As per the PubChem database, a wide range of physical and chemical characteristics of quercetin was determined depending on Lipinski’s rule of five (RO5). Following factors were considered for further characterization of quercetin; molecular weight or mass (MW) number of H-bond donor and acceptor, number of rotatable bonds, TPSA (topological polar surface area), and octanol-water partition coefficient (XLogP3). Swiss ADMET as well as the ADMET descriptor in Discovery Studio 2016 were used to examine the absorption characteristics of quercetin, namely the blood brain barrier (BBB) and human gastrointestinal absorption (HGIA) plots for clear understanding of the ADMET characteristics of quercetin (Daina et al., 2017; Vora et al., 2019).

To identify the targets of quercetin, researchers used a number of strategies relying mostly on structural similarities principle, reversible docking technique, and data integration as well as assimilation process. PharmMapper Server, BDB, TCMSP, Swiss Target Prediction, and STITCH database were used to compile the potential targets (Ru et al., 2014; Daina and Zoete, 2016; Wang et al., 2017; Daina et al., 2019). The name of the compound quercetin was used to acquire target genes from STITCH and TCMSP databases. We acquired anticipated genes by SMILES string of quercetin in the Swiss Target Prediction and BDB database focused on the structure resemblance criteria. Furthermore, the structure of quercetin in SDF file format (PubChem CID: 5280343) were uploaded into PharmMapper for potential gene identification, followed by the normalization of acquired genes performed by using the UniProt Database. Lastly, these obtained genes were integrated, after exclusion of redundant targets (Supplementary Table S2). This section of the gene set is referred to as “quercetin 182 targets” within subsequent summary.

EH-linked genes were obtained from the CooLGeN database Drug Bank database, MalaCards database, Therapeutic Target Database (TTD), and OMIM database, to determine the accuracy of disease-related genes. The scores of targets relatively greater than genes were extracted from the CooLGeN database. EH-associated targets were retrieved the OMIM and MalaCards databases. Redundant genes were removed from the datasets, and total 483 EH-associated targets were chosen for further study. This section of the gene array is referred to as “genes associated with EH.” Supplementary Table S2 provides more detailed information for the targets. The correlation within these two classes of targets could be one of the potential genes of quercetin. The above category of integrating targets is referred to as “intersection genes of quercetin and essential hypertension”.

The VarElect database leads to more accurate identification of genes that are often directly or indirectly linked to a specific phenotype or disease (Stelzer et al., 2016). VarElect’s ability to interpret and score gene lists dependent on phenotype keywords inputted is a critical feature. Elastic search computing is largely responsible for the VarElect analysis scores. The intensity of a term’s occurrence in particular GeneCards is relative to the rate via its expression within these genes (opposite report intensity rectification) to evaluate its score. The association between “the intersection genes of quercetin as well as EH” and also the phenotype of “Essential hypertension” was investigated using the VarElect.

In addition to creating PPI networks, the GeneMANIA platform can also find a set of targets similar to those entered. It is based on a variety of functionally relevant information, and compared differences between them, i.e., co-expression as well as co-localization, using a variety of analogous information (Ogris et al., 2018). The GeneMANIA database was used in research to generate protein-protein interaction (PPI) network of intersecting targets of quercetin and EH along with the corresponding genes. We may obtain not only the association between the entered intersecting targets, but also many other target genes that are tightly associated to these targets, using GeneMANIA examination. The above unique set of genes is termed “anticipated targets of quercetin anti-EH” within current analysis. The Network Analyzer tool in Cytoscape 3.7.2 was used to calculate topological characteristics of the PPI network, such as degree, closeness centrality, betweenness centrality, and average shortest path length.

Metascape is an effective gene annotation and enriched analysis program (Zhou et al., 2019). The predicted quercetin anti-EH targets were identified and categorized into various GO pathways. To obtain the biological processes as well as KEGG pathways relevant to the mechanism of EH, DAVID (version 6.8) tool was used.

The network was created using Cytoscape 3.7.2 for illustration of complicated networks. The nodes of the network identified potential quercetin targets for EH, biological processes, and signaling pathways discovered by enrichment analysis, and the edges represented associations between them.

