Unveiling the Anti-Obesity Potential of Thunder God Vine: Network Pharmacology and Computational Insights into Celastrol-like Molecules

Abstract

Obesity, characterized by abnormal or excessive fat accumulation, has become a chronic degenerative health condition that poses significant threats to overall well-being. Pharmacological intervention stands at the forefront of strategies to combat this issue. Recent studies, notably by Umut Ozcan's team, have uncovered the remarkable potential of Celastrol, a small-molecule compound derived from the traditional Chinese herb thunder god vine (Tripterygium wilfordii) as an anti-obesity agent. In this research, computational chemical analysis was employed, incorporating the "TriDimensional Hierarchical Fingerprint Clustering with Tanimoto Representative Selection (3DHFC-TRS)" algorithm to systematically explore 139 active small molecules from thunder god vine. These compounds were classified into six categories, with a particular focus on Category 1 molecules for their exceptional binding affinity to obesity-related targets, offering new avenues for therapeutic development. Using advanced molecular docking techniques and Cytoscape prediction models, six representative Celastrol-like molecules were identified, namely 3-Epikatonic Acid, Hederagenin, Triptonide, Triptotriterpenic Acid B, Triptotriterpenic Acid C, and Ursolic Acid. These compounds demonstrated superior binding affinity and specificity toward two key obesity targets, PPARG and PTGS2, suggesting their potential to regulate fat metabolism and mitigate inflammatory responses. To further substantiate these findings, molecular dynamics simulations and MM-PBSA free-energy calculations were applied to analyze the dynamic interactions between these small molecules and the enzymatic active sites of their targets. The results provide robust theoretical evidence that support the feasibility of these molecules as promising candidates for anti-obesity therapies. This study underscores the power of the 3DHFC-TRS algorithm in uncovering bioactive compounds from natural sources, such as thunder god vine, and highlights the therapeutic promise of PPARG and PTGS2 as novel obesity-related targets. Furthermore, it emphasizes the essential role of computational science in expediting drug discovery, paving the way for personalized and precision-based treatments for obesity and heralding a future of more effective healthcare solutions.

Keywords: celastrol-like molecules; molecular docking; molecular dynamics simulations; network pharmacology.

Figure 1

As illustrated in Figure 1, the network pharmacology workflow combines machine learning and quantitative computational methods to identify the key targets and related molecular mechanisms of the thunder god vine for the treatment of obesity. This process includes (A) Prediction of drug-disease targets, (B) Clustering, (C) Network pharmacology, (D) Molecular docking, (E) Molecular Dynamics Simulations.

Figure 2

Cluster analysis of 139 molecules of the thunder god vine.(A) The first cluster, highlighted in red. (B) The second cluster, highlighted in cyan. (C) The third cluster, highlighted in purple. (D) The fourth cluster, highlighted in orange. (E) The fifth cluster, highlighted in yellow. (F) The sixth cluster, highlighted in pink.

Figure 3

The intersection of the predicted targets and obesity targets of the representative six groups of small molecules. (A) Tripterygone, (B) Wilfotrine, (C) DBP, (D) Triptonide, (E) Tripteroside, and (F) Wilforlide B.

Figure 4

Using the Cytoscape analysis tool, we selected six molecules that correspond to the top ten hub genes based on their degree. (A) 3-Epikatonic Acid, (B) Hederagenin, (C) Triptonide, (D) Triptotriterpenic Acid B, (E) Triptotriterpenic Acid C, and (F) Ursolic Acid.

Figure 5

Visualization of molecular docking of six compounds with PPARG. (A) 3-Epikatonic Acid, (B) Hederagenin, (C) Triptonide, (D) Triptotriterpenic Acid B, (E) Triptotriterpenic Acid C, and (F) Ursolic Acid.

Figure 6

Visualization of molecular docking of six compounds with PTGS2 (A) 3-Epikatonic Acid, (B) Hederagenin, (C) Triptonide, (D) Triptotriterpenic Acid B, (E) Triptotriterpenic Acid C, and (F) Ursolic Acid.

Figure 7

Structural stability analysis. (3-Epikatonic Acid, highlighted in black. Hederagenin, highlighted in red. Triptonide, highlighted in blue. Triptotriterpenic Acid B, highlighted in green. Triptotriterpenic Acid C, highlighted in purple. Ursolic Acid, highlighted in yellow.) (A) Temporal evolution of the RMSD from their initial structures for four systems. (B) Radius of gyration of four systems over a 100 ns MD simulation. (C) Root-mean-square fluctuation (RMSF) of the four systems. (D) SASA during a 100 ns MD simulation.

Figure 8

Structural stability analysis. (3-Epikatonic Acid, highlighted in black. Hederagenin, highlighted in red. Triptonide, highlighted in blue. Triptotriterpenic Acid B, highlighted in green. Triptotriterpenic Acid C, highlighted in purple. Ursolic Acid, highlighted in yellow.) (A) RMSD from their initial structures for four systems. (B) Radius of gyration of four systems over a 100 ns MD simulation. (C) RMSF of the four systems. (D) SASA during a 100 ns MD simulation.

Figure 9

Planar structures of the six molecules, the red parts indicate the presence of oxygen atoms. (A) 3-Epikatonic Acid, (B) Hederagenin, (C) Triptonide, (D) Triptotriterpenic Acid B, (E) Triptotriterpenic Acid C, and (F) Ursolic Acid.

Figure 10

Screening of additional genes associated with PPARG in the GeneMANIA framework.

Figure 11

Screening of additional genes associated with PTGS2 in the GeneMANIA framework.