Analysis of cassava brown streak viruses reveals the presence of distinct virus species causing cassava brown streak disease in East Africa

Abstract

Cassava brown streak virus (CBSV) isolates were analysed from symptomatic cassava collected between 1997 and 2008 in the major cultivation regions of East Africa. An analysis of complete RNA genomes of seven isolates from Kenya, Tanzania, Mozambique, Uganda and Malawi revealed a common genome structure, but the isolates clearly clustered in two distinct clades. The first comprised isolates from Kenya, Uganda, Malawi, north-western Tanzania and the CBSV described previously, and shared between 87 and 95% nucleotide sequence identity, whilst the second included isolates from coastal regions of Mozambique and Tanzania, which shared only 70% nucleotide sequence identities with isolates of the first clade. When the amino acid sequences of viral proteins were compared, identities as low as 47% (Ham1) and 59% (P1) between the two clades were found. An antiserum obtained against the capsid protein of a clade 1 isolate identified a 43 kDa protein in clade 1 isolates and a 45 kDa protein in clade 2 isolates. Several cassava cultivars were susceptible to isolates of clade 2 but resistant to those of clade 1. The differences observed both in biological behaviour and in genomic and protein sequences indicate that cassava brown streak disease in East Africa is caused by at least two distinct virus species. It is suggested that those of clade 1 retain the species name Cassava brown streak virus, whilst those of clade 2 be classified as Cassava brown streak Mozambique virus.