Quercetin (Cat.no.Q4951-10G) was purchased from Sigma (St.Louis, MO, United States) with >95% purity (HPLC). The protein assay reagent kit bicinchoninic acid (BCA), trypsin-EDTA, fetal bovine serum (FBS) and Dulbecco’s Modified Eagle Medium (DMEM) were obtained from Thermo Fisher Scientific (Waltham, MA United States). Phosphate saline buffer was purchased from Maixin Biotechnology (Fuzhou, Fujian China), and polyvinylidene fluoride (PVDF) membrane was purchased from Millipore (Billireca, MA, United States). Antibodies against STAT3 (Cat no.41464), p-JAK2 (Cat no.11151) were purchased from Signaling antibody (SAB) (College Park, Maryland, United States). The α-SMA (Cat no.19245S), p-STAT3 (Cat no.9145S), JAK2 (Cat no. 3032S) antibodies were purchased from Cell Signaling Technology (Danvers, MA, United States). PCNA (Cat no. ab29), goat polyclonal secondary antibody to rabbit IgG H&L (Alexa Fluor^® 647; Cat no. ab150079), Ang II (Cat no. ab120183) were purchased from Abcam (Cambridge Science Park, Cambridge, United Kingdom) and anti-rabbit secondary antibody (Cat no. 7074) was purchased from Cell Signaling Technology (Danvers, MA, United States). Antibody against GAPDH (Cat no. Abp57259) and kit of CCK-8 (Cat no. ktc011001) were obtained from Abbkine (Wuhan, Hubei, China).

Wistar rats were anesthetized by using isoflurane; abdominal aorta was instantaneously dissected, washed with ice-chilled 1.5 mM of CaCl₂-HEPES-buffered salt solution (HBSS) for the removal of blood. After that, the aorta was cut longitudinally and the endothelial cells were removed with a cotton swab retrieved delicately. Cells were kept in 1.5 mM CaCl₂-HBSS at 4°C for 30 min and were treated with Ca²⁺ free-HBSS at room temperature for 20 min. Then, a digestion solution containing mixed enzymes (0.5 ml), collagenase (3.2 mg), BSA (2 mg), and papain (0.3 mg) was prepared, and the aorta was digested at 37°C for 20–30 min. After the removal of extra enzyme, Ca²⁺ free-HBSS was used to wash the cells and kept in T25 flask with 5 ml of 10% FBS containing DMEM medium supplemented with 100 μg/ml streptomycin and 100 IU/ml penicillin. Lastly, by pipetting up and down for 50 to 60 times, gently dispersed VSMCs were kept in incubator containing 5% CO₂. For each experiment fresh VSMCs were cultured and used for the corresponding experiments at third or fourth passage. All animal experiment protocols were approved by the Animal Ethics Committee of Fujian University of Traditional Chinese Medicine (No. FJTCM IACUC. 2021185).

Isolated primary VSMCs were seeded into glass bottom dishes at a density of 0.8 × 10⁵ cells/well. Following 24 h of cultivation, cells were permeabilized with 0.25% Triton X-100 for 10 min, fixed with 4% paraformaldehyde for 15 min, and then blocked with blocking buffer (containing of 5% BSA, 85% PBS and 10% goat serum) for 1 h. Subsequently, cells were incubated with antibody against α-SMA (1:200) at 4°C overnight, followed by washing with PBS for thrice and incubated with secondary antibody anti-rabbit (1:400) in the dark at room temperature for 1 h. Then, cells were washed thrice with PBS and incubated with Hoechst stain (Thermo Fisher Scientific) for 10 min and observed under confocal microscope (PerkinElmer Inc., Waltham, MA, United States).

