Metabolomic-guided diversity analysis reveals chemotypic groups and cosmetic-relevant biomarkers in shea trees (Vitellaria paradoxa C.F. Gaertn.) for industrial applications

Shea tree ( Vitellaria paradoxa C.F. Gaertn) is valued mainly for its butter, which is rich in bioactive compounds, making it important for the cosmetics, dermocosmetics, and pharmaceutical industries. While some studies have quantified these compounds, the metabolic diversity of shea trees in Benin remains largely underexplored. This study characterized the metabolic diversity of shea trees in Benin, using both targeted and untargeted metabolomics approaches to identify bioactive compounds with potential cosmetic and pharmaceutical applications. Freeze-dried shea kernel powders from 69 trees in nine localities of Benin were analysed by ultra-high-performance liquid chromatography. Multivariate statistical analyses, including Principal Component Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA), were used to identify metabolic diversity among the trees. A total of 4634 signals were detected, with 70 % annotated in silico , revealing a diverse chemical composition dominated by fatty acids (42 %), amino acids/peptides (22 %), alkaloids (15 %), and phenolics, polyketides and terpene compounds (14 %). PCA and K-means revealed three distinct metabolic groupings, and a targeted analysis identified six key metabolites contributing strongly to the variance, including gallic acid, gallocatechin, α-amyrin, epigallocatechin, α-tocopherol, and campesterol. PLS-DA confirmed these groupings, with eight metabolites showing high discriminatory power and potential as biomarkers. Notably, shea tree BG037 and BG143 exhibited particularly high concentrations of compounds of interest for cosmetic applications. This study highlights the substantial metabolic variability in Benin's shea trees, which offers opportunities for tailored applications in the dermocosmetic and nutraceutical industries. • Shea trees in Benin exhibit significant metabolic diversity, with three chemotypes. • Metabolic profiles within clusters do not show any obvious spatial organization. • Intra-population variability was independent of ‘’Shea parkland” as established. • Eight discriminating metabolites have been identified as potential biomarkers. • Shea tree BG037 and BG143 are rich in compounds of cosmetic interest.