A dissertation submitted to the Faculty of Science, University of the Witwatersrand, in fulfilment of the requirements for the degree of Doctor of Philosophy in Science at the School of Molecular and Cell Biology, 2017 / Different host genes playing a role in replication, transcription and movement of geminiviruses have been identified, allowing a better understanding of host response during infection. The cytoskeletal protein myosin has been shown to associate with RNA viruses movement protein and mediate its movement, however no geminivirus association with myosin has been established. Arabidopsis thaliana nitric oxide associated protein 1 (AtNOA1), once thought to be an enzyme involved in a nitric oxide (NO) production, has been reported to be differentially regulated in response to biotic and abiotic stress. In this study we sought to identify the role that myosins and NOA1 play in the development of disease by south african cassava mosaic virus (SACMV). Using a bioinformatics approach, 24 myosin transcripts were identified in Nicotiana benthamiana, and phylogeny analysis revealed that seven were class VIII myosins and 17 class XI. Five myosins silencing constructs M15.1 (transcript Niben101Scf11288g00015.1), MYOSIN XI-F (M11.F), MYOSIN XI-K (M11.K), MYOSIN XI-2 (M11.2) and MYOSIN VIII.B were selected for silencing using a virus induced gene silencing (VIGS) approach with SACMV and TRV-VIGS vectors. At 14 days post inoculation (dpi), both SACMV and TRV-VIGS vectors successfully silenced myosins with SACMV-VIGS silencing all five and TRV-VIGS silencing all but M11. F. At 28 dpi, SACMV-VIGS induced silencing of myosin of only two myosins and TRV-VIGS three. TRV-VIGS was found to be more efficient at silencing as the suppression of myosin induced by TRV-VIGS was stronger than that of SACMV-VIGS. To assess the effect of myosin silencing on SACMV infectivity in a separate experiment, 7 dpi of silencing, N. benthamiana plants were challenged with SACMV and reduction of myosin expression was assessed as well as viral accumulation. TRV-VIGS did not induce any silencing of myosin at 14 dpi, and at 28 dpi, the expression of M11.K and M11.F were silenced. SACMV-VIGS induced silencing of M11.F at both 14 and 28 dpi. In TRV-VIGS silenced M11.K, viral load at 28 dpi was not lower than the control, however the fold increase in viral load at 28 dpi compared to 14 dpi was 3-fold (p value 0.03) for M11.K silenced TRV-VIGS plants and 86-fold for the control 6-fold for the M11.K suggesting that silencing of M11.K decreases the spread of SACMV. In TRV-VIGS silenced M11.K, viral load at 28 dpi was lower than the control (9-fold p value 0.03) and the increase in viral load at 28 dpi compared to 14 dpi was insignificant, suggesting that spreading of SACMV was also hampered. The reduction in myosin M11.F expression induced iv by SACMV-VIGS resulted in an increase in viral load compared to the control. We hypothesise that the increase in viral load observed in M11.F silenced plants induced by SACMV-VIGS is due to the perceived resistance of SACMV-VIGS control (SACMV-challenged no silencing construct) to SACMV-challenge, and therefore results from the SACMV-VIGS study were inconclusive. From the TRV-VIGS study however, we have identified two candidate myosins in N. benthamiana myosin XI-K and myosin XI-F as potential interactor of SACMV during infectivity. Further research into their role in the development of SACMV disease is warranted. Nitric oxide associated 1 (NOA1) in plants is a cyclic GTPase involved in protein translation in the chloroplast and has been indirectly linked to nitric oxide (NO) accumulation. To understand the role played by NOA1 in response to (SACMV) infection, a bioinformatics approach was used to identify NOA1 homologues in cassava T200. Using the cassava genome data on Phytozome, a putative NOA1 namely cassava 4.1_007735m, was identified. Based on its protein sequence, cassava4.1_007735m shared a 69.6% similarity to Arabidopsis NOA1 (AtNOA1). The expression of cassava4.1_007735.m (MeNOA1) and N. benthamiana NOA1 (NbNOA1) and the accumulation of NO in leaf samples was compared between SACMV-infected and non-infected at early infection stage (14 dpi for N. benthamiana and 28 dpi for cassava T200) and full systemic stage (28 dpi for N. benthamiana and 56 dpi for cassava T200). Real-time PCR was used to measure SACMV viral load which increased significantly by 2-fold (p value 0.05) from 14 to 28 dpi for N. benthamiana and 8-fold from 28 to 56 dpi in cassava T200 (p value 0.04) as chlorosis and symptom severity concomitantly progressed. At 14 and 28 dpi, NbNOA1 expression was significantly lower than mock inoculated plants (2-fold lower at 14 dpi, p value 0.01 and 4 fold lower at 28, (p value 0.00) and the abundance of NO in infected N. benthamiana leaf tissue was 10% lower at 14 dpi and 40% lower at 28 dpi when compared to mock inoculated. In cassava T200, MeNOA1 expression was unchanged at 28 dpi and NO levels were decreased by 40% and at 56 dpi, MeNOA1 expression was 4-fold lower and NO accumulation was 37 % higher than that of mock inoculated leaf tissue. At 28 dpi for N. benthamiana and 56 for cassava T200, the decrease in NOA1 expression was accompanied by chloroplast dysfunction, evident from the significant reduction in chlorophylls a and b and carotenoids in SACMV-infected leaf samples. Furthermore, the expression of v chloroplast translation factors (chloroplast RNA binding, chloroplast elongation factor G, translation initiation factor 3-2, plastid-specific ribosomal protein 6 and) were found to be repressed in infected N. benthamiana and infected cassava T200 relative to mock inoculated plants. GC-MS analysis showed a decrease in fumarate and an increase in glucose in SACMV-infected N. benthamiana in comparison to mock samples suggesting a decrease in carbon stores. Collectively, these results provide evidence that in response to SACMV infection in N. benthamiana, decrease in photopigment and carbon stores, accompanied by an increase in glucose and decrease in fumarate, lead to a decline in NbNOA1 and NO levels. This is manifested by suppressed translation factors, and disruption of the chloroplast, resulting in chlorotic disease symptoms. In cassava T200 however, the link could not be established as the level of glucose was not significantly decreased and fumaric acid was not detected and although the concomitant decrease in the expression of MeNOA1 and chloroplast translation factors indicate dysfunction of the chloroplast, the link between MeNOA1 expression, carbon store, NO and chloroplast activity could not be established. / XL2017

Bemisia tabaci is the vector of cassava mosaic begomoviruses and cassava brown streak viruses which are main production constraints to cassava in sub-Saharan Africa. Current vector dynamics involved in the spread of both viruses in the region was established through comparison of the mitochondria cytochrome oxidase I DNA. Two distinct species were obtained: sub-Saharan Africa clade 1 (SSA1), comprising of two sub-clades (I & II), and a South West Indian Ocean Islands (SWIO) species. SSA1 sub-clade I whiteflies were widely distributed in East Africa. SSA1 sub-clade II whiteflies predominated the coast regions of Kenya, southern & coast regions of Tanzania and widespread in Uganda. SWIO whiteflies occurred in the coastal region of Kenya. This study also revealed that SSA1 sub-clade I haplotypes performed significantly better than SSA1 sub-clade II haplotypes with respect to mean number of eggs laid, developing instars and hatched adults on healthy, African cassava mosaic virus-[Tanzania:2001 ] and East African cassava mosaic Kenya virus-infected plants. There was no boost in whitefly numbers by the CMB-infected plants. The fecundity and development differences observed between SSA1 sub-clade I and II haplotypes have major epidemiology implications on the CMGs in the region