ABSTRACT

Cassava (Manihot esculenta L.), a starch-rich plant that has poor protein content and usually poor vitamin content, feeds about 600 million people each day. When cereals can no longer be grown because of soil fertility problems, it is often still possible to grow cassava. It is the third most important source of dietary calories in the tropics, and reliance on the crop is especially high in West and Central Africa. The International Laboratory for Tropical Agricultural Biotechnology is promoting research to improve cassava productivity and is a leader in developing genetic engineering to improve the quantitative and qualitative traits of this essential food crop. Improvements in the areas of resistance to insects and viral diseases, enhanced nutritional qualities, reduced cynogenic content and modifi ed starch characteristics are urgently needed. Traditional breeding is hampered by the nature of this crop, which has a high degree of heterozygosity, irregular fl owering, and poor seed set. Biotechnology has the potential to enhance crop improvement efforts, and genetic engineering techniques for cassava have been developed over the past decade. Selectable and scorable markers are critical to effi cient transformation technology, and must be evaluated for biosafety, as well as effi ciency and cost-effectiveness. The selectable marker genes nptII, hpt, bar/pat, and manA, and the scorable marker gene uidA, all have little risk in terms of biosafety. These appear to represent the safest options for use in cassava biotechnology available at this time (Petersen et al., 2005). In spite cassava providing food security in

marginal lands, its scientifi c breeding began only recently compared with other crops. However, signifi cant progress has been achieved, particularly in Asia where cassava is used mainly for industrial processing and no major biotic constraints affect its productivity. Cassava breeding faces several limitations that need to be addressed. The heterozygous nature of the crop and parental lines used to generate new segregating progenies make it diffi cult to identify parents with good breeding values. Breeding so far has been mainly based on a mass phenotypic recurrent selection. There is very little knowledge on the inheritance of traits of agronomic relevance. Several approaches have been taken to overcome the constraints in the current methodologies for genetic improvement of cassava. Evaluations at early stages of selection allows for estimates of general combining ability effect or breeding values of parental lines. Inbreeding by sequential selfpollination facilitates identifi cation of useful recessive traits, either already present in the Manihot gene pool or induced by mutagenesis (Ceballos et al., 2004). But the capacity to integrate transgenes into cassava has now been established and being utilized to generate plants expressing traits of agronomic interest. Tissue culture and gene transfer systems currently employed to produce these transgenic cassava have improved signifi cantly in this direction. Programs are underway to develop cassava with enhanced resistance to viral diseases and insect pests, improved nutritional content, modifi ed and increased starch metabolism and reduced cynogenic content of processed roots (Taylor et al., 2004). On the commercial front, Roble et al. (2003) produced l-lactic acid from raw cassava starch, by simultaneous enzyme production, starch saccharifi cation and fermentation in a circulating loop bioreactor with Aspergillus awamori and Lactococcus lactis ssp. lactis. Repeated fed-batch l-lactic acid production was performed for more than 400 hours and the average lactic acid yield and productivity from raw cassava starch were 0.76 gram lactic acid per gramstarch and 1.6 gram lactic acid per liter per hour, respectively.