Comparing the regional epidemiology of the cassava mosaic and cassava brown streak virus pandemics in Africa.

The rapid geographical expansion of the cassava mosaic disease (CMD) pandemic, caused by cassava mosaic geminiviruses, has devastated cassava crops in 12 countries of East and Central Africa since the late 1980s. Region-level surveys have revealed a continuing pattern of annual spread westward and southward along a contiguous 'front'. More recently, outbreaks of cassava brown streak disease (CBSD) were reported from Uganda and other parts of East Africa that had been hitherto unaffected by the disease. Recent survey data reveal several significant contrasts between the regional epidemiology of these two pandemics: (i) severe CMD radiates out from an initial centre of origin, whilst CBSD seems to be spreading from independent 'hot-spots'; (ii) the severe CMD pandemic has arisen from recombination and synergy between virus species, whilst the CBSD pandemic seems to be a 'new encounter' situation between host and pathogen; (iii) CMD pandemic spread has been tightly linked with the appearance of super-abundant Bemisia tabaci whitefly vector populations, in contrast to CBSD, where outbreaks have occurred 3-12 years after whitefly population increases; (iv) the CMGs causing CMD are transmitted in a persistent manner, whilst the two cassava brown streak viruses appear to be semi-persistently transmitted; and (v) different patterns of symptom expression mean that phytosanitary measures could be implemented easily for CMD but have limited effectiveness, whereas similar measures are difficult to apply for CBSD but are potentially very effective. An important similarity between the pandemics is that the viruses occurring in pandemic-affected areas are also found elsewhere, indicating that contrary to earlier published conclusions, the viruses per se are unlikely to be the key factors driving the two pandemics. A diagrammatic representation illustrates the temporal relationship between B. tabaci abundance and changing incidences of both CMD and CBSD in the Great Lakes region. This emphasizes the pivotal role played by the vector in both pandemics and the urgent need to identify effective and sustainable strategies for controlling whiteflies on cassava.

A distinct Bemisia tabaci (Gennadius) (Hemiptera: Sternorrhyncha: Aleyrodidae) genotype cluster is associated with the epidemic of severe cassava mosaic virus disease in Uganda.

During the 1990s, an epidemic of cassava mosaic virus disease caused major losses to cassava production in Uganda. Two factors associated with the epidemic were the occurrence of a novel recombinant begomovirus, EACMV-Ug, and unusually high populations of the whitefly vector, Bemisia tabaci. Here we present molecular evidence for the occurrence of two cassava-colonizing B. tabaci genotype clusters, Ug1 and Ug2, one of which, Ug2, can be consistently associated with the CMD epidemic in Uganda at the time of collection in 1997. By contrast, a second genotype cluster, Ug1, only occurred 'at' or 'ahead of' the epidemic 'front', sometimes in mixtures with Ug2. Comparison of mitochondrial cytochrome oxidase I gene sequences for Ug1 and Ug2 and well-studied B. tabaci reference populations indicated that the two Ugandan populations exhibited approximately 8% divergence, suggesting they represent distinct sub-Saharan African lineages. Neither Ugandan genotype cluster was identified as the widely distributed, polyphagous, and highly fecund B biotype of Old World origin, with which they both diverged by approximately 8%. Within genotype cluster divergence of Ug1 at 0.61 +/- 0.1% was twice that of Ug2 at 0.35 +/- 0.1%. Mismatch analysis suggested that Ug2 has undergone a recent population expansion and may be of nonUgandan origin, whereas Ug1 has diverged more slowly, and is likely to be an indigenous genotype cluster.