https://orcid.org/0000-0002-7773-0568
Figures
Abstract
Ischemic stroke (IS) is one of the major global causes of death and disability. Because blood clots block the neural arteries provoking ischemia and hypoxia in the brain tissue, IS results in irreversible neurological damage. Available IS treatments are currently limited. Curcumin has gained attention for many beneficial effects after IS, including neuroprotective and anti-inflammatory; however, its precise mechanism of action should be further explored. With network pharmacology, molecular docking, and molecular dynamics (MD), this study aimed to comprehensively and systematically investigate the potential targets and molecular mechanisms of curcumin on IS. We screened 1096 IS-related genes, 234 potential targets of curcumin, and 97 intersection targets. KEGG and GO enrichment analyses were performed on these intersecting targets. The findings showed that the treatment of IS using curcumin is via influencing 177 potential signaling pathways (AGE-RAGE signaling pathway, p53 signaling pathway, necroptosis, etc.) and numerous biological processes (the regulation of neuronal death, inflammatory response, etc.), and the AGE-RAGE signaling pathway had the largest degree of enrichment, indicating that it may be the core pathway. We also constructed a protein–protein interaction network and a component–target–pathway network using network pharmacology. From these, five key targets were screened: NFKB1, TP53, AKT1, STAT3, and TNF. To predict the binding conformation and intermolecular affinities of the key targets and compounds, molecular docking was used, whose results indicated that curcumin exhibited strong binding activity to the key targets. Moreover, 100 ns MD simulations further confirmed the docking findings and showed that the curcumin–protein complex could be in a stable state. In conclusion, curcumin affects multiple targets and pathways to inhibit various important pathogenic mechanisms of IS, including oxidative stress, neuronal death, and inflammatory responses. This study offers fresh perspectives on the transformation of curcumin to clinical settings and the development of IS therapeutic agents.
Citation: Wang Y, Zu G, Yu Y, Tang J, Han T, Zhang C (2023) Curcumin’s mechanism of action against ischemic stroke: A network pharmacology and molecular dynamics study. PLoS ONE 18(1): e0280112. https://doi.org/10.1371/journal.pone.0280112
Editor: Divakar Sharma, Maulana Azad Medical College, INDIA
Received: July 13, 2022; Accepted: December 7, 2022; Published: January 4, 2023
Copyright: © 2023 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting information files.
Funding: This work was supported by the first batch of outstanding research and innovation team of Shandong University of Traditional Chinese Medicine. Mechanism and effect evaluation of prevention of major diseases (220316) and Shandong Geriatrics Society 2021 scientific and technological research plan project (No.LKJGG2021Z018). HT and TJ are authors who received awards. And they trevised the manuscript in detail.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Stroke has been identified as a global cerebrovascular disease with increased incidence in younger individuals and high rates of disability and mortality [1]. Stroke can be roughly divided into ischemic stroke (IS) and hemorrhagic stroke. The American Heart Association reports that IS accounts for 87% of total stroke patients [2]. IS is often caused by cerebral artery blockage due to cerebral thrombosis or embolism, resulting in ischemic lesions in the arterial blood supply area of the brain and sensory impairment of patients’ limbs [3]. Thrombolytic therapy is currently the main treatment for IS, but it has some limitations [4]. Cerebral ischemia-reperfusion following thrombolytic therapy may trigger inflammatory responses, which can further lead to secondary brain injury such as neuronal necrosis and brain tissue edema [5]. The limited time window for thrombolytic therapy reduces the number of patients who can receive effective treatment [6]. Therefore, identifying new drugs or chemical components to treat IS has become a current focus in the field.
Curcumin, a food-derived chemical, is a plant polyphenol extracted from turmeric rhizome [7]. Modern medical research has shown that curcumin has various pharmacological properties, including being anti-inflammatory, antioxidant, platelet inhibition, and apoptosis regulation [8–10]. At present, several studies have indicated that curcumin can protect the brain tissue and ameliorate nerve injury, thereby making it a promising IS treatment [11]. However, the potential targets and mechanisms of curcumin in IS treatment are yet to be determined. Network pharmacology is an emerging pharmacological research method that integrates bioinformatics and traditional pharmacology knowledge. It can be used to identify the targets of candidate treatments, as well as the functions and mechanisms of bioactive ingredients in disease treatment [12].
Given that curcumin is a potential therapeutic candidate for IS, this study used network pharmacology to comprehensively and systematically explore the potential targets and intricate mechanisms of how curcumin exerts therapeutic benefits. Additionally, we run molecular docking and molecular dynamic (MD) simulations to verify the stability and affinities of key targets and compound binding. Our study has crucial implications on how curcumin is applied clinically and how well we understand the fundamentals of stroke treatments. The flow chart of this network pharmacology study is shown in Fig 1.
Materials and methods
Curcumin target protein prediction
The protein targets for curcumin were screened in the HERB database (http://herb.ac.cn/), which integrates multiple traditional Chinese medicine databases (including SymMap, TCMID, TCMSP, and TCM-ID) and contains the most comprehensive traditional Chinese medicine and chemical components [13]. Curcumin was also searched in the PubChem database (https://pubchem.ncbi.nlm.nih.gov/) to obtain its corresponding 3D structure and on the simplified molecular input line entry search (SMILES). The SMILES notation was entered into the SwissTargetPrediction database (http://www.swisstargetprediction.ch/) [14] to obtain curcumin’s predicted protein targets and their gene names, which were further screened for a >0.1 probability. The target proteins obtained from the two databases were combined and de-duplicated, and were used as curcumin’s potential targets.
