Introduction

Ulcerative colitis (UC) is a chronic nonspecific intestinal inflammatory disease of unknown aetiology and a subtype of inflammatory bowel disease (IBD). UC has a long disease course, serious complications, and a high recurrence rate. This condition can lead to toxic megacolon, intestinal perforation, intestinal bleeding, and cancer, which seriously affect the quality of life of patients (Kaplan 2015). The existing epidemiological data in China show that both the incidence and prevalence of IBD have been increasing. According to the statistics of the Chinese Center for Disease Control and Prevention in 2014, the total number of IBD cases in China from 2005 to 2014 was approximately 350,000. The number of IBD patients in China is expected to reach 1.5 million by 2025, with a younger age of onset (Kobayashi et al. 2020).

The aetiology of UC is unknown, and its pathogenesis may be related to the imbalance of intestinal immunity caused by the interaction of multiple factors, such as the environment, genetics, and intestinal microecology (Ungaro et al. 2017). Corticosteroids, 5-aminosalicylic acids, immunomodulators, Janus kinase inhibitors, and biologic agents targeting tumour necrosis factor-alpha (TNF-α), α4β7-integrin, and interleukin (IL)-12/23 have been widely applied in the clinical treatment of UC (Hernandez et al. 2020). However, an increasing number of patients eventually become refractory or intolerant to the side effects or complications of drugs, and some even require colectomy. In addition, the risk of colorectal cancer in UC is significantly increased due to the continuous infiltration of immune cells and repeated stimulation of the intestinal mucosal epithelium (Pike and Tremblay 2018). Traditional Chinese medicine (TCM) is regarded as a form of “complementary and alternative medicine” in the Chinese healthcare system and provides new options for standard therapy. Bletilla striata (B. striata) is one of the most valuable TCMs in China, was first recorded in the Shennong Herbal Scripture, and has a history of more than 2000 years (Chen et al. 2018). This herb has effects on haemostasis, detumescence, muscle generation, and sore convergence (Xu et al. 2019). Modern pharmacology has found that B. striata can stop bleeding, promote wound healing, and regulate immunity (Jiang et al. 2019a, b). Based on data mining, B. striata was found to be a common drug for the treatment of UC (Jia et al. 2021). Many clinical trials have also confirmed that B. striata can protect the intestinal mucosa, promote the healing of intestinal ulcers, and improve the symptoms of intestinal inflammation and blood in the stool (Zheng et al. 2010; Xu et al. 2014), but its mechanism of action is still unclear.

Network pharmacology combines pharmacology, bioinformatics, and other scientific and systematic network analyses to study the pharmacological mechanism of TCM. Its integrity and systematic characteristics are consistent with the overall concept of TCM and the principle of syndrome differentiation and treatment, which has been widely used in TCM research. In this study, a network pharmacology approach was used for the first time to screen the effective bioactive components and targets of B. striata and to analyse its key targets and signalling pathways for the treatment of UC, guiding further research of this formula. The detailed workflow of this study is shown in Fig. 1.

Fig. 1
figure 1

The workflow of network pharmacology, molecular docking, and in vivo experimental verification deciphering the underlying mechanisms of B. striata on the treatment of UC

Full size image

Materials and methods

Network pharmacology

B. striata ingredient collection and target gene prediction

Screening of active ingredients of B. striata was performed using the TCM Systems Pharmacology (TCMSP) database and analysis platform (https://tcmspw.com/tcmsp.php) (Ru et al. 2014). Then, we used pharmacokinetic information retrieval filters for absorption, distribution, metabolism, and excretion screening based on oral bioavailability (OB) ≥ 30% and drug-likeness (DL) ≥ 0.18. Target genes were predicted using the TCMSP database after identifying the active ingredients. Predictive target gene names were standardized from the UniProt database (https://www.uniprot.org/) (UniProt. 2019). Cytoscape software was used to draw the composition and target gene network of B. striata.

