1Department of Plant Biology, Genetic Centre, Swedish University of Agricultural Sciences (SLU), PO Box 7080, 5-750 07 Uppsala, Sweden; 2Kawanda Agricultural Research Institute, PO Box 7065, Kampala, Uganda; 3Natural Resources Institute, University of Greenwich, Kent ME4 4TB, UK

Background and objectives
Sweet potato feathery mottle potyvirus (SPFMV; family Potyviridae) is the commonest virus infecting sweet potato (Ipomea batatas; family Convolvulaceae). It is non-persistently transmitted by aphids. Certain isolates cause much economic damage by inducing cracking or internal corkiness in some cultivars grown outside Africa. In Africa, SPFMV causes a severe sweet potato virus disease (SPVD) complex in dual infection with the whitefly-transmitted sweet potato chlorotic stunt closterovirus (SPCSV). The economic damage caused by SPVD is most serious in East Africa where sweet potato can be the main starch staple [1]. SPVD-affected plants commonly produce less than half of the tuberous yield of symptomless ones.

Sweet potato clones from West Africa or outside Africa became severely diseased with SPVD when grown in the field in Uganda. Therefore, the purpose of this study was to examine whether isolates of SPFMV differ from those prevailing elsewhere.

Materials and methods
Sweet potato plants infected with SPVD in the field were collected from different parts of Uganda. Presence of SPFMV and SPCSV and absence of three other sweet potato viruses in these plants were confirmed by ELISA. Seven SPFMV isolates were obtained by allowing individual aphids (Myzus persicae) access to a sweet potato leaf before transferring to a lpomea setosa seedling. SPFMV infection was detected by development of the characteristic mottle symptoms and by ELISA using polyclonal antibodies obtained from the International Potato Center (CIP). The SPFMV-infected I. setosa plants and the SPVD-infected sweet potato plants were used as sources of scions to graft-inoculate virus free plants of the CIP sweet potato clone 420026, previously reported to be resistant to SPFMV.

SPFMV isolates were also tested with a monoclonal antibody (MAb) that detects West African SPFMV and was provided by Dr G. Thottapilly, International Institute for Tropical Agriculture. RT-PCR was used to amplify the coat protein (CP)-encoding region of the viral genomic RNA, and two independently amplified products from each SPFMV isolate were sequenced using standard techniques.

Results and conclusion
All seven SPFMV isolates infected the sweet potato clone CIP 420026 indicating that Ugandan SPFMV isolates can overcome resistance that is effective against SPFMV occurring elsewhere. Further evidence that Ugandan SPFMV isolates were distinct was obtained by CP sequence analysis, which showed that Ugandan SPFMV isolates formed a cluster distinct from the SPFMV isolates from other continents. CP analysis also revealed variability of SPFMV within Uganda, consistent with different reaction of the MAb to different isolates. Biological variability is also plausible because preliminary data indicate that some Ugandan SPFMV isolates cause cracking symptoms in sweet potato tubers. However, the economic impact of these symptoms in Africa needs to be examined in more detail.

1. Karyeija RF, Gibson RW, Valkonen JPT, 1998. Plant Disease 82, 4-15.