| dc.description.abstract | Cassava brown streak disease (CBSD) was first described in Tanganyika (now Tanzania) about 75 years ago and continues to be a serious threat to cassava production in East and Central Africa. The disease was associated with Cassava brown streak virus (CBSV; genus Ipomovirus; family Potyviridae). Early reports indicated the disease was endemic in the coastal lowlands of East Africa but a few years ago CBSD was observed in high altitude areas of Uganda, probably due to a virulent strain(s) of CBSV that could have evolved. It is from this premise that this study was initiated. The virus isolates associated with CBSD were collected from different regions in Uganda and Tanzania. Using the RT-PCR technique, the complete genomes of two isolates (MLB3 and KOR6) were cloned and sequenced. The 3’-proximal region of 16 isolates was also sequenced. Additional sequences were retrieved from validated sequence databases and analyzed using modern bioinformatics tools. Complete sequences of MLB3 and KOR6 isolates were compared with other nine complete genomes that were from other laboratories. The complete nucleotide (nt) and amino acid (aa) sequence identities ranged between 69.0-99.3% and 73.6-99.4%, respectively. At coat protein (CP) level, comparison between all isolates showed sequence identities that varied from 70.1 to 99.8% and 75.6 to 99.7% at nt and aa levels, respectively. A neighbour joining tree for non-recombinant CP sequences revealed two phylogenetic groups that were supported by high bootstrap values (100%), suggesting distinct species. For one group, the old name CBSV (Isolate KOR6) has been retained whereas for the second group the name Ugandan cassava brown streak virus (UCBSV; Isolate MLB3) has been assigned. The viruses belong to the same genus, Ipomovirus. The CBSV group constitutes isolates that are predominately found in the lowlands (<1000 masl) while the UCBSV isolates are more commonly found in the high altitude (>1000 masl) areas of East Africa. Within species CP sequence identities were found to be 90.7 to 99.8% and 91.0 to 98.5% at nt level for UCBSV and CBSV, respectively. At amino acid level, the identities were 93.1-99.7% for UCBSV and 92.8-99.2% for CBSV. When the comparison was made between the two phylogroups of isolates, the identities were 70.1 to 72.6% (nt) and 75.6% to 79.1% (aa).
The complete genomes of the isolate MLB3 (UCBSV) and isolate KOR6 (CBSV) consisted of 9069 and 8995 nt, which translated into 2902 and 2912 aa, respectively. The genomes contained nine of the ten expected proteins, namely the first protein (P1-Pro), third protein (P3), first 6 kilodalton protein (6K1), cytoplasmic Inclusion (CI), second 6 kilodalton protein (6K2), genome linked viral protein (VPg), nuclear inclusion a (NIa-Pro), nuclear inclusion b (NIb) and CP. The proteins P1-Pro, CI, VPg and CP and the untranslated regions (5’- and 3’-UTRs) differed in size between the isolates of the two viruses. The 3’-UTR of CBSV was shorter (131 nt) than that of UCBSV (225-227 nt). The N-terminal ends of the polyproteins of CBSV and UCBSV were evolutionarily intriguing. While other characterized members of Ipomovirus contain a single P1 protein and HC-Pro or two P1 proteins (P1a and P1b) and no HC-Pro at the N-terminus of polyproteins, CBSV and UCBSV contained a single P1 protein that was experimentally shown to function as an RNA silencing suppressor (isolate MLB3) and no HC-Pro. As a result, the polyproteins of CBSV and UCBSV were predicted to be processed by only two proteases, the serine P1 and cysteine NIa-Pro. Interestingly, the CBSV and UCBSV genomes contained a Maf/HAM1-like sequence (HAM1h) (678 nt; 226 aa) recombined between the NIb and CP domains in the 3’-proximal part of the genome, which is known to be highly conserved in the family Potyviridae. HAM1 is widespread in cellular organisms such as plants, man, fish and animals. HAM1h in CBSV and UCBSV, and as established by this study in Euphorbia ringspot virus (EuRSV), was flanked by consensus proteolytic cleavage sites for Ipomovirus NIa-Pro, a cysteine proteinase. HAM1h of CBSV and UCBSV share low homology (< 56%). Homology of viral HAM1h with cellular HAM1 pyrophosphatases suggests that it may intercept noncanonical nucleoside triphosphates to reduce mutagenesis of viral RNA.
There was evidence for intraspecific (UCBSV) but not interspecific recombination. Nucleotide diversity between the two viruses was high (0.323 ± 0.021). The capsids of both viruses were under strong purifying selection but some amino acid sites were under diversifying selection. Moreover, an RT-PCR based diagnostic tool was developed. It enabled for simultaneous detection of single and mixed infections of CBSV and UCBSV isolates. Employing this diagnostic tool, it was established that the two viruses co-infect cassava in East Africa and co-infection were as high as 67-100% in some places around Lake Victoria. This study contributes to the understanding of the genetic diversity and structure of CBSD associated viruses in East Africa and general evolution of viruses in the family Potyviridae. The study has clearly demonstrated the occurrence of two distinct viruses that are associated with CBSD, information that had hitherto not been apparent. The information generated will benefit and guide CBSD resistance breeding | en_US |
