Kangfuxin solution alleviates esophageal stenosis after endoscopic submucosal dissection: A natural ingredient strategy

Abstract

BACKGROUND

Esophageal stricture ranks among the most significant complications following endoscopic submucosal dissection (ESD). Excessive fibrotic repair is a typical pathological feature leading to stenosis after ESD.

AIM

To examine the effectiveness and underlying mechanism of Kangfuxin solution (KFX) in mitigating excessive fibrotic repair of the esophagus post-ESD.

METHODS

Pigs received KFX at 0.74 mL/kg/d for 21 days after esophageal full circumferential ESD. Endoscopic examinations occurred on days 7 and 21 post-ESD. In vitro, recombinant transforming growth factor (TGF)-β1 (5 ng/mL) induced a fibrotic microenvironment in primary esophageal fibroblasts (pEsF). After 24 hours of KFX treatment (at 1.5%, 1%, and 0.5%), expression of α-smooth muscle actin-2 (ACTA2), fibronectin (FN), and type collagen I was assessed. Profibrotic signaling was analyzed, including TGF-β1, Smad2/3, and phosphor-smad2/3 (p-Smad2/3).

RESULTS

Compared to the Control group, the groups treated with KFX and prednisolone exhibited reduced esophageal stenosis, lower weight loss rates, and improved food tolerance 21 d after ESD. After treatment, Masson staining revealed thinner and less dense collagen fibers in the submucosal layer. Additionally, the expression of fibrotic effector molecules was notably inhibited. Mechanistically, KFX downregulated the transduction levels of fibrotic functional molecules such as TGF-β1, Smad2/3, and p-Smad2/3. In vitro, pEsF exposed to TGF-β1-induced fibrotic microenvironment displayed increased fibrotic activity, which was reversed by KFX treatment, leading to reduced activation of ACTA2, FN, and collagen I. The 1.5% KFX treatment group showed decreased expression of p-Smad 2/3 in TGF-β1-activated pEsF.

CONCLUSION

KFX showed promise as a therapeutic option for post-full circumferential esophageal ESD strictures, potentially by suppressing fibroblast fibrotic activity through modulation of the TGF-β1/Smads signaling pathway.

Key Words: Kangfuxin solution; Natural component; Endoscopic submucosal dissection; Esophagus stricture; Fibrosis

Core Tip: This study investigated the therapeutic effects and mechanisms of Kangfuxin solution (KFX) on esophageal stenosis following endoscopic submucosal dissection (ESD). Pigs underwent full circumferential ESD and received KFX treatment for 21 days. Endoscopic examinations assessed esophageal repair, while the expression of α-smooth muscle actin-2, fibronectin, and collagen I was evaluated. Profibrotic signaling was analyzed, including transforming growth factor (TGF)-β1, Smad2/3, and phosphor-smad2/3. In vitro, TGF-β1 induced a fibrotic microenvironment in primary esophageal fibroblasts, and KFX treatment reduced the expression of fibrotic markers in these cells.

INTRODUCTION

Endoscopic submucosal dissection (ESD) extensively manages early esophageal cancer and precursor lesions[1,2]. The incidence of esophageal stricture increases with the increase in the area treated with ESD[3,4]. The esophageal stricture rate after full circumferential ESD can reach 100%[5,6]. Hence, preventing and treating esophageal stricture following extensive ESD have emerged as pressing clinical concerns demanding immediate attention.

In recent years, progress has been made in treating esophageal stricture after ESD, but it still falls short of clinical needs. Studies have shown that metallic and biodegradable stents have shown promise clinically, yet they are costly and carry risks such as perforation and bleeding[7-9]. Biomedical materials such as polyethylene glycol sheets[10], high-density collagen patches[11], and carboxymethyl cellulose membranes[12] aid in re-epithelialization but are still in pre-clinical stages. Finding safer, more effective treatments remains a challenge.

According to traditional Chinese medicine (TCM) principles, esophageal stricture post-ESD stems from Qi and Yin deficiency alongside toxic stasis obstruction. The Periplaneta americana, a TCM insect remedy, is recognized for its heat-clearing, detoxifying, and blood-stasis-resolving properties. The Kangfuxin solution (KFX) is the brand name of a Chinese patent medicine whose primary component is the ethanol extract of Periplaneta americana. It has been shown that KFX enhances mucosal healing in ulcers. Li et al[13] reported that KFX promotes full-thickness healing of experimental skin wounds by facilitating the regeneration of epithelial tissues and skin appendages. Recent studies have found that KFX shows dual regulatory effects on fibrosis[14]. Su et al[15] discovered that KFX promotes wound healing and suppresses scar formation by optimizing fibroblast function and inhibiting excessive collagen synthesis. In the late stage of wound healing, KFX effectively inhibits collagen production[16,17]. Given the histological structure and biological functions similarity between esophageal mucosa and skin[11], we postulated that KFX could enhance the remodeling of artificial ulcers and inhibit excessive fibrotic repair following esophageal ESD.