IS disease target screening
The occurrence and development of IS involve the co-regulation of multiple genes. IS’s targets were searched using “ischemic stroke” as the keyword in four databases: DrugBank (https://www.drugbank.ca/) [15], GeneCards (https://www.genecards.org/) [16], DisGeNET (https://www.disgenet.org/) [17], and the Therapeutic Target Database (TTD) (http://db.idrblab.net/ttd/) [18]. UniProt (https://www.uniprot.org/) [19] was used to convert each database’s target information into standardized gene names, which were summarized and de-duplicated.
Getting common targets
Curcumin’s potential targets and IS disease targets were imported into the drawing tool InteractiVenn (http://www.interactivenn.net/) [20]. Venn diagrams were drawn online to identify the common targets.
GO and KEGG enrichment analysis
The common targets were entered into Metascape (https://metascape.org∥) [21]. “Homo sapiens” was selected to conduct the GO and KEGG enrichment analyses to predict the pathways of action and function distribution of the common targets. The GO functional enrichment analysis included biological process, molecular function, and cellular component-related processes. Using the log10(P) as the sorting standard, the 20 most enriched KEGG pathways and their associated top 10 GO functions were selected. Bioinformatics online tools (http://www.bioinformatics.com.cn) were used for visual analysis.
Construction of a protein–protein interaction network and Curcumin–target–pathway network
The STRING database (https://string-db.org/) [22] can search online for protein-protein interaction (PPI) relationships for protein targets. The curcumin–IS common targets were entered into the STRING database. To obtain the network diagram of the interaction between curcumin’s potential targets and IS’s target proteins, terms such as “Homo sapiens,” “>0.9,” and “hide disconnected nodes in the network” were filtered. The results were exported in a “TSV”-formatted file. The TSV file was imported into Cytoscape 3.9.0 software [23]. We conducted network topology analysis, calculated attribute values such as node degree, combined scores, etc., and drew the network diagram of component–disease common target interaction. The PPI was adjusted according to the node degree and combined scores. The intersection targets were sorted according to node degree value, and the top 10 were selected as the candidate key targets for curcumin’s effects in IS treatment.
The 20 selected KEGG pathway and their enriched target genes were sorted. The files were converted into Network.xlsx and Type.xlsx documents, which were then imported into Cytoscape 3.9.0 software to make a curcumin–target–pathway network diagram. The node degree values of the network were computed and the top 10 with higher values were chosen as the potential primary targets for the curcumin treatment of IS. The final key targets were identified from the PPI network topology analysis and curcumin–target–pathway network diagram.
Molecular docking
The PDB database (https://www.rcsb.org/) was used to download the 3D crystal structures of key targets, which were then imported into Pymol 2.1 to remove water molecules and residue ligands and AutoDock Tools-1.5.6 to add hydrogen atoms. The 3D structure of curcumin obtained from the PubChem database was imported into Chem3D for energy optimization, and into AutoDock Tools-1.5.6 for hydrogen addition, atom type assignment, etc. The center of the grid box is established based on the interaction between the processed compound (ligand) and the target proteins (receptor) after their importation into AutoDockTools-1.5.6 [24]. Finally, molecular docking was performed by AutoDock software, and we selected the docking conformation with the lowest binding energy. Visualization was done with Pymol 2.1 software and Schrödinger maestro 2018.
Molecular dynamics
To evaluate the dynamic characteristics and stability of the compound–target protein complex, Gromacs 2020 software was used to run 100 ns MD simulations. In this investigation, the complex generated from the molecular docking results served as the starting conformation. Small molecule ligands were treated with the general force field settings; proteins were treated using AMBER99SB-ILDN force field parameters. Select a periodic stereo box where the atoms of the complex are at least 1.0 nm from the edge of the water box. Select the TIP3P-dominating water model and include salt or chloride ions to balance the charge of the simulated system. The maximum rate of the descent approach and the conjugate gradient method was used to minimize the energy of the solvated system. The system was gradually heated to 300 K in 50 ps and then equilibrated for 50 ps under the NPT (constant number of particles, pressure, and temperature) ensemble. Finally, MD simulations for each equilibrium system were run for 100 ns. The root mean square deviation (RMSD), root mean square fluctuation (RMSF), Hydrogen bond number, and the radius of gyration (RG) were performed on the trajectory data, which were stored every 10 ps.
The binding free energy of the complex was calculated using the MM/GBSA method with the g_MMPBSA script in Gromacs 2020. The binding free energy can be decomposed into molecular mechanical energy and solvation energy. Molecular mechanical energy comprises electrostatic and van der Waal’s interactions, whereas solvation energy comprises polar and nonpolar solvation-free energies [25, 26].
Results
Common targets of curcumin and IS
Curcumin’s 3D structure obtained from PubChem is shown in Fig 2A. The HERB and SwissTargetPrediction databases generated 257 and 64 targets, respectively. Of these, 87 duplicates were removed for a total of 234 potential curcumin targets.
(A) 3D structure of curcumin. (B) Venn diagram of the common target of curcumin and IS.
DrugBank, GeneCards (Median Relevance Score ≥ 2), DisGeNET (Relevance Score ≥0.1), and TTD generated 83, 925, 126, and 14 IS disease targets, respectively. During standardization of the target names, 3 targets were removed due to failure to find their control gene names, while 126 duplicate targets were also removed. In total, 1021 IS disease targets were collected.
The Venn diagrams showed that a total of 97 common targets were obtained between curcumin potential targets and IS disease targets (Fig 2B and S1 Table).
