Sage Journals: Discover world-class research

Abstract

Molecular docking and molecular dynamics simulations were employed to screen plant-derived compounds from Aralia armata for potential inhibition of tumor necrosis factor-alpha (PDB ID: 2AZ5). Aralia armata has a history of use in traditional medicine and contains bioactive compounds with anti-inflammatory potential, such as triterpene glycosides, as reported in previous studies. The docking protocol was validated using root mean square deviation, yielding a value of 1.827 Å, confirming its reliability. The binding affinity analysis revealed that compounds calenduloside E (1), calenduloside G (7), calenduloside G methyl ester (8), and 3β-(β-glucopyranosyloxy) olean-12-en-28-oic acid (38) exhibited the strongest interactions with tumor necrosis factor-alpha, with Tyr B:119 playing a critical stabilizing role. Molecular dynamics simulations demonstrated overall structural stability, with compound 8 exhibiting the lowest binding free energy, surpassing the reference inhibitor SPD304. Absorption, distribution, metabolism, excretion, and toxicity analysis indicated varying pharmacokinetic properties, with compound 8 showing the most balanced absorption, metabolism, and toxicity profile, whereas compound 7 exhibited potential toxicity concerns. These findings suggest that compound 8 is a promising candidate for further in vitro and in vivo validation as a tumor necrosis factor-alpha inhibitor.

Keywords

tumor necrosis factor-alpha inhibitors absorption distribution metabolism excretion and toxicity predictions molecular docking molecular dynamics

Introduction

Tumor necrosis factor-alpha (TNF-α) is a key pro-inflammatory cytokine that plays a vital role in various biological processes, including cell proliferation, differentiation, apoptosis, immune regulation, and inflammation induction.¹ Dysregulation of TNF-α has been implicated in the pathogenesis of numerous inflammatory and autoimmune diseases, such as rheumatoid arthritis (RA), psoriasis, and ankylosing spondylitis.^2,3 Due to its pivotal role in inflammation, TNF-α has become a major therapeutic target in drug development, leading to the introduction of TNF-α inhibitors such as infliximab, etanercept, adalimumab, and golimumab.^4,5 These biologics have significantly transformed the treatment of autoimmune diseases. However, biologic TNF-α inhibitors are often associated with high production costs and require parenteral administration, which can limit patient accessibility and compliance.^6,7 In contrast, small-molecule TNF-α inhibitors are being actively explored as a cost-effective alternative, offering the advantages of oral bioavailability, lower manufacturing costs, and ease of administration. TNF-α is a 17-kDa monomeric protein that forms a biologically active non-covalent homotrimer. Its activity is mediated through interactions with two receptors: TNFR1 and TNFR2. TNFR1 primarily triggers inflammatory signaling pathways, such as the mitogen-activated protein kinase (MAPK) cascade, extracellular signal-regulated kinase (ERK), and c-Jun N-terminal kinase (JNK), which contribute to the progression of autoimmune diseases, including RA and hepatitis.^8–10 In addition, TNFR1 signaling activates the nuclear factor-kappa B (NF-κB) pathway, leading to the transcription of pro-inflammatory cytokines and survival proteins.¹¹ In contrast, TNFR2, which is mainly expressed on immune and endothelial cells, plays a crucial role in immune modulation, tissue regeneration, and vascular homeostasis.¹² Beyond its involvement in autoimmune diseases, TNF-α is also a critical mediator of the inflammatory response during infections.¹³ It is rapidly produced by macrophages, dendritic cells, and T cells in response to pathogen-associated molecular patterns (PAMPs) recognized by Toll-like receptors (TLRs).¹⁴ TNF-α enhances immune defense by increasing vascular permeability, promoting leukocyte recruitment, and stimulating the production of other pro-inflammatory mediators such as interleukin-1 (IL-1) and interleukin-6 (IL-6).¹⁵ However, excessive TNF-α production can lead to systemic inflammation, as seen in conditions such as sepsis and cytokine storm syndromes. Understanding the molecular pathways governing TNF-α activity highlights its dual role in host defense and pathology, reinforcing the importance of targeted therapeutic interventions.

Aralia armata (Wall.) Seem., a member of the Araliaceae family, is a well-known medicinal plant in Vietnam traditionally used to treat hepatitis, arthritis, stomachache, malaria, and snake bites (Figure 1).¹⁶ The main components of A. armata’s leaves and roots have been identified as oleanane-type triterpene glycosides as a result of phytochemical investigations.^17,18 These compounds have demonstrated a variety of bioactivities, including antibacterial, antifungal, antioxidant, anti-inflammatory, anticancer, and anti-HIV effects.^19–25 However, the underlying mechanisms of action of these compounds, particularly in relation to inflammation-associated diseases, remain poorly understood. This suggests the potential of bioactive compounds from A. armata as inhibitors of key biological targets involved in inflammatory pathways, paving the way for the identification of promising anti-inflammatory drug candidates.