This research investigated the potential impact and mechanisms of KFX in preventing esophageal stenosis after full circumferential resection during esophageal ESD, utilizing both in vivo and in vitro experiments.

MATERIALS AND METHODS

Drugs and animals

KFX has been approved for market by the China Food and Drug Administration. The administration standard for KFX (produced by Good Doctor Panxi Pharmaceuticals, China) specifies a 30 mL/d dosage for adults (60 kg).

The study adhered to animal ethical guidelines and was approved by the Animal Ethics Committee of Southwest Medical University (SWMU20210381). Thirteen healthy male Chinese experimental minipigs, averaging 15.9 ± 1.122 kg, were acquired from Shanghai JiaGan Biotechnology Co., Ltd. and adaptively fed for 1 week before surgery and treatment.

ESD procedure

The disease model chosen for this study involved using a porcine model with full circumferential esophageal ESD. Before surgery, all animals underwent a 24-h fast and were deprived of water for 6 hours. After general anesthesia (Virbac, China), an upper gastrointestinal endoscope (Olympus, GIF-Q260J) with a transparent cap (Olympus, D-201-11, 804) was inserted into the esophagus of the pig in the lateral position. Circumferential marking, submucosal injection, initial incision, establishing two submucosal tunnels, and submucosal tissue dissection between the two tunnels were performed sequentially as previously described (Supplementary Figure 1)[12,18]. The pigs’ vital signs were continuously monitored using an electrocardiograph throughout the endoscopic procedure.

Treatment strategies

Twelve pigs were randomly divided into three groups: The esophageal ESD-induced stenosis model group (Control, n = 4), KFX treatment group after ESD (KFX, n = 4), and the prednisolone (PDNN) treatment group after ESD, n = 4. After the ESD surgery, animals in the KFX group were treated with KFX 0.74 mL/kg/d for 21 continuous days. The doses administered in the PDNN group were[19]: 0.74 mg/kg/d for the week 1, 0.49 mg/kg/d for week 2, and 0.25 mg/kg/d for week of 3 prednisone (Xinyi Pharmaceutical Co., Ltd., China).

Dysphagia score

The food tolerance of pigs was assessed utilizing a swallowing difficulty scoring system, as reported before[20]. It is based on a score of 0 for regular diet (ability to swallow solid foods), 1 for granular diet, 2 for semi-liquid diet, 3 for liquid diet, and 4 for inability to swallow any food.

Visual observation through the digestive endoscope

The pigs in each group underwent endoscopic re-examination on 7 and 21 days after ESD. The essential condition of the artificial ulcer, including bleeding, perforation, mucosal regeneration, and luminal patency, was visually observed.

Esophageal stricture rate

When esophageal stenosis develops, endoscopic evaluation may be hindered, making it difficult to objectively determine the degree of stenosis. The stricture rate was evaluated by the ratio of the narrowest part of the artificial ulcer to the normal mucosa[11].

Hematoxylin-eosin staining

The four um-thick paraffin-embedded esophageal tissue sections were processed according to Beyotime’s manual (cat# C0105M, China), including deparaffinization, nuclear and cytoplasm staining, dehydration, and mounting. Histological features were observed by experienced researchers using an optical microscope, and a semi-quantitative analysis of inflammatory cells and fibroblast count was performed using Image J software.

Masson staining

Following Masson’s trichrome staining protocol (Solarbio, China), esophageal tissue paraffin sections underwent sequential staining with hematoxylin solution, Masson’s Ponceau fuchsin solution, and aniline blue solution for cell nuclei, collagen fibers, and muscle fibers, respectively. Rinsing was conducted with a 0.2% aqueous acetic acid solution, and light microscopy was employed to capture the images (Leica, Germany, ICC50W).

Immunohistochemistry

The paraffin-embedded esophageal tissue blocks underwent sectioning, dehydration, rehydration, antigen retrieval in citric acid buffer at 100 °C, peroxidase blocking, and bovine serum albumin blocking. Primary antibodies were applied at appropriate concentrations (Table 1), followed by overnight incubation at 4 °C[21-24]. The following day, the sections were treated with a secondary antibody conjugated to horseradish peroxidase (HRP). A brown precipitate formed within the sections upon exposure to a 3,3’-Diaminobenzidine chromogenic solution (Bioss, Cat# C-0010, China). Subsequently, sections were examined under a light microscope, and nuclear fast red counterstaining was applied. Semi-quantitative analysis was performed using Image-Pro Plus software.

Table 1 Primary antibodies used in this study.

