Protective Effects of Aqueous Extracts of the Herb of Paederia scandens (Lour.) Merr. against HCl/EtOH-Induced Gastric Ulcer in Rats: Involvement and Inhibitors’ Identification of NF-κ B Signaling

,


Introduction
Gastric ulcer, one of the most constant digestive disorders with an increasing incidence and prevalence, is afecting millions of people yearly [1,2]. Various stimulus including alcohol and nonsteroidal anti-infammatory medications can cause increasing incidences of gastric ulcer [3]; the imbalance between these stimulating factors and gastric mucosal protective factors (mucin and prostaglandins) leads to gastric mucosal lesions. Te commonly used antiulcer therapies in clinic include antiacid drugs and H 2 receptor inhibitors, but these therapies are always accompanied with adverse efects such as low healing rates and high recidivism rates [4,5]. Accordingly, it is urgently needed to explore an efective and safe alternative therapy for treating gastric ulcer. Numerous studies have indicated that infammatory response has an important role in the pathogenesis of gastric ulcer [6]. Te development of gastric ulcer is accompanied by a signifcant release of infammatory factors that can lead to activation of nuclear factor-κB (NF-κB) and the translocation of NF-κB into the nucleus. Te entry of NF-κB into the nucleus triggers the transcription of various infammatory mediators (tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), and interleukin-6 (IL-6)), which trigger local or systemic infammation and cause cellular or organ dysfunction [7,8,9]. Terefore, inhibiting infammatory response is an efective method to treat gastric ulcer. Traditional Chinese medicine has obvious advantages in treating gastric infammatory response [10,11].
Paederia scandens (Lour.) Merr., "Jishiteng" in Chinese, recorded in Chinese Pharmacopoeia (1977 Edition) [12], has been used as a traditional medicine to treat gastritis for hundreds of years. Various ingredients have been reported to occur in P. scandens, including favonoids and iridoids [13,14]. Modern pharmacological studies have revealed that P. scandens have anti-infammatory and antioxidative activities [15,16] and has been used to treat gastric ulcer, rheumatoid arthritis, and diarrhea [15,17,18]. However, its antigastric ulcer mechanistic and efective substances studies have not been fully elucidated.
In this work, we investigated the protective efects of P. scandens on HCl/EtOH-induced gastric ulcer in rats and explored the involvement of NF-κB signaling in the efects; furthermore, we explored its efective substances by using pharmacophore modeling and molecular docking, which provides pharmacological justifcations for the functional food use of P. scandens for treating gastric disease.

Preparation of Extract.
PS-AE was extracted as described in [13]. Specifc operations are as follows: P. scandens (1.0 kg) was extracted twice with 10 L water (2 h each time) under refux condition. After fltering the extract, the fltrate was collected and concentrated to 200 mL by rotary evaporation at 50°C under reduced pressure. Ten, the concentrate was lyophilized to obtain powders; the yield of PS-AE was 12.65%.
MS data were acquired in fast chromatography MS 2 mode; the mass spectrometer parameters were set as follows [19]: the ESI was used in negative ion mode (ESI−) and in positive ion mode (ESI+). Te following settings were used: the spray voltage was 2.5 kV(−) and 3.5 kV(+), the UHPLC-MS/MS mode was applied with an Orbitrap resolution of 12,0000 for full-MS and 15,000 for dd-MS 2 , the isolation window (m/z) was 2, the RF Lens% was 70, the sheath gas pressure was 45 Arb, the auxiliary gas pressure was 15 Arb, the sweep gas pressure was 0 Arb, the capillary temperature was 320°C, vaporizer temperature was 350°C, the scanning range was m/z 80∼800, and the stepped normalized collision energies (NCE) were 15, 30, and 45 eV.

