ABSTRACT
Literature cited.............................................................................................................................182
One approach to improving plant resistance to insects is to alter the plant’s chemical composition so that it will deter or intoxicate target herbivores. Traditional breeding to achieve this effect is restricted to the transfer/exchange of “desired” traits among sexually compatible, i.e., related, species. The crossing partners in this process already share genomes that are in large part homologous. Groups of functionally linked genes are moved; they include the relevant promoters, regulatory sequences, and associated genes involved in the regulated expression of the character of interest. The breeding process typically is slow, the change in traits often minimal, and the new levels of resistance to pests rarely high. During lengthy breeding programs, unintended effects can be identified and eliminated to the extent possible before a new cultivar is commercially released. In contrast,
genetic engineering typically uses genes and gene constructs derived from completely unrelated organisms and adds them to the existing genome in an entirely novel genetic context (Regal 1994). This technology involves the random insertion of genes without the relevant promoter sequences and associated regulatory genes (Antoniou 1996). Currently, viral promoter genes from reproductively incompatible species — coding, for example, for antibiotic resistance — are inserted instead. Because genetic engineering is expected to yield commercial cultivars more readily than does conventional breeding, subtle unexpected effects may escape the attention of the seed producers and be discovered at the commercial stage, i.e., by farmers. Events like the “boll drop” of transgenic cotton or “stem splicing” of transgenic soybeans, both of which occurred in areas of the southeastern United States, may be the first such cases.
