ABSTRACT
In Chapter 7 we considered different aspects of vector design that can be used to control transgene expression. The salient point here is that vector design allows transgene expression to be controlled by the investigator, and the effects of vector modification are usually predictable. However, there are also many influences on transgene expression which are unpredictable, and most of these occur inside the cell where the investigator has little influence. As stated at the beginning of this book, the integration of exogenous DNA by illegitimate recombination is regarded as an approximately random process depending to a large degree on the presence of pre-existing chromosome breaks 1 . We say approximately random because there does appear to be a certain bias for insertion in all genomes, probably reflecting the normal state of chromatin architecture and its accessibility to foreign DNA. With transpositional vectors such as P-elements and retroviruses, there may even be a degree of target site choice at least at the sequence level, although the consensus target sequences are generally weak with the result that insertions viewed on a genomic scale remain more or less random. The position of integration is therefore one aspect of gene transfer than cannot be controlled unless specialized gene-targeting strategies are used (Chapter 6). Another is the structure of the resulting transgenic locus—the number of transgene copies, their arrangement, their structural integrity—all of these lie outside the control of the investigator again unless gene targeting is used to introduce precise modifications.