Antibody name	Host	Brand, category No.	Country	Applications and dilution
				Western blot	IHC	IF
ACTA2	Rabbit	Proteintech	China	1:5000	1:800	1:100
FN	Rabbit	Proteintech	China	1:1000	1:200	1:100
Collagen Ⅰ	Rabbit	Proteintech	China	1:1000	1:500	1:100
TGF-β1	Mouse	Novus	United States	1:1000	-	-
Smad2	Rabbit	CST	United States	1:1000	-	-
Smad3	Rabbit	CST	United States	1:1000	-	-
P-Smad2	Rabbit	Affinity Biotech	United States	1:1000	1:100	1:50
P-Smad3	Rabbit	Affinity Biotech	United States	1:1000	1:100	1:50
Vimentin	Rabbit	Proteintech	China	-	-	1:200
Cytokerin	Rabbit	Abcam	China	-	-	1:200
GAPDH	Rabbit	CST	United States	1:1000	-	-

ACTA2: Alpha-smooth muscle actin; FN: Fibronectin; p-Smad2/3: Phosphor-Smad2/3; IHC: Immunohistochemistry; IF: Immunofluorescence; CST: Cell signaling technology.

Quantitative real-time polymerase chain reaction

For RNA extraction, tissue or cellular samples were processed using the Trizol reagent kit (Simgen, cat# 5003050, China). RNA samples’ absorbance was measured with a UV spectrophotometer at 260 and 280 nm, ensuring an OD₂₆₀/OD₂₈₀ ratio greater than 1.8 for suitability. RNA that was extracted underwent reverse transcription to produce cDNA, utilizing a reverse transcription kit (Takara, cat# RR036Q, Japan), followed by RT-PCR analysis on the Light Cycler 480 system (Roche, Switzerland) using the Green® Premix EX Tap TM II kit (Takara, cat# RR820A, Japan) according to manufacturer instructions. The amplification process had an initial denaturation at 95 °C for 30 seconds, then 40 cycles of denaturation at 95 °C for 5 seconds, annealing at 60 °C, and extension for 30 seconds. The expression levels of mRNA were normalized against GAPDH, and the relative expression levels were evaluated by applying the 2 ^-ΔΔCt method. The primer sequences used in this analysis are listed in Table 2.

Table 2 Primers used in this study.

Primer name	Forward 5’ to 3’	Reverse 5’ to 3’
ACTA2	TAGAACACGGCATCATCACCAACTG	TGGGGCAACACGAAGCTCATTG
FN	GTGTGGATGTCCTGCCTGTCAAC	GCTTGCTTTCCCTGCCCTGATG
Collagen I	CTCAAGATGTGCCACTCCGACTG	GTCTCGCCTGTCTCCATGTTGC
TGF-β1	GACCGCAACAACGCAATCTATGAC	GGCACTGCTTCCCGAATGTCTG
Smad2	ATGTCGTCCATCTTGCCATTCACTC	TGCTTTCTCACACCACTTCTCTTCC
Smad3	TGTCGTCCATCCTGCCCTTCAC	ACTTCTCCTCCTGCCCGTTCTG
GAPDH	ATGATTCCACCCACGGCAAGTTC	AGCACCAGCATCACCCCATTTG

ACTA2: Alpha-smooth muscle actin; FN: Fibronectin; TGF-β1: Transforming growth factor.

Western blot

The tissue or cell samples were lysed using RIPA lysis buffer to isolate total proteins, and the protein concentrations were assessed with a BCA protein quantification kit (Epizyme, cat# ZJ102 L, China). Following denaturation at 100 °C, 35 µg of protein was subjected to separation via 12% SDS-PAGE and subsequently transferred onto a PVDF membrane. The membrane was subsequently blocked using BSA for 2 hours, followed by overnight incubation with primary antibodies at 4 °C. Post washing with TBST, a specific secondary antibody conjugated with HRP was applied. The detection of signals was facilitated through the High-Sensitive ECL Chemiluminescence Detection Kit (Vazyme, Cat# E412, China) and detected by the ChemiScope 600 Exp system (ClinX, China).

Isolation and culture of primary esophageal fibroblasts

The submucosal tissue of the normal pig esophagus was isolated and preserved in pre-cooled sterile phosphate-buffered saline (PBS) (containing 100000 U/L penicillin G and 100 mg/L streptomycin). After washing the tissue clean of blood in a centrifuge tube, it was cut into 1-3 mm³ fragments. The tissue fragments were positioned on the base of a culture dish, and high-glucose DMEM (Gibco, United States) supplemented with 10% fetal bovine serum (Gibco) was gradually introduced. The cells were co-cultured in a CO₂ incubator (Thermo, United States).

CCK-8 assay

The primary esophageal fibroblasts (pEsF) cells in the logarithmic growth phase were seeded in a 96-well plate at a density of 3 × 10 ³ cells per well. Varying concentrations of the KFX (0%-5%) were incorporated into experimental wells at equivalent volumes of the culture medium. Additionally, blank and control wells were established. Following a 24 hours incubation, the culture medium was refreshed, and serum-free medium supplemented with 10% CCK-8 reagent (Dojindo, Japan) was administered to each well at 100 µL. The absorbance at 450 nm was subsequently measured using a microplate reader. The reagent instructions determined Cell viability by applying the absorbance values to the formula.