Screening for Potential NF-κB Inhibitors in P. scandens
Using Pharmacophore Modeling. Tis study used the common feature pharmacophore generation module in Discovery Studio software to construct a small molecule ligandbased model of the HipHop pharmacophore. 6 compounds with anti-NF-κB activity from references are selected as the training set and used to generate HipHop pharmacophore model. From this approach, 10 pharmacophore models were successfully generated. Ten, except for some compounds with no activity or unclear activity, 30 compounds with anti-NF-κB activity from references are selected as the testing set. Te training set was used to train the NF-κB pharmacophore model, while the test set was used to evaluate its predictive performance. 10 pharmacophore models were then evaluated using the ligand profler module in Discovery Studio to select the model with the best predictive power. A total of 99 components of P. scandens were collected from the TCMSP (https://tcmsp-e.com/) database. Based on this preprocessing, the anti-NF-κB activity of 99 compounds in P. scandens was predicted and other parameters were set as default. Te structures of both training set and test set compounds are provided in Supplementary Materials (the training set is marked by * ).

Molecular Docking Studies.
A molecular docking study of small molecule compounds predicted (pharmacophore model) using the CDOCKER module in Discovery Studio software. Te 3D structure of the compound built by ChemOfce software was saved in * mol2 format, and its energy was minimized. Te 3D structure of the target protein was downloaded from the PDB data (https://www.rcsb.org/ ), and Discovery Studio 2020 software was used to perform operations such as water removal and hydrogenation of the protein and generate an efective single 3D conformation by minimizing the energy.
2.6. Experimental Animals. Male Wistar rats (3-8 weeks old with 180 ± 20 g body weight) purchased from Jinan Pengyue Experimental Animal Breeding Co., Ltd. (SYXK (Lu) 20190003) were used to study the protective efects of PS-AE on HCl/EtOH-induced gastric ulcer in rats. Te experimental unit used license number SYXK (Yu) 2020−0004, and all animals were kept under environmentally controlled conditions (22 ± 1°C and 50 ± 10% relative humidity) with a 12 h light/dark cycle and were allowed to eat, and drink, at will during the experimental period. All experimental procedures including the rat model adopted in this study conformed to the Basel Declaration and the ethical guidelines of the International Council for Laboratory Animal Science (ICLAS). Te experimental protocol was approved by the Animal Ethics Committee of Henan University of Traditional Chinese Medicine, ethics number: DWLL202009059. Te rats were allowed to acclimatize to the laboratory environment for one week before the experiment.

Animal
Modeling, Grouping, and Drug Administration. 40 rats were randomly divided into 5 groups of 8 rats each. Gastric ulcer was induced in rats by HCl/EtOH (take an appropriate amount of HCl and dissolve it in 60 : 40 ethanol : water (v : v) solution to get a concentration of 150 mmol/L HCl/EtOH.) as described previously [20]. Rats in the administration group received PS-AE (50 and 100 mg/kg) or ranitidine (35 mg/kg) twice a day for three days, whereas the control group and model group received vehicle (1% CMC-Na) twice a day for three days, the experimental doses of PS-AE and ranitidine were determined according to references and the result of pre-experiments [20]. Te animals were fasted overnight with free access to water before the experiment, 30 minutes after the last PS-AE treatment; 2 mL of HCl/EtOH was administrated via oral gavage. 1 h after the HCl/EtOH treatment, the rats were anesthetized, the stomachs were removed and immediately washed in ice-cold saline, images were recorded with a digital camera, and the area (mm 2 ) of each mucosal erosion lesion was measured using ImageJ software.

Gastric Histopathology and Mucosal Glycoprotein Assay in
Rats with Gastric Ulcer. Hematoxylin-eosin (HE) staining was used for histopathological observation of the gastric mucosa, and periodic acid-Schif (PAS) staining was used to assess gastric mucosal glycoprotein. Gastric tissues from each group were fxed in 4% paraformaldehyde solution for 24 h, dehydrated through an ethanol (75-100%), and then cleared with xylene. Te gastric samples were embedded in parafn, selected for typical ulcerated areas, cut into 5.0 μm, and stained using HE and PAS, and then placed under a light microscope for observation.

