Exploring the chemogeographical variation of a commercially important medicinal tree (Prunus africana (Hook.f.) Kalkman) using a metabolomics approach

The stem-bark of Prunus africana (Hook.f.) Kalkman (African cherry) is traditionally used for the treatment of stomach ache, chest pain, malaria, fever, diabetes and high blood pressure. A large number of constituents have been isolated from the stem-bark including β-sitosterol and ursolic acid, which were reported to have anti-inflammatory activity. The aim of the study was to establish chemotypic variation among Prunus africana samples collected from three African countries, Cameroon, Democratic Republic of Congo (DRC) and Zimbabwe. Ninety-five stem-bark samples were collected from a total of nineteen populations, 11 in Cameroon (n = 55), 4 in Democratic Republic of Congo (n = 20) and 4 in Zimbabwe (n = 20). The samples were extracted with ethyl acetate, dichloromethane and methanol (1 g in 10 mL, 1 g in 10 mL and 2 mg in 5 µL, respectively). Chemometric analysis of chromatographic data obtained from high-performance thin layer chromatography (HPTLC), ultra performance liquid chromatography coupled with mass spectrometry (UPLC–MS), one-dimensional gas chromatography-time-of-flight-mass spectrometry (1D GC–ToF–MS) and spectroscopic data obtained from proton nuclear magnetic resonance (¹H NMR) were done using MetaboAnalyst 4.0 software to explore chemotypic variation. A sample from each country was analysed by two-dimensional gas chromatography-time-of-flight-mass spectrometry (2D GC–ToF–MS) to resolve compounds that co-eluted in the 1D GC analysis. Quantification of selected compounds (β-sitosterol and ursolic acid) were done using a validated ultra performance liquid chromatography-photodiode-array detection (UPLC–PDA) method. The chemical profiles obtained from each of the four techniques were very similar for the samples from each country. The HPTLC profiles for Zimbabwe samples were distinctly different from the other two countries with a unique band at R_f value of 0.80. Multivariate analysis of the HPTLC data (using rTLC software), UPLC–MS, 1D GC–ToF–MS and ¹H–NMR data revealed three clusters that were country specific. Chemometric analysis through the construction of PCA and a PLS–DA scores plots was performed, followed by the construction of variable important in projection plots, whereby marker compounds contributing significantly to the separation of the three clusters were selected and tentatively identified. Quantitative analysis revealed that samples from Zimbabwe contained higher levels of β-sitosterol (29.2 – 119 µg/g dry weight (DW) of the extract) compared to DRC and Cameroon, while samples from Cameroon were found to contain the highest levels of ursolic acid (1.80 – 141 µg/g DW of the extract). HPTLC, UPLC–MS, 1D and 2D GC–ToF–MS and ¹H–NMR are valid tools for quality control of P. africana plant materials after revealing three clusters of samples collected from the three countries. Cameroon samples had high content of ursolic acid, while Zimbabwe samples showed high levels of β-sitosterol. The number of compounds detected across the three countries by 1D GC–ToF–MS was less compared to 2D GC–ToF–MS, which enhanced separation and resolution.