Interactions between phenolic constituents of Scutellaria salviifolia and key targets associated with inflammation: network pharmacology, molecular docking analysis and in vitro assays

Abstract Scutellaria salviifolia Benth. (SS), an endemic plant for Turkey, is used for gastric ailments as folk medicine. In this study, we aimed to uncover the underlying molecular mechanisms with the help of network pharmacology and molecular docking analysis in the inflammation processes of gastric ailments. Gene enrichment analysis and target screening were carried out. Experimental validation was performed via cytokines of nitric oxide (NO) and interleukin-6 (IL-6) in LPS stimulated RAW 264.7 cells. Furthermore, antioxidant activity studies were performed by radical scavenging effects on different radicals. A total of 144 targets were listed for the isolated compounds where 26 of them were related to selected inflammation targets. According to the gene enrichment analysis, HIF1 signaling pathway and TNF signaling pathway were found to be involved in inflammation. We also defined AKT1, TNF, EGFR, and COX2 as key targets due to the protein-protein interactions of 26 common targets. The extract inhibited NO and IL-6 production at 100 and 200 µg/mL, while flavonoid-rich fraction possessed significant anti-inflammatory activity at the concentration of 50 and 100 µg/mL via NO and IL-6 production, respectively. It is thought that the anti-inflammatory effects of extracts, fractions and pure compounds were achieved by reducing NO and IL-6 levels via regulating the NF-κB pathway or reducing NO production by suppressing iNOS through the HIF-1 pathway when evaluated together with the results of network analysis and literature. Anti-inflammatory activities of the extract and fractions were promising and comparably with S. baicalensis, commonly used for its anti-inflammatory activity. Graphical Abstract Communicated by Ramaswamy H. Sarma


Introduction
Genus Scutellaria L. (Lamiaceae), which contains about 350 species, is widespread throughout the world (Shang et al., 2010). Scutellaria species are one of the most prominent plants used in traditional therapy and have many uses in the treatment of inflammation, cancer, hypertension, cardiovascular diseases, bacterial and viral infections, gastrointestinal disturbances such as diarrhea, cramps, nausea, vomiting (Baytop, 2021;Liu & Cheng, 2012;Li-Weber, 2009;Meng et al., 2013). Phytochemical studies on Scutellaria genus showed the presence of flavonoids, iridoid glucosides, phenylethanoid glycosides, diterpenoids, triterpenoids, alkaloids (Shang et al., 2010). While S. baicalensis is generally used in the treatment of inflammatory diseases such as dermatitis, gingivitis, and ulcer for its anti-inflammatory activities (Arweiler et al., 2011;Lin et al., 2016;Seok et al., 2016), S. barbata is used to treat cancer due to its antitumor properties (Perez et al., 2010;Rugo et al., 2007). S. baicalensis was listed as a nationally protected plant (class III conserved plant) due to the marked decrease of wild resources in China (Yuan et al., 2010). This fact has led to the investigation of other Scutellaria species with similar biological activity (Salini et al., 2013). Scutellaria species are represented with 25 taxa in the flora of Turkey and 14 taxa of which are endemic (Cicek, 2008). S. salviifolia, the endemic species was selected for some biological activity studies. S. salviifolia is used for gastric ailments in folk medicine (Quattrocchi, 2012;Tuzlaci, 2016). In Turkey, some Scutellaria species are used traditionally because of their anti-inflammatory activity. S. albida is used in the treatment of cold, and S. orientalis is used in the treatment of rheumatism and for wound healing (Tuzlaci, 2016). Additionally, it has been reported that S. salviifolia contains various phenylpropanoids, which have cytotoxic activity against different types of cancer cells as dRLh-84, HeLa, S-180, P-388/D1 (Dogan, 2018;Saracoglu et al., 1995). It has been identified that the polyphenolic compounds contributed to the antioxidant behaviors of medicinal plants. Therefore, there is a correlation between polyphenolic contents and antioxidant activities of plants (Cai et al., 2004;Tang et al., 2004). It is believed that the antioxidant properties of plant extracts are related to their therapeutic effects in various diseases such as inflammation and cancers. Inflammation is a common pathophysiological sign of complex diseases and an important part of immune pathogenesis (Ye et al., 2021;Yoon et al., 2009;Yu et al., 2018).
In this study, we aimed to use network pharmacology, a rising trend for drug discovery studies or reconstruction of current drugs that review relationships between molecules, diseases, and signaling pathways by predicting molecules and target interactions. The mechanisms of pharmacological effects of the current drugs could also be clarified by this method . Network pharmacology is a very encouraging method for discovering the mechanisms of herbal products, considering that herbal products used in therapy contain multiple compounds and their therapeutic effects are the results of a combination of these molecules (Kutluay & Diker, 2020).
According to traditional usage of S. salviifolia and other Scutellaria species, in the present study, we aimed to discuss the role of targets in gastric ailments and the anti-inflammatory effects of the plant using network pharmacology and molecular docking analysis. To validate the results obtained from network pharmacology analysis, in vitro anti-inflammatory activity studies were assessed on LPS-stimulated RAW 264.7 cells via NO and IL-6 cytokines. On the other hand, we investigated antioxidant activity of the extract and fractions through ABTS (2,2 0 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt), DPPH (2,2-diphenyl-1-picrylhydrazyl), NO and SO (superoxide) radicals. Furthermore, we expected to find out interactions between usage of the plant, inflammatory process, and key targets of the molecules isolated from S. salviifolia.

