least prominent in the control group, whereas the α-SMA–positive mesothelial monolayer was most prominent in the control group and least prominent in the 20 mg EW-7197 group (original magnification 200 ×). (Color version of figure is available online.)

Figure 6. Discontinuation of EW-7197 was not associated with rebound effects. Bar graph of (A) incidence, (B) quality score, and (C) tenacity score of peritoneal adhesions.∗P < .05. Bar, mean; error bar, standard deviation; NS, not significant. Peritoneal adhesion formation is a complex pathologic process. The underlying mechanisms include exacerbated inflammatory response, impaired fibrinolysis, enhanced proliferation of myofibrob- lasts, and increased ECM production.30,31 Myofibroblasts are the main cellular source of ECM and key effector cells in the patho- genesis of peritoneal adhesions. The cellular origins of myofibrob- lasts include resident fibroblasts, circulating fibrocytes, endothe- lial cells, adipocytes, and mesothelial cells.9,35 MMT, a process in which mesothelial cells transform into myofibroblasts, has re- cently been found to play a key role in peritoneal adhesion for- mation and could be a novel therapeutic target.9 This process is mainly mediated by TGF-β signaling in a Smad-dependent and - independent manner.9 TGF-β is a profibrotic factor produced by activated inflammatory cells and stromal cells. EW-7197 is a TGF-β type I receptor kinase inhibitor with high specificity and low toxi- city.11,12 In the present study we found that EW-7197 significantly inhibited TGF-β1/Smad2/3–induced MMT and decreased peritoneal adhesion formation via oral administration. In addition, EW-7197– related complications were not identified. These findings suggest that EW-7197 could be a potential therapeutic option for peritoneal adhesion prevention. Figure 7. Diagram illustrating possible mechanism of action EW-7197 in prevention of peritoneal adhesion formation. TGF-β, transforming growth factor β. This study had a number of important limitations. First, no hematologic or histologic examinations were performed to assess the organ-specific toxicity of EW-7197. However, a rigorous toxic- ity study in rats reported that the no-observed-adverse-effect level after oral EW-7197 once daily for 4 weeks was 20 mg/kg for fe- males.11 In addition, a phase I clinical trial designed to evaluate the safety and tolerability of EW-7197 at multiple ascending doses is ongoing (ClinicalTrials.gov identifier: NCT02160106). Second, the possible adverse effects of TGF-β type I receptor kinase inhibitors (eg, delayed wound healing) were not assessed, although no com- plications were identified in this study. Third, only a few markers were assessed using western blot analysis to determine the mecha- nism by which EW-7197 prevented peritoneal adhesion formation. In conclusion, EW-7197 prevented peritoneal adhesion forma- tion potentially by inhibiting TGF-β1/Smad2/3–induced MMT in a rat model. Therefore it warrants further evaluation as a potential therapeutic agent for preventing postoperative peritoneal adhesion formation in clinical trials Vactosertib after the completion of human safety studies.