7.5. TARGETED TRANSGENE INSERTIONS
Once a recombinant T-DNA vector has been generated, with features designed to provide
stable integration and gene expression, the DNA enters the plant cell and integrates ran-
domly within the genome. The position of integration is uncontrolled and can often
result in variable levels of transgene expression. A number of factors influence the level
of transgene expression in plants, including the number of transgenes inserted into the
genome, localcis-acting elements, and RNA silencing. Nontranscribed, A/T-rich regions
in eukaryotic genomes, known asmatrix attachment regions(MARs), have been used to
flank genes in T-DNA vectors (Butaye et al. 2004). These sequences have been reported
to result in more reliable transgene expression shielding transgenes from RNA silencing
(Mlyna ́rova ́et al. 2003). However, targeting transgenes to predetermined chromosomal
sites by homologous recombination would perhaps provide greater control and reduce
potential positional effects. Until relatively recently, such approaches have been very inef-
ficient in plants. Advances in the production of synthetic transcription factors [zinc-finger
proteins (ZFPs)] designed to recognize specific DNA target sequences have now made it
possible to increase the efficiency of targeted homologous recombination in plants, creating
the potential to engineer precise deletions, insertions, or mutations within specific chromo-
somal regions. The production of customized ZFPs will provide a variety of precision tools
to alter genomes, changing the expression of endogenous genes and transgenes in future
generations of genetically engineered plants.
7.6 Safety Features in Vector Design
The use of plants as bioreactors for the manufacture of polymers, antibodies, vaccines, hor-
mones, and a variety of other therapeutic agents also presents new challenges in vector
design and construction. The effective use of plants as bioreactors, involves not only the
careful selection of host plants (i.e., food crop or a nonfood crop), but also innovative
vector design to ensure high levels of gene expression with safety features that ensure
that products will not enter the food chain. For pharmaceutical production, plants have
many advantages, the most significant of which is their eukaryotic protein synthesis
pathway, capable of the posttranslational modification and assembly steps required to
produce active eukaryotic proteins, such as antibodies. Unfortunately, plants glycosylate
proteins differently to mammals, however, recent and future advances in “humanizing”
plant glycosylation pathways (for a recent review see Joshi and Lopez 2005) will make
the production of “humanize” proteins feasible. A great advantage of plants is that they
can be grown in huge numbers to produce very large quantities of protein, they are free
of mammalian pathogens, and many plant varieties are edible, providing an easy means
of administering medication. However, to reduce the risk of nonedible products entering
the food chain, plant expression vectors for such products must be engineered with
robust safety features. One such safety mechanism is to incorporateinteinsequences that
permit the transsplicing of proteins. This means that genes encoding the transsplicing
protein fragments do not need to be located in the same genome; one can be contained
in the nuclear genome and another in the chloroplast genome, for example. In this case,
the nuclearly encoded protein fragment is engineered to target the chloroplast, where it is
transspliced to the second protein fragment encoded by the chloroplast. This type of
split-gene technology requires two types of vector construction: a T-DNA vector, for
186 RECOMBINANT DNA, VECTOR DESIGN, AND CONSTRUCTION