GO and KEGG enrichment analysis
GO functional enrichment analysis was performed for the 97 intersection curcumin–IS targets, revealing a total of 1703 biological processes (BP), 120 molecular functions (MF), and 54 cellular components (CC) involved. The top 10 GO enrichment results were visualized and analyzed (Fig 3A). The BP primarily involved neuron death regulation, inflammatory response and positive regulation of cytokine production, cytokine production, and positive regulation of cell migration. The MF primarily involved oxidoreductase activity, acting on single donors with incorporation of molecular oxygen, amyloid-β binding, cytokine receptor binding, and other processes. The CC primarily involved platelet alpha granule, membrane raft, transcription repressor complex, and organelle outer membrane.
(A) Histogram of GO functional enrichment analysis. (B) Classification diagram for the top 20 KEGG pathways. (C) Bubble Diagram of KEGG enrichment for the top 20 pathways.
The KEGG enrichment results showed that a total of 177 pathways were involved in curcumin treatment of IS. The top 20 pathways can be roughly divided into five categories: human disease, organismal systems, cellular processes, environmental information processing and metabolism (Fig 3B). Curcumin can regulate the AGE-RAGE signaling pathway in diabetic complications, Th17 cell differentiation, the p53 signaling pathway, apelin signaling pathway, necroptosis, and other pathways to treat IS, suggesting that curcumin may affect IS outcomes through intervention of multiple pathways (Fig 3C, S2 Table). Among these, the AGE-RAGE signaling pathway had the highest enrichment, suggesting that it is the most significant pathway involved in curcumin–IS common targets (Fig 4, S2 Table).
Generation of PPI network, curcumin–target–pathway network and key targets
Fig 5A shows the PPI network constructed for the 97 common targets using STRING and Cytoscape 3.9.0. Since free targets were hidden during the screening process, the PPI network contained a total of 81 nodes and 688 edges. The average node degree value was 8.494. The size and color of the node reflect the magnitude of degree value; larger and darker nodes indicate a greater target point degree value than smaller and lighter nodes. The thickness and color of the connection between nodes reflect the magnitude of the tightness value; thicker and darker connections indicate greater tightness values. Sorted by node degree value, the top 10 curcumin–IS candidate key targets were EP300, STAT3, MAPK3, TP53, NFKB1, AKT1, IL6, TNF, MAPK1, and IL1B (Table 1).
(A) PPI network diagram. (B) Curcumin–pathway–target network diagram. The green quadrilateral represents the KEGG pathway, the purple oval represents the target gene, and the pink hexagon represents curcumin.
The curcumin–target–pathway network diagram contained 94 nodes, including 73 common targets, 20 KEGG enrichment pathways, and 1 chemical component, curcumin (Fig 5B). The diagram had 462 edges. Ranked by degree value, the top 10 targets were NFKB1, MAPK8, BAX, TP53, AKT1, STAT3, SMAD3, PIK3CB, CCND1, and TNF (Table 1). Among them, NFKB1, TP53, AKT1, STAT3, and TNF also play important roles in the PPI network structure; thus, they are considered the key targets of curcumin for IS treatment.
Molecular docking results
The five identified key targets—NFKB1(PDB:2O61), TP53(PDB:5BUA), AKT1(PDB:5BUA), STAT3(PDB:6NUQ), TNF (PDB:7KPA)—and curcumin were used for molecular docking. Five repetitions of molecular docking results showed that the binding energies were each <−5.0 kcal/mol, indicating that curcumin has good binding activity with the five targets (Table 2). The docking score (−10.07 kcal/Mol) of curcumin and TNF was statistically considerably lowest than that of other complexes, indicating that they had the strongest binding activity. The visualization results showed that curcumin primarily relied on hydrogen bonding, π–π conjugation, and water transport interactions to stabilize the docking into the binding pocket of the target protein (Fig 6). Curcumin formed hydrogen bonds with residues SER-72 and GLY-66 in the NF-κB binding site, π–π conjugation with residues PHE-53, and hydrophobic interactions with PHE-53, TYR-79 and PRO-68 (Fig 6A). Curcumin formed hydrogen bonds with residues GLU-224, ASN-200, CYS-229 in TP53 binding site, π–π conjugation with HIS-233, and hydrophobic interactions with CYS-229, ILE-232, and so on (Fig 6B). Curcumin formed strong hydrogen bonding interactions with GLU-234, ALA-230, GLU-228, ASP-439 amino residues on AKT1 protein, π–π conjugation interactions with PHE-442, PHE-236, and hydrophobic interactions with PHE-237, PHE-236, VAL-164 and other residues (Fig 6C). Curcumin formed hydrogen bonds with GLU-644, TYR-657, SER-611 and hydrophobic interactions with residues TYR-740, PRO-639 of STAT3 protein structure (Fig 6D). Curcumin forms strong hydrogen bonding interactions with amino acid residues SER-60 and TYR-151 of the TNF protein structure, π–πconjugation interactions with TYR-59 and TYR-119, and hydrophobic interactions with TYR-151, TYR-119, and other residues (Fig 6E).
(A) NF-κB–curcumin. (B) TP53–curcumin. (C) AKT1–curcumin. (D) STAT3–curcumin. (E) TNF-curcumin. Yellow dash represents hydrogen bond distance or π-stacking.
Molecular dynamics results
We calculated the RMSD to explain the structural conformations of the complexes and the system’s stability during the simulation. The average RMSD values of curcumin with the key target AKT1, NF–κB, STAT3, TNF, and TP53 were 1.4, 3.7, 2, 1.5, and 2.4 Å, respectively. All molecules had essentially reached dynamic equilibrium with minor fluctuations after 40 ns. Additionally, there were no faults in the RMSD curves. All the above information suggests that the molecules could firmly attach to the protein during the simulation and not dissociate from the protein pocket (Fig 7).