Prediction of UC-related pathogenic genes

GeneCards (https://www.genecards.org/) (Safran et al. 2010), DisGeNET (https://www.disgenet.org/) (Piñero et al. 2019), OMIM (http://www.omim.org/) (Amberger et al. 2020), TTD (https://idrblab.org/ttd/) (Wang et al. 2020), and DrugBank (https://www.drugbank.ca/) (Wishart et al. 2018) were selected to obtain UC-related pathogenic genes. All the databases used “ulcerative colitis” as the keyword.

Protein–protein interaction (PPI) network and topological analysis

To clarify the interaction between the predicted target of the main active components of B. striata and the target of UC, we investigated the intersection of the above target genes. Then, the common target genes were uploaded to the Search Tool for the Retrieval of Interacting Genes (STRING) database (http://string-db.org/) (Szklarczyk et al. 2021). Furthermore, the condition was restricted to “Homo sapiens,” and the free points were hidden. In this study, a medium confidence score above 0.4 was selected to obtain PPI data. Next, the interaction files were downloaded and imported into Cytoscape to visualize the PPI network and analyse the topological value. Furthermore, the topology parameter of the degree value was determined by the “NetworkAnalyzer”, an analytical tool in Cytoscape 3.9.0. The top 10 ranked proteins were defined as hub targets based on the degree level.

Enrichment analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG)

The biological process (BP), molecular function (MF), and cell component (CC) enrichment analyses were carried out using the Metascape system (https://metascape.org/) (Zhou et al. 2019), in which the background as an organism was selected as “Homo sapiens” for customized analysis with a cut-off P value of  < 0.05. The KEGG database (https://www.kegg.jp/kegg/) (Kanehisa et al. 2021) was applied for KEGG enrichment analysis.

Component-target molecular docking

Molecular docking is a method of placing the ligand in the binding area of the receptor through computer simulation and calculating its physical and chemical parameters to predict the binding affinity and conformation of the ligand and receptor. Receptor and ligand files were downloaded from the PubChem (https://pubchem.ncbi.nlm.nih.gov/) and PDB (https://www.rcsb.org) databases. AutoDock Vina software (Eberhardt et al. 2021) was used for molecular docking.

Animal experimental study

Dose and preparation of B. striata decoction

According to the 2020 edition of the Pharmacopoeia of the People’s Republic of China (National Pharmacopoeia Committee 2020), the maximum dosage of B. striata for human adults is 15 g per day. The raw material was purchased from The Third Affiliated Hospital of Zhejiang Chinese Medical University. The raw material was soaked in 8 times the volume of distilled water for 1 h, then boiled for 20 min, and decocted 2 times. The two decoctions were mixed and filtered with gauze and centrifuge. Finally, the filtrate is concentrated by rotary evaporation to 100% of the solution (i.e. 1 g of raw drug per 1 mL of solution) and stored at 4 ℃ for backup.

Animal modelling and drug treatment

Male C57BL/6 mice (8–10 weeks of age) were randomly divided into six groups (n = 10/group): control group, model group, low­dose (1 g/kg•day−1) B. striata decoction group (BS­L group), medium­dose (2 g/kg•day−1) B. striata decoction group (BS­M group), high­dose (4 g/kg•day−1) B. striata decoction group (BS­H group), and mesalazine group. Colitis was induced in mice by administering 3% (w/v) dextran sulphate sodium (DSS) in their drinking water for 7 days. On days 8 to 16, the mice were switched to normal water and treated with low, medium, and high doses of B. striata decoction or mesalazine (0.5 g/kg•day−1). The doses of 1 g/kg•day−1, 2 g/kg•day−1, and 4 g/kg•day−1 B. striata decoction are equivalent to 0.5 times, 1.0 times, and 2.0 times the human adult dose, respectively. All animal experiments were performed with the approval of the local ethical committee of Zhejiang Chinese Medical University and conducted in accordance with the relevant guidelines of the Animal Center of Zhejiang Chinese Medical University.