2.9.
Determination of SOD, MDA, GSH-Px, TNF-α, IL-1β, and IL-6 in Gastric Tissues. Te stomach tissue was rinsed with cold PBS and the typical ulcerated area was homogenized in phosphate bufer (pH 7.4); the homogenate was centrifuged at 10,000 g for 15 min at 4°C. Te levels of SOD, MDA, GSH-Px, TNF-α, IL-1β, and IL-6 in the supernatant were detected by biochemical detection kits.

Cell
Culture. Te RAW264.7 macrophage cell line was purchased from Procell Life Science and Technology Co., Ltd. RAW264.7 cells were cultured in high-glucose DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 100 U/mL streptomycin. Ten, they were placed in a 37°C thermostat incubator and in a humid atmosphere with 5% CO 2 .

Measurements of NO, TNF-α, and IL-6 Production.
RAW264.7 cells (2 el10 6 cells/mL), seeded in 6-well plates and cultured overnight for adhere, were then treated with various concentrations of PS-AE (0, 12.5, and 25 μg/mL) and paederoside (0, 12.5, and 25 μM) in the presence of LPS (1 μg/mL) for 24 h. Supernatants were collected, and NO content was then measured by the Griess reagents, and TNFα and IL-6 levels were then measured by the ELISA kits.
2.14. Statistical Analysis. Statistical analysis was performed using GraphPad Prism 8.0.1 (GraphPad Software, CA, USA). Results are expressed as mean ± standard deviation (SD); multiple group comparisons were performed using one-way ANOVA, followed by the Tukey multiple-comparison test. Signifcance was set at p < 0.05 and p < 0.01.

Identifcation of the Constituents of PS-AE.
A total of 24 compounds were tentatively identifed based on accurate mass, retention times, major fragment ions, online databases and published articles ( Figure 1 and Table 1). Among these, compounds 2, 4, 9, 11, 16, and 18 were found to be consistent with the retention times, secondary fragmentation, and other information of the standard paederosidic acid, rutin, quercetin, paederoside, linarin, and kaempferol. Figure 2(c), the lesion was barely visible in the control gastric tissue. Compared to the control group, the HCl/EtOH-induced model group showed obviously severe lesions and signifcantly larger ulcerated areas (50.35 ± 12.41 mm 2 ). PS-AE showed a signifcant reduction in the lesion site and ulcer area compared to the model group (50 mg/kg: 8.801 ± 4.22 mm 2 , 79.14% ulcer healing rate; 100 mg/kg: 2.22 ± 2.66 mm 2 , 95.47% ulcer healing rate) in a dose-dependent manner. Statistical analyses of the ulcer area are shown in Figure 2(b). HE staining results ( Figure 2(d)) showed that the gastric mucosa in the control group was intact without any visible bleeding or ulceration. Among the model group, the stomachs presented with severe mucosal ulcers, bleeding, edema, and infammatory cell infltration. Te PS-AE treatment attenuated this damage in a dose-dependent manner. In fact, gastric mucosal glycoprotein form a viscoelastic mucus gel layer that protects and lubricates the lower epithelial layer of the gastric tissue and therefore plays a vital defensive role in preventing gastric ulcer [22]. PAS staining results ( Figure 2(d)) showed that the staining intensity of gastric mucosal glycoprotein was potent in the control group, where the intensity of PAS staining in the model group was obviously weakened, indicating the injuries of mucosal glycoprotein. Compared to the model group, PS-AE treatment markedly reversed the weakened staining intensity, indicating the protective efects on glycoprotein in the gastric mucosa. To sum up, PS-AE signifcantly attenuated the HCl/ EtOH-induced gastric ulcer in rats.