The fluctuation of amino acid residues in a protein after small molecule binding is reflected by RMSF. The results show that most amino acid conformation changes during the simulation are minor. Only a few residues underwent large conformational changes, most likely due to their location in the protein’s hinge region. Moreover, TNF–curcumin and AKT1–curcumin had RMSF values less than 2.5 Å (Fig 8).
(A) AKT1–curcumin. (B) NF-κB–curcumin. (C) STAT3–curcumin. (D) TNF–curcumin. (E) TP53–curcumin.
The number of hydrogen bonds in a complex can reflect its binding strength. The ligands and residues in all five protein pockets formed one or more hydrogen bonding interactions. AKT1–curcumin had the highest hydrogen bond density and size, followed by TP53–curcumin, TNF–curcumin, NF-κB–curcumin, and STAT3–curcumin (Fig 9).
(A) AKT1–curcumin. (B) NF-κB–curcumin. (C) STAT3–curcumin. (D) TNF–curcumin. (E) TP53–curcumin.
Rg evaluated the compactness of the protein structures. In descending order of compactness, the complexes are TP53–curcumin, AKT1–curcumin, TNF–Curcumin, NF-κB–Curcumin, and STAT3–curcumin (Fig 10).
The binding free energy can be used to determine the change in binding pattern and stability of the ligands and proteins. The free energy formed by the binding of AKT1, NF-κB, STAT3, TNF, and TP53 to curcumin was −94.891±12.951, −86.48±7.536, −60.019±13.929, -114.599±11.442 and −53.592±18.824 kJ/mol, respectively (Table 3). And TNF–curcumin has the lowest binding free energy and is the most strongly bound complex, which is consistent with the docking results. The main reason is that the active site of TNF is long, narrow, and mainly composed of hydrophobic amino acids. The long-form compound has a better chance of forming a solid bond with the TNF protein pocket, allowing the tiny molecule to bind to the protein. Additionally, the energy decomposition results show that van der Waal’s forces, followed by electrostatic interactions, significantly contribute to the stabilization of small molecules (Table 3). Nonpolar solvation energy is only a minor contributor.
Discussion
Cranial injury in patients with IS is caused by various pathological processes, such as inflammatory response, oxidative stress, platelet activation, and blood-brain barrier disruption [27]. Curcumin has multifunctional characteristics, which can reduce nerve tissue damage through multiple target effects with few side effects [28]. This study systematically investigated the potential targets and molecular mechanisms of curcumin on IS.
The results showed that curcumin and IS had 97 common targets. The KEGG enrichment analysis of common targets indicated that curcumin might play a role in IS through regulation of the AGE-RAGE signaling pathway, Th17 cell differentiation, the p53 signaling pathway, the apelin signaling pathway, necrotic apoptosis, and other pathways. Among these, the AGE-RAGE pathway was the most enriched, indicating that it may be the core pathway in curcumin–IS effects. AGE-RAGE plays an important role in diabetic complications, and stroke has been identified as one of the major vascular complications of diabetes [29]. AGE is a harmful molecule formed by macromolecular glycosylation, and RAGE is one of its transmembrane receptors [30]. AGE-RAGE can activate NFKB1 through effects on the MAPK, PI3K/Akt, JAK/STAT, and Nox/ROS signaling pathways, as well as others. The activated NFKB1 enters the nucleus to promote expression of TNF, IL-6, and other inflammatory factors and oxidative stress-related molecules. Ultimately, NFKB1 increases inflammation in the injured neurons and induces a continuous oxidative stress reaction which damages the vascular endothelium, thus further increasing IS severity [31–34]. The inflammatory reaction may, in turn, increase plasma AGE levels and activate the AGE-RAGE pathway to form a continuous cycle of damage [35]. Although AGE-RAGE was the most enriched pathway, Th17 cell differentiation, the p53 signaling pathway, the apelin signaling pathway, and necrotic apoptosis also contribute to inflammation, which may be mediated by curcumin following IS. Curcumin may also improve brain injury outcomes by regulating Th17 cell differentiation. T cells, a harmful component of the IS response, enter the ischemic injury area 24 hours after stroke onset. Th17 is a subtype of T lymphocytes [36] which can secrete IL-17, a pro-inflammatory factor. IL-17 plays a key role in the inflammation caused by ischemic brain injury [37]. The p53 signaling pathway can interact with Bcl.2 family multi-domain members and directly participate in the endogenous apoptosis pathway [38]. Xie et al. demonstrated for the first time that curcumin could increase Bcl-2 expression and inhibit Bax activation, promoting neuronal survival and exerting antiapoptotic biological activity [39]. P53 channels are a promising therapeutic target to reduce stroke injury; they can induce a strong neuroprotective effect against cerebral ischemia-reperfusion injury and significantly reduce brain injury [40]. The apelin signaling pathway has neuroprotective effects against aspartate-mediated excitotoxicity and can reduce inflammatory factor levels, especially TNF expression levels [41]. Necroptosis is one mechanism of neuronal death following stroke. Necrosis causes cell lysis, which releases potential immune inflammatory cytokines to induce a strong neuroinflammatory response, thereby aggravating brain tissue injury from cerebral ischemia processes and cerebral ischemia-reperfusion [42].