Disease activity index (DAI)

DAI was assessed based on weight loss, diarrhoea, and rectal bleeding. Weight loss was defined as the difference between the initial and final weights. Diarrhoea was defined as the absence of faecal pellet formation and the presence of continuous fluid faecal material in the colon. Rectal bleeding was defined based on the presence of diarrhoea containing visible blood and on the presence of gross rectal bleeding. The DAI value was calculated using the following formula: DAI = [(body weight loss score) + (diarrhoea score) + (haematochezia score)]/3 (Table 1) (Liu et al. 2021).

Table 1 Criteria for DAI
Full size table

Histological colitis score

Colon sections were fixed in 10% formalin, embedded in paraffin, and stained with haematoxylin and eosin for histological scoring. Histological colitis score = inflammation + depth of lesions + destruction of crypt + width of lesions (see Table 2 for details (Dieleman et al. 1998)).

Table 2 Histological score to quantify the degree of colitis
Full size table

Real-time quantitative PCR

One-Step TB Green PrimeScript PLUS Perfect Real-Time (TaKaRa Bio, Shiga, Japan) and ABI 7500 Fast Real-Time PCR (Thermo Fisher Scientific) were used for reverse transcription and amplification of RNA according to the manufacturer’s instructions. The primer sequences are listed in Table 3. The relative expression of target genes was calculated using the 2­ΔΔ CT method with β-actin as a normalized control.

Table 3 Primer sequences and amplification length
Full size table

Enzyme-linked immunosorbent assay (ELISA)

EGFR, PIK3CA, p-AKT, TNF-α, and IL-6 protein expression levels were measured by mouse EGFR, PIK3CA, p-AKT, TNF-α, and IL-6 ELISA kits (Shanghai Fanke Industrial Co., Ltd., Shanghai, China) according to the manufacturer’s instructions.

Western blot

The protein expression levels of EGFR, PIK3CA, and p-AKT were tested again by Western blot, and the bands were analysed by ImageJ software. The primary antibodies used were as follows: EGFR (#4267: Cell Signaling Technology, Danvers, MA, USA), PIK3CA (#4249: Cell Signaling Technology, Danvers, MA, USA), p-AKT (#9271: Cell Signaling Technology, Danvers, MA, USA), and GAPDH (#5174: Santa Cruz Biotechnology, CA, USA). Secondary antibodies used were goat anti-rabbit IgG-HRP (#SC-2004, Santa Cruz Biotechnology, Dallas, TX, USA).

Statistical analysis

All graphs and statistical analyses were prepared and performed using GraphPad Prism version 9.0.0 (GraphPad Software, La Jolla, CA, USA). One-way analysis of variance (ANOVA) was used for comparisons between groups, followed by Bonferroni post hoc tests for multiple comparisons; otherwise, the Tamhane T2 method was used. All data are expressed as the mean ± SD. P < 0.05 was considered statistically significant.

Results

Screening of active ingredients and target genes of B. striata

In the TCMSP database, B. striata contains 36 active ingredients in total. These components were then screened with OB ≥ 30% and DL ≥ 0.18 as the conditions. Consequently, 9 active ingredients were obtained. However, 4 active ingredients of B. striata, 2,7-dihydroxy-4-methoxyphenanthrene-2,7-O-diglucoside (MOL005759), 3,7-dihydroxy-2,4-dimethoxyphenanthrene-3-O-glucoside (MOL005766), bletlol A (MOL005770), and blespirol (MOL005773), were eliminated because they could not be found in PubChem (https://pubchem.ncbi.nlm.nih.gov/). Therefore, a total of 5 active ingredients were included in the next step of the study (Table 4) (Supplementary file 1).

