ABSTRACT
BACKGROUND: Techniques to study plant viral diseases under controlled growth conditions are required to fully understand their biology and investigate host resistance. Cassava brown streak disease (CBSD) presents a major threat to cassava production in East Africa. No infectious clones of the causal viruses, Cassava brown streak virus (CBSV) or Ugandan cassava brown streak virus (UCBSV) are available, and mechanical transmission to cassava is not effective. An improved method for transmission of the viruses, both singly and as co-infections has been developed using bud grafts. FINDINGS: Axillary buds from CBSD symptomatic plants infected with virulent isolates of CBSV and UCBSV were excised and grafted onto 6-8 week old greenhouse-grown, disease-free cassava plants of cultivars Ebwanateraka, TME204 and 60444. Plants were assessed visually for development of CBSD symptoms and by RT-PCR for presence of the viruses in leaf and storage root tissues. Across replicated experiments, 70-100% of plants inoculated with CBSV developed CBSD leaf and stem symptoms 2-6 weeks after bud grafting. Infected plants showed typical, severe necrotic lesions in storage roots at harvest 12-14 weeks after graft inoculation. Sequential grafting of buds from plants infected with UCBSV followed 10-14 days later by buds carrying CBSV, onto the same test plant, resulted in 100% of the rootstocks becoming co-infected with both pathogens. This dual transmission rate was greater than that achieved by simultaneous grafting with UCBSV and CBSV (67%), or when grafting first with CBSV followed by UCBSV (17%). CONCLUSIONS: The bud grafting method described presents an improved tool for screening cassava germplasm for resistance to CBSD causal viruses, and for studying pathogenicity of this important disease. Bud grafting provides new opportunities compared to previously reported top and side grafting systems. Test plants can be inoculated as young, uniform plants of a size easily handled in a small greenhouse or large growth chamber and can be inoculated in a controlled manner with CBSV and UCBSV, either singly or together. Disease symptoms develop rapidly, allowing better studies of interactions between these viral pathogens, their movement within shoot and root systems, and how they induce their destructive disease symptoms.