The results of the PPI and component–target–pathway network using network pharmacology showed that NFKB1, TP53, AKT1, STAT3, and TNF were key targets in IS regulated by curcumin. NFKB1 is a key transcription factor in the NF-κB signaling pathway and is involved in the release of pro-inflammatory factors such as TNF-α, IL-6, IL-1β, and NLRP3. Curcumin’s anti-inflammatory effects are closely related to its inhibition of the NF-κB signaling pathway [43]. TP53, as a tumor suppressor, can activate pro-apoptotic factors and participate in the apoptotic process, thus further increasing IS severity [44]. AKT1 is highly expressed in the nerve cytoplasm and is a key growth factor, inducing the survival of neurons affected by stroke [45]. AKT1 activation can initiate a downstream cascade reaction of the PI3K/Akt signaling pathway, and it further phosphorylates a series of downstream substrates such as Bad, Caspase-3, and GSK-3β, thereby promoting cell survival and anti-apoptosis [46]. Some studies have suggested that STAT3 activation can not only enhance the microglia’s anti-inflammatory response and inhibit their pro-inflammatory response, but that it can also promote endogenous angiogenesis and neurogenesis in an ischemic brain to effectively inhibit neuronal apoptosis [47, 48]. At present, experiments have demonstrated that curcumin can improve neuronal survival rate by activating JAK2/STAT3 signals, as well as reduce cerebral ischemia-reperfusion injury [49]. Some studies suggest that STAT3 activation might cause neuronal damage [50]. Microglia are resident macrophages of the central nervous system and are known to be involved in the maintenance of the central nervous system stability under normal conditions. They can be roughly divided into pro-inflammatory (M1) and anti-inflammatory phenotypes (M2) [51, 52]. Following ischemia, M1 are rapidly activated and release a variety of inflammatory mediators, such as IL-1 β, IL6, and TNF [53]. Pro-inflammatory factors lead to an inflammatory cascade reaction, aggravate inflammatory expression in the injured area, and participate in a variety of stroke-induced pathophysiological changes, with persistent inflammation causing irreversible damage to the brain cells [54]. At present, studies have shown that curcumin indirectly promotes functional recovery by regulating the polarization of pro-inflammatory microglia/macrophages to anti-inflammatory ones; reducing the release of pro-inflammatory factors such as IL-1, IL-6, and TNF; and reducing inflammatory response [55].
We used molecular docking to predict the binding conformation and intermolecular affinities of the five key targets and compounds. The docking results showed that curcumin can more firmly bind to key proteins through hydrogen bonds, π–π conjugation, and hydrophobic interactions, indicating that curcumin can be used for the treatment of IS by interacting with multiple targets. In this study, the Autodock program was selected for molecular docking because it is now a commonly used tool that can accurately predict the conformation of small molecule ligands within the predicted target binding region [24, 56]. When a protein recognizes or binds a ligand, its conformation changes, but molecular docking ignores the flexibility of this target binding site [57]. This limitation can be overcome by molecular docking.
MD can identify the trajectory and temporal changes of the complex, as well as discover atomic interactions between ligands and protein amino acid residues, which can validate and supplement the results of molecular docking [58]. In this study, the analysis of various data based on MD simulation trajectories shows that curcumin has a strong affinity for the five proteins, which promotes the formation of stable complexes between small molecules and proteins, thereby exerting curcumin’s active role. The insufficient collection of molecular conformations and the considerable processing cost necessary for simulations should not be neglected [59].
Numerous biological properties of curcumin have been reported, including antioxidant, anti-inflammatory, antibacterial, and neuroprotective effects [11, 60]. Our study revealed the pharmacological mechanism of curcumin multipotency in IS. However, there are still some challenges in its therapeutic application. Curcumin’s limited solubility in water, low bioavailability, and brief half-life have prevented it from reaching its full potential for clinical use [61]. Drug delivery systems such as nanoparticles, hydrogels, and other carriers provide solutions [62]. However, the synthesis protocols of curcumin and delivery systems should still be optimized to ensure safety, efficacy, and stability [62, 63].
Conclusions
In conclusion, this study used network pharmacology to identify and analyze potentially relevant curcumin mechanisms for the treatment of IS. The results showed that, in the context of IS-related processes, curcumin may exert anti-inflammatory and apoptosis-inhibiting pharmacological effects through multiple targets—NFKB1, TP53, AKT1, STAT3, and TNF—and by regulating multiple pathways such as the AGE-RAGE signal pathway, Th17 cell differentiation, the p53 signal pathway, the apelin signal pathway, and necrotic apoptosis. Of these, the AGE-RAGE signal pathway was the most enriched pathway of the common curcumin–IS targets, indicating that it may have a critical role in IS treatment. However, the complex development of diseases and pharmacodynamic processes of bioactive ingredients are dynamic, whereas computer biotechnology is a static network analysis, has limited simulation time, etc. Therefore, further experiments and clinical trials are still needed for validation. In the future, dynamic network data analysis or simulation with sufficient time can be developed to drive the data results as close to the actual results as possible, minimizing needless trial-and-error experiments and assisting in the acceleration of the clinical translation of drugs.
Supporting information
S1 Table. The common targets of curcumin and IS.
https://doi.org/10.1371/journal.pone.0280112.s001
(DOC)
S2 Table. The top 20 pathways of KEGG enrichment analysis.
https://doi.org/10.1371/journal.pone.0280112.s002
(DOC)
Acknowledgments
The authors are grateful to Yinglun Pavilion (www.enago.cn) for English language editing.
References
- 1. Ekker MS, Verhoeven JI, Vaartjes I, van Nieuwenhuizen KM, Klijn CJM, de Leeuw FE. Stroke incidence in young adults according to age, subtype, sex, and time trends. Neurology. 2019;92(21):e2444–e54. Epub 2019/04/26. pmid:31019103.
- 2. Virani SS, Alonso A, Aparicio HJ, Benjamin EJ, Bittencourt MS, Callaway CW, et al. Heart Disease and Stroke Statistics-2021 Update: A Report From the American Heart Association. Circulation. 2021;143(8):e254–e743. Epub 2021/01/28. pmid:33501848.