PS-AE via NF-κB Pathway Plays a Terapeutic Role and
Implicates Infammation and Oxidative Stress. As shown in Figure 3(a), the levels of TNF-α, IL-6, and IL-1β in the model group were signifcantly higher than those in the control group. Compared to the model group, PS-AE signifcantly inhibited the increase in TNF-α, IL-1β, and IL-6 levels. As shown in Figure 3(b), versus control group, the MDA levels in the model group were signifcantly increased, but the SOD and GSH-Px levels were decreased. Compared to the model group, PS-AE signifcantly inhibited the increase in MDA levels but the decrease in SOD and GSH-Px levels. As shown in Figure 3(c), versus control group, the phosphorylated protein levels of IKKα, IκBα, and P65 and the protein level of COX-2 in the gastric tissue of the model group were abnormally up-regulated. PS-AE signifcantly decreased the HCl/EtOH-induced upregulation on phosphorylation protein levels of IKKα, IκBα, and P65 and protein levels of COX-2 in gastric tissues.

Screening of 13 Potential Inhibitors of NF-κB in P. scandens
Using Pharmacophore Modeling and Molecular Docking. In this study, pharmacophore models were constructed based on 6 training sets of molecules. A total of 395 pharmacophore elements, including 217 hydrogen bond acceptors, 159 hydrogen bond donors, 15 hydrophobic groups, 2 positive charge centers, and 2 aromatic ring centers, were extracted by feature mapping, and the results are shown in Figure 4(a). Tese characteristic elements were selected, and a total of 10 pharmacophore models were generated using the HipHop method, the resulting parameters of which are shown in Table 2. Te scores refer to the match between the molecules in the training set and the pharmacophore model, and it can be seen that the 10 pharmacophores generated all scored over 60, demonstrating that the molecules selected for the test set share common features, with pharmacophore 01 scoring 74.808 and the rest of the pharmacophores scoring relatively close to each other. Combining the results of the pharmacophore thermogram, where the warm shades represent a high response to ligands and the cool shades represent a low response to ligands, it can be seen that pharmacophore 01 is more responsive to the active compounds in the test set and less responsive to the inactive compounds (Figure 4(b)), so pharmacophore 01 was chosen as the subject for subsequent studies. From the results, it can be observed that the preferred pharmacophore 01 consists mainly of 4 hydrogenbonded receptors, 1 hydrophobic center, and 2 hydrogen bond donors (Figure 4(a)).

PS-AE and Paederoside Could Signifcantly Inhibit the Activation of NF-κB p65 and the Production of Infammatory
Factors in Downstream Cells. As shown in Figure 5(b), PS-AE showed no cytotoxic efects at 12.5-50 μg/mL (cell viability was above 90%), and the viability was slightly decreased at a high concentration (100 μg/mL). Tus, the concentrations 12.5 and 25 μg/mL were used in subsequent experiments. In addition, at all concentrations of 6.25-100 μM, the cell viability of paederoside was above 90%, so 12.5 and 25 μM were used in the subsequent experiments. As shown in Figure 5(c), PS-AE and paederoside signifcantly decreased the LPS-stimulated upregulation on phosphorylation protein levels of NF-κB p65 in RAW264.7 cells. As shown in Figure 5