Figure 1.

Aralia armata (Wall.) Seem.

Furthermore, computational approaches such as molecular docking combined with molecular dynamics (MD) simulations have proven effective in the preliminary screening of natural compounds, allowing the identification of potential inhibitors of the key target protein. These methods offer a cost-effective and efficient strategy to accelerate the discovery of bioactive molecules for further biological evaluation.^26,27

Therefore, in this study, we conducted in silico investigations on compounds derived from A. armata to support the discovery of potential TNF-α inhibitors for the treatment of inflammatory diseases.

Materials and methods

Molecular docking

The structures of the compounds derived from A. armata were obtained from previous reports and drawn using ChemSketch software.^{17,18,28–31} Then, their structures were energy-optimized through MMFF94s using Avogadro software and prepared for docking using AutoDockTools software, saved in PDBQT format.^32,33 The three-dimensional (3D) protein structure of TNF-α was downloaded from the RCSB Protein Data Bank (PDB ID: 2AZ5).³⁴ Water molecules and co-crystallized ligands were removed using PyMOL software. In addition, polar hydrogen atoms were added to the protein structure, Kollman charges were calculated, and the structure was saved in PDBQT format using AutoDockTools software. The docking process was performed using AutoDock Vina v1.2.3 program with the exhaustiveness parameter set according to previous studies.^35,36 The grid box parameters were set within the active site of the receptor, specifically with center coordinates at x = −19.2 Å, y = 74.5 Å, z = 33.8 Å, a grid size of 24 Å × 24 Å × 24 Å, and a grid spacing of 1 Å. The binding affinity (∆G) of the compounds was ranked, and the most potent compounds were selected. The best docking poses of these compounds were visualized in two-dimensional (2D) and three-dimensional (3D) interaction images using Discovery Studio Visualizer software.

MD

To gain deeper insights into the configurational dynamics of the TNFα system upon ligand binding, we conducted all-atom MD simulations of protein-ligand complexes for 100 ns using GROMACS v.2023.3.³⁷ First, the topology files for the ligand structures were generated using AmberTools and the AnteChamber PYthon Parser interfacE (ACPYPE).^38,39 The initial protein topology was parameterized with the AMBER99SB-ILDN force field.⁴⁰ Each protein–ligand complex was solvated in a triclinic box filled with TIP3P water molecules, ensuring a minimum 4.0-nm buffer between the solute and the box boundaries. The system was then neutralized by adding counterions. Energy minimization was carried out using the steepest descent algorithm, followed by sequential equilibration under NVT and NPT ensembles at 300 K and 1 bar for 100 ps. Temperature and pressure were controlled using the Berendsen thermostat and the Parrinello–Rahman barostat, respectively. Finally, MD simulations were performed for 100 ns with a 2-fs integration timestep. All trajectory coordinates were recorded in the MD simulation files, and the results were visualized and analyzed using Excel software. Furthermore, the binding free energy was calculated using the MMGBSA approach with the gmx_MMPBSA program.⁴¹

ADMET prediction

The online web server pkCSM (https://biosig.lab.uq.edu.au/pkcsm/prediction) was used to predict the pharmacokinetic properties (absorption, distribution, metabolism, excretion, and toxicity) of the selected compounds.⁴² The predicted models include water solubility, Caco-2 permeability, intestinal absorption (human), skin permeability, P-glycoprotein substrate, P-glycoprotein I inhibitor, P-glycoprotein II inhibitor, VDss (human), fraction unbound (human), BBB permeability, CNS permeability, CYP2D6 substrate, CYP3A4 substrate, CYP1A2 inhibitor, CYP2C19 inhibitor, CYP2C9 inhibitor, CYP2D6 inhibitor, CYP3A4 inhibitor, total clearance, renal OCT2 substrate, AMES toxicity, maximum tolerated dose (human), hERG I inhibitor, hERG II inhibitor, oral rat acute toxicity (LD₅₀), oral rat chronic toxicity (LOAEL), hepatotoxicity, skin sensitization, Tetrahymena pyriformis toxicity, and minnow toxicity.