- 3. Tuo QZ, Zhang ST, Lei P. Mechanisms of neuronal cell death in ischemic stroke and their therapeutic implications. Med Res Rev. 2022;42(1):259–305. Epub 2021/05/07. pmid:33957000.
- 4. Ma Y, Li L, Niu Z, Song J, Lin Y, Zhang H, et al. Effect of recombinant plasminogen activator timing on thrombolysis in a novel rat embolic stroke model. Pharmacol Res. 2016;107:291–9. Epub 2016/04/04. pmid:27038532.
- 5. Cao BQ, Tan F, Zhan J, Lai PH. Mechanism underlying treatment of ischemic stroke using acupuncture: transmission and regulation. Neural Regen Res. 2021;16(5):944–54. Epub 2020/11/25. pmid:33229734.
- 6. Mendelson SJ, Prabhakaran S. Diagnosis and Management of Transient Ischemic Attack and Acute Ischemic Stroke: A Review. Jama. 2021;325(11):1088–98. Epub 2021/03/17. pmid:33724327.
- 7. Roy S, Priyadarshi R, Ezati P, Rhim JW. Curcumin and its uses in active and smart food packaging applications—a comprehensive review. Food Chem. 2022;375:131885. Epub 2021/12/26. pmid:34953241.
- 8. Singh L, Sharma S, Xu S, Tewari D, Fang J. Curcumin as a Natural Remedy for Atherosclerosis: A Pharmacological Review. Molecules. 2021;26(13). Epub 2021/07/20. pmid:34279384.
- 9. Rathore P, Dohare P, Varma S, Ray A, Sharma U, Jagannathan NR, et al. Curcuma oil: reduces early accumulation of oxidative product and is anti-apoptogenic in transient focal ischemia in rat brain. Neurochem Res. 2008;33(9):1672–82. Epub 2007/10/24. pmid:17955367.
- 10. Rukoyatkina N, Shpakova V, Bogoutdinova A, Kharazova A, Mindukshev I, Gambaryan S. Curcumin by activation of adenosine A(2A) receptor stimulates protein kinase a and potentiates inhibitory effect of cangrelor on platelets. Biochem Biophys Res Commun. 2022;586:20–6. Epub 2021/11/26. pmid:34823218.
- 11. Fan F, Lei M. Mechanisms Underlying Curcumin-Induced Neuroprotection in Cerebral Ischemia. Front Pharmacol. 2022;13:893118. Epub 2022/05/14. pmid:35559238.
- 12. Xu T, Ma C, Fan S, Deng N, Lian Y, Tan L, et al. Systematic Understanding of the Mechanism of Baicalin against Ischemic Stroke through a Network Pharmacology Approach. Evid Based Complement Alternat Med. 2018;2018:2582843. Epub 2019/01/17. pmid:30647760.
- 13. Fang S, Dong L, Liu L, Guo J, Zhao L, Zhang J, et al. HERB: a high-throughput experiment- and reference-guided database of traditional Chinese medicine. Nucleic Acids Res. 2021;49(D1):D1197–d206. Epub 2020/12/03. pmid:33264402.
- 14. Daina A, Michielin O, Zoete V. SwissTargetPrediction: updated data and new features for efficient prediction of protein targets of small molecules. Nucleic Acids Res. 2019;47(W1):W357–w64. Epub 2019/05/21. pmid:31106366.
- 15. Wishart DS, Feunang YD, Guo AC, Lo EJ, Marcu A, Grant JR, et al. DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Res. 2018;46(D1):D1074–d82. Epub 2017/11/11. pmid:29126136.
- 16. Stelzer G, Rosen N, Plaschkes I, Zimmerman S, Twik M, Fishilevich S, et al. The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. Curr Protoc Bioinformatics. 2016;54:1.30.1–1..3. Epub 2016/06/21. pmid:27322403.
- 17. Piñero J, Ramírez-Anguita JM, Saüch-Pitarch J, Ronzano F, Centeno E, Sanz F, et al. The DisGeNET knowledge platform for disease genomics: 2019 update. Nucleic Acids Res. 2020;48(D1):D845–d55. Epub 2019/11/05. pmid:31680165.
- 18. Zhou Y, Zhang Y, Lian X, Li F, Wang C, Zhu F, et al. Therapeutic target database update 2022: facilitating drug discovery with enriched comparative data of targeted agents. Nucleic Acids Res. 2022;50(D1):D1398–d407. Epub 2021/11/01. pmid:34718717.
- 19. UniProt: the universal protein knowledgebase in 2021. Nucleic Acids Res. 2021;49(D1):D480–d9. Epub 2020/11/26. pmid:33237286.
- 20. Heberle H, Meirelles GV, da Silva FR, Telles GP, Minghim R. InteractiVenn: a web-based tool for the analysis of sets through Venn diagrams. BMC Bioinformatics. 2015;16(1):169. Epub 2015/05/23. pmid:25994840.
- 21. Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523. Epub 2019/04/05. pmid:30944313.
- 22. Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019;47(D1):D607–d13. Epub 2018/11/27. pmid:30476243.
- 23. Franz M, Lopes CT, Huck G, Dong Y, Sumer O, Bader GD. Cytoscape.js: a graph theory library for visualisation and analysis. Bioinformatics. 2016;32(2):309–11. Epub 2015/09/30. pmid:26415722.
- 24. Morris GM, Huey R, Lindstrom W, Sanner MF, Belew RK, Goodsell DS, et al. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J Comput Chem. 2009;30(16):2785–91. Epub 2009/04/29. pmid:19399780.
