Periplaneta Americana Extract may Attenuate 2,4,6-Trinitrobenzenesulfonic Acid-Induced Intestinal Fibrosis by Inhibiting TGF-β/Smad Signaling Pathway

Purpose Intestinal brosis is an incurable digestive disease accompanied by stricture formation, and it has an increasing incidence in recent years. Periplaneta americana is one of the medicinal insects with a long history. There are few reports on the effect of intestinal brosis. This study aims to evaluate the inhibitory effect of PA treatment on intestinal brosis. Methods TNBS was used to establish intestinal brosis model by enema in BALB/c mice. The mice were treated with PA (50, 100, 200 mg/kg body weight) and 5-aminosalicylic acid (5-ASA) (40mg/kg) by gavage once a day for 6 weeks. At the end of the last week, the mice were sacriced. Colon samples were collected for H&E and Masson staining. The mRNA and protein expression of α-smooth muscle actin (α-SMA), collagen I and the transforming growth factor-β (TGF-β) / Smad signaling pathway were conducted by real-time PCR and western blot analysis. In vitro, TGF-β1 was used to induce intestinal brosis at human colon broblasts (CCD-18Co). And using real-time PCR and western blot methods to detect the expression of α-SMA and collagen I. TGF-β/Smad vivo, and the same results had not been obtained in vitro. Conclusion: PA may attenuate intestinal brosis by inhibiting TGF-β/Smad signaling pathway, but more experiments were needed in vitro.


Introduction
Intestinal brosis is currently a common clinical unsolved disease, and is more common in complications of IBDs [1,2]. Fibrosis occurs in more than one-third of CD patients. About 50% of patients with CD have brotic stenosis, and 75% of them will eventually receive surgery [3]. With the extension of the onset time, intestinal brosis will also have other complications, such as stenosis [4], stula, and even colon cancer [5]. The progress of intestinal lesions varies greatly and may last for a long time, and brosis more easily occurs in deep ulcers or transmural ssures [6]. Despite the signi cant progress in the in ammation treatment of IBDs, the incidence of intestinal stenosis and brosis has not been signi cantly reduced [7].
Normally, excessive accumulation of scar tissue in the intestinal wall is a manifestation of intestinal brosis, accompanied by the distribution of in ammation, this accumulation may run through the thickness of the entire intestinal wall and may cause stenosis [8].
Fibrosis is the result of long-term local chronic in ammation, mainly manifested in ECM protein deposition, which is produced by activated broblasts [9]. TGF-β1 isoform is not only closely related to brosis of various organs (including intestinal brosis), but also has a signi cant relationship with ECM synthesis. Moreover, the TGF-β/Smad signal pathway is generally considered to be related to brosis, Smad protein is a typical TGF-β signal pathway [10]. TGF-β1 is the central pathway of organ brosis in most if not in all organs, including gastrointestinal tract and heart. Despite the important clinical impact, disease pathogenesis is not fully understood, and no targeted therapies able to revert brosis are currently available.
TCM is widely used in the treatment of various clinical diseases in Asia and has a long history. In addition, many TCM exhibit special pharmacological effects in the treatment of human chronic diseases [11]. The extract of PA contains many ingredients that can kill bacteria, fungi, viruses, protozoa and even cancer cells. PA has been studied as an alternative therapy for many diseases, such as heart disease, burn wounds [12][13], cancer [14], etc. Studies have also shown that PA could resist liver brosis [15], renal brosis [16], and may reduce the accumulation of ECM components by inhibiting the expression of TGF-β1 and α-smooth muscle actin (α-SMA). Even, PA can improve pulmonary brosis through TGF-β/Smad signaling pathway [17]. Although the research on PA has made great progress in recent years, the role of the drug in intestinal diseases is rarely reported, especially in intestinal brosis. In view of these ndings, this study aims to evaluate whether PA could inhibit the activation of broblasts and the progression of