Plant material
The aerial parts of Scutellaria salviifolia Benth. were collected from Mamak, Ankara in June 2012 in stony slopes and were determined by Prof. Dr. Hayri Duman (Gazi University, Faculty of Science, Department of Biology). Voucher specimen is deposited in the Herbarium of the Faculty of Pharmacy, Hacettepe University, Ankara, Turkey [HUEF 12003]. The plant material was dried in shade, for protection from direct sunlight to minimize enzyme activity and to prevent the possible changes in phytochemical content for approximately one week (until dryness).

Preparation of the extract and fractions
Preparation of the aqueous extract, fractions and isolation process of pure compounds were described previously (Dogan et al., 2019). Briefly, air-dried aerial parts of S. salviifolia (SS, 682 g) were extracted seven times with MeOH (100%; with purity of 97%, GC) at 40 C. The combined extracts were evaporated under vacuum to give 143 g of crude methanol extract. The crude methanolic extract was dissolved in H 2 O (500 mL) and was extracted with petroleum ether to remove chlorophylls and other lipophilic compounds. 35 g of aqueous fraction of methanolic extract (SS, aqueous extract) was applied to polyamide column for main fractionation due to their phenolic profile. Frs. A-E were eluted with an increasing ratio of methanol in distilled water as 0, 25, 50, 75, 100% methanol, respectively. Obtained 5 main fractions (Frs. A-E) were subjected to several column chromatographies to get apigenin (1), apigenin 5-O-b-glucopyranoside (2), luteolin 4 0 -O-b-glucopyranoside (3), hispidulin (4), martynoside (5), tadehaginoside (6), scusalvioside A (7) (Figure 1). These compounds were the major constituents in the fractions and they were obtained with high purity in sufficient amounts. Therefore, they were selected for further analysis in order to determine the responsible compounds for the anti-inflammatory activity of the plant.

Determination of total phenolic contents in the aqueous extract and fractions
The total phenol contents of the extract were determined using the spectroscopic Folin-Ciocalteu method (Cespedes et al., 2008). Gallic acid solution in water (0-400 mg/mL) was used as a reference (Harput et al., 2012). 10 lL of the extract or fractions (1 mg/mL) and gallic acid solutions were transferred to wells of 96-well plate and 150 lL of Folin reagent (1:4 Folin-Ciocalteu reagent: water) was added. After three minutes of incubation, 50 lL Na 2 CO 3 (2:3 Na 2 CO 3 : water) solution was added and incubated for 2 hours at room temperature. The absorbance was measured at 725 nm. Total phenolic contents were expressed as gallic acid equivalents (mg/g dry extract).