Results and discussion

Molecular docking simulation plays a crucial role in screening plant-derived natural compounds to identify promising candidates for in vitro and in vivo experiments.⁴³ This computational approach predicts the binding affinity between compounds and target proteins, significantly accelerating the screening process while reducing time and cost compared to traditional biological assays.^44,45 In this study, a set of compounds from A. armata was docked into the active site of the TNF-α protein (PDB ID: 2AZ5). Prior to screening, docking protocol validation was conducted to ensure reliability. The accuracy of the protocol was assessed using the root mean square deviation (RMSD) value, which measures the difference between the redocked co-crystallized ligand structure and its experimentally determined conformation. An RMSD value below 2.0 Å typically indicates high reliability, demonstrating that the method can accurately reproduce the ligand’s actual binding position.^35,46 Conversely, a high RMSD suggests the need for adjustments, such as refining simulation parameters, selecting an optimized docking algorithm, or improving input structure quality. The RMSD value obtained in this study was 1.827 Å, which is below 2.0 Å, with minimal deviation as shown in Figure 2, confirming the reliability of the docking protocol. Therefore, this validated approach was applied to the subsequent screening process.

Figure 2.

Superimposition of the co-crystallized ligand (cyan) and the redocked ligand (magentas) in the TNF-α protein with an RMSD value of 1.827 Å.

The binding affinities and molecular interactions of A. armata compounds with TNF-α reveal significant variations in binding strength and interaction patterns among the studied compounds. The binding affinities ranged from −4.741 to −10.27 kcal/mol, as shown in Table S1. Notably, compound 7 exhibited the strongest binding affinity (−10.27 kcal/mol), followed by compounds 38 (−10.12 kcal/mol), 1 (−9.58 kcal/mol), and 8 (−9.54 kcal/mol), all of which had values below −9.5 kcal/mol (Table 1).

Table 1.

Binding Affinities and Molecular Interactions of Top-Hit Compounds With Amino Acid Residues Within the Active Site Pocket of the TNF-α Protein (PDB ID: 2AZ5).

ID	Compound	Binding affinity (kcal/mol)	Interaction type	Amino acid residue Interactions
1	Calenduloside E	-9.58	Conventional Hydrogen bond	Ser B:95
			π-Sigma	Tyr B:59
			Alkyl, π-Alkyl	Leu A:57, Leu B:57, Leu D:55, Tyr B:119, Tyr A:119
7	Calenduloside G	-10.27	Conventional Hydrogen bond	Tyr B:119
7	Calenduloside G	-10.27	Alkyl, π-Alkyl	Leu A:57, Leu B:57, Leu D:55, Tyr A:119
8	Calenduloside G methyl ester	-9.54	Conventional Hydrogen bond	Tyr B:119
8	Calenduloside G methyl ester	-9.54	Alkyl, π-Alkyl	Leu A:57, Leu B:57, Tyr A:119, Tyr B:59
38	3β-(β-glucopyranosyloxy) olean-12-en-28-oic acid	-10.12	Conventional Hydrogen bond	Tyr B:119, Gly A:121
38	3β-(β-glucopyranosyloxy) olean-12-en-28-oic acid	-10.12	Alkyl, π-Alkyl	Leu A:57, Leu B:57, Tyr A:119
Ref.	SPD304	-8.73	Halogen (Fluorine)	Gly A:121
Ref.	SPD304	-8.73	Alkyl, π-Alkyl	Tyr B:119, Tyr B: 59, Tyr B: 151, Leu A:57

These compounds demonstrated strong interactions with TNF-α, suggesting a higher inhibitory potential compared to the others, making them prime candidates for further molecular interaction analysis (Figure 3). Compounds 7, 8, and 38 formed conventional hydrogen bonds with Tyr B:119, underscoring its critical role in stabilizing the compound–TNF-α complex (Table 1). In addition, compound 38 formed an extra hydrogen bond with Gly A:121, potentially enhancing its binding stability and explaining its stronger affinity compared to compounds 1 and 8. Meanwhile, compound 1 exhibited a hydrogen bond with Ser B:95, a unique interaction absent in other compounds, which may contribute to its distinct binding characteristics. Beyond hydrogen bonding, the compounds also engaged in hydrophobic interactions (Alkyl, π-Alkyl) with key amino acids, including Leu A:57, Leu B:57, and Tyr A:119. These residues were consistently involved across all selected compounds, indicating their fundamental role in TNF-α interactions. However, compound 1 showed an additional interaction with Tyr B:59 (pi-sigma interaction), whereas compound 8 lacked interaction with Leu D:55. These differences could explain the slightly lower binding affinity of compound 8 compared to compounds 7 and 38. When compared to the reference inhibitor SPD304 (ref.), the results revealed key interaction similarities and differences. SPD304 formed alkyl and π-alkyl interactions with Tyr B:119, Tyr B:59, Tyr B:151, and Leu A:57, along with a fluorine interaction with Gly A:121. Among the screened compounds, compound 38 exhibited the most similar interaction pattern to SPD304, as it also interacted with Tyr B:119 and Gly A:121. In addition, compound 1 showed similarities with SPD304 in its interaction with Tyr B:59, while compound 7 displayed overlapping interactions with Tyr B:119 and Leu A:57. These findings suggest that compounds share key binding features with SPD304 and may serve as promising inhibitors of TNF-α.

