Unleashing the potential of traditional Chinese medicine: a computational approach to discovering drug targets utilizing the CSLN and molecular dynamics

The diverse chemical components of traditional Chinese medicine (TCM) exhibit significant therapeutic potential; however, the action mechanisms of these compounds often remain unclear. The use of drug-target prediction can aid in identifying the specific targets of TCM, thereby revealing their bioactivity and mechanisms. The efficiency, cost-effectiveness, and powerful predictive capabilities of artificial intelligence algorithms have led to their emergence as effective tools for accelerating drug-target interaction analysis. To systematically investigate TCM interaction mechanisms, we integrated cosine‑correlation and similarity‑comparison of local network (CSLN) and molecular dynamics (MD) simulations. The CSLN algorithm predicts that 11-beta-hydroxysteroid dehydrogenase-1 (HSD11B1) serves as a common target for the synergistic effects of triptolide (TP) and glycyrrhizic acid (GA). MD simulations indicate that both TP and GA can maintain stable interactions with HSD11B1 and form a common binding hot region. Surface plasmon resonance (SPR) experiments reveal that both TP and GA can effectively bind to HSD11B1, with binding constants of 29.21 μM and 31.75 μM, respectively. When used in combination, the binding constant is 5.74 μM. The combination of CSLN and MD simulations represents an effective tool for the initial analysis and simulation of interaction patterns between TCM and their targets at the computational level. These findings enhance our understanding of the interaction mechanisms between drugs.

Artificial intelligence; Glycyrrhizic acid; Molecular dynamics; Traditional Chinese medicine; Triptolide.