Experimental protocols in Mice
Female BALB/c mice (8-10 weeks old. body weight 18-22 g.) were kept under the Animal Center of Anhui Medical University (Hefei, China). All mice mice were housed under standard environmental conditions at controlled temperature (22 ± 2°C), humidity (50 ± 10%), and light (12 h light/dark cycle), they were free access to standard diet and water. A total of 110 mice were randomized into six groups, listed as follow: Control group, TNBS group, PA groups (50mg/kg, 100mg/kg, 200mg/kg), 5-ASA group (40 mg/kg). The dosage of PA was selected according to the references [18]. The amount of TNBS per week was: 1.0, 1.0, 1.5, 1.5, 2.0, and 2.0 mg, prepared with 45% ethanol. The mice were fasted overnight and gently anesthetized with ether. A 3.5-F catheter was connected to a 1 ml syringe, the animals were anesthetized, 100 μl/20g of TNBS was aspirated with an enema. Slowly insert the catheter 4 cm from the anus. The key step: proceed very slowly, so as not to damage or destroy the colon wall. If any resistance you feel during catheter insertion, remove the catheter and try to reinsert it gently. Slowly inject 100 µl / 20g of 45% ethanol + TNBS solution into the colon cavity. Slowly remove the catheter from the intestine and place the mouse head down for 1 min. TNBS was given to mice once a week, physiological saline, PA and 5-ASA were given to mice by gavage every day. The whole experiment lasted for 6 weeks. The mice were executed at the seventh week, and blood was taken from the eyeballs before execution.

Hematoxylin and Eosin (H&E) Staining
All mice colon samples were washed and immersed in a 4% formaldehyde solution and then embedded in para n at 4% paraformaldehyde. Then the tissue was dehydrated in an ascending series of ethanol, embedded in para n. Serial sections (3.5mm) were stained with hematoxylin and eosin (H&E). The stained sections were observed under a Nikon Eclipse E800 (Nikon Corporation, Tokyo, Japan).

Table1: The histological activity index (HAI)
Masson staining The aforementioned para n embedded slices (5 µm) were stained with Weigert solution (Sigma-Aldrich, Merck KGaA) for 5-10 min. After being fully washed, sections were treated with Ponceau fuchsin acid solution for 5-10 min, immersed in 2% acetic acid aqueous solution for 1min, then differentiated in 1% phosphomolybdic acid aqueous solution for 3-5 min. Without washing with water, the sections were treated with aniline blue for 5 min then immersed in 0.2% acetic acid aqueous solution for 1 min. Slices were permeabilized with xylene and mounted with neutral resin.  Proteins were extracted from colon samples using RIPA lysis buffer, PMSF and protease inhibitors, and they were separated on 12% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), and then transferred to PVDF membranes. Membranes were blocked with 5% non-fat dried milk for 1h at room temperature, and then incubated with primary antibodies overnight at 4℃, including anti-β-actin, anti-collagen I, anti-TGF-beta1, anti-Smad2, anti-phospho-Smad2, anti-Smad3, anti-phospho-Smad3 and anti-Smad7, anti-α-SMA. Subsequently, the membrane was incubated with secondary antibodies for 1 h and washed them three times with PBS. The blots were visualized using Super Signal West Pico (Bridgen, Beijing, China).

Statistical Analysis
Statistical signi cance was determined by analysis of variance (ANOVA) using GraphPad Prism version 8.0 (GraphPad Prism, CA, USA). Data were expressed as mean ± standard deviation (SD). A value of P<0.05 was considered statistically signi cant.

PA Attenuates Intestinal Fibrosis in Mice
Severity of colonic in ammation was evaluated by HE staining assay. As shown in Figure 1, colonic tissues from the control group was almost normal with little in ammatory in ltration necrosis and edema, lamina propria and mucosal muscle layers were well arranged, there is almost no loss of crypts and cups cell. In TNBS group, colonic crypts and cups cell loss was more, and a large number of in ammatory cells in ltrated in all layers of the colon. With the increase dose of PA, the protective effect on the colon structural integrity became more obvious, especially in the 200mg/kg PA group, and the 5-ASA group also showed a protective effect on the colon.

