An image-based fingerprint-efficacy screening strategy for uncovering active compounds with interactive effects in...

Yindan Xinnaotong soft capsule (YDXNT) is a clinically effective herbal prescription used for the treatment of cardiovascular and cerebrovascular diseases. Since Chinese medicines (CMs) exert their effects via a "multiple-components and multiple-targets" mode, discovery of the active compounds with interactive effects may contribute to reveal their mechanisms of action.

This study aimed to establish an image-based fingerprint-efficacy screening strategy to identify active compounds with interaction effects from CM prescription, using YDXNT to inhibit microglia-mediated neuroinflammation as an instance.

A multi-component random content-oriented chemical library of YDXNT was constructed by uniform design, and their chemical fingerprint was profiled by liquid chromatography-mass spectrometry (LC-MS) and gas chromatography-mass spectrometry (GC-MS) methods. Then the neuroinflammation activities of chemical library members of YDXNT were determined by image-based dual phenotypic quantification. Subsequently, fingerprint-efficacy correlation and random forest analysis were applied to predict the potentially active compounds with interactive effects. Finally, the interactive effects among the active compounds were confirmed by quantitative polymerase chain reaction (qPCR) and apoptosis analysis, and network pharmacology was applied to explore the possible mechanisms.

Image-based fingerprint-efficacy correlation analysis revealed that six tanshinones (TNs) and four flavonoids (FAs) were potential anti-neuroinflammatory compounds. The inter-family of TNs and FAs possessed obvious interactive effects (combination index ≤ 0.825). Moreover, the combination of scutellarein and tanshinone I (2:1, w/w) was discovered as the possible interactive combinatorial components, which, comparing with individual scutellarein or tanshinone I, shown more powerful effects on anti-inflammatory and anti-apoptotic effects in lipopolysaccharide (LPS)-induced BV2 cells. Network pharmacology showed that the active compounds might suppress microglia-mediated neuroinflammation via multiple targets in the T cell receptor, Jak-STAT, and Toll-like receptor signaling pathways.

The image-based fingerprint-efficacy strategy simplifies the screening process of efficacious component combinations in CMs for complex diseases, which also offers a promising approach to explore the integrative therapeutic mechanisms of CMs.