ABTS 1 radical scavenging effect
The antioxidant capacity assay was carried out by the improved ABTS þ method as described by Re et al. (1999). ABTS þ radical cation was generated by reacting 7 mM ABTS and 2.45 mM potassium persulphate after incubation at room temperature in dark for 16 hr. The ABTS þ solution was diluted with 80% ethanol to give an absorbance of 0.700 ± 0.050 at 734 nm.
130 mL of sample solutions diluted with 80% ethanol at different concentrations (0-200 mg/mL) was added to 50 mL ABTS þ solution (absorbance of 0.700 ± 0.050) and mixed thoroughly. The mixture was incubated at room temperature for 6 minutes; the absorbance was measured at 734 nm in a microplate reader. Trolox standard solution in ethanol (0-200 mmol) was used as a reference. Results were expressed in terms of mM Trolox equivalent antioxidant capacity (TEAC) in 100 g extract or fraction (Cai et al., 2004).

2,2-Diphenyl-1-picrylhydrazyl (DPPH) radical scavenging effect
The DPPH radical scavenging effect was assessed by the discoloration of methanol solution of 2,2-diphenyl-1-picrylhydrazyl (DPPH) spectroscopically. Quercetin, a natural antioxidant was used as a reference compound. Reaction mixtures containing DPPH (50 mL, 1 mM) solution and MeOH solution (200 mL) of the extract or fractions at different concentrations (20-500 mg/mL for the extract and 20-200 mg/mL for the fractions) were placed in a 96 well plate and incubated for 30 minutes in dark. The absorbance of the remaining DPPH was measured at 517 nm (Harput et al., 2012). The radical scavenging effect was determined by comparing the absorbance with that of the blank.

Superoxide (SO) radical scavenging effect
The superoxide radical was generated non-enzymatically by alkaline DMSO (dimethyl sulfoxide). The superoxide radical and NBT (nitroblue tetrazolium) give colored NBT-diformazan. The assay was assessed by reducing the concentration of the superoxide radical that was measured by the formation of NBT-diformazan. Quercetin was used as a reference compound. The reaction mixture containing 10 mL of NBT (1 mg/ mL in DMSO), DMSO solution (30 mL) of the extract or fractions at different concentrations (20-500 mg/mL for the extract and 20-200 mg/mL for the fractions), and 100 mL of alkaline DMSO (1 mL DMSO containing, 5 mM NaOH in 0.1 mL water) were placed in a 96 well plate. The absorbance was measured at 560 nm using a microplate reader (Harput et al., 2012). The radical scavenging effect was determined by comparing the absorbance with that of the blank.

Nitric oxide (NO) scavenging effect
The concentration of nitric oxide in the test medium was determined spectrophotometrically by colorization with Griess reagent. Quercetin was used as a reference compound. 60 lL of 10 mM sodium nitroprusside in phosphate buffered saline (PBS) was added to aqueous solutions of the extract and fractions (60 lL) at different concentrations (20-500 mg/mL for the extract and 20-200 mg/mL for the fractions). It was incubated under light at room temperature for 150 minutes. After incubation 120 lL of Griess reagent (1% sulphanilamide, 0.1% naphthylethylenediamine dihydrochloride, 2.5% phosphoric acid) was added to the medium to measure the content of nitrite. After 10 minutes, the absorbance was measured at 577 nm in a microplate reader (Harput et al., 2012).