Figure 3.

2D and 3D interactions of top-hit compounds within the active site pocket of the TNF-α protein (PDB ID: 2AZ5).

MD

In MD simulations, root mean square deviation (RMSD) is a key parameter used to assess the structural stability of a system over time. A low and stable RMSD value indicates that the system maintains its structural integrity, while a high or fluctuating RMSD suggests significant conformational changes.^47,48 This stability assessment is evident in Figure 4, where the RMSD values of the protein backbone in complexes 2AZ5-1, 2AZ5-7, 2AZ5-8, and 2AZ5-38 demonstrate overall structural stability, with average values of 0.2913, 0.3546, 0.2598, and 0.2466 nm, respectively. In comparison, the reference system (ref.) and apo-protein exhibit average RMSD values of 0.3033 and 0.2643 nm, respectively. Beyond protein stability, ligand conformational changes can also be monitored through RMSD values. As shown in Figure 3, ligand 1 in the 2AZ5-1 complex initially decreases from approximately 0.2 to 0.1 nm around 20 ns, then increases again near 60 ns, fluctuating until the end of the simulation. Meanwhile, ligands 7 and 38 remain highly stable, with average RMSD values of 0.1678 and 0.225 nm, respectively. In contrast, ligand 8 exhibits more significant fluctuations, with its RMSD rising from approximately 0.1 to 0.35 nm between 20 and 30 ns. Overall, ligands 1, 7, and 38 undergo structural adaptations within the TNF-α active site, showing similarities to the reference compound SPD304.

Figure 4.

Root mean square deviation (RMSD) analysis of the backbone (left) and ligand (right) atoms of TNF-α complexes.

While RMSD provides insights into the structural stability of protein–ligand complexes, a more comprehensive understanding of their binding strength requires energetic analysis. To achieve this, the MMGBSA (Molecular Mechanics/Generalized Born Surface Area) method is employed to estimate the binding free energy between small molecules and proteins.^49,50 Compared to conventional docking methods, MMGBSA offers superior accuracy by accounting for solvent effects and system flexibility. While docking primarily relies on a rigid model of the protein–ligand complex and evaluates only static interactions, MMGBSA recalculates binding energy across multiple conformational states derived from MD simulations. This dynamic evaluation provides a more realistic representation of binding interactions in biological environments, making MMGBSA particularly valuable for systems with significant conformational changes or strong solvent interactions.^49,50 The data in Table 2 present the binding free energy of various compounds with TNF-α, a crucial target in pharmaceutical research. The total binding free energy (ΔG_total) is derived from several energy components, including van der Waals interactions (E_VDWAALS), electrostatic interactions (E_EL), polar solvation energy (E_GB), and surface energy (E_SURF). A more negative ΔG_total indicates a stronger binding affinity between the compound and the protein. The results show that compound 8 exhibits the lowest binding free energy (−45.2288 kcal/mol), suggesting the strongest interaction with TNF-α. This is likely due to a significant contribution from van der Waals interactions (−60.5256 kcal/mol) and an optimal balance between gas-phase and solvation energies. Compared to the reference compound (ref, −36.3253 kcal/mol), compound 8 displays a binding affinity nearly 9 kcal/mol stronger, highlighting its potential as a more effective TNF-α inhibitor. Compound 1 (−37.5591 kcal/mol) exhibits a binding affinity comparable to the reference, indicating stable interactions. In contrast, compounds 7 (−29.3309 kcal/mol) and 38 (−29.4251 kcal/mol) show significantly weaker binding than the reference, suggesting they may be less effective as TNF-α inhibitors. The structural correlation among compounds 1, 7, 8, and 38 reveals that they all share the same olean-12-en-28-oic acid core, differing only in the substituents at the C-3 position. Notably, the β-d-galactopyranosyl-(1→3)-β-d-glucuronopyranoside-6-O-methyl ester group in compound 8 significantly enhances binding free energy compared to the other substitutions. Previous studies have highlighted that compounds with derivatives of olean-12-en-28-oic acid exhibit strong in vitro activity against TNF-α, including 3β,6β,23-trihydroxyolean-12-en-28-oic acid 28-O-β-d-glucopyranoside, nipponogenin E, 3β,6β,23-trihydroxyolean-12-en-28-oic acid, and 2α,3β,19α-trihy-droxy-24-oxo-olean-12-en-28-oic acid.^51,52 These findings provide a foundation for further studies aimed at optimizing molecular structures to enhance binding efficiency with the target protein.

Table 2.