Cell treatment

Recombinant transforming growth factor (TGF)-β1 (5 ng/mL) (Peprotech, United States) was used to induce a fibrotic environment based on pEsF. After 1 hour of TGF-β1 stimulation, different doses of the KFX selected through CCK-8 assay were added and co-incubated for 24 hours, with solvent treatment as the negative control (NC).

Immunofluorescence

Following the specified treatment, pEsF were fixed in 4% paraformaldehyde for 10 minutes and then washed with PBS. Cells were permeabilized using 0.1% Triton X-100 for 10 minutes. After blocking with a protein-free blocking solution, primary antibodies at suitable concentrations were added and incubated for 14 hours. In a darkroom, specific secondary antibodies labeled with fluorescent markers were added and incubated for 1 hour. DAPI dye (Byotime, Cat# DAPI, China) was diluted 1:1000 to stain cell nuclei. The observations are made using a Leica DM4B fluorescence microscope equipped with a Leica DMC 6200 camera.

Statistical analysis

Statistical analysis was conducted using SPSS 27.0 software. Normally distributed continuous data are presented as mean ± SD. One-way analysis of variance was performed for intergroup comparison. Post hoc testing was performed using the least significant difference method, with P < 0.05 considered significant for detecting differences.

RESULTS

KFX treatment alleviated the degree of esophageal stricture after full circumferential ESD

We established a porcine model of full circumferential esophageal ESD and treated animals orally with KFX and PDNN for 21 days post-ESD (Figure 1A).

Figure 1 Kangfuxin solution treatment improved the esophageal stricture after full circumferential endoscopic submucosal dissection. A: The schematic graph shows the design of animal experiments (n = 4); pigs in each group were treated for 21 days with saline, Kangfuxin solution (KFX), and PDNN, respectively; B: The weight loss of pigs in each group; reduction in weight loss was observed post-treatment with KFX and PDNN (n = 4); C: Swallowing difficulty score for pigs in each group on day 21 post- endoscopic submucosal dissection (ESD) (n = 4); after KFX and PDNN treatment, pigs showed improved tolerance to semi-liquid diets compared to the Control group; D: Representative digestive endoscopic images of esophageal artificial ulcers on the 7 and 21 days following full circumferential ESD (n = 4); E: Macroscopic observation of esophageal gross specimens (n = 4); F: Cartoon diagram measuring the length of the short axis at the narrowest part of the artificial esophageal ulcer and adjacent normal mucosa (n = 4); G: The histogram shows the esophageal stricture rate (n = 4). Data are presented as mean ± SE. ^aP < 0.001 vs the control group, ^bP < 0.01 vs the control group. ESD: Endoscopic submucosal dissection; KFX: Kangfuxin solution; PDNN: Prednisolone.

Before ESD, the groups had no significant differences in body weight. However, by day 21 post-ESD, the Control group showed substantial weight loss, with the KFX and PDNN groups also experiencing weight reduction, albeit less severe (Figure 1B, Table 3).

Table 3 Summary of statistical results for the Control, Kangfuxin solution-treated, and PDNN-treated groups.

	Control	KFX	PDNN	P value
Weight loss, kg	4.08 ± 0.69	1.98 ± 0.79b	2.03 ± 1.07b	0.011
Dysphagia score	3.50 ± 0.58	1.75 ± 0.50b	1.25 ± 0.96a	0.004
Esophageal stricture rate	87.68 ± 1.90	59.67 ± 5.05a	53.90 ± 6.29a	< 0.001
Number of inflammatory cells	1976.25 ± 116.39	950.25 ± 276.92a,e	1651.50 ± 153.97c	< 0.001
Number of fibroblasts	5736.25 ± 691.56	1537.00 ± 118.89a, f	710.00 ± 347.14a	< 0.001
Fibrosis area ratio	66.20 ± 8.95	34.63 ± 3.64a,f	23.47 ± 2.65a	< 0.001
Thickness of submucosal fibrosis, um	853.93 ± 20.16	672.20 ± 28.69a,d	512.83 ± 42.73a	< 0.001

^aP < 0.001 vs the control group.

^bP < 0.01 vs the control group.

^cP < 0.05 vs the control group.

^dP < 0.001 vs the prednisolone (PDNN) group.

^eP < 0.01 vs the PDNN group.

^fP < 0.05 vs the PDNN group.

Data are presented as mean ± SE. KFX: Kangfuxin solution; PDNN: Prednisolone.

Swallowing difficulty was assessed using a scoring system (Figure 1C, Table 3). After KFX and PDNN treatment, pigs showed improved tolerance to semi-liquid diets compared to the Control group, which struggled with liquid sustenance.