Table 4 Active component network information
Full size table

In addition, we collected targets of potential active ingredients in B. striata using the TCMSP database and annotated the target genes using UniProt data. There were 410 corresponding targets of B. striata. After removal of duplicate targets, a total of 216 human-derived target proteins were obtained. The active compound-target network of B. striata is shown in Fig. 2 (Supplementary file 1).

Fig. 2
figure 2

Network of active components and targets of B. striata. Blue diamonds indicate 216 human-derived target protein target nodes. The pink circle represents B. striata. Five green hexagons represent five active ingredients in B. striata (BJ1: MOL005755; BJ2: MOL005756; BJ3: MOL005761; BJ4: MOL005768; BJ5: MOL005776)

Full size image

Common targets of B. striata and UC

To confirm the therapeutic effect of B. striata on UC, we collected 2047 different target genes of UC and 216 target genes of B. striata from various databases. Through their crossover, we obtained 78 common target genes that are shown in Fig. 3 (Supplementary file 1).

Fig. 3
figure 3

Venn diagram of common targets between B. striata and UC. B. striata shares 78 putative targets with known UC targets

Full size image

Construction of the PPI network and core target screening

Seventy-eight overlapping genes of B. striata and UC were submitted to the STRING database. Moreover, the free points (HTR7, AKR1B10, CACNA2D1, and BDKRB2) were hidden. Finally, 74 target genes with moderate connectivity (degree ≥ 0.400) in the PPI network were obtained (Fig. 4). The tsv file was downloaded in STRING, and the PPI network was analysed with Cytoscape v3.9.0. According to the degree values, the top 10 core targets were identified as EGFR, SRC, ESR1, CCND1, ERBB2, PTGS2, PIK3CA, MMP9, MDM2, and MCL1 (Fig. 5), whose full names are provided in Table 5 (Supplementary file 1).

Fig. 4
figure 4

PPI network of targets for B. striata against UC. Each node represents a target, and the lines between nodes represent interconnections between genes

Full size image
Fig. 5
figure 5

The top 10 core targets of B. striata in treating UC. Each node represented a target for B. striata in the treatment of UC. The larger size means the higher degree value

Full size image
Table 5 The top 10 core targets of B. striata in treating ulcerative colitis
Full size table

GO/KEGG analysis

Furthermore, GO and pathway enrichment analyses were performed to explore the biological processes of 74 putative targets of B. striata in UC. As shown in Figs. 6, 7, and 8, the top 10 most significantly enriched GO biological processes, cellular components, and molecular functions were identified with their q-value and gene count. Biological processes were mainly related to the regulation of protein kinase activity, transmembrane receptor protein tyrosine kinase signalling pathway, regulation of protein serine/threonine kinase activity, and peptidyl-tyrosine phosphorylation. The cellular components were mainly related to the transferase complex, protein kinase complex, and cyclin-dependent protein kinase holoenzyme complex. Moreover, the molecular functions were mainly related to protein kinase activity, kinase activity, phosphotransferase activity, and protein tyrosine kinase activity. KEGG pathway enrichment analysis screening resulted in 211 signalling pathways. As Fig. 9 shows, the top 10 signalling pathways of B. striata in the treatment of UC were concentrated in pathways in cancer, the PI3K-Akt signalling pathway, metabolic pathways, microRNAs in cancer, endocrine resistance, proteoglycans in cancer, prostate cancer, human papillomavirus infection, the Rap1 signalling pathway, and chemical carcinogenesis (Supplementary file 1).