Discussion
Paederia scandens (Lour.) Merr., as an edible and medicinal plant, can be used to treat many digestive diseases. Several historical documents have described the protective efects of P. scandens on gastric ulcer. However, the mechanism of action of this herb for treating gastric ulcer remains unclear. As our results indicated, PS-AE treatment signifcantly attenuated HCl/EtOH-induced gastric tissue damage, which was probably associated with inhibition activities of oxidative stress and pro-infammatory mediators regulated by NF-κB signaling pathway.
As many studies reported, HCl/EtOH-induced gastric ulcer model has been used as a robust model resembling pathogenesis of gastritis or gastric ulcer [23,24]. Ethanol is a noxious factor that can negatively infuence gastric mucosa [25]; numerous studies have demonstrated that the pathological characteristics of gastric mucosa in alcohol-treated experimental animals are similar to those in humans consuming excessive alcohol such as abnormal gastric acidity and stomach bleeding [26]. Hydrochloric acid is a strong acid that can directly damage the gastric mucosa [20].
In this study, we prepared aqueous extracts (decoction) of P. scandens using the traditional decocting method for Chinese medicine. Tese decoctions are mixed solvents that can take on various forms, such as turbidity, suspension, gel, colloid, and solvent. Te colloid and gel components can adhere to the surface layer of the gastric mucosa, which in turn can help block the erosive efects of gastric acid [27]. In addition, the decoctions are generally weakly acidic, which allows them to bufer the acidic gastric juice and achieve an acid-suppressing efect. Te decoction contains nutrients like protein and polysaccharides that can provide direct   Table 1.  Journal of Food Biochemistry nourishment to the epithelial cells of gastric mucosa and promote metabolism [28]. In conclusion, PS-AE suspension may physically protect the gastric mucosa, but its medicinal substances still need to be explored in our further study. In our pre-experiment, we investigated the protective efect of PS-AE at various doses (25,50,100, and 200 mg/kg) on HCl/EtOH-induced ulcers in rats. Results showed that PS-AE at 50 and 100 mg/kg exerted a signifcant protective efect on HCl/EtOH-induced gastric ulcers in rats in a dosedependent manner. Terefore, corresponding assays at doses of 50 and 100 mg/kg in the subsequent experiments were performed, and administration of the herbal extract twice daily for three days was performed following routine pharmacological studies as literatures extensively reported [20,29,30]. Clinical treatment drugs for gastric ulcer include proton pump inhibitors and H 2 receptor antagonists, which are efective in reducing gastric acid secretion and protecting the gastric mucosa when administered 30 minutes before a meal [22]. Terefore, our study followed this approach by modeling with HCl/EtOH 30 minutes after the last dose to evaluate the protective efect of PS-AE on gastric ulcers. Te results showed that HCl/EtOH-induced severe gastric tissue damage in the model group, accompanied by a great number of visible hemorrhage spots. Compared with the model group, the ulcer area and bleeding points were signifcantly decreased in PS-AE-treated rats, indicating the efectiveness of PS-AE in reducing HCl/EtOH-induced gastric tissue damage.