PA reduces collagen expression in the colon of mice with intestinal brosis
As shown in gure 2, the collagen expression in control group was less, but it signi cantly increased in TNBS group. In the PA groups, the decreased collagen expression could be observed obviously. The most signi cant inhibitory effect of collagen expression appeared in 5-ASA group. It was not di cult to nd that the 100mg/kg and 200mg/kg PA had almost the same degree of inhibition as the 5-ASA group.

PA Suppresses α-SMA and Collagen I Expression in Mice with Intestinal Fiborsis
Real time-PCR and western blot assays were performed to evaluate the expression levels of brosisrelated mRNA and proteins in intestinal tissue. As shown in Figure 3, compared with control group, mRNA and protein levels of collagen I and α-SMA were obviously up-regulated in TNBS group, while they were signi cantly down-regulated in 100mg/kg, 200mg/kg PA groups and 5-ASA group.
PA Attenuates CCD-18Co Cells Fibrosis and Inhibits the Expression of α-SMA and collagen I MTT assay was used to determine the safe concentration of PA's effect on CCD-18Co cells. In gure 4A, compared with control group, TGF-β1 group was not observed proliferative effect on CCD-18Co cells. TGF-β1 stimulated and induced CCD-18Co cells to produce ECM, but the proliferation did not be promoted. It could be observed that 8-32mg/ml PA inhibited cell proliferation obviously, which might be a toxic concentration. The concentration of 1, 2, 4 mg/ml PA was selected in this study. As shown in Figure  4B-E, mRNA expression of α-SMA and collagen I in the TGF-β1 group were higher than control group.
Compared with the TGF-β1 group, both PA groups and the 5-ASA group were able to reduce the mRNA expression of α-SMA and collagen I. The protein expression of α-SMA was signi cantly higher in TGF-β1 group, and was lower in PA and 5-ASA groups. Compared with control group, the protein expression of collagen I showed a signi cant increase in TGF-β1 group, and it was reduced in PA groups.

PA Protects against Intestinal Fibrosis by Inhibiting TGF-β/Smad signaling pathway in Mice
In order to detect the expression of TGF-β/Smad signaling pathway, RT-PCR and Western blot were used to detect the expression of TGF-β1, Smad2, p-Smad2, Smad3, p-Smad3, and Smad7. As shown in gure 5, in the gene expression results ( gure A-D), the expression of Smad2, Smad3 and TGF-β1 were all upregulated in TNBS group, and down-regulated after 100mg/kg and 200mg/kg PA treatment. Although 5-ASA group could also inhibit their expression, but the effect was not as good as PA. In addition, the expression of Smad7 showed the opposite result. In the protein expression results ( gure E-J), the expression of p-Smad2, Smad2 and p-Smad3, Smad3 were up-regulated in TNBS group, and downregulated in PA groups and 5-ASA group. The expression result of Smad7 and TGF-β1 was basically consistent with gene expression.

