2.4.1 Collection of inflammatory and V. Jatamansi related-targets

SwissTargetPrediction database (http://www.swisstargetprediction.ch/), TCMSP database (https://old.tcmsp-.com/tcmsp.php), TCMIP database (http://www.tcmip.cn/ETCM/index.php/Home/Index/index.html), Bionet database (https://bionet.ncpsb.org.cn/batman-tcm/) was used to predict the targets of compounds of V. jatamansi. GeneCards (https://www.genecards.org/) databases were searched for “inflammatory”. V. jatamansi predicted targets in common with inflammatory targets as potential targets.

2.4.2 Protein-protein interaction (PPI) analysis

The potential targets were imported into the online mapping platform (https://www.bioinformatics.com.cn/) to construct Venn diagrams. The intersection targets were uploaded to the STRING 11.5 platform (https://strin g-db.org/). Select the organism as Homo sapiens. The minimum score for interaction is 0.9.

The topological properties of the PPI network were analyzed using the CytoHubba plug in Cytoscape 3.7.2. CytoHubba analysis including degree, edge percolated component (EPC), maximal clique centrality (MCC), maximum neighborhood component (MNC) calculations.

2.4.3 Pathway analysis

The pathways of the potential targets were enriched and analyzed by the DAVID database (https://david.ncifcrf.gov/tools.jsp). The species was set as Homo sapiens. Signaling pathways were analyzed using Kyoto Encyclopedia of Genomes (KEGG) (p < 0.01). Enrichment dot bubble was plotted by https://www.bioinformatics.com.cn.

2.5.1 HPLC-DAD condition

HPLC analysis was performed on a Thermo Scientific™ Dionex™ UltiMate™ 3000 system (Thermo Fisher Scientific, Bremen, Germany). The chromatographic column, gradient elution program, flow rate and temperature conditions were same with 2.4. The detection wavelength was at 254 nm. The injection volume was set as 10 μL.

2.5.2 Method validation

Precision was verified by evaluating the same sample solution for six consecutive injections. Repeatability was evaluated by six sample solution. The stability of the same samples (0, 2, 4, 6, 8, 10, 12, 17, 19, 21, 24, 24 h) was analyzed by the above method. Relative standard deviation (RSD) was used to calculate the area of each characteristic peak with relative accuracy.

2.5.3 Establishment of fingerprints

The chromatographic fingerprints of 24 batches of V. jatamansi were imported into the Similarity Evaluation System (Version 2012A; Beijing, China) according to the literature (Li et al., 2021).

2.5.4 Quantitative analysis of bioactivity markers

The three markers quantification was conducted by a HPLC System. The chromatographic conditions were the same as in 2.5.1. The linearity, limit of quantitation, repeatability, precision, stability and recovery of the method were verified. The linearity of the method was calculated by plotting the peak area (y) against the concentration (x, μg /mL). The LOD (S/N = 3) and LOQ (S/N = 10) were calculated by the concentration of the mixed standard.

2.10.1 Gray relational analysis (GRA)

The grey correlation theory was used to calculate the grey correlation coefficient and rank of each common peak of 24 samples. The specific operation was to use the IC₅₀ value as the reference sequence and the peak area of the 17 common peaks as the comparison sequence.

2.10.2 Bivariate correlation analysis (BCA)

The Pearson model was used for bivariate correlation analysis (BCA) analysis. The independent variable was the common peak area, and the dependent variable was the IC₅₀ value of the anti-inflammatory activity.

3.2.1 PPI network construction analysis

A total of 36 compounds identified from V. jatamans and out of which 6 compounds could not be assessed from their targets database. 426 cross-targets that were obtained from 478 targets of 30 compounds and 13,697 anti-inflammation targets (Fig. 2A). The PPI network was constructed by uploading 426 targets to the string platform (Fig. 2B), which was then imported into cytoscape for further visualization. The degree value represents the number of connections of a node, it indicates the importance of a node in the network. The higher degree of nodes connected implies higher centrality and influence of a node. A total of 28 targets were considered potential targets with degree > 20 (Fig. 2D). In addition, the key targets were screened from the intersection of the top 20 targets of the four parameters, which are EPC, MCC, MNC and degree parameters in cytoHubba plug (Fig. 2C) as well as 13 key targets were identified from 28 targets based on EPC, MCC and MNC parameters (Table S3 in Supplementary materials). The results showed that 13 key targets screened from 28 potential targets were key targets for the anti-inflammatory activity of V. jatamansi, including STAT3, LCK, EGFR, JAK2, PTK2, PIK3CA, MAPK1, FYN, JAK1, MAPK14, PIK3R1, SRC, and HRAS (Fig. 2D).