Determination of cytotoxic activity by MTT method
Cytotoxic activity assays were implemented by using a modified version of the MTT [3-(4,5-dimethylthiazol-2-yl) À2,5diphenyl tetrazolium bromide] method developed by Mossman (Harput et al., 2012;Mosmann, 1983;Szliszka et al., 2013). The dehydrogenase enzyme in living cells reduces MTT to purple formazan and thus MTT reacts only in living cells. The method is based on the colorimetric measurement of the color produced by living cells (Mosmann, 1983).
DMEM medium containing 10% FBS, 1% L-glutamine and 0.5% penicillin-streptomycin solution was used for RAW 264.7 cell lines. Suspension of RAW 264.7 cells was prepared as 1 Â 10 6 cells/mL. 100 lL of the prepared cell suspensions were dispensed into 96-well plates and incubated in an incubator containing 5% CO 2 at 37 C. After 24 hours of incubation, the medium in the wells was removed and 100 lL of the solutions containing different concentrations of the samples (50-200 mg/mL for the extract and fractions and 20-100 mM for isolated compounds) in the medium was added and incubated for 48 hours. Following the incubation, 10 lL MTT solution (5 mg MTT/1 mL PBS) was added to the wells and incubated for 4 hours. After incubation, the culture medium was aspirated and 100 lL of DMSO (dimethylsulfoxide) solution was added to the wells and the resulting formazan crystals were dissolved. The absorbance was measured at 577/655 nm after shaking for 5 minutes.
Determination of anti-inflammatory activity on LPS stimulated RAW 264.7 cells The suspension of RAW 264.7 cells (1 Â 10 6 cells/mL) was seeded in 96 well plates and incubated in an incubator containing 5% CO 2 at 37 C for 24 hours. Then, the culture medium was aspirated and 100 mL of sample solutions (50-200 mg/mL for the extract and fractions and 20-100 mM for isolated compounds) with LPS solutions (2 mg/mL) were added. After 24 hours of incubation, 100 ml of supernatant was used for measurement of NO and IL-6 levels. Griess reagent was mixed with supernatant equally for determination of NO level. The absorbance was measured at 577 nm after shaking for 5 minutes. Quantification was done based on the standard curve of sodium nitrite (Genc et al., 2019). The production of IL-6 was determined by ELISA (Biolegend-431307).

Statistical analysis
Statistical analyses have been computed by IBM SPSS Statistic 23 software. The statistical significance has been determined by independent samples Student's t-test. Results of experiments have been expressed as mean ± standard error of mean (SEM). The graphs have been created using GraphPad Prism 6.00 software (GraphPad Software Inc., San Diego, CA, USA).

Results
Target identification and construction of 'Compound-Target', 'Disease-Target', 'Compound-Target-Disease' and 'protein-protein interaction' networks Compounds isolated from S. salviifolia; apigenin, apigenin 5-O-b-glucopyranoside, luteolin 4 0 -O-b-glucopyranoside, hispidulin, martynoside, tadehaginoside, and scusalvioside A were scanned through SYMMAP, STITCH, and SWISSTARGET databases to list the potential targets. 'Compound-Target' network was constructed using these 144 targets (Figure 2). 'Disease-Target' network was constructed using inflammation associated 467 targets obtained from Disgenet database. 'Compound-Target' and 'Disease-Target' networks were merged using Cytoscape 3.8.0 to give 'Compound-Target-Disease' network. The 26 common targets in 'Compound-Target-Disease' network were subjected to STRING database to obtain protein-protein interaction (PPI) network, which was consisted of 26 nodes and 47 edges (Figure 3). For topological analysis, degree distributions were selected to determine the hub genes as these proteins interact with more proteins and play a critical role. For this purpose, two-fold of the mean of the degree in PPI network was determined as threshold. The proteins with a higher degree were defined as key targets. RAC-alpha serine/threonine-protein kinase (AKT1), epidermal growth factor receptor (EGFR), prostaglandin G/H synthase 2 (PTGS2), and tumor necrosis factor (TNF) were selected as the key targets in this study.