The Bind Free Energy (in kcal/mol) Analysis of TNF-α and Selected Compounds Complexes, Determined via a MMGBSA Method.

Energy component (kcal/mol)	Compounds
Energy component (kcal/mol)	1	7	8	38	ref
E_VDWAALS	−48.85	−45.04	−60.53	−44.19	−48.71
E_EL	133.42	149.95	30.43	65.40	−5.90
E_GB	−116.08	−128.37	−7.78	−45.52	24.10
E_SURF	−6.05	−5.87	−7.35	−5.11	−5.81
ΔG_gas	84.57	104.91	−30.09	21.20	−54.61
ΔG_solv	−122.13	−134.24	−15.13	−50.63	18.29
ΔG_Total	−37.56	−29.33	−45.23	−29.43	−36.33

ADMET profiles

Based on the ADMET data of the four selected compo-unds (Table 3), their absorption, distribution, metabolism, excretion, and toxicity can be assessed to determine their pharmacokinetic properties and safety profiles. All four compounds exhibit low water solubility (−2.96 to −3.249), which may negatively impact their oral bioavailability. Their permeability across the Caco-2 membrane is also poor (−0.109 to −0.404), suggesting limited passive diffusion through the intestinal epithelium. However, intestinal absorption varies significantly, with compound 38 having the highest absorption (40.731%), followed by compound 8 (34.086%), while compound 7 shows the lowest (5.582%), indicating potential difficulties in gastrointestinal uptake. In addition, all compounds are substrates of P-glycoprotein, a transporter that actively pumps drugs out of cells, which could reduce intracellular drug concentrations and impact therapeutic efficacy. In terms of distribution, all four compounds have negative volume of distribution (VDss) values, meaning they are primarily retained in plasma rather than extensively distributed into tissues. Among them, compound 8 shows the highest VDss (−0.172), suggesting slightly better tissue penetration. The fraction of unbound drug in plasma is highest for compound 7 (0.31), which may enhance its biological effects. However, none of the compounds efficiently cross the blood–brain barrier, particularly compounds 7 and 8 (−2.072 and −2.07), making them unlikely to cause central nervous system effects an advantage when avoiding neurological side effects. Regarding metabolism, compounds 8 and 38 are substrates of CYP3A4, a crucial enzyme in drug metabolism, implying that their pharmacokinetics may be affected by CYP3A4 inhibitors or inducers. However, none of the compounds inhibit major CYP enzymes, suggesting a low risk of drug–drug interactions due to metabolic inhibition. The elimination of these compounds is relatively slow, with total clearance values ranging from −0.158 to −0.028, indicating prolonged systemic circulation. None of the compounds are substrates of OCT2, suggesting that renal excretion is not a significant pathway for elimination. In terms of toxicity, none of the compounds tested positive in the AMES assay, indicating a low risk of genotoxicity. However, the maximum tolerated dose in humans varies, with compound 7 showing a positive value (0.183), while compound 8 has a negative value (−0.835), suggesting potential tolerability concerns for compound 8. Encouragingly, none of the compounds inhibit the hERG channel, minimizing the risk of cardiotoxicity. Acute toxicity (LD₅₀) in mice is relatively similar across compounds (2.489–2.69), but compound 7 has the highest chronic toxicity (LOAEL: 8.064), raising concerns about long-term accumulation and toxicity risks. Environmental toxicity data reveal that compound 7 poses the greatest ecological risk, with the highest Minnow toxicity value (6.708), whereas compound 38 has the lowest (−0.665), suggesting a lower environmental impact. Overall, compounds 8 and 38 demonstrate superior intestinal absorption compared to compounds 1 and 7. However, compound 7 presents the highest toxicity risks, including chronic toxicity and significant environmental concerns. Compound 38 has the slowest clearance rate, potentially leading to prolonged therapeutic effects. Among the four, compound 8 appears to offer the most balanced profile in terms of absorption, metabolism, and toxicity, whereas compound 7 raises the greatest safety concerns.

Table 3.

ADMET Profiles of Selected Compounds.

