β-Sitosterol Suppresses LPS-Induced Cytokine Production in Human Umbilical Vein Endothelial Cells via MAPKs and NF-κB Signaling Pathway

Atherosclerosis (AS) is an inflammatory disease, whose occurrence and development mechanism is related to a great number of inflammatory cytokines. β-sitosterol (BS), a natural compound extracted from numerous vegetables and plant medicines, has been suggested to improve AS, but the underlying mechanism remains vague. This work focused on investigating how BS affected the lipopolysaccharide (LPS)-treated human umbilical vein endothelial cells (HUVECs) and further exploring the potential targets and mechanisms through network pharmacology (NP) and molecular docking (MD). According to in vitro experiments, LPS resulted in an increase in the expression of inflammatory cytokines like tumor necrosis factor-α (TNF-α), cyclooxygenase-2 (Cox-2), and interleukin-6 (IL-6). Besides, secretion of IL-6, interleukin-1β (IL-1β), and TNF-α also increased in HUVECs, whereas BS decreased the expression and secretion of these cytokines. NP analysis revealed that the improvement effect of BS on AS was the result of its comprehensive actions targeting 99 targets and 42 pathways. In this network, MAPKs signaling pathway was the core pathway, whereas MAPK1, MAPK8, MAPK14, and NFKB1 were the hub targets. MD analysis also successfully validated the interactions between BS and these targets. Moreover, verification test results indicated that BS downregulated the abnormal expression and activation of MAPKs and NF-κB signaling pathways in LPS-treated cells, including p38, JNK, ERK, NF-κB, and IκB-α phosphorylation expressions. Furthermore, p65 nuclear translocation was also regulated by BS treatment. In conclusion, the BS-related mechanisms in treating AS are possibly associated with inflammatory response inhibition by regulating MAPKs and NF-κB signaling pathways.


Introduction
Atherosclerosis (AS) is a special pathological condition involving endothelial dysfunction, infammatory infltration, and plaque formation [1]. Te chronic build-up of plaques causes stenosis and thrombosis in sick vessels, while the acute rupture of atheromatous plaques leads to vessel occlusion and critical tissue hypoxia. Typically, myocardial infarction (MI) and stroke, the most fatal complications caused by plaque disruption and thrombosis, have become the major causes of cardiovascular mortality in the elderly, worldwide [2,3]. Te pathophysiological mechanisms of AS are complicated, but evidence has indicated that endothelial cells are the major player in the initiation of AS. Atherosclerotic lesions are frequently formed in certain regions with fow perturbation and endothelial dysfunction [4]. Te infammatory responses of endothelium and vascular leakage have provided the areas for the entry of leukocytes and smooth muscle cells, leading to fatty streak development, which is the earliest visible lesion in AS [5]. Te anti-infammatory strategies of the endothelium seem to be the promising treatments to reduce the risk of atherosclerotic diseases [6]. However, the clinically used therapies for preventing AS are limited to drugs that lower cholesterol levels, while treatments targeting infammatory properties of AS deserve further investigation.
Phytosterols, a kind of natural dietary micronutrient, are widely distributed in vegetables and fruits such as legumes, seeds, and nuts, which exhibit their potent efects on promoting human wellness in many diseases such as cancers, diabetes, and cardiovascular diseases (CVDs) [7]. β-sitosterol (BS), one of the most common components of phytosterols, is the Food and Drug Administration (FDA)approved potential herbal nutraceutical used to treat AS [8]. Since the chemical structure of BS is similar to cholesterol, the administration of BS can competitively suppress animal cholesterol absorption and reduce serum lipid levels, and it has no deleterious side efects [9,10]. Some study also indicates that BS improves the viability and morphology of human endothelial cells exposed to oxidized low density lipoprotein (ox-LDL) [11]. In addition, BS inhibits the monocyte chemotactic protein-1 (MCP-1) and intercellular adhesion molecule-1 (ICAM-1) levels in a cell model induced by ox-LDL and TNF-α, thus reducing the migration and adhesion of THP-1 monocytes to endothelial cells (ECs) [12]. Tese results suggest that BS may not only regulate lipid metabolism, but also improve the infammatory responses of ECs during the development of AS, however, its associated mechanisms remain largely unclear.
With the development of computational methodologies, network pharmacology (NP) has been a novel discipline that can identify drug targets and predict the underlying mechanisms of chemical components [13]. As a branch of NP, molecular docking (MD) can predict the quantitative structure activity relationship (QSAR) between the compound and target, thus achieving the high-throughput screening of herb monomer and providing new insights in the drug discovery and drug design [14]. Recently, these computational methodologies are applied to reveal the regulatory mechanism between the compound and target. For example, Huang et al. successfully predicted and validated the anticancer efects of gentiopicroside by NP [15], and Mu et al. used NP for exploring antiviral actions in rhizoma polygonati [16].