- 25. Qayed WS, Ferreira RS, Silva JRA. In Silico Study towards Repositioning of FDA-Approved Drug Candidates for Anticoronaviral Therapy: Molecular Docking, Molecular Dynamics and Binding Free Energy Calculations. Molecules. 2022;27(18). Epub 2022/09/24. pmid:36144718.
- 26. Sun H, Li Y, Shen M, Tian S, Xu L, Pan P, et al. Assessing the performance of MM/PBSA and MM/GBSA methods. 5. Improved docking performance using high solute dielectric constant MM/GBSA and MM/PBSA rescoring. Phys Chem Chem Phys. 2014;16(40):22035–45. Epub 2014/09/11. pmid:25205360.
- 27. Paul S, Candelario-Jalil E. Emerging neuroprotective strategies for the treatment of ischemic stroke: An overview of clinical and preclinical studies. Exp Neurol. 2021;335:113518. Epub 2020/11/05. pmid:33144066.
- 28. Bhat A, Mahalakshmi AM, Ray B, Tuladhar S, Hediyal TA, Manthiannem E, et al. Benefits of curcumin in brain disorders. Biofactors. 2019;45(5):666–89. Epub 2019/06/12. pmid:31185140.
- 29. Yan Y, Sun C, Rong X, Han R, Zhu S, Su R, et al. Mechanism of Action of Dengzhan Shengmai in Regulating Stroke from an Inflammatory Perspective: A Preliminary Analysis of Network Pharmacology. Evid Based Complement Alternat Med. 2021;2021:6138854. Epub 2021/11/11. pmid:34754318 conflicts of interest.
- 30. Chaudhuri J, Bains Y, Guha S, Kahn A, Hall D, Bose N, et al. The Role of Advanced Glycation End Products in Aging and Metabolic Diseases: Bridging Association and Causality. Cell Metab. 2018;28(3):337–52. Epub 2018/09/06. pmid:30184484.
- 31. Ramasamy R, Yan SF, Schmidt AM. Receptor for AGE (RAGE): signaling mechanisms in the pathogenesis of diabetes and its complications. Ann N Y Acad Sci. 2011;1243:88–102. Epub 2012/01/04. pmid:22211895.
- 32. Senatus LM, Schmidt AM. The AGE-RAGE Axis: Implications for Age-Associated Arterial Diseases. Front Genet. 2017;8:187. Epub 2017/12/21. pmid:29259621.
- 33. Sorci G, Riuzzi F, Giambanco I, Donato R. RAGE in tissue homeostasis, repair and regeneration. Biochim Biophys Acta. 2013;1833(1):101–9. Epub 2012/10/30. pmid:23103427.
- 34. Wang Z, Zhang J, Chen L, Li J, Zhang H, Guo X. Glycine Suppresses AGE/RAGE Signaling Pathway and Subsequent Oxidative Stress by Restoring Glo1 Function in the Aorta of Diabetic Rats and in HUVECs. Oxid Med Cell Longev. 2019;2019:4628962. Epub 2019/04/05. pmid:30944692.
- 35. Shen CY, Lu CH, Wu CH, Li KJ, Kuo YM, Hsieh SC, et al. The Development of Maillard Reaction, and Advanced Glycation End Product (AGE)-Receptor for AGE (RAGE) Signaling Inhibitors as Novel Therapeutic Strategies for Patients with AGE-Related Diseases. Molecules. 2020;25(23). Epub 2020/12/03. pmid:33261212.
- 36. Planas AM, Chamorro A. Regulatory T cells protect the brain after stroke. Nat Med. 2009;15(2):138–9. Epub 2009/02/07. pmid:19197285.
- 37. Waisman A, Hauptmann J, Regen T. The role of IL-17 in CNS diseases. Acta Neuropathol. 2015;129(5):625–37. Epub 2015/02/27. pmid:25716179.
- 38. Hong LZ, Zhao XY, Zhang HL. p53-mediated neuronal cell death in ischemic brain injury. Neurosci Bull. 2010;26(3):232–40. Epub 2010/05/27. pmid:20502500.
- 39. Xie CJ, Gu AP, Cai J, Wu Y, Chen RC. Curcumin protects neural cells against ischemic injury in N2a cells and mouse brain with ischemic stroke. Brain Behav. 2018;8(2):e00921. Epub 2018/02/28. pmid:29484272.
- 40. Xu ZQ, Yang MG, Liu HJ, Su CQ. Circular RNA hsa_circ_0003221 (circPTK2) promotes the proliferation and migration of bladder cancer cells. J Cell Biochem. 2018;119(4):3317–25. Epub 2017/11/11. pmid:29125888.
- 41. Kleinz MJ, Davenport AP. Emerging roles of apelin in biology and medicine. Pharmacol Ther. 2005;107(2):198–211. Epub 2005/05/24. pmid:15907343.
- 42. Frank D, Vince JE. Pyroptosis versus necroptosis: similarities, differences, and crosstalk. Cell Death Differ. 2019;26(1):99–114. Epub 2018/10/21. pmid:30341423.
- 43. Dong HJ, Shang CZ, Peng DW, Xu J, Xu PX, Zhan L, et al. Curcumin attenuates ischemia-like injury induced IL-1β elevation in brain microvascular endothelial cells via inhibiting MAPK pathways and nuclear factor-κB activation. Neurol Sci. 2014;35(9):1387–92. Epub 2014/03/22. pmid:24651933.
- 44. Gomez-Sanchez JC, Delgado-Esteban M, Rodriguez-Hernandez I, Sobrino T, Perez de la Ossa N, Reverte S, et al. The human Tp53 Arg72Pro polymorphism explains different functional prognosis in stroke. J Exp Med. 2011;208(3):429–37. Epub 2011/03/02. pmid:21357744.
