Proanthocyanidins and other phenolics in Acacia leaves of Southern Africa

Four provenances each of Acacid karroo, Acacia nilotica, Acacia tortilis, A cacia senegal, Acacia erioloba and Faidherbia albida were sampled at Cyrene , Lucydale and Mahiye in the Matopos Research Station. Provenances were not replicated within site. The leaves and leaf stalks sampled were dried at 5 5 degreesC and then analysed for total phenolics and proanthocyanidins (PAs ). Variations in content of PAs and total phenolics of the Acacias were stu died. Three assay methods for phenolics were evaluated. Total phenolics were estimated by precipitation with trivalent ytterbium af ter extraction with acetone:water (7:3 v/v) or colorimetrically by the Foli n-Ciocalteu method. PAs were estimated colorimetrically by the butanol-HCl method. The results from the colorimetric methods are reported as absorbanc e units (au). A. senegal contained the lowest level of ytterbium precipitated phenolics ( Ybppt) (mean = 47 g/kg DM), while A. nilotica contained the highest (mean = 298 g/kg DM). Amounts of Ybppt for A. karroo, A. erioloba and F. albida we re not significantly different (P > 0.05) from one another. The amount of Y bppt for A. tortilis was significantly different (P < 0.05) from the amount for A. karroo but not different from the amount for F. albida and A. eriol oba (P > 0.05). A. senegal was again the lowest in total phenolics detected by the Folin-Ci ocalteu method (A(675) = 0.370 au/g DM), while A. nilotica was the highest (A(675) = 5.261 au/g DM). A. karroo, F. albida and A. tortilis were not sig nificantly different (P > 0.05) from one another. PAs were lowest for A. senegal (A(550) =0.040au/g DM) and highest for A. ka rroo (A(550) = 2.011au/g DM), Values for A. tortilis, F, albida and A. erio loba were not significantly different (P > 0.05) from one another. For all the three assay methods, there were no significant differences (P > 0.05) b etween Cyrene and Mahiye sites. The regression analysis of PAs on Ybppt showed the relationship to be highl y significant (P < 0.05) for A. karroo, F. albida and A. erioloba (R-2=0.90 4, 0.892, 0.884, respectively, and significant for A. tortilis (R-2 = 0.680 ). It was poor for A. senegal and A. nilotica (R-2 = 0.347 and 0.460, respe ctively). The relationship between total phenolics and Ybppt was similarly good for F. albida, A. erioloba, A. nilotica and A. karroo (R-2 = 0.957, 0. 808, 0.792, and 0.725, respectively) but poor for A. senegal, and A. tortil is (R-2 = 0.413 and 0.004, respectively). (C) 2001 Published by Elsevier Sc ience B.V.