Use of Roche 454 Sequencing Technology to Understand Diversity of Cassava Mosaic Begomovirus Genomes

A total of 16 full length DNA-A sequences of Cassava Mosaic Begomoviruses (CMBs), the causal agents of Cassava Mosaic Disease (CMD) were obtained using the Roche 454 sequencer in 2012 from begomovirus (CMB) amplicons of the length of 2.8 kb collected from Tanzania (4), Rwanda (4), Uganda (1), Zambia (2), Kenya (3), Malawi (1) and Mozambique (1).  Sequence independent single primer amplification (SISPA) was used in order to amplify the viral nucleic acid. Once denatured, the dsDNAs were labeled and randomly amplified through Klenow reaction (Biolabs). The Klenow products were then PCR amplified using the Accuprime kit (Invitrogen). Libraries construction, quantification, quality control, emulsion PCR and sequencing were carried out according to the 454 manual recommendations. Two pools of tagged amplicons sequenced on the 454 were deconvoluted in two phases and assembled using a software pipeline consisting of Perl scripts to automate two alignment algorithms, Newbler (provided by Roche) and CLC Genomics workbench, a commercial package for assembling 454 short reads. Both de novo and reference mapping was performed. Four full-length DNA-A and DNA-B nucleotide sequences were obtained from the GenBank and used for reference mapping. For the 16 samples the average sequence coverage depth across the analyzed base pairs following de novo assembly was 49× while reference assembly yielded an average sequence coverage depth of 54× across the analyzed base pairs. CMBs sequence analysis was performed using Geneious® Pro 5.6.5, Biomatters, ltd. CMBs genome organization structure was typical of the bipartite CMBs with AV2, AV1, AC1, AC2, AC3 and AC4 open reading frames (ORFs). Phylogenetic relationship analysis grouped three of the CMBs with the published African cassava mosaic virus (ACMV) and thirteen with East African cassava mosaic virus (EACMV) genetic groups. With species attributions in begomoviruses made from full length DNA-A, where a sequence of 89% identity is the species distinction cut-off point, two new CMB species were identified: one each from Uganda and Zambia. The two species have been named here as East African cassava mosaic Uganda virus (EACMUV) and East African cassava mosaic Zambia virus (EACMZmV), respectively. The highest percentage nucleotide sequence homology to the reference CMB sequences in the GenBank was 72% for EACMUV and 75% for EACMZmV. Our results highlight the need for further research to understand and develop durable management strategies for the emerging CMBs for improved cassava productivity in sub-Saharan Africa.