Property	Model name	Compounds
Property	Model name	1	7	8	38
Absorption	Water solubility	−3.249	−2.898	−2.96	−2.995
Absorption	Caco2 permeability	−0.193	−0.404	−0.109	−0.347
Absorption	Intestinal absorption (human)	19.405	5.582	34.086	40.731
Absorption	Skin Permeability	−2.735	−2.735	−2.735	−2.735
Absorption	P-glycoprotein substrate	Yes	Yes	Yes	Yes
Absorption	P-glycoprotein I inhibitor	No	No	No	No
Absorption	P-glycoprotein II inhibitor	No	No	No	No
Distribution	VDss (human)	−1.047	−0.691	−0.172	−1.612
Distribution	Fraction unbound (human)	0.189	0.31	0.223	0.224
Distribution	BBB permeability	−1.306	−2.072	−2.07	−1.166
Distribution	CNS permeability	−2.74	−3.566	−3.483	−3.266
Metabolism	CYP2D6 substrate	No	No	No	No
Metabolism	CYP3A4 substrate	No	No	Yes	Yes
Metabolism	CYP1A2 inhibitior	No	No	No	No
Metabolism	CYP2C19 inhibitior	No	No	No	No
Metabolism	CYP2C9 inhibitior	No	No	No	No
Metabolism	CYP2D6 inhibitior	No	No	No	No
Metabolism	CYP3A4 inhibitior	No	No	No	No
Excretion	Total Clearance	−0.158	−0.127	−0.069	−0.028
Excretion	Renal OCT2 substrate	No	No	No	No
Toxicity	AMES toxicity	No	No	No	No
Toxicity	Max. tolerated dose (human)	−0.149	0.183	−0.835	−0.036
Toxicity	hERG I inhibitor	No	No	No	No
Toxicity	hERG II inhibitor	No	No	No	No
Toxicity	Oral Rat Acute Toxicity (LD₅₀)	2.659	2.489	2.69	2.63
Toxicity	Oral Rat Chronic Toxicity (LOAEL)	3.713	8.064	7.352	3.026
Toxicity	Hepatotoxicity	No	No	No	No
Toxicity	Skin Sensitisation	No	No	No	No
Toxicity	Tetrahymena pyriformis toxicity	0.285	0.285	0.285	0.285
Toxicity	Minnow toxicity	1.021	6.708	5.003	−0.665

Conclusion

This study utilized molecular docking, MD simulations, and MMGBSA binding energy calculations to identify promising TNF-α inhibitors from A. armata compounds. The docking protocol was validated with an RMSD value of 1.827 Å, ensuring accuracy in predicting binding interactions. Among the screened compounds, compound 7 exhibited the strongest binding affinity (−10.27 kcal/mol), followed by compounds 38, 1, and 8. Molecular interactions revealed that hydrogen bonding with Tyr B:119 and hydrophobic interactions with key residues played a crucial role in stabilizing the protein–ligand complexes. MD simulations confirmed the stability of these interactions, with compound 8 demonstrating the lowest binding free energy (−45.2288 kcal/mol), surpassing the reference inhibitor SPD304. However, ADMET analysis highlighted differences in pharmacokinetic properties, with compound 8 showing the most balanced absorption, metabolism, and toxicity profile. In contrast, compound 7, despite its high binding affinity, presented significant toxicity concerns, including chronic toxicity and environmental risks. Overall, compound 8 emerged as the most promising candidate for further experimental validation due to its strong binding interactions, stability, and favorable pharmacokinetic profile. Future studies should focus on optimizing its molecular structure to enhance bioavailability and minimize potential safety risks.

Supplemental Material

sj-docx-1-chl-10.1177_17475198251339509 – Supplemental material for Phytochemicals from Aralia armata as potential tumor necrosis factor-alpha inhibitors: A computational study

Supplemental material, sj-docx-1-chl-10.1177_17475198251339509 for Phytochemicals from Aralia armata as potential tumor necrosis factor-alpha inhibitors: A computational study by Vu Thi Thu Le, Luc Quang Tan, Do Khac Hung and Nguyen Xuan Ha in Journal of Chemical Research

Footnotes

ORCID iD

Nguyen Xuan Ha

Ethical Considerations

Ethical approval is not applicable to the article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was financially supported under the project with the code: B2024-TNA-28 by the Ministry of Education and Training, Vietnam.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Statement of human and animal rights

This article does not contain any studies with human or animal subjects.

Statement of informed consent

There are no human subjects in this article, and informed consent is not applicable.

Supplemental material

Supplemental material for this article is available online.

References

Wijbrandts

Dijkgraaf

MGW

Kraan

, et al. The clinical response to infliximab in rheumatoid arthritis is in part dependent on pretreatment tumour necrosis factor α expression in the synovium. Ann Rheum Dis 2008; 67(8): 1139–1144.

Hyrich

Lunt

Watson

, et al. Outcomes after switching from one anti-tumor necrosis factor α agent to a second anti-tumor necrosis factor α agent in patients with rheumatoid arthritis: results from a large UK national cohort study. Arthritis Rheum 2007; 56: 13–20.

Schulz

Dotzlaw

Neeck

. Ankylosing spondylitis and rheumatoid arthritis: serum levels of TNF-alpha and its soluble receptors during the course of therapy with etanercept and infliximab. Biomed Res Int 2014; 2014: 675108.

Jin

Chang

Wei

. Clinical application and evaluation of anti-TNF-alpha agents for the treatment of rheumatoid arthritis. Acta Pharmacol Sin 2010; 31: 1133–1140.