Fig. 6
figure 6

The results of the Gene Ontology (GO) biological process analysis. The X-axis represents the gene count, while the Y-axis represents the categories of biological processes (q-value < 0.05)

Full size image
Fig. 7
figure 7

The results of the Gene Ontology (GO) cellular component analysis. The X-axis represents the gene count, while the Y-axis represents the categories of cellular components (q-value < 0.05)

Full size image
Fig. 8
figure 8

The results of the Gene Ontology (GO) molecular function analysis. The X-axis represents the gene count, while the Y-axis represents the categories of molecular functions (q-value < 0.05)

Full size image
Fig. 9
figure 9

KEGG relational regulatory network. This network shows the relationship between the top 10 enriched pathways and 57 genes, and the size of the graph shows the number of pathways or genes connected

Full size image

Verification with molecular docking

To further validate the results of network pharmacological screening, we linked the five active compounds of B. striata to the hub targets EGFR and PIK3CA by molecular docking. Generally, the Vina score is negative; the lower the score is, the better the binding activity between the ligand and receptor is. As shown in Table 6 and Figs. 10 and 11, except for 2,3,4,7-tetramethoxyphenanthrene, the active components of B. striata have good binding activities to the pivotal targets EGFR and PIK3CA, and the binding activity to EGFR is better than that of PIK3CA (Supplementary file 1).

Table 6 The results of molecular docking
Full size table
Fig. 10
figure 10

Molecular docking analysis of the binding affinity of the four active compounds towards the hub target EGFR. A EGFR and 1-(4-hydroxybenzyl)-4-methoxy-9,10-dihydrophenanthrene-2,7-diol (MOL005755), Vina score =  − 8.6; B EGFR and 3-(p-hydroxybenzyl)-4-methoxy-9,10-dihydrophenanthrene (MOL005761), Vina score =  − 8.5; C EGFR and 4,7-dihydroxy-1-hydroxybenzyl-2-methoxy-9,10-dihydrophenanthrene (MOL005768), Vina score =  − 8.5; D EGFR and 1-(2,7-dihydroxy-4-methoxy-1-phenanthryl)-4-methoxyphenanthrene-2,7-diol (MOL005776), Vina score =  − 8.4

Full size image
Fig. 11
figure 11

Molecular docking analysis of the binding affinity of the four active compounds towards the hub target PIK3CA. A PIK3CA and 1-(4-hydroxybenzyl)-4-methoxy-9,10-dihydrophenanthrene-2,7-diol (MOL005755), Vina score =  − 6.5; B PIK3CA and 3-(p-hydroxybenzyl)-4-methoxy-9,10-dihydrophenanthrene (MOL005761), Vina score =  − 6.6; C PIK3CA and 4,7-dihydroxy-1-hydroxybenzyl-2-methoxy-9,10-dihydrophenanthrene (MOL005768), Vina score =  − 7.1; D PIK3CA and 1-(2,7-dihydroxy-4-methoxy-1-phenanthryl)-4-methoxyphenanthvene-2,7-diol (MOL005776), Vina score =  − 7.2

Full size image

Results of the animal experimental study

B. striata ameliorated DSS-induced colitis symptoms

To investigate the protective effect of B. striata on UC, we generated a mouse model of DSS-induced colitis. Mesalazine was used as a positive control in this study. As Fig. 12A shows, on the 7th day of modelling, the mice in the control group had shiny hair, normal diet and activity, dry stools, no anal bleeding, and normal weight gain. In contrast, the mice in the BS-L group, BS-M group, BS-H group, and mesalazine group had dull hair, depression, blood in the stool, anal bleeding, and significant weight loss (Fig. 13). However, after 10 days of treatment, all the remaining groups showed relief of symptoms after each dose compared to the model group.

Fig. 12
figure 12

B. striata ameliorated DSS-induced body weight loss, colonic shortening, DAI, and colon histopathology scores. A Anal lesions of mice in each group. B Change in body weight. C DAI score. D Colon length. E Colon histopathology scores. Values are presented as the mean ± SD. **P < 0.01 vs. the control group; ***P < 0.001 vs. the control group; #P < 0.05 vs. the model group; ##P < 0.01 vs. the model group; ###P < 0.001 vs. the model group