Cell
Culture and Treatment. HUVECs were obtained from Chinese Academy of Sciences Cell Bank (Shanghai, China). After acquisition, cells were cultivated within the endothelial cell medium (ECM), basal medium (Carlsbad, USA) that contained 10% fetal bovine serum (FBS, Gibco, USA) and incubated under 37°C and 5% CO 2 conditions. Tis work adopted the thawed cells in the initial 5 passages. Lipopolysaccharide (LPS), an infammatory agent, was provided by Sigma (St. Louis, USA), followed by dilution to 20 μg/mL with ECM. In the meantime, BS of over 98% purity was obtained in Chengdu Must Bio-technology Co. Ltd (Sichuan, China), followed by dilution to indicated concentrations with ECM before experiments. Ten, HUVECs were subject to 24 hours BS pretreatment and later 24 hours LPS treatment.

Enzyme-Linked Immunosorbent Assay (ELISA).
HUVECs were pretreated under diferent conditions. In line with specifc instructions (Dakewe, China), the supernatant was collected and incubated with biotinylated antibody and streptavidin-HRP. Te absorbance (OD) values determined at 450 nm were used to calculate the contents of infammatory cytokines in the supernatant.

2.5.
Real-Time Quantitative PCR (qRT-PCR). Infammatory cytokine mRNA expression was measured by qRT-PCR. TRIzol reagent was added in cells to extract the total RNA, while RNA quality was determined based on the A260/A280 ratio. According to specifc protocols, Prime-Script RT reagent kit was utilized to collect cDNA by using gDNA Eraser (Takara, Japan). Te SYBR Green (Takara, Japan) was employed for qRT-PCR, while the 2 −ΔΔCt approach was utilized to measure gene levels. Table 1 displays the sequences of all primers.
2.6. Immunofuorescence (IF) Analysis. P65, the NF-κB subunit, is related to IκB/NF-κB pathway activation via nuclear import. Tis work used IF analysis to measure the translocation of p65. In line with specifc protocols (Beyotime, China), HUVECs were subject to fxation, washing, blocking, incubation, and staining by diverse reagents. A laser scanning confocal microscope was employed for result observation.

Statistical Analysis.
Results were represented by mean ± SD. Statistical analysis was completed with one-way ANOVA, while the least-signifcant diference (LSD) test was utilized to compare two groups. p < 0.05 stood for statistical signifcance.
For analyzing target interactions, some bioinformatic methods were applied. Tis work also built the proteinprotein interaction (PPI) network based on diferent evidence channels via STRING (https://string-db.org/). A high pooled score greater than 0.7 was set and the tsv format fle was downloaded. Tis fle was thereafter imported into Cytoscape 3.7.2 for visualization. Gene functions in the PPI network were analyzed through DAVID (https://david. ncifcrf.gov/). With DAVID, Gene Ontology (GO) as well as Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was conducted for investigating biological progresses together with signaling pathways enriched by BSrelated genes against AS.
In addition, the plunge-in CytoNCA of Cytoscape was employed to calculate the "degree centrality (DC)" within genes. Te top 30 targets with the highest DC values were selected as the core targets. With the comprehensive analysis of targets, some core targets closely correlated with AS were considered as the hub targets. Te 3D PDB fles of these hub targets were provided by RCSB PDB (https://www.rcsb.org/ search), and then utilized in the MD analysis.

Validation of Hub Genes with MD.
In this study, AutoDock Vina was applied to display the binding conformation between BS and hub genes. Te crystal structures of hub genes were prepared by AutodockTools to separate local ligand, remove water, add hydrogen, and regulate atom distribution prior to MD, respectively. Te 3D structure of BS was also prepared as the ligand. According to the tutorial of AutoDock Vina, the results of semifexible docking superimposed on the crystal structures of hub genes were obtained. Results with the binding energy <7 kcal were considered as strong combination [17].

Efects of BS on Infammatory Factor Levels.
Te contents of infammatory cytokines were measured by WB, qRT-PCR, and ELISA assays. As a result, IL-1β, IL-6, TNF-α, and Cox-2 levels increased after LPS stimulation. Compared with the LPS group, HUVECs treated with BS produced less infammatory cytokines, indicating the potential antiinfammatory efects of BS (Figures 1(a)-1(f )).