Close

Fig. 2

Network pharmacology analysis. (A) Veen diagram. (B) PPI network. (C) Four parameters (degree, EPC, MCC, MNC) were used to calculate the intersection analysis of the top 20 targets (D) Core target network. The size of a node in the network indicates how important the node is in the network. The yellow circles refer to the potential targets (degree＞20), and the purple circles refer to the core targets (four parameter intersection targets). (E) KEGG pathway enrichment analysis.

Export to PPT

3.2.2 Pathway enrichment analysis

To further explore the anti-inflammatory mechanisms of V. jatamansi, KEGG enrichment analysis of the potential targets were performed using the DAVID database. Based on “P＜0.01”and “signaling pathway”, the top 20 pathways were screened (Fig. 2E). The name of the signaling pathway is displayed on the Y-axis, while the enrichment factor is displayed on the X-axis. The number of genes in each pathway is proportional to the bubble size, and the bubble color reflects how significant the gene enrichment in that pathway is. The results suggested that V. jatamansi may exert anti-inflammatory activity by modulating JAK-STAT signaling pathway, PI3K-Akt signaling pathway, NOD-like receptor signaling pathway, Chemokine signaling pathway, TNF signaling pathway, IL-17 signaling pathway, etc.

3.3.1 Cell cytotoxicity analysis

The cytotoxicity of V. jatamansi extract on RAW264.7 cells was measured by the CCK-8 assay (Fig. 3A). It showed that the cell survival rate was reduced significantly by higher concentrations of the extract, and the concentrations below 50 μg/mL had no cytotoxic effect.

Close

Fig. 3

Anti-inflammatory effects of V. jatamansi against RAW264.7 cells. (A) RAW264.6 cells were treated with different concentrations of V. jatamansi for up to 24 h to examine the cell viability. (B) NO concentration in cell supernatant. (C-G) The mRNA levels of IL-6, IL-1β, TNF-α, JAK1, STAT3 in cells. (H-I) The protein expression of JAK1, STAT3 in cells; (H)Western blot image and (I) densitometric analysis. Data are shown as mean ± SD. *p < 0.05, **p < 0.01 and ns, no significant.

Export to PPT

3.3.2 Anti-inflammatory effects of V. Jatamansi in vitro

To induce an inflammation model in macrophages, 1 μg/mL LPS was stimulated for 24 h (Cho et al., 2023). The nitric oxide (NO) levels were measured in the cell supernatant using Griess reagents. The results showed that LPS significantly enhanced the NO production (P < 0.01), indicating that the inflammation model was successfully established. In parallel, V. jatamansi (0.78 ∼ 50 μg/mL) decreased the level of NO production on LPS-induced RAW264.7 cells in a dose-dependent manner (Fig. 3B). In inflammatory diseases, pro-inflammatory factors were elevated, including IL-6, IL-1β, TNF-α (Jing et al., 2021). The mRNA expression of IL-6, IL-1β and TNF-α was examined. It was shown that the mRNA levels of IL-6, IL-1β and TNF-α were significantly increased by LPS than control group (P < 0.01), while V. jatamansi treatment significantly reversed the elevated expression levels caused by LPS. (P < 0.05) (Fig. 3C-E). This indicated that V. jatamansi extracts have good anti-inflammatory activity in our experiment.

3.3.3 V. Jatamansi regulate key targets in vitro

The anti-inflammatory activity of V. jatamansi may involve the JAK-STAT signaling pathway, according to the KEGG pathway enrichment analysis results. To measure the mRNA and protein levels of JAK1 and STAT3, two key targets of this pathway, qRT-PCR and western blotting analysis were performed. It was found that LPS-treated RAW264.7 cells had higher JAK1 and STAT3 expression than control group (P < 0.01). V. jatamansi reduced JAK1 and STAT3 expression compared with the model group (P < 0.05) (Fig. 3F-I). These results indicate that V. jatamansi may have anti-inflammatory activity by modulating JAK1 and STAT3 mRNA and protein expression.

3.4.1 Similarity analysis

The HPLC fingerprint chromatogram similarity analysis was used to analyze the variability among different batches. Firstly, good precision and reproducibility of the HPLC instrument and stability of the sample solution within 24 h were shown by the method validation results (Table S5 in Supplementary materials). The fingerprint method was suitable for fingerprint analysis.