Inman

Baraliakos

Hermann

, et al. Serum biomarkers and changes in clinical/MRI evidence of golimumab-treated patients with ankylosing spondylitis: results of the randomized, placebo-controlled GO-RAISE study. Arthritis Res Ther 2016; 18: 304.

Kirchhoff

Mittendorf

Schmidt

, et al. Cost-effectiveness of TNF-α inhibition in active ankylosing spondylitis: a systematic appraisal of the literature. Expert Rev Pharmacoecon Outcomes Res 2012; 12(3): 307–317.

Nash

Florin

. Tumour necrosis factor inhibitors. Med J Aust 2005; 183(4): 205–208.

Esposito

Cuzzocrea

. TNF-alpha as a therapeutic target in inflammatory diseases, ischemia-reperfusion injury and trauma. Curr Med Chem 2009; 16(24): 3152–3167.

Mukai

Nakamura

Yoshikawa

, et al. Solution of the structure of the TNF-TNFR2 complex. Sci Signal 2010; 3(148): ra83.

10.

Mancini

Toro

Mabilia

, et al. Inhibition of tumor necrosis factor-a (TNF-a)/TNFa receptor binding by structural analogues of suramin. Biochem Pharmacol 1999; 58(5): 851–859.

11.

Hayden

Ghosh

. Regulation of NF-κB by TNF family cytokines. Semin Immunol 2014; 26(3): 253–266.

12.

Yang

Islam

, et al. TNFR2: role in cancer immunology and immunotherapy. Immunotargets Ther 2021; 10: 103–122.

13.

Jang

Lee

Shin

, et al. The role of tumor necrosis factor alpha (TNF-α) in autoimmune disease and current TNF-α inhibitors in therapeutics. Int J Mol Sci 2021; 22(5): 2719.

14.

Yang

Wang

Brand

, et al. Role of TNF–TNF receptor 2 signal in regulatory T cells and its therapeutic implications. Front Immunol 2018; 9: 784.

15.

Sprague

Khalil

. Inflammatory cytokines in vascular dysfunction and vascular disease. Biochem Pharmacol 2009; 78(6): 539–552.

16.

Chi

. Dictionary of Vietnamese medicinal plants. Hanoi, Thailand: Medicine Publishing House, 2012, 956–957.

17.

Miao

Sun

Yuan

, et al. Herbicidal and cytotoxic constituents from Aralia armata (Wall.) Seem. Chem Biodiversity 2016; 13(4): 437–444.

18.

Ogawa

Sashida

, et al. Triterpenoid glucuronide saponins from root bark of Aralia armata. Phytochem 1995; 39(1): 179–184.

19.

Sonfack

Fossi Tchinda

Simo

, et al. Saponin with antibacterial activity from the roots of Albizia adianthifolia. Nat Prod Res 2021; 35(17): 2831–2839.

20.

Chen

Jiao

, et al. Triterpenoid saponins from the pulp of Sapindus mukorossi and their antifungal activities. Phytochem 2018; 147: 1–8.

21.

Liu

Yang

Feng

, et al. Eight new triterpenoid saponins with antioxidant activity from the roots of Glycyrrhiza uralensis Fisch. Fitoterapia 2019; 133: 186–192.

22.

Rho

Jeong

Hong

, et al. Identification of a novel triterpene saponin from Panax ginseng seeds, pseudoginsenoside RT8, and its antiinflammatory activity. J Ginseng Res 2020; 44(1): 145–153.

23.

Liu

Deng

Zhou

, et al. 3,4-seco-Dammarane Triterpenoid saponins with anti-inflammatory activity isolated from the leaves of Cyclocarya paliurus. J Agric Food Chem 2020; 68(7): 2041–2053.

24.

Konoshima

Yasuda

Kashiwada

, et al. Anti-AIDS agents, 21. Triterpenoid saponins as anti-HIV principles from fruits of Gleditsia japonica and Gymnocladus chinesis, and a structure-activity correlation. J Nat Prod 1995; 58(9): 1372–1377.

25.

Desai

Joshi

. Anticancer activity of saponin isolated from Albizia lebbeck using various in vitro models. J Ethnopharmacol 2019; 231: 494–502.

26.

Singh

Baker

Singh

. Molecular docking and molecular dynamics simulation. In: Singh

Pathak

(eds) Bioinformatics. Amsterdam: Elsevier, 2022, pp. 291–304.

27.

Santos

Ferreira

Caffarena

. Integrating molecular docking and molecular dynamics simulations. Methods Mol Biol 2019; 2053: 13–34.

28.