Endoscopic monitoring post-ESD showed no abnormalities on day 7. However, by day 21, the Control group developed a pinpoint-sized ulcer, causing water passage difficulty and premature termination of observation. In contrast, KFX and PDNN treatment significantly improved esophageal luminal contraction (Figure 1D).

Macroscopic examination of esophageal specimens (Figure 1E) revealed reduced strictures and contractions in the KFX and PDNN groups compared to the Control group. Additionally, the KFX group exhibited thicker newly formed esophageal mucosa than the PDNN group, with both treatment groups showing a lower incidence of esophageal mucosal stricture (Figure 1F and G, Table 3). These findings indicate that KFX and PDNN alleviate esophageal stricture after ESD.

KFX treatment improved pathological structure remodeling after full circumferential ESD of the esophagus

ESD led to a fragmented distribution of regenerated epithelium, a lack of muscular mucosae, and atrophy of the muscular propria in the esophageal tissue. Compared to the Control group, KFX treatment significantly decreased the infiltration of inflammatory cells and the atrophy of the intrinsic muscle layer in the artificial ulcers. Consistent with the delayed healing observed in the esophageal specimens, compared to the PDNN group, hematoxylin-eosin (HE) staining showed a significant inflammatory cell infiltration in the KFX group (Figure 2A and B, Table 3). The morphologically polygonal or stellate fibroblasts in the submucosal layer of the esophagus were significantly reduced in the KFX and PDNN groups (Figure 2C, Table 3). These findings indicate that KFX positively influences the regeneration of submucosal in the esophagus following ESD.

Figure 2 Kangfuxin solution administration improved pathological structure and collagen remodeling of the esophagus after full circumferential endoscopic submucosal dissection. A: Representative images of hematoxylin-eosin staining of the esophageal tissue, magnification 20 × , 50 × , and 200 × , respectively (n = 4); the blue arrows and the yellow arrows indicate the inflammatory cells and fibroblasts, respectively; B: Quantitation of inflammatory cells in (A) (n = 4); C: Quantitation of fibroblasts in the regenerative tissue beneath the mucosa in (A) (n = 4); the number of inflammatory cells and the presence of fibroblasts in the esophageal submucosa were significantly reduced in both the Kangfuxin solution (KFX) and prednisolone (PDNN) groups; D: Representative images of Masson staining of the esophageal tissue, magnification 20 × and 50 × respectively (n = 4); in contrast to the Control group, both KFX and PDNN groups showed decreased fibrotic areas and more loosely arranged collagen structure; E: Quantitation of the total fibrotic area in (A) (n = 4); F: Quantitation of the newly formed fibrosis thickness beneath the mucosa in (A) (n = 4); the yellow arrow indicates the distance of fibrosis thickness measurement under the mucosa of the esophagus. Data are presented as mean ± SE. ^aP < 0.001 vs the control group, ^cP < 0.05 vs the control group, ^dP < 0.001 vs the PDNN group, ^eP < 0.01 vs the PDNN group, ^fP < 0.05 vs the PDNN group. KFX: Kangfuxin solution; PDNN: Prednisolone.

KFX treatment reduced ectopic fibrosis after full circumferential ESD of the esophagus

The collagen fiber structure of the esophageal tissue was observed after Masson staining (Figure 2D-F, Table 3). The Control group displayed larger and denser fibrotic regions, with thicker collagen fibers filling the submucosal layer. In contrast to the Control group, the KFX and PDNN groups exhibited decreased fibrotic areas and a more loosely arranged collagen structure. However, compared to PDNN treatment, KFX demonstrated a less pronounced effect on fiber inhibition. These results suggest that KFX may benefit the reformation of collagen structure following full circumferential ESD in the esophagus.

Consistent with the reduced Masson staining, the ulcer site of the esophagus in the KFX group showed decreased distribution of α-smooth muscle actin-2 (ACTA2)-positive myofibroblast, as revealed by immunohistochemical (Figure 3A and B). KFX treatment also reduced the protein levels of extracellular matrix components fibronectin (FN) and collagen I (Figure 3A, C, and D). Similarly, western blotting and RT-PCR validated the suppressed expression of ACTA2, FN, and collagen I (Figure 3E-K). However, KFX inhibited ACTA2 and collagen I expression slightly less than PDNN treatment. These data indicate that administration of KFX significantly suppressed the fibrosis-related remodeling of the esophageal ulcer. The positive drug PDNN exhibited a similar anti-fibrotic effect.