Then, the fingerprint chromatograms of different batches were established (Fig. 4). The common peaks with relatively high intensity that appeared in each sample were selected as the characteristic peaks of V. jatamansi. The 17 characteristic peaks were determined by peak alignment. The similarity values among the 24 batches varied from 0.705 to 0.997, suggesting that their chemical characteristics were highly similar and chemical fingerprints alone were insufficient to differentiate the quality (Table S6 in Supplementary materials).

Close

Fig. 4

HPLC fingerprints of 24 batches of V. jatamansi extracts.

Export to PPT

3.4.2 NO inhibitory activity of V. Jatamansi

NO inhibition rate and IC₅₀ value were used to evaluate the anti-inflammatory activity of 24 batches of V. jatamansi. Celecoxib has the effect of inhibiting NO as a reference for this experiment (Bekkouch et al., 2023; Cui et al., 2023). In this work, 24 batches of V. jatamansi extracts showed good inhibition of NO production (IC₅₀ = 5.57–60.19 μg/mL) (Table 2). Surprisingly, S18, S20-S22, S24 batches showed excellent inhibition (IC₅₀ = 5.57–7.625 μg/mL) than the positive control celecoxib (IC₅₀ = 20.9 μM = 7.97 μg/mL). The value of IC₅₀ indicating that there were significant differences in activity among different batches.

3.4.3 Bivariate correlation analysis

The area of 17 common peaks and the IC₅₀ value of the anti-inflammatory activity were analyzed by BCA to determine their correlation degree and direction. The Pearson’s model correlation coefficient showed that 13 common peaks had a negative correlation with the IC₅₀ value, indicating that the 13 common peaks may possess anti-inflammatory activity. As shown in Table 3, among 9 common peaks, P6, P7, P12, P14, P17, P23, P26, P30 and P32 were showed significant correlation between peaks area and IC₅₀ value (P＜0.05), suggesting that 9 compounds may have anti-inflammatory activity.

3.4.4 Grey relational analysis

To screen the potential anti-inflammatory components, the spectrum-effect relationship between IC₅₀ and the area value of 17 common peaks were studied by the GRA method. As can be seen from Table 3, the correlation values between 17 common peaks and inhibition of NO production were 0.6883 ∼ 0.8549. The results indicated that 16 common peaks may be highly correlated with anti-inflammatory activity (with GRA＞0.7). To prevent bias caused by a single screening method, we combined the methods of GRA and BCA in a comprehensive analysis, and the results showed that P6, P7, P12, P14, P17, P23, P26, P30 and P32 were considered as potentially bioactive compounds with anti-inflammatory activity.

Based on principle of measurability, easily acquired standard and specificity, we predict that P6 (isochlorogenic acid B), P7 (hesperidin), P23 (acevaltrate) and P32 (valtrate) may be potential combinatorial quality markers for the anti-inflammatory activity of V. jatamansi.

3.5.1 In silico ADME prediction

The pharmacokinetic parameters of the potential combinatorial quality markers were investigated to assess their bioavailability and absorption. Previous studies showed that hesperidin has good absorption properties in vivo (Li et al., 2023), but acevaltrate and valtrate have not been reported. The pharmacokinetic and physicochemical parameters of acevaltrate and valtrate were predicted using SwissADME website. The prediction results indicated that both compounds had favorable lipophilicity (WLogP = 2.35, 2.66) and water solubility (class: soluble) (Table S7 in Supplementary materials). According to different reporting filters provided, acevaltrate complied with the Lipinski, Egan and Muegge rules, while valtrate satisfied the Lipinski, Veber, Egan, and Muegge rules. The abbott bioavailability score for both compounds were 0.55, suggesting moderate bioavailability. Both compounds also exhibited high gastrointestinal absorption properties and good central nervous system safety profiles (no BBB permeability). Moreover, neither compound was predicted to be a substrate for CYP1A2, CYP2C19, CYP2C9 and CYP3A4, which could reduce the possibility of drug-drug interactions. The potential combinatorial quality markers exhibited favorable pharmacokinetic properties.