Yen

Chuong

NTH

Lien

GTK

, et al. Oleanane-type triterpene saponins from Aralia armata leaves and their cytotoxic activity. Nat Prod Res 2021; 36(1): 142–149.

29.

Chuong

NTH

Thuy Van

Kim Lien

, et al. Aramatosides C and D, 2 previously undescribed triterpene glycosides isolated from the roots of Aralia armata. Nat Prod Commun 2021; 16(7): 1934578X211033686.

30.

Chuong Nguyen

Kim Lien

Yen

, et al. Molluscicidal activity of compounds from the roots of Aralia armata against the golden apple snail (Pomacea canaliculata). Nat Prod Commun 2022; 17(12): 1934578X221144573.

31.

Yen

Cuc

Huong

PTT

, et al. Araliaarmoside: a new triterpene glycoside isolated from the leaves of Aralia armata. Nat Prod Commun 2020; 15(9): 1934578X20953300.

32.

Halgren

TA.

MMFF

. MMFF94s option for energy minimization studies. J Comput Chem 1999; 20(7): 720–729.

33.

Hanwell

Curtis

Lonie

, et al. Avogadro: an advanced semantic chemical editor, visualization, and analysis platform. J Cheminform 2012; 4: 1–17.

34.

Smith

Oslob

, et al. Small-molecule inhibition of TNF-α. Science 2005; 310(5750): 1022–1025.

35.

Trang

HTT

Xuan Ha

Hong Le

, et al. In silico molecular docking, DFT, and toxicity studies of potential inhibitors derived from Millettia dielsiana against human inducible nitric oxide synthase. J Chem Res 2024; 48(4): 17475198241263837.

36.

Hoang Minh Chau

Thi Tra Giang

Thi Thuy Tram

, et al. In silico molecular docking and molecular dynamics of Prinsepia utilis phytochemicals as potential inhibitors of phosphodiesterase 4B. J Chem Res 2024; 48(6): 17475198241305879.

37.

Abraham

Murtola

Schulz

, et al. GROMACS: high performance molecular simulations through multi-level parallelism from laptops to supercomputers. Softwarex 2015; 1: 19–25.

38.

Sousa da Silva

Vranken

. ACPYPE-Antechamber python parser interface. BMC Res Notes 2012; 5: 1–8.

39.

Case

Aktulga

Belfon

, et al. AmberTools. J Chem Inf Model 2023; 63(20): 6183–6191.

40.

Lindorff

Larsen-K

Piana

Palmo

, et al. Improved side-chain torsion potentials for the Amber ff99SB protein force field. Proteins 2010; 78(8): 1950–1958.

41.

Valdés-Tresanco

Valiente

, et al. gmx_MMPBSA: a new tool to perform end-state free energy calculations with GROMACS. J Chem Theory Comput 2021; 17(10): 6281–6291.

42.

Pires

Blundell

Ascher

. pkCSM: predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. J Med Chem 2015; 58(9): 4066–4072.

43.

Asiamah

Obiri

Tamekloe

, et al. Applications of molecular docking in natural products-based drug discovery. Sci Afr 2023; 20: e01593.

44.

Morris

Lim-Wilby

. Molecular docking. Methods Mol Bio 2008; 443: 365–382.

45.

Fan

Zhang

. Progress in molecular docking. Quant Biol 2019; 7: 83–89.

46.

. In silico molecular docking and ADMET study of Isodon coetsa phytochemicals targeting TNF-α in inflammation-mediated diseases. Vietnam J Chem 2024; 62: 387–393.

47.

Shukla

Tripathi

. Molecular dynamics simulation of protein and protein–ligand complexes. In: Singh

(ed.) Computer-aided Drug Design. Singapore: Springer, 2020, pp. 133–161.

48.

Liu

Watanabe

Kokubo

. Exploring the stability of ligand binding modes to proteins by molecular dynamics simulations. J Comput Aided Mol Des 2017; 31: 201–211.

49.

Wang

Sun

Wang

, et al. End-point binding free energy calculation with MM/PBSA and MM/GBSA: strategies and applications in drug design. Chem Rev 2019; 119(16): 9478–9508.

50.

Genheden

Ryde

. The MM/PBSA and MM/GBSA methods to estimate ligand-binding affinities. Expert Opin Drug Discov 2015; 10(5): 449–461.

51.

Quang

Ngan

NTT

Van Minh

, et al. Inhibitory effects of oleanane-type triterpenes and saponins from the stem bark of Kalopanax pictus on LPS-stimulated pro-inflammatory cytokine production in bone marrow-derived dendritic cells. Arch Pharm Res 2013; 36: 327–334.

52.

Huang

Wang

, et al. Anti-inflammatory oleanolic triterpenes from Chinese Acorns. Molecules 2016; 21(5): 669.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.34 MB