- 45. Zenke K, Muroi M, Tanamoto KI. AKT1 distinctively suppresses MyD88-depenedent and TRIF-dependent Toll-like receptor signaling in a kinase activity-independent manner. Cell Signal. 2018;43:32–9. Epub 2017/12/16. pmid:29242168.
- 46. Du S, Liu J, Liu T. Effect of leonurine on pathological changes of cerebral tissue in ischemic stroke rats based on PI3K/AKT/NF-κB signaling pathway. Chin J Arteriosclerosis. 2019; 27: 853–861.
- 47. Liu ZJ, Ran YY, Qie SY, Gong WJ, Gao FH, Ding ZT, et al. Melatonin protects against ischemic stroke by modulating microglia/macrophage polarization toward anti-inflammatory phenotype through STAT3 pathway. CNS Neurosci Ther. 2019;25(12):1353–62. Epub 2019/12/04. pmid:31793209.
- 48. Hou Y, Wang K, Wan W, Cheng Y, Pu X, Ye X. Resveratrol provides neuroprotection by regulating the JAK2/STAT3/PI3K/AKT/mTOR pathway after stroke in rats. Genes Dis. 2018;5(3):245–55. Epub 2018/10/16. pmid:30320189.
- 49. Li L, Li H, Li M. Curcumin protects against cerebral ischemia-reperfusion injury by activating JAK2/STAT3 signaling pathway in rats. Int J Clin Exp Med. 2015;8(9):14985–91. Epub 2015/12/03. pmid:26628981.
- 50. Choi JS, Kim SY, Cha JH, Choi YS, Sung KW, Oh ST, et al. Upregulation of gp130 and STAT3 activation in the rat hippocampus following transient forebrain ischemia. Glia. 2003;41(3):237–46. Epub 2003/01/16. pmid:12528179.
- 51. Qin C, Fan WH, Liu Q, Shang K, Murugan M, Wu LJ, et al. Fingolimod Protects Against Ischemic White Matter Damage by Modulating Microglia Toward M2 Polarization via STAT3 Pathway. Stroke. 2017;48(12):3336–46. Epub 2017/11/09. pmid:29114096.
- 52. Hu X, Li P, Guo Y, Wang H, Leak RK, Chen S, et al. Microglia/macrophage polarization dynamics reveal novel mechanism of injury expansion after focal cerebral ischemia. Stroke. 2012;43(11):3063–70. Epub 2012/08/31. pmid:22933588.
- 53. Williams L, Bradley L, Smith A, Foxwell B. Signal transducer and activator of transcription 3 is the dominant mediator of the anti-inflammatory effects of IL-10 in human macrophages. J Immunol. 2004;172(1):567–76. Epub 2003/12/23. pmid:14688368.
- 54. Iadecola C, Anrather J. Stroke research at a crossroad: asking the brain for directions. Nat Neurosci. 2011;14(11):1363–8. Epub 2011/10/28. pmid:22030546.
- 55. Liu Z, Ran Y, Huang S, Wen S, Zhang W, Liu X, et al. Curcumin Protects against Ischemic Stroke by Titrating Microglia/Macrophage Polarization. Front Aging Neurosci. 2017;9:233. Epub 2017/08/09. pmid:28785217.
- 56. Wang Z, Sun H, Yao X, Li D, Xu L, Li Y, et al. Comprehensive evaluation of ten docking programs on a diverse set of protein-ligand complexes: the prediction accuracy of sampling power and scoring power. Phys Chem Chem Phys. 2016;18(18):12964–75. Epub 2016/04/26. pmid:27108770.
- 57. Lin JH. Accommodating protein flexibility for structure-based drug design. Curr Top Med Chem. 2011;11(2):171–8. Epub 2010/10/14. pmid:20939792.
- 58. Baig MH, Ahmad K, Rabbani G, Danishuddin M, Choi I. Computer Aided Drug Design and its Application to the Development of Potential Drugs for Neurodegenerative Disorders. Curr Neuropharmacol. 2018;16(6):740–8. Epub 2017/10/20. pmid:29046156.
- 59. Durrant JD, McCammon JA. Molecular dynamics simulations and drug discovery. BMC Biol. 2011;9:71. Epub 2011/11/01. pmid:22035460.
- 60. Pluta R, Furmaga-Jabłońska W, Januszewski S, Czuczwar SJ. Post-Ischemic Brain Neurodegeneration in the Form of Alzheimer’s Disease Proteinopathy: Possible Therapeutic Role of Curcumin. Nutrients. 2022;14(2). Epub 2022/01/22. pmid:35057429.
- 61. Kunnumakkara AB, Harsha C, Banik K, Vikkurthi R, Sailo BL, Bordoloi D, et al. Is curcumin bioavailability a problem in humans: lessons from clinical trials. Expert Opin Drug Metab Toxicol. 2019;15(9):705–33. Epub 2019/07/31. pmid:31361978.
- 62. Mohi-Ud-Din R, Mir RH, Wani TU, Shah AJ, Mohi-Ud-Din I, Dar MA, et al. Novel Drug Delivery System for Curcumin: Implementation to Improve Therapeutic Efficacy against Neurological Disorders. Comb Chem High Throughput Screen. 2022;25(4):607–15. Epub 2021/07/07. pmid:34225614.
- 63. Zheng B, McClements DJ. Formulation of More Efficacious Curcumin Delivery Systems Using Colloid Science: Enhanced Solubility, Stability, and Bioavailability. Molecules. 2020;25(12). Epub 2020/06/21. pmid:32560351.