The cell viability assay was determined using cell counting assay. The extracted VSMCs in suspension at 3 × 10⁴ cells/mL density was seeded into 96 well plates. When VSMCs confluency was reached to 30%–40%, cells were placed in serum free media for starvation for 6–8 h and after that VSMCs were treated with various concentrations of quercetin (3.125, 6.25,12.5, 25, 50, and 100 μM) for 24 h or Ang II (0.01, 0.1, 1, and 10 μM) for 24 h. Besides, 0.1 μM of Ang II stimulated VSMC_S were treated with quercetin (6.25, 12.5, and 25 μM) for 24 h. Following treatment, each well loaded with 10 µl of the CCK8 solution, which was then incubated at 37°C for 2 h. By using micro plate reader (Multiskan FC, Thermo Fisher Scientific) absorbance was measured at 450 nm.

For the extraction of total proteins, Western cell lysis buffer (Beyotime Biotechnology, Shanghai, China) supplemented with 1 mM phenylmethylsulfonyl fluoride (PMSF) and other protease inhibitors including cocktail (MedChemExpress, Monmouth Junction, NJ, United States), and phosStop (Roche; Basel, Switzerland) was used. Briefly, cells were lysed with the lysis buffer for 20 min in ice and centrifuged at 14,000 g at 4°C for 20 min as described previously (Liu et al., 2021). Followed by the collection of supernatants, total concentration of protein was assessed by BCA protein kit assay and equal amount of each sample were subjected for separation through 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were transferred to PVDF membranes using wet transfer system. Further, 5% skimmed milk was used for the blockage of membranes for 2 h at room temperature and then primary antibodies; anti-p-JAK2, anti-JAK2, anti-p-STAT3, anti-STAT3, anti-PCNA or anti-GAPDH (dilution factor 1:1000) were incubated at 4°C for overnight. After that, membranes were washed thrice with TBST buffer and membranes were incubated with anti-rabbit or mouse secondary antibody conjugated to horse radish peroxidase (dilution factor 1:5000) for 1 h at room temperature. The protein expression was detected by chemiluminescence kit and ImageJ Software.

Statistical Package for the Social Sciences (SPSS 25.0) software (Chicago, IL, United States) was used for statistical analysis. All data were provided as means ± SD the experiment. The Shapiro-Wilk test was employed to determine normality in experiments with three or more groups. When the data meet a normal distribution, one-way analysis of variance (ANOVA) was used, followed by the Bonferroni post-hoc test for pairwise comparisons. The non-parametric Kruskal–Wallis test was used to analyze the non-normally distributed data, pursued by the Dunnett’s t-test for paired comparisons. The significant difference was considered with p < 0.05.

For health impact assessment (HIA) and blood brain barrier (BBB), analysis of absorption, distribution, metabolism, excretion, and toxicity (ADMET) provides four levels of prediction between 95% and 99% of ellipsoids confidence. Quercetin was well absorbed via intestine and also had very high BBB level of penetration, as seen in Table 1. Supplementary Figure S1A illustrates the chemical structure, while Supplementary Figure S1B demonstrates the suitable physicochemical space for oral bioavailability (OB) as well as pharmacokinetic characteristics including TPSA 131.36Å, nine rotatable bonds, and so on. In addition, the BOILED-EGG analysis revealed that quercetin penetration into the blood-brain [−1.098 (log BB)] and 77.207% absorption in the human gastrointestinal tract was illustrated in orange and colored zones respectively, as given in Supplementary Figure S1C plot.

The intersection of targets from various databases was depicted by Circos plot (Figure 1). To identify gene overlaps, relatively similar enriched ontology term(s) and their functions or shared pathways were used to significantly improve gene overlaps as given in Supplementary Table S1. To study the genotype-phenotype correlations, a number of 54 overlapping genes of quercetin and EH were entered into the VarElect interactive platform. The results are presented in Table 2. The GJA1 gene on chromosome six encodes the gap junction alpha-1 protein (GJA1), widely recognized as connexin 43 (Cx43). Connexin (Cx) protein isoforms give a shape to gap junctions (GJs), which acts as a pathway for information exchange between cells and expressed in blood vessels in four different forms: Cx37, Cx40, Cx43, and Cx45. MAPK1 commonly recognized as p42MAPK and ERK2 is a mitogen-activated protein kinase enzyme that is denoted via MAPK1 gene in humans. MAP Kinases are involved in several cellular activities, like differentiation, proliferation, expression control, and growth, as well as acting as an integrating site for various biochemical signals. STAT3 is an essential intercellular signaling circuit and acts on several downstream genes such as CyclinD1, c-myc, and VEGFs. Activation of STAT3 depends on stimulation of the signal transduction protein JAK2 by various cytokines such as interleukins and growth factors (HGF and EGF) and results in phosphorylation of STAT3 at 705 tyrosine amino acid. STAT3, which is continuously activated, promotes tumor cell proliferation and survival, stimulates angiogenesis, and inhibits the immune response to malignancy.