Figure 3 Kangfuxin solution inhibited the excessive deposition of fibrous components in artificial esophageal ulcers. A: Representative images of immunohistochemical staining to detect Α-smooth muscle actin-2 (ACTA2), fibronectin (FN), and collagen I in the artificial ulcer of esophageal tissue, magnification 200 × (n = 4); B-D: Quantitation of the staining intensity of ACTA2, FN, and collagen I in (A) (n = 4); Kangfuxin solution (KFX) treatment reduced the distribution of ACTA2-positive myofibroblasts and suppressed the expression of extracellular matrix components FN and collagen I; E: The protein levels of ACTA2, FN, and collagen I were determined by western blot; α-tubulin was used as the loading control (n = 4); F-H: Quantitation of relative expression of ACTA2, FN, and collagen I in (E), respectively (n = 4); I-K: The mRNA levels of ACTA2, FN, and collagen I were quantitated by RT-PCR (n = 4); administration of KFX significantly suppressed the protein and gene expression of ACTA2, FN, and collagen I. Data are presented as mean ± SE. ^aP < 0.001 vs the control group, ^bP < 0.01 vs the control group, ^eP < 0.01 vs the PDNN group, ^fP < 0.05 vs the PDNN group. KFX: Kangfuxin solution; PDNN: Prednisolone; ACTA2: Α-smooth muscle actin-2; FN: Fibronectin; IHC: Immunohistochemistry; WB: Western blotting.

KFX treatment inhibited the excessive activation of the TGF-β1/Smads signaling pathway after full circumferential ESD of the esophagus

We studied the underlying mechanism by which KFX improves the esophageal stricture. The expression levels of TGF-β1, Smad2, and Smad3 were reduced in the KFX and PDNN groups when compared to the Control group (Figure 4A-C). p-Smad2 and p-Smad3 proteins were also decreased by KFX treatment (Figure 4D-I). Compared with the PDNN group, Smad3 expression in esophageal tissue was down-regulated to a lesser extent following KFX treatment. However, there was no notable difference in p-Smad 2/3 expression between the KFX and PDNN groups. Immunohistochemical staining indicated reduced nuclear expression of p-Smad2 and p-Smad3 in the KFX group (Figure 4J-L). These data suggest that KFX inhibits the activation of the TGF-β1/Smad2/3 signaling pathway in the esophagus following full circumferential ESD.

Figure 4 Kangfuxin solution suppressed upregulation of TGF-β1/Smads signaling in the artificial ulcer tissue of the esophagus. A-C: Relative expression of transforming growth factor β1 (TGF-β1), Smad2, and Smad3 mRNA was determined by RT-PCR (n = 4); D: Western blotting to detect the protein expression of TGF-β1, Smad2, p-Smad2, Smad3, and p-Smad3 in each group (n = 4); E-I: Quantitation of proteins relative expression in D (n = 4); compared to the control group, the expression of TGF-β1, Smad2, and Smad3 was suppressed in both the Kangfuxin solution and prednisolone groups; J: Representative images of immunohistochemical staining of p-Smad2 and p-Smad3, magnification, 200 × (n = 4); K and L: The quantitation of relative expression of p-Smad2 and p-Smad3 in J (n = 4). Data are presented as mean ± SE. ^aP < 0.001 vs the control group, ^bP < 0.01 vs the control group, ^cP < 0.05 vs the control group, ^dP < 0.001 vs the PDNN group, ^eP < 0.01 vs the PDNN group. KFX: Kangfuxin solution; PDNN: Prednisolone; TGF-β1: Transforming growth factor β1; WB: Western blotting.

Phenotyping of isolated primary esophageal cells and determination of KFX dose in vitro

We isolated and cultured primary cells from the submucosal tissue of the esophagus (Supplementary Figure 2). Immunofluorescence revealed that most primary cells exhibited specific expression of vimentin. Cytokeratin, an epithelial marker, was used as a NC (Figure 5A).

Figure 5 Phenotyping of isolated primary esophageal cells and dose determination of Kangfuxin solution. A: Representative images of immunofluorescence staining of vimentin and cytokeratin, magnification 400 ×; primary cells exhibited specific expression of vimentin while showing minimal expression of cytokeratin protein; B: The proliferation rate of primary esophageal fibroblasts after stimulation with different concentrations of Kangfuxin solution using the CCK-8 assay. Data are presented as mean ± SE. ^gP < 0.001 vs the zero tube.

The CCK-8 assay revealed that KFX concentrations < 1.5% exhibited minimal inhibitory effects on the viability of pEsF after 24 hours of co-culture (Figure 5B). Consequently, KFX concentrations of 0.5%, 1%, and 1.5% were chosen for subsequent experiments.

KFX suppressed excessive fibrotic activity of pEsF in the TGF-β1-induced fibrotic microenvironment

Compared to the NC group, TGF-β1 significantly increased ACTA2, FN, and collagen I expression, while KFX intervention dose-dependently suppressed it (Figure 6A-E). KFX treatment reduced the green fluorescence expression of ACTA2, FN, and collagen I in the pEsF cytoskeleton and cytoplasm (Figure 6F). These findings suggest that KFX can modulate the fibrotic function of pEsF within the TGF-β1-induced fibrotic microenvironment.