Gene enrichment analysis
KEGG gene enrichment analysis has been used often by researchers to evaluate the relation of the different genes and proteins. KEGG pathway database provides networks including metabolism, cellular processes, organismal systems and human diseases that help researchers to reduce the complexity of biological processes by grouping thousands of genes through pathways (Khatri et al., 2012). In this study we aimed to gain insights into the role of the target proteins of the tested compounds in inflammation process through the pathways determined.
The proteins in 'Compound-Target-Disease' network were subjected to DAVID Bioinformatics Resources 6.8 for gene enrichment analysis. The results were listed due to the p-values. 12 pathways with a lower p-value than 0.05 were shown in Figure 4. HIF1 signaling pathway and TNF signaling pathway were found to be involved in the potential mechanisms of the anti-inflammatory activity of the S. salviifolia and the isolated compounds.

Molecular docking
Molecular docking studies were performed to validate the findings of the network analysis. The 4 key targets (AKT1, TNF, EGFR, COX2) and IL-6, an important cytokine in inflammation, were chosen to verify the reliability of the compounds binding abilities. Apigenin, apigenin 5-O-b-glucopyranoside, luteolin 4 0 -O-b-glucopyranoside, hispidulin, martynoside, tadehaginoside, and scusalvioside A were tested for their binding abilities to the selected 5 proteins ( Figure 5). Tested compounds except scusalvioside A showed good binding abilities to AKT1 with a binding energy in a range of À7.38 and À9.13 kcal/mol. Tadehaginoside has the lowest binding energies to TNF and EGFR among the tested compounds. Apigenin, apigenin 5-Ob-glucopyranoside, luteolin 4 0 -O-b-glucopyranoside, and hispidulin showed good binding abilities to all proteins.

Determination of total phenolic contents and the radical scavenging effect
Total phenolic contents of the extract were expressed as gallic acid (mg) equivalent in 1 g dry extract and were found as 69.93 ± 0.65 mg gallic acid/g extract. Radical scavenging effect of the aqueous extract and the fractions screened against DPPH, SO, NO, ABTS þ radicals. For ABTS þ radical scavenging value, TEAC value has been calculated and data have been expressed as mM Trolox equivalent in 100 mg of sample. TEAC value of the extract was found as 198.76 ± 0.94, while these were in the range of 357.73 ± 2.51-402.4 ± 1.15 for the fractions (Tables 1  and 2). DPPH, NO and SO scavenging effects of the extract were examined in the concentrations of 20-500 mg/mL (Table 1). Extract and fractions showed concentration-dependent manner radical scavenging effect against all tested radicals. Fr. C, Fr. D, and Fr. E showed the strongest activity against DPPH and SO radicals among all tested samples (Tables 2 and 3). These fractions are rich in flavonoids, phenylethanoid glycosides, and phenolic acids with the total phenolic contents of 249.12-341.96 mg GA/g sample. There is a correlation between antioxidant activity and phenolic content ( Table 4). The factors such as different mechanisms in the radical-antioxidant reaction and the stereoselectivity of the radical/antioxidant may cause the different radical scavenging values of the compounds/fractions against different radicals (Yabanoglu Ciftci et al., 2011).

Determination of anti-inflammatory activity
Anti-inflammatory activities of extract, fractions, and pure compounds were determined on LPS-stimulated macrophage cells. Macrophage activation by lipopolysaccharides (LPS) derived from gram-negative bacterial cell walls causes the release of various inflammatory mediators, including NO, IL-6, TNF-a, and prostaglandin E2 (PGE2) (Yoon et al., 2009). The concentrations of 50-200 mg/mL were selected for the detection of anti-inflammatory activities of extract, fractions, and isolated compounds; due to lack of cytotoxicity against RAW 264.7 cell at these concentrations (Table 5). The extract and all tested fractions significantly reduced the production of NO at 200 mg/mL compared to positive control LPS alone [LPS (þ)] (P < 0.001); furthermore, the fractions except Fr. A and Fr. C decreased the NO levels significantly at 100 mg/mL (P < 0.001). Compounds 1, 3, and 4 were determined as the most active compounds, that they significantly reduced NO production at 100 mg/mL (P < 0.001) (Figure 6). IC 50 values of compounds 1, 3, and 4 were 84.74, 85.60, and 80.92 mg/mL, respectively. SS significantly reduced the production of IL-6  at 200 mg/mL compared to positive control LPS alone [LPS (þ)] (P < 0.001). Fr. D and Fr. E significantly reduced the production of IL-6 at 100 and 200 mg/mL (P < 0.001) (Figure 7). Isolated compounds could not be tested for IL-6 inhibition because of their limited amounts.