Discussion
Studies have shown that PA has many pharmacological activities, and even found that PA has potential effect on inhibiting liver brosis [18] and renal brosis [16]. However, the effect of PA on intestinal brosis is unknown. In this experiment, a model of intestinal brosis was constructed rstly. Rectal administration of TNBS and ethanol to cause colon in ammation in susceptible mouse strains is a rapid method [19].When TNBS-induced chronic colitis model of BALB/c mice was used, it showed the characteristics of brosis. Its histological characteristics are mainly manifested in: transmural in ammation, in ammatory cell in ltration (macrophages, neutrophils and lymphocytes) [20]. In the TNBS group, intestinal section staining showed a large number of in ammatory cells, indicating signi cant in ammation of the colon. However, in ammatory cells were relatively reduced in the PA groups, which may be the result of anti-in ammatory effects of PA. This might have a positive effect on inhibiting intestinal brosis.
In addition, we had mentioned earlier that another manifestation of intestinal brosis was the deposition of ECM [21], the deposition of ECM was manifested in a large amount of α-SMA and collagen (typically type I collagen) expression. ECM is produced by activated broblasts and is regulated by pro-/anti-brotic factors [22]. Cells associated with intestinal brosis, typically colonic subepithelial myo broblasts, have been shown to be the major producers of ECM [23]. When colon tissue is damaged, myo broblasts are activated and contract by shrinking the expression of α-SMA. Then, ECM multiply and migrate to the wound site and secrete large amounts of ECM components such as collagen and bronectin. Therefore, in our experiments, the genes and proteins expression of α-SMA and collagen I showed similar results in vivo and in vitro. Wherefore, PA might be able to inhibit intestinal brosis by reducing ECM. In our experiments, the gene and protein expression of α-SMA and collagen I increased in the TNBS group, which might be the cause of intestinal brosis. Under the treatment of PA, the expression of α-SMA and collagen I were decreased, and the 100mg/kg and 200mg/kg of PA showed signi cant inhibitory effects, suggested that PA could reduce intestinal brosis at this concentration.
According to statistics, the most important pro brotic mediators include TGF-β, insulin-like growth factor (IGF)-1 and -2, activins, connective tissue growth factor (CTGF), various cytokines, such as Interleukin (IL)-1, -4, -6, -13, -17), angiogenic factors and so on [9,24]. TGF-β is expressed in a large number of cell types and organs in mammals and is linked to ECM. The TGF-β1 isoform can promote ECM synthesis and broblast contraction [25]. In this experiment, TGF-β1 was used to stimulate human colonic broblasts (CCD-18Co), which caused ECM secreted. The results showed that the up-regulation of α-SMA and collagen I expression in the TGF-β1 group, it was the evidence of intestinal brosis. After PA treatment, their expression were down-regulated, further indicated that PA could also attenuate intestinal brosis in vitro.
TGF-β is a multifunctional cytokine involved in the brosis of almost all organs and tissues (including the gastrointestinal tract). One of the most typical signaling pathways is the Smad pathway. Activated TGF-β receptor mediate Smad protein expression and regulate the expression of type I collagen [26][27][28].
Overexpression of TGF-β in mice can cause intestinal brosis and obstruction, preventing intestinal TGFβ/Smad signaling mice from surviving colonic brosis. Additionally, Smad7 can inhibit TGF-β signaling. Disrupting the TGF-β/Smad pathway or increasing the expression of Smad7 can effectively reduce the occurrence of intestinal brosis. The increase in p-Smad2/3 or the decrease in Smad7 is consistent with the pro brotic effect of the TGF-β/Smad pathway [29]. Despite the risks, TGF-β is still a very important target in the progress of intestinal brosis. Consequently, TGF-β signaling is a potential strategy for the treatment and alleviation of intestinal brosis [10]. In vivo results, PA treatment caused a decrease in the expression of Smad2, p-Smad2, Smad3, p-Smad3, and TGF-β1 protein molecules in the TGF-β/Smad signaling pathway, and an increase expression of Smad7. Therefore, PA could inhibit intestinal brosis through the TGF-β/Smad signaling pathway in vivo. However, PA was not a monomer, and the results in vitro experiments had not been consistent with the results in vivo of TGF-β/Smad signaling pathway inhibition. In this experiment, we also observed that some protein expression results were inconsistent with mRNA expression. This may be due to the posttranscriptional processing, transcriptional products undergoes degradation, translation, and posttranslational modifcation [30].

Conclusions
This result indicated that PA might have a potential role in inhibiting intestinal brosis by regulating the TGF-β/Smad signaling pathway. In the next experiment, we will deeply study and consider exploring other factors related to brosis effects of signal pathways. We hope that our experiment will contribute to the treatment of intestinal brosis. The animal study was approved by the Committee for Experimental Animal Use and Care of Anhui Medical University.

Consent for publication
Not applicable.
Availability of data and materials