Figure 6 Kangfuxin solution inhibited the excessive fibrotic activity in primary esophageal fibroblast. A-C: The expression levels ofΑ-smooth muscle actin-2 (ACTA2), fibronectin (FN), and Collagen Ⅰ mRNA were determined by RT-PCR; D: Western blot detected the protein expression of ACTA2, FN, and collagen Ⅰ in primary esophageal fibroblasts (pEsF); E: Quantitation of relative expression of fibrosis protein in (D); Kangfuxin solution intervention dose-dependently suppressed the expression of ACTA2, FN, and collagen Ⅰ in pEsF; F: Representative images of immunofluorescence staining of ACTA2, FN, and collagen Ⅰ in the pEsF, magnification, 400 ×. Data are presented as mean ± SE. ^hP < 0.001 vs the negative control group, ⁱP < 0.001 vs the TGF-β1 group, ^jP < 0.01 vs the TGF-β1 group, ^kP < 0.05 vs the TGF-β1 group. NC: Negative control group; KFX: Kangfuxin solution; TGF-β1: Transforming growth factor β1; ACTA2: Α-smooth muscle actin-2; FN: Fibronectin; IHC: Immunohistochemistry; WB: Western blotting.

We further discovered the expression of Smad2/3 and phosphor-smad2/3 (p-Smad2/3) in pEsF. TGF-β1 stimulation upregulated expression of Smad2 and Smad3, whereas KFX intervention resulted in a dose-dependent decrease in Smad2 and Smad3 mRNA transcription (Figure 7A-C). Compared to the TGF-β1 group, KFX significantly downregulated p-Smad2/3 protein expression (Figure 7D-I). Immunofluorescence demonstrated a dose-dependent decrease in the fluorescence signal in the KFX group (Figure 7J). These findings suggest that KFX similarly inhibits the hyperactivation of TGF-β1/Smads signaling in pEsF cells.

Figure 7 Kangfuxin solution inhibited upregulation of TGF-β1/Smads signaling in primary esophageal fibroblast. A-C: RT-PCR to detect gene transcription level of Transforming growth factor β1 (TGF-β1), Smad2, and Smad3; D: The western boltting determined the protein expression of TGF-β1, Smad2/3, and p-Smad2/3; E-I: Quantitation of proteins expression in D; compared to the TGF-β1 group, KFX significantly downregulated endogenous p-Smad2/3 expression; J: Representative images of immunofluorescence staining of p-Smad2 and p-Smad3, magnification, 400 ×. Data are presented as mean ± SE. ^hP < 0.001 vs the negative control group, ⁱP < 0.001 vs the TGF-β1 group, ^jP < 0.01 vs the TGF-β1 group. NC: Negative control group; KFX: Kangfuxin solution; TGF-β1: Transforming growth factor β1; WB: Western blotting.

DISCUSSION

Esophageal strictures following ESD often entail structural abnormalities and functional impairment of the esophagus, inflicting significant distress upon patients. While the pathogenesis of ESD-induced esophageal stricture remains incompletely understood, numerous factors contributing to submucosal fibrous hyperplasia are widely acknowledged[25]. Our previous reports indicated that the typical pathological features of esophageal stenosis post-ESD involve myofibroblast proliferation and excessive extracellular matrix deposition, constituting an exaggerated fibrotic repair process[26]. Hence, addressing the inhibition of exaggerated fibrotic repair in the submucosa is crucial to prevent and treat esophageal strictures.

Corticosteroids are recognized as effective agents that can improve scarring by decreasing collagen synthesis and inhibiting fibroblast proliferation[27]. However, the oral corticosteroid route is unreliable and may cause delayed repair, immunosuppression, and abnormal glucose metabolism[28]. In addition, after esophageal full circumferential ESD, the mucosal defect area is extensive, and continuous use of steroid therapy significantly increases the risk of infection. Patients who receive systemic steroid treatment for 21 days or take > 700 mg PDNN are at increased risk of infection[29]. Other therapeutic agents, including thymosin β4, botulinum toxin type A, and mitomycin C, are still in the nascent stages of research and necessitate multicenter, large-sample, prospective randomized controlled trials to substantiate their efficacy[30-32]. Therefore, it is still the focus of researchers to study more safe and reliable drugs against esophageal strictures after ESD.

KFX is a Chinese patent medicine approved for clinical use in China (GYZ51021834). The main ingredient of KFX is the ethanol extract of the natural insect medicine Periplaneta americana, which contains abundant active compounds, including amino acids (such as leucine, alanine, proline, and lysine)[33-35], polyols (such as a polyol, glycerol, and 1, 2, 3-propanetriol)[36], nucleosides (such as uracil, guanosine, and xanthosine)[37,38], and peptides[36]. The positive role of KFX in repairing the original structure of mucosal injury is widely recognized. Studies have demonstrated that KFX not only facilitates mucosal repair by promoting the organized arrangement of new stroma but also enhances the anti-irritation function of the stratified squamous epithelium[39]. During wound healing, it stimulates the repair of dermal ultrastructures such as cell mitochondria, endoplasmic reticulum, and desmosomes[15]. A recent investigation underscores the potential merit of KFX in ameliorating glandular atrophy within the digestive tract[40]. Although some studies have documented the efficacy of biomaterials in facilitating esophageal mucosal tissue regeneration, these findings remain confined to the short-term animal experimental phase[10-12]. Furthermore, clinical studies have indicated that long-term oral administration of KFX does not exhibit significant toxic side effects[41]. Hence, in this study, KFX was selected to provide potential insights into the development of natural remedies for esophageal stenosis.