Full size image
Fig. 13
figure 13

Representative images of colonic tissues with HE staining

Full size image

Evaluation of the effectiveness of B. striata on UC based on weight loss, colon length, DAI score, and colon histopathology score was performed. On day 7 of DSS induction, the body weight of the model group decreased significantly compared to that of the control group and decreased daily thereafter. Compared with the treatment in the model group, the administration of medium-dose B. striata decoction, high­dose B. striata decoction, and mesalazine significantly alleviated body weight loss (P < 0.001, Fig. 12B). The DAI and colon histopathology scores are shown in Fig. 12C, E, and we found that compared with the control group, the model group had significantly increased values, whereas compared with those in the model group, the DAI and colon tissue histopathology scores in the mesalazine group and B. striata groups were decreased. In addition, the DAI and colon tissue histopathology scores in the B. striata groups decreased with an increase in the dose. DSS resulted in a significant shortening of the colon length. As shown in Fig. 12D, the colon length in the model group was significantly shorter than that in the control group (4.98 ± 0.25 vs. 6.73 ± 0.18, P < 0.001). However, compared with that in the model group, the colon length in the medium-dose and high-dose B. striata decoction, and mesalazine groups was significantly restored (6.58 ± 0.14 vs. 4.98 ± 0.25; 7.53 ± 0.20 vs. 4.98 ± 0.25; 7.43 ± 0.27 vs. 4.98 ± 0.25; P < 0.001) (Supplementary file 2).

B. striata inhibited the proinflammatory cytokines in colon tissues

Proinflammatory factors play a vital role in the pathogenesis and progression of UC. To determine whether B. striata exerts anti-inflammatory effects in DSS-induced colitis, we examined the expression levels of the proinflammatory cytokines TNF-α and IL-6 in colon tissues by ELISAs. We found that DSS significantly triggered the expression of proinflammatory cytokines, including TNF-α (394.01 ± 21.06 vs. 120.06 ± 2.12, P < 0.001) (Fig. 14A) and IL-6 (464.06 ± 38.29 vs. 160.69 ± 10.94, P < 0.001) (Fig. 14B), compared with those in the control group. However, the administration of medium­dose and high­dose B. striata decoction, and mesalazine inhibited the content of TNF-α (185.92 ± 20.26 vs. 394.01 ± 21.06; 145.69 ± 12.36 vs. 394.01 ± 21.06; 124.83 ± 5.38 vs. 394.01 ± 21.06; P < 0.001) (Fig. 14A) and IL-6 (206.95 ± 19.81 vs. 464.06 ± 38.29; 162.59 ± 12.35 vs. 464.06 ± 38.29; 177.20 ± 13.04 vs. 464.06 ± 38.29; P < 0.001) (Fig. 14B) significantly compared with that of the model group. These data demonstrated that medium­dose and high­dose B. striata decoction, and mesalazine could restrain the proinflammatory reaction in UC (Supplementary file 2).

Fig. 14
figure 14

The levels of the proinflammatory cytokines TNF-α (A) and IL-6 (B) were changed in each group. Values are presented as the mean ± SD. ***P < 0.001 vs. the control group; ###P < 0.001 vs. the model group

Full size image

B. striata ameliorated DSS-induced inflammation through the EGFR/PI3K/AKT signalling pathway

Inflammation plays an important role in the pathogenesis and course of UC, and the inflammatory response may be mediated in part by the EGFR/PI3K/AKT signalling pathway (Liu et al. 2021). Based on network pharmacology analysis and molecular docking technology, we found that the therapeutic effect of B. striata on UC may be related to the EGFR/PI3K/AKT signalling pathway. Therefore, we demonstrated through animal experiments that B. striata could inhibit inflammation in UC by regulating the expression of EGFR, PI3K, and p-AKT. Finally, RT-qPCR, Western blot, and ELISA results showed that the mRNA and protein expression levels of EGFR, PI3K, and p-AKT were markedly increased in the colon tissue of the model group (Fig. 15), which indicated that the EGFR/PI3K/AKT signalling pathway was activated in UC. Noteworthily, the rise of these targets was prevented by B. striata treatment. Combining the above results, we drew the following conclusion: B. striata may treat UC by inhibiting the activation of the EGFR/PI3K/AKT signalling pathway (Supplementary file 2).