3.5.2 Anti-inflammatory effects of potential quality markers in vitro

CCK-8 assay was used to detect the cytotoxic effects of potential quality markers at different concentrations on RAW264.7 (Fig. 5A-C; Fig. S4 in Supplementary materials). The cell viability of 800 μM isochlorogenic acid B, 300 μM hesperidin, 6 μM valtrate, and 6 μM acevaltrate were used as experimental concentrations for subsequent experiments. NO, IL-6, IL-1β, and TNF-α are important inflammatory mediators involved in the inflammatory response. Firstly, the IC₅₀ of hesperidin, acevaltrate, and valtrate were, 131.6 μM, 4.33 μM, and 2.58 μM, respectively (Fig. 5D-F). Unfortunately, isochlorogenic acid B with weak inhibition of NO production (IC₅₀ = 637.9 μM) (Fig. S4 in Supplementary materials). A compound is considered weakly activity if its IC₅₀ value is more than an order of magnitude higher than that of a positive control. Thus, isochlorogenic acid B was excluded as a quality marker. Subsequently, the mRNA expression of IL-6, IL-1β and TNF-α were decreased in three compounds treated with model group (P < 0.05) (Fig. 5G-O). In conclusion, these compounds may exert anti-inflammatory effects in LPS-stimulated macrophages by inhibiting NO, IL-6, IL-1β and TNF-α.

Close

Fig. 5

Anti-inflammatory effects of potential quality markers on RAW264.7 cells. (A-C) Cell viability of hesperidin, valtrate and acevaltrate. (D-F) NO concentration in cell supernatant. (G-O) The mRNA levels of IL-6, IL-1β and TNF-α in cells. Data are shown as mean ± SD. *p < 0.05, **p < 0.01 and ns, no significant.

Export to PPT

3.5.3 Potential quality markers regulate key targets

Under inflammatory conditions, the inhibitory effects of the three compounds on JAK1 and STAT3 expression were equally intriguing. Surprisingly, compared with the model group, the three compounds were able to significantly reduce the mRNA level of JAK1 and STAT3 (P＜0.05) (Fig. 6A-F), and also significantly reduce the protein expression levels of JAK1 and STAT3 (P＜0.05) (Fig. 6G-L). It was further confirmed that the three compounds (hesperidin, acevaltrate and valtrate) were shown to modulate the same anti-inflammatory targets as V. jatamansi.

Close

Fig. 6

V. jatamansi potential quality markers regulate the key targets. (A-F) The mRNA levels of JAK1 and STAT3 in cells. (G-L) The protein expression of JAK1, STAT3 in cells; (G-I) Western blot image and (J-H) densitometric analysis. Data are shown as mean ± SD. *p < 0.05, **p < 0.01 and ns, no significant.

Export to PPT

3.5.4 Quantitative analysis of potential quality markers

In order to verify the potential quality markers of V. jatamansi, these three compounds (hesperidin, acevaltrate and valtrate) were analyzed quantitatively. The HPLC results of V. jatamansi samples and the three compounds showed good separation (Fig. 7A and 7B). The verification method showed that the RSD of each compound precision, repeatability, 24 h stability, and recovery were all less than 3.84 %, indicating that the method was feasible. The linear results showed that the three compounds all met the expected linear requirements within the target concentration range (Table S8 for Supplementary materials). The results showed that the contents of the three potential quality markers in different batches were different (Table 4). There was a positive correlation between the total content of the three compounds in the batches and their anti-inflammatory activity as measured by the IC₅₀ values, indicating that the batches with lower total concentration of the three compounds had lower potency in inhibiting inflammation. (Fig. 7C).

Close

Fig. 7

Quantitative analysis and contribution of V. jatamansi. (A) HPLC chromatogram of V. jatamansi. 1: hesperidin; 2: acevaltrate 3: valtrate. (B) HPLC of chromatogram mixed standard solution. 1: hesperidin; 2: acevaltrate 3: valtrate. (C) Correlation of anti-inflammatory activity and combinatorial quality markers content of different batches of V. jatamansi. (D) Inhibition rates of anti-inflammatory activity of V. jatamansi and combinatorial quality markers at different dilutions.

Export to PPT

3.5.5 Contribution of combinatorial quality markers

To evaluate the contribution rate of the combinatorial quality markers, we compared the NO inhibition rate of a certain batch of V. jatamansi extracts (S2) and a mixture of three combinatorial quality markers, with 1644.31 μg/g of hesperidin, 680.93 μg/g acevaltrate, and 1406.20 μg/g valtrate. Contribution rate = AUC _{combinatorial quality markers inhibition rate} /AUC _{V. jatamansi extracts inhibition rate} * 100 %. The results showed that the inhibition trend of the combinatorial quality markers and the V. jatamansi extracts was consistent at different dilution factors, and the contribution of the combinatorial quality markers reached 45.0 % of the V. jatamansi (Fig. 7D). This suggests that the combinatorial quality markers are important in evaluating the anti-inflammatory activity of V. jatamansi.