Target Identifcation and Analysis.
BS showed its excellent ability to attenuate the expression and secretion of proinfammatory cytokines. In the following analysis, NP was conducted for exploring the BS-related mechanisms against infammation. Tis work acquired altogether 139 targets in BS in Pharm Mapper, TCMSP, and BATMAN databases, while 4942 targets related to AS were acquired. After overlapping the targets via Venn diagram, 99 potential targets were fnally identifed and chosen in the PPI analysis (Figure 2(a)).
Te disconnected nodes were hidden, and PPI analysis results are shown in Figure 2(b), with 93 nodes and 314 edges in the network. Nodes were labeled as targets while edges were defned as interactions among nodes. Besides, the font size of the node represented the importance degree in the network. For elucidating BS-related mechanism against AS, GO, and KEGG analyses were performed via     Figure 2(e). Some genes such as MAPK8, MAPK1, MAPK14, and NFKB1 were not only the important nodes in the PPI network, but also the major players in biological pathways, which were identifed as the hub genes and validated in further study.

Efects of BS in NF-κB and MAPKs Pathway-related
Protein Levels. As previously reported, MAPK and IκBα/NF-κB pathways have a critical efect on regulating infammatory cytokine production. After pretreatment with BS (1, 5, and 25 μmol/mL), the phosphorylation levels of IκBα and NF-κB were reduced in HUVECs exposed to LPS (Figures 4(a) and 4(b)). Furthermore, the activation of MAPKs family was enhanced in HUVECs after LPS induction. However, BS declined p-p38, p-ERK, and p-JNK expression within MAPKs family to varying degrees dosedependently (Figures 4(c) and 4(d)). IF staining revealed the bright red-stained nuclei following LPS treatment, while BS inhibited the abovementioned alterations, indicating that nuclear translocation of NF-κB (p65) was blocked by the BS treatment (Figure 4(e)). All these fndings were consistent with their prediction.

