The mechanisms of Qizhu Tangshen formula in the treatment of diabetic kidney disease: Network pharmacology, machine learning, molecular docking and experimental assessment

Abstract

Background: Qizhu Tangshen Formula (QZTS) has been shown therapeutic effects on diabetic kidney disease (DKD). However, to date, the pharmacological mechanisms remain vague.

Methods: To explore the underlying mechanisms of QZTS in treating DKD using network pharmacology, machine learning, molecular docking and experimental assessment.

Results: First, we found that QZTS improved glycolipid metabolism disorder, decreased proteinuria and alleviated kidney tissue injury in DKD model KKAy mice. Then, by integrating multiple databases, a total of 96 targets of 74 active compounds in QZTS and 759 DKD-related genes were acquired. Next, we identified 13 hub targets of QZTS in DKD by three rank algorithms, including functional similarity, topological similarity and shortest path. Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses demonstrated that the pathways mainly centered on the processes of glycolipid metabolism disorder, inflammation and angiogenesis. Among them, VEGF signaling pathway was significantly enriched. Molecular docking showed that key active compounds of QZTS all had relatively good binding affinity with predicted hub targets. Finally, animal experiments found that QZTS significantly inhibited the secretion of plasma VEGF and downregulated the protein and mRNA expression levels of AKT, p38MAPK and VEGFR2.

Conclusion: Our results indicated that QZTS treated DKD via multiple targets and pathways and the VEGF signaling pathway may be highly involved in this process.

Keywords: Diabetic kidney disease; Machine learning; Network pharmacology; VEGF.