Genetic variation in the threatened medicinal tree prunnus africana (Hook. F.) in Kenya and Cameroon: implications for the genetic management of the species.

Abstract

Prunus africana (Hook. f.) Kalkman is a geographically widespread tree restricted to highlands forests of mainland Africa and outlying islands. The species is commercially important for its bark, which is used in the treatment of prostate gland disorders. It also produces high quality timber used locally for building poles destruction of the species in natural forests, leading to concerns on the long-term sustainability of harvesting and the conservation of the species. As a result, P. africana is listed under Appendix 11 of CITES. Cultivation of P. africana can increase production of the species and ease exploitation from natural forests. At the same time, planting in agroforestry systems can provide revenue for small-scale farmers. Although knowledge of genetic variation and origin of planting material are vital for the sustainable management of the species under cultivation, little information on these factors is currently available. Low genetic variation may lead to unsustainable production through inbreeding depression or an inability to adapt to changing environmental or user requirements. An inappropriate origin may lead to poor adaptation and associated low performance. in order to determine the risks associated with current planting practices and suggest remedial actions, surveys of genetic composition are required. Random amplified polymorphic DNA (RAPD) analysis was used to assess genetic variation in natural populations, nursery stocks and planted materials of P. africana from Cameroon and Kenya. Reference natural population from Uganda and Ethiopia were also included. In total, 175 individuals from 20 populations were assessed at 41 RAPD loci. Results showed that most genetic variation (56%) occurred among sample populations. Significant differences existed among populations from western Kenya were more similar to both Uganda and Cameroon than to eastern Kenya populations, which were more similar to the stand sampled from Ethiopia. In Kenya and Cameroon, no clear differences were observed in diversity estimates between natural, nursery and planted categories of materials, although several caveasts concerned with the sampling of the material make definitive conclusion difficult. In at least one case in Kenya, high diversity in a nursery population appeared due to mixing of material from eastern and western Kenya origins, indicating the potential for long distance germplasm transfer by humans. A part from this case, the genetic composition of the planted and nursery materials in Kenya corresponded well to that of geographically proximate natural populations. In Cameroon, however, this correspondence was not observed, possibly reflecting a lack of resolution with the scored RAPD markers and /or the relatively long historical background of germplasm transfer.