heavy rain and windy conditions. The dwarf plants of the so-called Green Revolution are
short because they respond abnormally to the plant growth hormone, gibberellin.
Attempts to generate transgenic dwarf rice, by misexpressing theArabidopsis gibberellin-
insensitive(GAI) gene, resulted in short plants that unfortunately also produced low seed set
(Fu et al. 2001; Tomsett et al. 2004). Subsequent experiments have shown that this problem
could be resolved, at least inArabidopsis, by constructing a vector that places theGAItrans-
gene under the control of an inducible promoter (Ait-ali et al. 2003).
Although many endogenous (originating from within the organism) plant promoters that
can rapidly respond to the application of inducers have been identified, these often also
respond to environmental factors, such as water, salt stress, temperature, illumination,
wounding, or infection by pathogens. Other nonendogenous inducible systems have been
developed, but these rely on DNA sequences of foreign origin [for a more recent review,
see Curtis and Grossniklaus (2006)]. Since endogenous promoters can be triggered inappro-
priately by environmental factors, and inducers may modify native gene expression
(perhaps altering the physiology and development of the plant), an alternative approach
that restricts transgene activity to specific tissue types to produce the desired trait would
be more profitable. In the case ofGAIexpression, a construct with a promoter that is
active in vegetative tissues only (and not reproductive tissues) may result in dwarf plants
that do not have reduced seed set.
7.4.6. Vectors for Multigenic Traits
The construction of vectors for crop improvement can rely on the insertion of a single gene,
as is the case with the production ofBttoxin to protect crops against insects, or on the inser-
tion of several genes, as is likely to be required to engineer the vast majority of agronomi-
cally important traits. Currently, multigenic traits are obtained either through sequential
sexual crossing of transgenic plant lines that allows the accumulation of three or four inde-
pendent transgenes in a single plant (see Chapter 3), or by the use of different transgenes
held on distinct T-DNAs that are used to cotransform plants (see Chapter 10). The former
approach is laborious, and the latter is technically challenging. Careful consideration of the
design and construction of plant transformation vectors can resolve many of the technical
difficulties, allowing polygenic traits to be expressed from a single T-DNA. One such
design relies on a collection of auxiliary vectors (Fig. 7.21). Using such approaches, a con-
struct capable of expressing up to six transgenes from a single location within the nuclear
genome has been generated (Goderis et al. 2002) (Fig. 7.22).
There are many alternative approaches to “stacking” multiple genes into acceptor
vectors. These make use of site-specific recombination systems and homing endonucleases
that allow the sequential and indefinite delivery of expression cassettes to an acceptor
vector, thereby allowing the expression of many transgenes from a single locus in the
genome.
Figure 7.22 (Continued)Agrobacterium tumefaciensnopaline synthase gene (tNOS), the terminator
of gene 7 of theA. tumefaciensplasmid Ti15955 (tG7), the terminator of the mannopine synthase
gene fromA. tumefaciens(tMAS), or the terminator from theA. tumefaciensoctopine synthase
gene (tOCS). (b) These six gene cassettes are arranged between the LB and RB of the T-DNA
vector pTRANS3458.
7.4. VECTOR DESIGN 185