Species- and Organ-dependent Variation in Phytochemical Composition and Antioxidant Activity of Selected Medicinal Plants from West Pokot, Kenya

Abstract

Medicinal plants remain important sources of bioactive compounds; however, comparative evidence on how plant species and organs influence phytochemical composition and antioxidant potential remains limited for several traditionally used taxa in sub-Saharan Africa. This study evaluated the phytochemical composition, thin-layer chromatographic (TLC) profiles and antioxidant activity of Lannea fulva, Ochna insculpta, Pittosporum viridiflorum, Schrebera alata and Teclea pilosa across leaves, stem bark and roots collected from West Pokot County, Kenya. Methanolic extracts were prepared using Soxhlet extraction, followed by qualitative and quantitative phytochemical analyses, TLC profiling and 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging assays. The half maximal inhibitory concentration (IC50) values were estimated using a four-parameter logistic model and data analyzed using analysis of variance at p < 0.05. Qualitative screening revealed tannins, flavonoids, alkaloids, saponins, terpenoids and steroids, with clear interspecific and inter-organ variation. The TLC analysis showed 2-6 separated bands with distinct retention factor (Rf) values (0.18-0.92), indicating variability in phytochemical complexity among extracts. Quantitative analysis demonstrated significant differences in secondary metabolite concentrations across species and plant parts (p < 0.001), with flavonoids and alkaloids generally higher in leaves and stems than in roots. Antioxidant activity increased in a concentration-dependent manner across all extracts (p < 0.001). IC50 analysis showed that P. viridiflorum (32.19 µg/mL) and S. alata (34.99 µg/mL) exhibited the strongest antioxidant activity among the plant extracts, whereas L. fulva, T. pilosa and O. insculpta showed higher IC50 values of 53.73, 53.79 and 57.93 µg/mL, respectively. Root extracts consistently exhibited higher IC50 values than leaves and stems. Hierarchical clustering grouped the species into three antioxidant activity clusters corresponding to IC50 patterns, with P. viridiflorum and S. alata clustering with the standard antioxidant (ascorbic acid), while T. pilosa formed a distinct low-activity group. The findings demonstrate that phytochemical composition and antioxidant activity in the studied medicinal plants are strongly influenced by species identity and plant part, providing baseline evidence for future isolation of bioactive compounds and pharmacological evaluation.