<br/>Cassava <em>(Manihot esculenta</em> Crantz) is a tropical crop grown for its starchy thickened roots, mainly by peasant farmers, in the tropics, for whom it is a staple food. There is an increasing demand for the use of cassava in processed food and feed products, and in the paper and textile industries amongst others. This thesis describes research on the cloning of the genes encoding ADP-glucose pyrophosphorylase small and large subunits (AGPase B and S, respectively) and granule bound starch synthase II (GBSSII). These genes and their products were extensively characterised to determine their role in starch biosynthesis in cassava. Functional verification of the genes was carried out by transforming potato and cassava followed by analysis of the starch produced by the transgenic plants.<p>In Chapter 1 cassava production in the world in general and in Zimbabwe in particular is examined against the backdrop of new cloning and transformation strategies to improve starch quality and quantity. The development of cassava cultivars whose starches have novel physico-chemical properties by genetic modification of the process of starch biosynthesis is examined therein. The main criteria for these new cultivars to emerge are set forth as being: the availability of cloned and characterised starch biosynthesis genes, a universally applicable transformation and regeneration procedure for cassava, transfer to appropriate cassava cultivars, and biosafety analysis of transgenic cassava plants before disbursement to farmers.<p>The cloning of the cassava starch biosynthesis genes encoding granule bound starch synthase II (GBSSII) and the large and small subunits of ADP-glucose pyrophosphorylase (AGPase) is described in Chapters 2 and 3. The cloning of GBSSII reveals that there is indeed a second isoform of this enzyme in cassava as in other plants species. While sharing very little amino acid sequence homology with cassava GBSSI the GBSSII isophorm shares high amino acid sequence homology to other GBSSII genes from pea and potato. Cassava GBSSII seems to be more important in leaf tissue where it is more highly expressed than in tuber tissue where GBSSI predominates. Mapping of GBSSII revealed that this is a single copy gene located on the male derived linkage group T of the cassava mapping population.<p>Cloning of the cassava genes coding for the small (B) and large subunit (S) of AGPase revealed interesting aspects about the cassava enzyme. The cassava AGPase is likely to be heterotetrameric in constitution as had been found in other plant species. Comparison of the cassava AGPase sequences with those of already cloned AGPases revealed that AGPase B is more similar to small subunit genes from other plants than to cassava AGPase S coding for the large subunit (Chapter 3). Segregation analysis of a cassava mapping population revealed that AGPase S is a single copy gene that is localised on the female derived linkage group E of the cassava genetic map. Both genes are expressed in all cassava tissues but AGPase B was shown to have a higher steady state mRNA level than AGPase S especially in leaf and tuber tissue. Post-transcriptional control of small subunit polypeptide levels could be inferred from the discrepancy between AGPase B mRNA and polypeptide levels. The AGPase enzyme activity was much higher in young cassava leaves than older leaves and tubers. Cassava leaf AGPase activity was increased 3 fold by the addition of 3-PGA (3-phospho-glycerate) and inhibited by up to 90% in the presence of inorganic phosphate (Pi). The tuber enzyme was relatively unaffected by 3PGA, but was highly inhibited by Pi.<p>In order to verify the biological role of the AGPase B gene antisense constructs were made of the cassava AGPase B behind a CaMV35S promoter (chapter 3). This was transferred into potato plants by <em>Agrobacterium tumefaciens.</em> While the 224 transgenic antisense AGPase B potato plants did not differ in appearance from normal potato plants, 45 transgenic plants, however, had more numerous and smaller tubers than control plants. Antisense plants with reduced AGPase B mRNA levels had 1.5 to 3 times less starch than tubers from the control plants. The levels of the soluble sugars in the antisense plants increased significantly (up to 10 times more glucose, 6 times the amount of fructose, and 5 times the amount of sucrose) when compared to those found in control plants. These results show that a heterologous gene from cassava can have an antisense effect in potato, but that the number of plants required to find plants exhibiting maximum antisense effect has to be very large. This is probably due to sequence homology differences between the cassava AGPase B and potato AGPase B genes which share only 68% amino acid sequence homology.<p>Chapter 5 describes the further development of an efficient, time and labour saving protocol for transforming cassava based on stringent selection of the luciferase (firefly) marker gene. In addition the first reported transformation of cassava with a gene (AGPase B) other than a marker gene is described. An antisense construct was made for transforming cassava. This consisted of the cassava AGPase B gene which was placed in antisense orientation behind the CaMV35S promoter. This was then coupled to the luciferase gene driven by another CaMV35S promoter. After particle bombardment of cassava FEC transgenic tissue was selected using three different selection regimes: non stringent luciferase selection, stringent luciferase selection and combined chemical (phosphinothrycin) and luciferase selection. Stringent luciferase selection whereby luciferase positive FEC units were precisely pinpointed, isolated and cultured was found to be the most effective and time saving method. It was possible to generate cultures having more than 90% luciferase positive FEC tissue after 12 weeks of stringent LUC selection, compared to 45% and <1 % for combined selection and non stringent selection respectively. The number of luciferase positive mature embryos generated was directly proportional to the percentage of luciferase positive tissue in the original FEC culture. Stringent luciferase selection enabled the time taken for production of transgenic cassava plants to be reduced to 28-36 weeks as compared to 8 months to a year with no stringent selection or LUC/PPT selection.<p>Cassava plants carrying the AGPase B antisense gene had extremely low levels of starch, compared to control plants, as shown by iodine staining of in vitro induced thick stems. In plants exhibiting the highest AGPase B antisense effect, starch formation was limited only to the epidermal layer. These results functionally confirm the identity of cassava AGPase B as well as emphasising the critical role of AGPase in starch formation in cassava.<p>A discussion about the significance and implications of cloning cassava genes and producing transgenic cassava for culture in developing countries is carried out in Chapter 6. While there are clearly many economic and nutritional benefits to producing transgenic cassava, for resource poor farmers, many people in the South are not aware of the biosafety implications of growing transgenic crops. It is further emphasised that discussions and debate should be initiated to make local communities aware of the issues surrounding transgenic crops and their products. In addition it is recommended that some form of international legal framework be set up to ensure that resource poor farmers are not disadvantaged by the patenting of material originating from their communities by individuals and companies in the North. This thesis clearly demonstrates how it will be possible in the near future to produce new cassava cultivars carrying the appropriate genes to affect pronounced changes on tuber productivity and starch quality.
|Qualification||Doctor of Philosophy|
|Award date||23 Dec 1997|
|Place of Publication||S.l.|
|Publication status||Published - 1997|
- manihot esculenta