Discussion
Scutellaria species have several usages in traditional medicine worldwide. Some of these usages have been proved by clinical research including studies on anti-inflammatory, anticancer,   neurological effects (Shen et al., 2021). However endemic plants are still waiting to be studied due to their limited and local use. S. salviifolia is used in the treatment of gastric ailment in Anatolia (Quattrocchi, 2012;Tuzlaci, 2016). Gastric ailments are consisting of gastritis, dyspepsia, gastric ulcer, and inflammatory bowel syndrome (Tafti et al., 2017). All of these diseases are associated with the inflammatory process. Progress of gastritis is closely related to inflammatory response and may be cause of gastric cancer (Yin et al., 2021;Yu et al., 2018). The occurrence of intestinal metaplasia after chronic atrophic gastritis is considered to be the preliminary stage of gastric cancer formation. Chronic Helicobacter pylori infection is also considered a cause of gastric cancer (Yin et al., 2021). Gastritis could be triggered by damage in gastric mucosa and inflammation in epithelial cells. Induced NF-jB signaling pathway stimulates the releasing of pro-inflammatory cytokines as NO, IL-6, TNF-a in the progress of gastritis (Yu et al., 2018). TNF-a and NF-jB levels have been shown to increase in gastric ulcer (AbdelAziz et al., 2021). According to the traditional usages of Scutellaria species, in the present study anti-inflammatory potential of the titled plant was investigated by network pharmacology analyses and via NO and IL-6 cytokines. According to network pharmacology analysis, KEGG pathway analysis, the anti-inflammatory effect of the extract was found to be related to the effect on TNF and HIF-1 signaling pathways (Figure 4). The role of these pathways in gastric ailments was clarified and discussed. The HIF-1 signaling pathway plays an important role in inflammation and injury of gastric mucosa contributed to the increase of HIF-1, COX-2 and VEGF (Yin et al., 2019). In the biopsies of Helicobacter pylori positive-gastritis patients, HIF-1 was expressed in all samples (Matak et al., 2015). HIF-1 induces the production of iNOS (inducible nitric oxide synthase) and proinflammatory genes as IL-6 with IL-1b. NOS enzyme activation is also related to the generation of NO (Muniandy et al., 2018). Chronic gastritis is associated with increasing iNOS levels with IL-6, IL-1b and TNF-a in the gastric mucosa (Matak et al., 2015). Additionally, NF-jB signaling pathway is expressed in gastritis and gastric ulcer (AbdelAziz et al., 2021;Matak et al., 2015;Siriviriyakul et al., 2020;Yu et al., 2018). Activation of NF-jB signaling pathway results with releasing of cytokines as TNF-a and IL-6 causing to gastric mucosal injury and gastric epithelial cell apoptosis (Siriviriyakul et al., 2020). Furthermore, expression of TNF-a and IL-6 and the formation of reactive oxygen species (ROS) can be demonstrated as reasons of development of stress ulcers ). In the current study, IL-6 levels were significantly lower in the fractions D and E, which  ¼ 3). Frs. A-E: Polyamide column fractions of aqueous extract of S. salviifolia eluted with methanol in water (0, 25, 50, 75, and 100% for Frs. A-E, respectively). A-E, respectively). a The calculated IC 50 is outside the tested concentration range. The estimate is at least the maximum concentration, so the value was given as >200 or >500 mg/mL. compounds 1-4 and 6 were isolated from, comparing with LPS treated group (P < 0.001), these findings were also confirmed by docking studies. Compounds 1-4 showed high binding abilities to IL-6 among all tested compounds. Pumklin et al. (2016) described apigenin (compound 1) as HIF-1 inhibitor. In another study, apigenin inhibits expression  of HIF-1 and VEGF in ovarian cancer cells (Fang et al., 2005). Moreover, hispidulin (compound 4) down-regulates HIF-1a signaling pathway by activation of AMPK (Gao et al., 2015). Apigenin and hispidulin may be leading the inhibitory effect of Fr. D and E on production of IL-6 through HIF-1 pathway, respectively. Moreover, molecular docking results also support this subject. Based on the findings of our study, AKT1, TNF, EGFR, PTGS2 (COX-2) proteins are the key targets of isolated compounds related with inflammation process. In a previous study, it was observed that AKT1 and TNF were expressed in chronic atrophic gastritis, significantly. AKT-1 is also significantly upregulated in gastric cancer. EGFR is located in gastric mucosa and it will be released in the circumstance of injury (Yu et al., 2018). Chronic atrophic gastritis induced HIF-1a upregulates COX-2 formation (Yin et al., 2021). COX-2 expression may be induced in Helicobacter pylori infection (Liu et al., 2010). Hispidulin (4) inhibits cell proliferation on gastric cancer cells via downregulation of COX-2 (Hao et al., 2020;Yu et al., 2013). Hispidulin decreases the levels of NO, IL-6, COX-2, TNFa by suppressing NF-jB