Fig. 15
figure 15

The mRNA and protein levels of EGFR, PIK3CA, and p-AKT in colon tissues. The mRNA levels of EGFR (A) and PIK3CA (B) in each group was determined relative to the level in the control group (defined as 100%). The protein levels of EGFR, PIK3CA, and p-AKT are presented in C–I. Western blot–detected protein levels in each group was determined relative to the level in the control group (defined as 100%). Values are presented as the mean ± SD. **P < 0.01 vs. the control group; ***P < 0.001 vs. the control group; #P < 0.05 vs. the model group; ##P < 0.01 vs. the model group; ###P < 0.001 vs. the model group

Full size image

Discussion

UC is a chronic nonspecific intestinal inflammatory disease of unknown aetiology, a subtype of IBD, and is refractory and persistent, showing recurrence and the risk of intestinal perforation, toxic megacolon, and cancer, which strongly limit patients’ daily life and work. B. striata, as a classic Chinese herbal medicine, has astringent effects and effects on haemostasis, swelling, and muscle growth. Based on data mining of TCM for the treatment of UC, it was found that B. striata is a common Chinese herbal medicine for the treatment of UC that can protect the intestinal mucosa, promote the healing of intestinal ulcers, and improve intestinal inflammation and symptoms of blood in the stool, but its mechanism of action is unclear. This study analysed the active compounds, potential targets, and relevant pathways of B. striata for the treatment of UC through network pharmacology. Finally, 5 compounds and 74 target genes were identified and were significantly enriched in multiple UC-related pathways, such as the cancer pathway, PI3K-AKT signalling pathway, and metabolic pathways. Furthermore, the molecular docking results showed that the key active compounds in B. striata showed good binding affinity to EGFR and PIK3CA. This study comprehensively clarified the putative active components and multitarget mechanism of B. striata in UC and provided a theoretical basis for the clinical application of B. striata in the treatment of UC.

Among the identified compounds associated with this network, the majority of UC-related chemical components in B. striata are dihydrophenanthrene and phenanthrene compounds, especially 9,10-dihydrophenanthrene. Dihydrophenanthrene and phenanthrene are the main active components of B. striata, and numerous studies have shown that they have significant anti-inflammatory activity (Kovács et al. 2008; Lin et al. 2013; Wang et al. 2017). The anti-inflammatory mechanism of dihydrophenanthrene and phenanthrene may be related to the inhibition of inflammatory factors as well as the PI3K/AKT, NF-κB, and MAPK signalling pathways (Jiang et al. 2019a, b; Lin et al. 2013). In addition, many studies have demonstrated that dihydrophenanthrene and phenanthrene compounds have anticytotoxic activity and have significant curative effects on cancer-related diseases (Wang et al. 2017; He et al. 2017; Sun et al. 2016). This finding indicates that they not only play a role in the treatment of UC but also inhibit the carcinogenesis induced by colon inflammation, thereby controlling the transformation of UC to colon cancer. Interestingly, recent research identified 9,10-dihydrophenanthrene derivatives as potent noncovalent SARS-CoV-2 3CLpro inhibitors for the treatment of COVID-19 (Zhang et al. 2022). Thus, dihydrophenanthrene may also have antiviral effects.

