Engineering Cassava Mosaic Disease (CMD) Resistance in a Ghanaian Cassava Cultivar

Abstract

Cassava is an important staple crop for an estimated 800 million people worldwide. Cassava production in Africa is constrained by cassava mosaic disease (CMD) and cassava brown streak disease (CBSD). Both viral diseases cause annual yield losses estimated around $2.1 billion. CMD is caused by nine distinct virus species known as cassava mosaic geminiviruses (CMGs, genus: Begomovirus family: Geminiviridae). This disease is widespread across all cassava-growing regions in Africa. CBSD is caused by Cassava brown streak virus (CBSV) and Ugandan cassava brown streak virus (UCBSV) (genus Ipomovirus; family Potyviridae). Unlike CMD, CBSD is prevalent in low to mid altitude regions of Eastern and Central Africa. The necrotic lesions on storage roots, which make them unfit for consumption or processing, further exacerbate economic losses due to CBSD. Efforts aiming at controlling incidence of both CMD and CBSD in Africa have focused on screening for resistance in both cultivated and wild species of cassava and subsequent introgression of resistance into farmer-preferred cultivars. Three sources of resistance (CMD1, CMD2 and CMD3) have been identified for CMD and have been introgressed into local germplasm. Similarly, a number of CBSD "tolerant" varieties or clones such as "Kiroba", "Kaleso" KBH 2006/18 and MM 06/0082 have been identified. However, high heterozygosity and outcrossing in cassava and the limited diversity of these sources of resistance challenge development of durable resistance to CMD and CBSD via conventional breeding approach. The main goal of this project is to engineer broad-spectrum resistance against cassava mosaic disease (CMD) into a farmer-preferred cassava cultivar in Ghana using expertise and tools developed at the Plant Biotechnology laboratory, ETH Zürich. In the first chapter, I present results characterizing genetic diversity of cassava mosaic geminiviruses (CMGs) identified in farmers' fields in Ghana. Extending this characterization to include other CMG species identified elsewhere in Africa, highly conserved genomic regions were identified. CMGs have a high propensity for mutation and recombination that can influence their evolution. Besides, the relative contribution of natural selection in shaping genetic variation and thus evolution of geminiviruses remains poorly studied. Due to the high genetic variation present in CMGs, developing durable and broad-spectrum resistance in the field is challenging. Molecular characterization of CMG species in farmers' fields in Ghana resulted in the identification of an isolate of East African cassava mosaic virus (EACMV), closely related to the EACMV-Kenya virus (EACMKV). This is the first report showing that an EACMKV species is associated with CMD in Ghana, which warrants attention as this could have epidemiological implications for control of CMD. Using current phylogenetic approaches that infer evolutionary rate variation at individual sites along protein-coding sequences, we identified the replication-associated protein (Rep/AC1) and the coat protein (CP/AV1) of CMGS to be evolving under negative selection. Evidence of strong negative selection in the Rep and CP genes of cassava mosaic geminiviruses makes them suitable targets for engineering resistance to geminiviruses in cassava. In the second chapter, I present results on screening for CBSD in four cassava-producing regions in Ghana. Furthermore, the response of eleven selected farmer-preferred cassava cultivars to mixed infections of CBSV (TAZ-DES-01) and UCBSV (TAZ-DES-02) isolates is reported. Although CBSD has not been reported in West Africa, reports in Burundi and Democratic Republic of Congo indicate a westward drift of the disease into West Africa. In this chapter, we confirmed the absence of CBSD in farmer fields in the four cassava-producing regions in Ghana. However, susceptibility of all eleven cultivars to both CBSV and UCBSV infection is alarming. It is expected that these results will encourage extensive screening for CBSD across Ghana and West Africa. Furthermore, screening of other local cassava cultivars could lead to identification of sources of CBSD resistance. The use of complementary approaches such as transgenic technology is encouraged to speed up development of CBSD-resistant cassava for farmers. In the third chapter, I present results on the induction of friable embryogenic calli (FEC) in eleven cassava cultivars from Ghana. FEC induction in cassava is highly genotype-dependent making it difficult to achieve in several African cassava cultivars. We successfully induced quality FEC tissues in ADI 001, a farmer-preferred cultivar. Using the standard protocol for cassava transformation developed in our laboratory, we successfully transformed and regenerated plantlets of ADI 001. The application of flow cytometry for routine assessment of FEC ploidy levels was carried out to ensure that quality FEC tissues are used for genetic transformation. Furthermore, over 40% of recovered lines had a single insertion of the transgene, making them suitable for downstream molecular characterization and confined field trials. In the fourth chapter, I present results showing development and successful transfer of CMD resistance into a farmer-preferred cassava cultivar, ADI 001. Using the RNAi-based approach that has achieved some success in developing resistance against both RNA and DNA viruses, we tested the single and multiple target approach for engineering resistance against mixed CMG infections present in cassava fields. Based on the genetic diversity study of CMGs identified in farmers' fields, the replication-associated protein (AC1) and the movement protein (BC1) were selected as targets for silencing. Double targeting of AC1 and BC1 resulted in a shift in viral population in cuttings multiplied from scions inoculated with mixed CMG population. However, expression of high levels of hairpin-derived small RNAs homologous to only the AC1 gene sufficed efficient silencing of mixed CMG population. Near immunity to mixed CMG infection achieved in transgenic dsAC1 line-42 was maintained in cuttings multiplied from scions inoculated with field infected rootstocks. The work presented in this thesis demonstrates successful transfer of CMD resistance through transgenic technologies into a cassava cultivar preferred by farmers in Ghana. The identification of high levels of resistance against mixed CMG infection in this local cultivar, ADI 001 (dsAC1 line-42) encourages field screening to evaluate durability of resistance engineered. More importantly, transfer of transgenic technologies to laboratories in Africa such as the Biotechnology and Nuclear Agriculture Research Institute (BNARI) in Ghana through building of capacity is necessary to encourage and sustain adoption of transgenic technologies in Africa.