A Novel Network Pharmacology Strategy for Retrieving a Key Functional Component Group and Mechanisms in the Di-Huang-Yin-Zi Treatment of Parkinson's Disease.

Abstract

Introduction: Parkinson's Disease (PD) is a common and difficult-to-cure chronic neurodegenerative disorder. Current medications often target a single pathway and can have certain side effects. In contrast, traditional Chinese medicine formulas, such as Di-Huang-Yin-Zi (DHYZ), with their multi-component and multi-target characteristics, offer potential advantages by addressing these limitations, making them worthy of in-depth study.

Methods: Components of DHYZ were collected from public databases and literature. After screening, the remaining components underwent target prediction, and the predicted component-target pairs were used to construct the complex component-target network. A novel node importance algorithm, known as the fusion model, was applied to construct an effective space from the component-target network, thereby reducing redundancy. Meanwhile, the pathological genes were extracted from DisGeNET and GeneCards to judge the quality of effective space. The effective space was compared with other widely used network parameters to validate its efficiency, and the Key Functional Compound Group (KFCG) was inferred from the effective space. Finally, the protective mechanism of DHYZ was inferred based on the KFCG and was validated in the in vitro PD model.

Results: Compared to other commonly used algorithms, the effective space identified by the fusion model more accurately represented the full spectrum of DHYZ's targets and demonstrated stronger correlation with PD. Additionally, we utilized the component contribution ratio algorithm to identify the KFCG within the effective space. Through enrichment analysis, we hypothesized that KFCG may exert its anti-PD effects via the PI3K-Akt, MAPK, and AMPK pathways and validated these mechanisms in vitro.

Discussion: Collectively, the results of this study not only deepen our understanding of the therapeutic potential of DHYZ in the treatment of PD but also enhance the clinical translatability of DHYZ through formula optimization. However, this study has certain limitations. For instance, the pathogenic genes of PD were not incorporated into the network in this study, and the use of an undirected network may offer lower biological interpretability compared to a directed network.

Conclusion: This robust and precise algorithm allowed us to optimize Di-Huang-Yin-Zi. This provided preliminary insights into its potential molecular mechanisms for treating PD, laying a foundation for the secondary development of other formulas.