Discussion
Many studies regarding vascular infammation in atherosclerotic plaques have reported that the imbalance between anti-infammatory and proinfammatory responses is crucial for AS development. As the most common proinfammatory factors in infammatory responses, IL-1β, IL-6, TNF-α, and Cox-2 are recognized to be atherogenic factors. Tere is evidence supporting that the deletion of TNF-α gene reduces the atherosclerotic areas in apoE −/− mice without afecting their serum cholesterol level [18]. Similarly, the plaques narrow in IL-1β −/− /apoE −/− mice or IL-6 −/− /apoE −/− mice [19,20]. Some studies have demonstrated that Cox-2 depends on the previous release of IL-1β and plays an important role in modulating atherosclerotic plaque stability or instability [21]. Other studies report that these cytokines have a great impact on the vascular homeostasis, particularly in the regulation of vascular permeability, coagulation, leukocyte adhesion, and aggregation, which aggravate the disorders of ECs [22]. Blocking proinfammatory factor levels has been recognized as a candidate approach for preventing AS-related CVDs [6]. Treatments using monoclonal antibodies specifcally targeting molecules like IL-1β, IL-6, and TNF-α markedly decrease the relapsed cardiovascular events among patients developing stable CADs [23]. BS, the dietary phytosterol with bioactivity within plants, is used as a pharmaceutical product for a long history. Tough only 5% of BS is obtained from daily intake, it can still lower the serum lipid levels [24,25]. Besides, BS has displayed its potential in vivo and in vitroantiinfammatoryand immunomodulatory activities [8]. To elucidate how BS afected proinfammatory factors, this work measured IL-1β, IL-6, TNF-α, and Cox-2 levels within LPS-treated ECs; as a result, BS declined these infammatory cytokines to varying degrees. Tese results laid the possible pharmacological foundation for the application of BS in treating AS, but detailed mechanisms should be further explored.
With NP, the potential targets of BS to exert anti-AS efects, the biological progresses, signaling pathways, and networks were displayed. A total of 99 targets were screened, which participated in 43 GO biological processes and 42 KEGG pathways. Te GO biological processes were roughly divided into three parts, including protein kinase activity, signaling pathway transduction, and hormone regulation. Most biological processes were closely related to the treatment of AS, such as IκB-α kinase/NF-κB pathway negative regulation, cholesterol homeostasis regulation, and steroid signaling pathway. Some studies have demonstrated the protective efects of estrogen (a major kind of steroid hormone) in the development of CVDs [26]. High-fat diet can induce atherosclerotic lesion, oxidative stress, ICAM-1, and NF-κB in mice, while these sicknesses of atherogenic diet-fed mice are improved after the estrogen treatment [27]. Our KEGG analysis also showed that the estrogen pathway had a critical efect on treating AS by BS. Other studies have suggested that estrogen receptors (ESRs) encoded by ESR1 and ESR2 may exert favorable efects on cells and molecules implicated in vascular infammation, which can impede the development of AS [28]. In addition, KEGG analysis also identifed other pathways associated with immune and infammatory responses, such as MAPK, T cell receptor, TNF, and NOD-like receptor pathways. In addition, hub genes were obtained by network analysis. Some targets, such as MAPK8, MAPK1, MAPK14, and NFKB1, were selected as the hub genes for further MD because they were not only the important nodes in the network, but also the major players in the infammatory signaling pathway. Our MD results showed that BS had a strong binding activity to MAPK8, MAPK1, MAPK14, and NFKB1, respectively. It has been reported that hydrogen bonds and the role of intermolecular forces can be found between small molecule and the active region of target protein. As a result, the small molecule creates inhibition on the target [29], which enables us to conclude that BS can inhibit the activation of MAPK8, MAPK1, MAPK14, and NFKB1. Based on the NP data, it was speculated that the powerful anti-infammatory efects of BS on AS might be correlated with the inhibition of MAPK8, MAPK1, MAPK14, and NFKB1. NP ofers the fast and effective way for drug research; nonetheless, experiments should be conducted to validate targets.
According to our results of in vitro study, IκB-α and NF-κB activation in HUVECs was remarkably impeded after the BS treatment. Moreover, LPS also activated the subunit of NF-κB (p65), resulting in its translocation from cytoplasm to nucleus. However, BS attenuated this process. Te IκB-α/NF-κB pathway accounts for a critical pathway that is activated upon infammatory stress response [6]. Te phosphorylation of IκB contributes to NF-κB dimers release from cytoplasmic IκB-NF-κB complex, their subsequent nuclear import, and regulation of genes that encode proinfammatory factors like IL-1β, IL-6, and TNF-α, as well as the encoding enzymes such as Cox-2 in ECs [30]. Numerous studies have suggested that IκB-α/NF-κB pathway suppression can protect ECs from various stimuli in the development of AS. Song et al. reported that IκB overexpression in ECs signifcantly inhibited the NF-κB activity, mitigated the spontaneous atherosclerotic lesions, and alleviated AS resulting from chronic intermittent hypoxia and cholesterol diet, confrming that the internal endothelial NF-κB pathway was related to atherogenic response [31]. A pharmacologic inhibitor of NF-κB, BAY11-7082 has been proven to suppress proinfammatory factors such as TNF-α and IL-6 and inhibit prothrombotic efects for protecting the endothelium [32].
Te authors' results also suggested that the activation of JNK, ERK, and p-38 was inhibited in HUVECs exposed to LPS after the BS treatment. In addition to NF-κB pathway, the MAPK pathway is important for regulating cell growth, survival, as well as infammatory stimulation responses [33]. JNK, ERK, and p-38, which are members of MAPKs family, regulate the proinfammatory genes encoding IL-1, IL-6, TNF-α, and Cox-2 during ECs activation and contribute to AS progression [34,35]. Yamawaki et al. indicated that the normal blood fow restored TNF-α-induced infammation by the inhibition of JNK and p38 MAPK in ECs [36]. Moreover, some studies have suggested that there is crosstalk between MAPKs and NF-κB pathways. Inhibiting MAPKs can inhibit NF-κB phosphorylation, fnally blocking the amplifcation of endothelial infammation [37], indicating that BS acts via the MAPKs and NF-κB pathways to terminate the infammatory cascade in ECs.

. Conclusion
By conducting NP and experiments, this study verifes that BS may alleviate the infammatory cascade via inactivating MAPK and NF-κB pathways within ECs, which may be possible mechanisms of BS in treating AS. However, only in vitro assays about infammation are conducted, some targets related to metabolism and hormone detected in this work must be further explored. Nevertheless, our research not only provides new insights into the pharmacological actions of BS but also lays the scientifc foundation for applying BS in treating AS.

Data Availability
Te data can be found through the open databases mentioned above.

Conflicts of Interest
Te authors declare that they have no conficts of interest.