Comprehensive comparison of two colour varieties of Perillae folium by UHPLC-Q-TOF/MS analysis combining with feature-based molecular networking

Perillae Folium (PF), the leaf of Perilla frutescens (L.) Britt is extensively used as a culinary vegetable and medicinal herb in many countries. Purple PF (P. frutescens var. arguta) and green PF (P. frutescens var. frutescens) are two of the main varieties. In this study, UHPLC-Q-TOF/MS assisted with feature-based molecular networking (FBMN) was applied for chemical characterization. In total, 82 metabolites, mainly phenolic acid derivatives were identified from PF. With the help of FBMN, five organic acid glucosides together with three N-phenylpropenoyl-L-amino acids (NPAs) were identified in PF for the first time. Multivariate statistical methods were utilized for comprehensive comparison of purple and green PF profiles. 12 compounds with their relative contents varied significantly between purple and green PF were screened out. Overall, the present study offers valuable insights for chemical elucidation of PF which would be helpful for comprehensive utilization of PF resource.