Journal of Food Biochemistry
Te gastric tissue could defend against various endogenous and exogenous injuries [31]. Te layer of mucus, as the   Figure 4: Screening of 13 potential inhibitors of NF-κB in P. scandens using pharmacophore modeling and molecular docking studies: (a) results of pharmacodynamic characteristic elements. Te green sphere represents the hydrogen bond acceptor, the purple sphere represents the hydrogen bond donor, the blue sphere represents the hydrophobic group, the red sphere represents the positive charge center, and the yellow sphere represents the aromatic ring center; preferred pharmacophore model 01. It contains mainly 4 hydrogen-bonded receptors, 1 hydrophobic center, and 2 hydrogen bond donors. (b) Ligand profler heat map. (c) Te matching pattern diagram between pharmacophore model 01 and 14 potential NF-κB inhibitors in P. scandens. A-M are linarin, scandoside methyl ester, asperuloside, paederoside, paederosidic acid methyl ester, kaempferol 7-O-β-D-glucopyranoside, deacetyl asperulosidic acid methyl ester, rutin, geniposide, astragalin, quercimeritrin, quercetin, hirsutrin, and deacetyl asperulosidic acid. (d) Te interaction of NF-κB p65 with paederoside was studied by molecular docking, and the docking results were expressed in 3D and 2D formats. Note: Each row represents a pharmacophore model. Te meaning of each column is as follows (using 01 as an example). Features: HDDA, this pharmacophore contains 1 hydrophobic feature, 2 hydrogen bond donors, and 1 hydrogen bonding acceptor feature. Rank: 74.808, the scoring value for this pharmacophore model is 74.808. Direct hit: 111111, the potency profle of this pharmacophore model is matched to all 6 small molecules. Partial hit: 000000, this pharmacophore model matches the 6 small molecule fractions with a number of potency profles of 0. Max ft: 4, maximum match value is 4, i.e., all 4 potency characteristics can be matched. essential defense, protects the tissue from attacking substances such as stomach acid. As a vital defensive role in protecting against gastric ulcer, mucus glycoproteins form a viscoelastic mucus gel layer function to protect and lubricate the underlying epithelium of the gastric tissue [32]. HE staining results showed that HCl/EtOH caused edema and infammatory cell infltration in the gastric mucosa; these phenomena in the gastric mucosa were signifcantly improved after PS-AE treatment. Te PAS staining results showed that HCl/EtOH caused by great defciency of mucosal glycoprotein in gastric tissues. When treated with PS-AE, the condition of mucosal glycoproteins in gastric tissues was obviously improved, which indicated the protective efects of PS-AE on the gastric mucus layer. It has been shown that the pathogenesis of gastric ulcer is closely associated with oxidative stress [33]. Oxidative stress is produced by a large amount of reactive oxygen species (ROS) which then causes lipid peroxidation and production of MDA [34]. SOD and GSH-Px are two vital antioxidants that can clear free radicals in the body, protecting the body from oxidative stress damage [35][36][37]. Te results obtained in our work showed that PS-AE could protect against HCl/ EtOH-induced gastric tissue damage in rats by improving oxidative stress.
NF-κB is the most prominent pathway regulating infammation, and abnormal regulation of NF-κB may cause autoimmune diseases, chronic infammation, and some cancers [38]. It has been shown that HCl/EtOH stimulation results in activating NF-κB which leads to phosphorylation of IκBα and P50/P65 heterodimer, followed by translocation to the nucleus, thus prompting cells to secrete infammatory factors (e.g., TNF-α, IL-1β, and IL-6) and inducing an infammatory response [39]. Elevated levels of TNF-α could stimulate leukocytes to secrete more IL-1β and IL-6, thus causing even more potent infammatory response [40]. IL-1β is involved in ulcer formation [41], and elevated IL-1β exacerbates ulcer formation. Increased secretion of IL-6 stimulates neutrophil sites, and triggers oxidative stress, which leads to gastric mucosal damage [42]. COX-2 regulates the levels of PGE2, which enhances gastric mucosal repair [43]. Te results of this study showed that PS-AE signifcantly decreased the levels of pro-infammatory mediators such as TNF-α, IL-6, and IL-1β, suggesting that the protective efects of PS-AE against HCl/EtOH-induced gastric ulcer and involvement of NF-κB signaling pathway in the efects.
In recent years, molecular simulation technologies such as pharmacophore model and molecular docking have been utilized for new drug research and development [44,45]. Te HipHop model is used to identify chemical features common to a range of small ligand molecules and automatically generates pharmacophore models based on the superposition of these common characteristic structures; ligand-based pharmacophore strategy has been universally used to screen small molecule-leading compounds in drug development [46]. Molecular docking is a potent method for analysing molecular interaction [47]. It was found that proinfammatory cytokines are secreted at the onset of gastric ulcer and play a key role in acute phase infammation as well as in regulating the severity of gastric ulcer [22,48,49]. NF-κB is the main transcription factor that amplifes the infammatory response such that the NF-κB infammatory cascade in gastric ulcer highlights that inhibition of NF-κB activity plays a key role in the development of gastric ulcer [50]; [51,52]. Pharmacophore and molecular docking results show that there are 13 compounds included the major active substances in P. scandens [53]; [15], including 7 iridoids and 6 favonoids. Interestingly, paederoside, a substantial component of P. scandens, has a high ft value to the optimal pharmacophore model and has powerful binding abilities to NF-κB p65, so it was selected for in vitro validation in this study.
LPS-stimulated RAW264.7 macrophages are considered a canonical model for infammatory research. Upon LPS binding to toll receptor 4 (TLR4) in immune cells, such as macrophages and neutrophils, a great amount of infammation related factors (e.g., NO, TNF-α, IL-1β, and IL-6) was released [54]. NO, as an essential infammatory factor, is catalysed by inducible nitric oxide synthase (iNOS), which often causes oxidative stress and cellular damage [5]. TNF-α could trigger an infammatory response by promoting neutrophil infltration [55]; TNF-α at high concentration   NO production was measured using Griess assay, and levels of cytokines including TNF-α and IL-6 were analysed using ELISA kits (n � 6). Results are expressed as the mean ± SD, * * p < 0.01 vs. model group, and ## p < 0.01 vs. control group.
promote the secretions of IL-1β, IL-6, and other cytokines, triggering a series of downstream infammatory response [56,57]. Te results showed that PS-AE and paederoside signifcantly inhibited the LPS-induced activation of NF-κB p65 in RAW264.7 cells and suppressed the production of downstream cytokines NO, TNF-α, and IL-6. Our fndings demonstrated the protective efect of PS-AE on HCl/EtOH-induced gastric ulcer in rats and provide experimental evidence of P. scandens against gastric ulcer. Gastric ulcer in animals induced by diverse reagents/ means including stress, pylorus ligature, acetic acid, and indomethacin have been widely reported and used to investigate the efcacy of potential agents [58]. Te HCl/ EtOH-induced gastric ulcer is commonly used and considered as a reliable model for studying the pathogenesis of gastritis and gastric ulcers, for its superiority in rapid induction of ulcer which greatly resembles gastric mucosal damage in human beings [59]. Although inconsistent with therapeutic treatment as common human-used drugs, prophylactic treatment, and dosing arrangement are widely accepted in recent experimental studies for primary efcacy evaluation [60,61]. In our further study, a therapeutic treatment in model animals is warranted to evaluate the pharmacodynamic characteristics of PS-AE. In addition, we investigated the involvement of NF-κB signaling in the protective efect of PS-AE on gastric ulcer in vitro and in vivo. To further elucidate the mechanism of action of PS-AE against gastric ulcer, bioactive fraction/ingredient and crucial target/pathway of the PS-AE will be studied by using chemical analyses and biological technologies such as RNAsequencing and knock-out animals.

