Herbal medicines deliver enhanced therapeutic effects through multi‑ingredient synergy, yet identifying synergistic pairs from thousands of candidates remains a major bottleneck due to system complexity. A research team led by Prof. Yinyin Wang and Prof. Ninghua Tan from China Pharmaceutical University has developed HerbSyner_Finder, a tailored network framework that constructs a multidimensional combinatorial atlas to quantify herb-herb, ingredient-ingredient, herb-disease, and ingredient-disease interactions.
The team validated HerbSyner_Finder using five classic clinical prescriptions for cough variant asthma (CVA), a chronic airway inflammatory disease often treated with inhaled corticosteroids (ICS) that carry long‑term safety risks. By integrating network proximity calculation and Louvain community detection (modularity Q > 0.3), the model automatically prioritized two synergistic ingredient pairs from thousands of candidates: kaempferol-quercetin and berberine-luteolin.
In vitro experiments in RAW264.7 macrophages and airway smooth muscle cells (ASMCs) confirmed strong synergy: low‑dose berberine-luteolin inhibited LPS‑induced NO release and ASMC hyperproliferation with efficacy comparable to high‑dose dexamethasone. Multiple synergy models (ZIP, Loewe, HSA, Bliss) verified robust synergistic scores. In OVA‑induced CVA rat models, combined berberine and luteolin significantly reduced cough frequency, relieved airway inflammation, decreased collagen deposition and mucus hypersecretion, and outperformed single compounds in suppressing pro‑inflammatory cytokines (TNF‑α, IL‑1β, IL‑6).
Mechanistic studies revealed that berberine-luteolin synergistically suppresses the NLRP3/NF‑ κ B signaling pathway, reducing phosphorylation of p65 and I κ Bα and downregulating key CVA‑related chemokines (CXCL10, CXCL3, CCR10).
Notably, HerbSyner_Finder is openly accessible via GitHub and a dedicated web server (herbcomb.com), supporting broad application. As a universal tool, it can be extended to other complex diseases to accelerate the discovery of safe, synergistic herbal combinations with lower side effects.
This work entitled “ HerbSyner_Finder: a network community-based model for identifying synergistic combinations from herbal medicines and complex systems ” was published online March 26, 2026, in Targetome .
Experimental study
Not applicable
HerbSyner_Finder: a network community-based model for identifying synergistic combinations from herbal medicines and complex systems
26-Mar-2026