The integrated PPI network was constructed by uploading 54 genes to identify their functional role. The source networks were classified into different categories (e.g., co-expression). The weight of each network along with the number was listed within each category. The weight of interaction correlations in the network is represented by the percentage in the results. According to the outcomes, 55.84% of the target interactions in the network had co-expression and 16.23% had physical interactions. In addition, there were interactions in the form of co-localization (11.27%), predicted (7.37%), pathways (3.92%), genetic interactions (3.31%), and shared protein domains (2.06%) (Figure 2A). Figures 2B–E and Supplementary Table S3 illustrate the estimated average closeness centrality, betweenness centrality, degree, and shortest path length of the nodes in the network. Analysis of the topological variables of the network revealed that genes from the STAT family (STAT1, STAT3) and cytokine-related genes (IL-6, IL-β, and IL-12α) were ranked highly throughout the network.

We used GO evaluation to further investigate the role of a total of 54 quercetin targets. In the category of EH pathogenesis, these targets were involved in the development of the circulatory system, regulation of endothelial and epithelial cell proliferation, muscle cell proliferation, blood vessel development, and cytokine response (Figure 3A). It also had implication in the plasma membrane raft, RNA Pol-II transcription regulator complex, and IL-6 receptor complex (Figure 3B). In addition, the category of molecular functions included signaling receptor binding, protein kinase binding, cytokines activity, and protein kinase regulator activity (Figure 3C). The effect of quercetin over potential signaling pathways was investigated using KEGG pathway assessment. The analysis showed that “PI3K-AKT signaling pathway,” “cytokine-cytokine receptor interactions,” “JAK-STAT signaling pathway,” “MAPK signaling pathway” and “pathways in cancer” were highly enriched (Figure 3D).

We generated the function network/target-pathway after performing extensive network analysis on a number of representative biological processes, molecular function, and signal pathways. Multiple targets were systematically implicated in several biological processes, as given in Figure 4. For instance, STATs, and JAKs were involved in processes like “Wnt signaling pathway,” “JAK/STAT signaling pathway” “protein kinase activity,” “PI3K-AKT signaling pathway,” “vascular smooth muscles contraction,” and “protein homodimerization activity” processes. STAT1 and STAT3 were observed in a number of GO processes, including “circulatory system development,” “cell proliferation regulation,” “cytokine mediated signaling pathway,” “positive regulation of phosphorus metabolic process,” and “cytokine activity”.

The primary VSMCs were isolated from abdominal aorta and identification was performed with α-SMA-antibody and 95% cells were detected positive by immunofluorescence staining (Supplementary Figure S2A). CCK8 assay showed that quercetin treatments (3.125–100 μM) did not affect the cell viability significantly (Supplementary Figure S2B), while Ang II (0.01 and 0.1 μM concentrations) significantly increased cell viability of primary VSMCs (Supplementary Figure S2C). CCK8 assay showed that 6.25, 12.5, and 25 µM concentrations of quercetin treatment attenuated Ang II-induced elevated cell density significantly (Figure 5A), cell number (Figure 5B) and viability of VSMCs (Figure 5C). Ang II stimulation increased the expression levels of PCNA, which was significantly reduced with quercetin treatment (Figure 5D).

Western blot analysis (Figure 6) indicated that stimulation of Ang II activated JAK2 (Figure 6A) and STAT3 (Figure 6B) via their phosphorylation in VSMCs, which were attenuated after quercetin (12.5 μM) treatment.