| ||||||||||||||||||||
Crying to fit in
It is very frustrating: despite nearly 20 years of use, we still do not know how T-DNA integrates into plant genomes, or what, if anything, controls the sites of integration. This ignorance is all the more significant because both position effects and the number of integration sites affect the level of expression of the foreign gene. Jin and colleagues (pp. 31-36) have investigated the distribution of integrated genes in the rice genome using FISH (fluorescence in situ hybridization). The foreign gene construct was chimaeric but I focus on just one component, the gene encoding the Bacillus thuringiensis toxin [the Cry1A(b) gene]. Nine transgenic lines were grown to the T3 generation and were assessed for expression of Cry1A(b). The lines were extremely variable in this respect, the highest level of expression being about 25 times the lowest, while two lines showed no detectable expression. The detection by FISH of the genes themselves showed first that no integration occurred closer to the centromere than approx. 26 % of the total distance between centromere and the end of the chromosome arm (i.e. at FL values of less than 26); secondly, that the majority of integrations were at sites with FL values over 75; and thirdly, that in some lines there was more than one integration site. In terms of 'cold' and 'hot' spots for integration, the region proximal to the centromere is clearly cold, whereas the region distal to the centromere is 'hot'. However, the specific position of integration did not define the level of expression. Thus, one line in which the integration was a long way from the centromere showed no detectable expression. Clearly we need more information on the specific molecular characteristics of the integration sites, but this may need to await our elucidation of the actual integration mechanism.
Professor J. A. Bryant
University of Exeter, UK
j.a.bryant{at}exeter.ac.uk
| ||||||||||||||||||||