Through network pharmacology, we finally obtained 74 potential target genes for UC, such as EGFR, SRC, ESR1, CCND1, ERBB2, PTGS2, PIK3CA, and other key targets. Through GO enrichment analysis, we discovered that the targets of B. striata for the treatment of UC were associated with various biological processes, such as regulation of protein kinase activity, regulation of protein serine/threonine kinase activity, peptidyl-tyrosine phosphorylation, and peptidyl-tyrosine modification. According to KEGG pathway enrichment analysis, we found that the PI3K-AKT signalling pathway, pathways in cancer, and metabolic pathways were the most important. Moreover, EGFR is a key target for prediction. This molecule initiates the PI3K/AKT signalling pathway, which is closely linked to the regulation of cellular functions such as cell proliferation, differentiation, and apoptosis. During intestinal inflammation in experimental colitis mouse models and the colon of UC patients, colonic EGFR expression is upregulated (Lu et al. 2014). EGFR signalling is also an important marker of colitis-related colorectal carcinogenesis (Ko et al. 2019; Singh et al. 2018). PI3K/AKT is a key downstream signalling pathway of EGFR, and abnormal activation of the PI3K/AKT signalling pathway is associated with the development of UC, as well as promoting colon cancer cell proliferation, prolonging cell survival, inhibiting cell apoptosis, and participating in angiogenesis, leading to invasion and metastasis of colon cancer (Jiang et al. 2019a, b; Zhang et al. 2021). In addition, the PI3K-AKT pathway is closely related to the regulation of the inflammatory response during UC progression. The expression levels of inflammatory factors such as TNF-α, IL-1β, and IL-6 in UC were increased by abnormal activation of the PI3K/AKT signalling pathway (Li et al. 2019a, b; Li et al. 2019a, b). Furthermore, TNF-α is an important proinflammatory cytokine involved in the pathogenesis of UC, inducing the pathological opening of the intestinal tight junction barrier, increasing intestinal epithelial permeability, and disrupting intestinal barrier function (Xiang et al. 2021; Shin et al. 2020). Therefore, anti-TNF-α therapy has been widely used in clinical practice as a suitable method of UC treatment. In addition, our molecular docking results confirmed that the active ingredients of B. striata are effectively bound to the hub genes EGFR and PIK3CA. Therefore, we hypothesized that B. striata might have a therapeutic effect on UC by regulating the EGFR/PI3K/Akt signalling pathway to reduce the inflammatory response.

To investigate the effect of B. striata on the inflammatory response and the role of the EGFR-regulated PI3K/AKT signalling pathway in UC, we established a mouse model of colitis. The model group showed more severe damage and acute colitis, and these abnormalities were alleviated by B. striata or mesalazine. In addition, B. striata and mesalazine ameliorated body weight loss, shortening of the colon length, DAI, pathological changes, and secretion of inflammatory cytokines, and there were no significant differences between the BS-M and BS-H groups and the mesalazine group. To verify whether the mechanism by which B. striata treats UC is related to the EGFR/PI3K/AKT signalling pathway, we detected the mRNA expression levels of EGFR and PIK3CA by RT-qPCR and detected the protein expression levels of EGFR, PIK3CA, and p-AKT by Western blot and ELISA. The results indicated that the EGFR/PI3K/AKT signalling pathway is abnormally activated in the DSS-induced colitis model. However, the expression levels of EGFR, PI3K, and p-AKT in colon tissue in the BS-M and BS-H groups were significantly lower than those in the model group. Combined with network pharmacology analysis, molecular docking, and in vivo experiments, the results showed that B. striata can effectively downregulate the expression of EGFR, PI3K, and p-AKT, which may help to improve UC.

Conclusion

In this study, we initially screened 5 active compounds of B. striata. The key targets, including EGFR and PIK3CA, were obtained and verified by molecular docking, and the top EGFR/PI3K/AKT signalling pathway was further analysed. We found that B. striata could inhibit inflammatory cytokines, reduce the inflammatory response to UC, and alleviate the symptoms of UC in mice. Furthermore, B. striata reduced the mRNA expression of EGFR and PIK3CA and the protein expression of EGFR, PIK3CA, and p-AKT. Therefore, we believe that B. striata can affect the EGFR/PI3K/Akt signalling pathway, resulting in an anti-UC effect.