signaling pathway and suppresses AKT1 phosphorylation in in vitro neuroinflammation model (Yu et al., 2020). Apigenin (1), casein kinase 2 inhibitor decreases the phosphorylation of AKT1 in chronic lymphocytic leukemia cells (Shehata et al., 2010). Apigenin inhibits the expression of proinflammatory factors including COX-2 and IL-6 in H. pylori induced gastric carcinoma via activation of IjBa signaling pathway and inhibition of NF-jB (Wang & Huang, 2013). Several studies indicated that apigenin has cancer preventive effect by downregulation of COX associated with inhibition of NF-jB signaling (Yarla et al., 2016). Apigenin significantly inhibited EGFR, which play a key role in the growth of brain tumor (Stump et al., 2017). According to Cicek et al. (2021), apigenin inhibits production of TNFa and IL-6 in sepsis induced rat injury. Tadehaginoside (6) isolated from Fr. D, possessed anti-hepatotoxic effect by inhibiting NF-jB signaling and downregulating TNFa (Tang et al., 2014). Tadehaginoside also showed to have low binding energies for AKT1, TNF and EGFR in the current study. In a recent study, it was observed that the compounds responsible for the activity of S. barbata, which is used as a drug in the treatment of many cancer types, may be flavonoids, luteolin and quercetin, and that these compounds show their antiangiogenic and immunoregulatory effects on AKT1, MAPK1, IL6, EGFR, and VEGFA . This information indicates that Scutellaria flavonoids are the main compounds responsible for the activity and that S. salviifolia is a good alternative to the currently used Scutellaria species.
The result of molecular docking studies suggested that compounds 1-4 might play a critical role in the anti-inflammatory activity of S. salviifolia as they have good binding abilities to the key targets obtained from network pharmacology analysis.
S. baicalensis radix has been included in numerous monographs and widely used in Traditional Chinese Medicine due to the effects on inflammation, respiratory infections, diarrhea (Shen et al., 2021). In previous studies, the aqueous extract of S. baicalensis root was investigated for its anti-inflammatory activities on LPS-stimulated RAW 264.7 cells (Yoon et al., 2009;Zhang et al., 2011). One of the study indicated that the extract inhibited 50% of NO production at the concentration of 40 mg/mL (Zhang et al., 2011); while the extract showed 40% of inhibition at 100 mg/mL in another study (Yoon et al., 2009). According to Yoon et al. (2009), S. baicalensis inhibited also IL-6 production with the percentage of 8.5% at 200 mg/mL, while it was 16.6% for the extract of S. salviifolia on LPS-stimulated RAW 264.7 cells.
Fr. C, Fr. D and Fr. E rich in phenolic compounds, were determined to possess higher antioxidant and anti-inflammatory activities compared to the other fractions, while Fr. A rich in terpenoids, especially iridoids, has lower biological activities (Tables 2 and 3, Figure 6). These results were confirmed by the isolation of compounds 1-6 from the active fractions (Dogan et al., 2019). We compared activity of compound 2 and compound 3 considering structure-activity relationship. The presence of ortho-dihydroxy groups in the B ring contributed to the anti-inflammatory activity of flavonoids has been reported previously (Mora et al., 1990;Odontuya et al., 2005;Tatli et al., 2008). Additionally, hydroxyl group at the position of 5 is important for the activity (Odontuya et al., 2005), therefore this may be the reason why the activity of compound 2 lower than compound 1. Our in vitro anti-inflammatory activity results for the isolated compounds have supported to the literature data.
Overproduction of free radicals can cause oxidative damage to biomolecules, eventually leading to many chronic diseases such as atherosclerosis, chronic inflammation, dementia, and cancer (Harput et al., 2012). It is known that patients with gastric ulcer have low levels of gastric antioxidants compared to normal mucosa (Sangiovanni et al., 2013). Many plants, rich in flavonoids and other phenolic compounds are very important source of antioxidant agents. Phenolic hydroxyl groups are known as good hydrogen donors, and these antioxidants can react with reactive oxygen and nitrogen species. Thus they are prevented the formation of new radicals (Valentao et al., 2002). In the present study, the extract showed moderate radical scavenging effect on DPPH radical and weak radical scavenging effects on SO and ABTS radicals (Table 1). According to Zengin et al. (2019), ABTS þ radical scavenging effect of SS was determined as 200.74 mg trolox equivalent (TE) in g extract which was similar to our result (198.76 mg TE/g extract). According to Senol et al. (2010), DPPH radical scavenging effect of SS was reported as 33.09% inhibition at 500 mg/mL, which was 88.97% in the current study. The differences may be due to various factors as collecting place, collecting time, extraction method, etc.