The anatomy of the swine esophagus is more akin to that of humans than that of conventional rodent models. Furthermore, the esophageal size in pigs is particularly suitable for models of ESD-associated diseases. The full circumferential ESD procedure was successfully conducted on porcine esophagi as described previously[12,18]. Subsequent endoscopic reevaluations revealed no bleeding, perforation, mucosal re-generations, or obstruction in any esophagi. Under the conditions of this study, the degree of esophageal stricture in pigs decreased after 21 days of KFX treatment. In the KFX group, the new tissue in the esophagus exhibited a closer resemblance to normal tissue, with increased softness, enhanced tension resistance, and improved mucosal integrity. Masson staining revealed decreased fibrotic areas and a more relaxed collagen arrangement in the esophageal submucosa following KFX treatment. Consistent with this reduction, KFX treatment suppressed the expression of ACTA2, FN, and collagen I at both mRNA and protein levels. Moreover, in vitro experiments demonstrated that KFX inhibited the expression of ACTA2, FN, and collagen I in pEsF in a dose-dependent manner. While PDNN treatment also exhibited a significant anti-esophageal fibrosis effect, a delay in esophageal ulcer healing was observed in the PDNN group. After PDNN therapy, a gross examination of the esophagus revealed artificial ulcers not fully covered by epithelium, along with areas of dark red ecchymosis in the middle. Honda et al[20,42] found that steroid treatment deepened and delayed the healing of esophageal ulcers induced by endoscopic mucosal resection, with no improvement seen with systemic administration. Additionally, in contrast to the KFX group, HE staining revealed significant infiltration of inflammatory cells in the esophageal submucosa of the PDNN group. This observation may be attributed to heightened inflammation in regions where external stimuli (such as saliva, food, and microorganisms) contribute to delayed healing of the esophagus. Therefore, KFX effectively inhibits excessive fibrotic repair of esophageal ESD ulcers and promotes healing.

We further investigated the mechanism of action of KFX against esophageal stricture. TGF-β1, known for promoting organ fibrosis, activates Smad upon receptor binding[26,43]. Phosphorylated Smad forms a complex with other Smads proteins, translocating into the nucleus to induce fibrosis-related gene transcription. Research indicates a notable increase in serum TGF-β1 levels in esophageal stricture, correlating with its severity[12]. Previous findings have highlighted heightened TGF-β1 expression in ESD-induced esophageal segments, enhancing Smad 2/3 phosphorylation and nuclear translocation, thereby amplifying fibrotic signaling. Consequently, targeting TGF-β1/Smads or their downstream effectors is a promising therapeutic avenue for inhibiting fibrosis in esophageal injury[26]. In this study, we observed that administering KFX in a porcine post-ESD esophageal stricture model significantly suppressed TGF-β1-driven fibrosis and downstream fibrotic molecule Smad2/3. In vitro, our findings consistently show that KFX treatment effectively curbs the overactivation of pEsF via TGF-β1/Smad2/3-dependent mechanisms within the fibrotic microenvironment induced by TGF-β1, thus regulating fibrotic function.

Our study had some limitations. We did not compare the treatment response differences between direct application and oral administration of KFX, a Chinese patent medicine approved for both routes. It is well established that the dosage form of KFX is a Chinese herbal liquid. Researchers may consider the incorporation of KFX with a reliable biomaterial to enhance its concentration within the esophageal mucosa. Other assessment methods for esophageal strictures, such as esophagography, could be utilized to obtain more comprehensive and objective evidence. Our findings suggest that KFX mitigates excessive esophageal fibrosis post-full circumferential ESD. Combining KFX with fasting, acid suppression, and physical expansion might enhance therapeutic outcomes. Further clinical and safety evaluations of KFX to prevent strictures post-esophageal ESD are planned.

CONCLUSION

In summary, KFX holds promise in ameliorating esophageal stricture following full circumferential ESD by mitigating excessive fibrous deposition. Its mechanism potentially involves the downregulation of the TGF-β1/Smads signaling pathway.

ACKNOWLEDGEMENTS

We thank Dr. Yang WX and Dr. Li L for their excellent technical assistance. We also thank Mr. Su SC for the valuable discussions on the anesthesia methods and dosage for the experimental pigs.