Conclusion
Te fndings of this study demonstrated that the aqueous extract of the herb of Paederia scandens exerts a protective efect on HCl/EtOH-induced gastric ulcer in rats, which is probably associated with inhibition of the activities of oxidative stress and pro-infammatory mediators regulated by NF-κB signaling pathway ( Figure 6). In addition, 13 potential NF-κB-targeting inhibitors are probably efective substances contributing to the herb's pharmacological efect. It provides pharmacological justifcations for the functional food use of Paederia scandens for treating gastric diseases.

Data Availability
Te data used to support the fndings of this study are available from the corresponding authors upon reasonable request.

Conflicts of Interest
Te authors declare that they have no conficts of interest associated with the manuscript.

Authors' Contributions
Te published version of this work has been reviewed and approved by all authors. Qingxia Li and Hui Guo conducted experimental research and wrote the manuscript. Man Gong, Yang Zhang, and Lianhe Yang collected and analysed the data. Jing Wang and Pei Wang conducted the manuscript revision. Zhimin Wang, Erping Xu, and Liping Dai designed and supervised the study. Qingxia Li and Hui Guo contributed equally to this work.  Figure S1: UPLC chromatograms of PS-AE and the 6 components in mixed reference solution (the numbers in the fgure indicate the number of chromatographic peaks: (1) paederoside acid, (2) rutin, (3) paederoside, (4) quercetin, (5) linarin, and (6) kaempferol). Figure S2: PS-AE dose dependently decreased the ulcer area in model rats. Table S1: the structures of both training and test set compounds. (Supplementary Materials)