Conclusion
In the present study, the anti-inflammatory effect of S. salviifolia aqueous extract was investigated by in silico and in vitro methods based on the traditional usage of the plant. To the best of our knowledge, this is the first anti-inflammatory activity and network pharmacology studies on S. salviifolia. Obtained findings by in silico analyses were partially confirmed by in vitro anti-inflammatory effect studies and literature. Flavonoid-rich fractions, Fr. D and E, possessed anti-inflammatory activity via reducing NO and IL-6 levels in LPS-stimulated RAW 264.7 cells at 100 mg/mL (P < 0.001). Flavonoids such as apigenin (1), luteolin-4 0 -O-b-glucopyranoside (3) and hispidulin (4) isolated from Fr. D and Fr. E were the most active compounds among all isolates at 100 mg/mL (P < 0.001) inhibiting the production of NO at RAW 264.7 cells. Therefore, flavonoids may be leading the anti-inflammatory activity in gastric ailment. It is thought that the anti-inflammatory effects of extracts, fractions and pure compounds were achieved by reducing NO and IL-6 levels via regulating the NF-jB pathway or reducing NO production by suppressing iNOS through the HIF-1 pathway.
Anti-inflammatory activities of the extract and fractions are promising as they showed similar efficacy with S. baicalensis commonly used species due to its anti-inflammatory effect all over the world. Therefore, S. salviifolia may be used and recommended instead of S. baicalensis that is in danger of extinction and nationally protected species in China. For this reason, the availability of Scutellaria species with similar biological effects gained importance. Consequently, it is important to discover different Scutellaria species, which have similar biological effects for developing new medicinal products.

Consent for publication
The manuscript is approved by all authors for publication.

Disclosure statement
The authors declare that they have no conflict of interest.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.