Agriculture Reference
In-Depth Information
transgenic wheat lines by Jackson et al. (2001),
visualization of transgene loci by fi ber-FISH
showed that the most common structures (8 of
13) contained one to two transgene copies. The
next most common structure (4 of 13) consisted
of large tandem arrays of transgenes spanning
32 to 125 kb and interspersed with unknown
DNA. A single transgenic event contained a
77-kb tandem array of plasmids with one 7-kb
interruption (Jackson et al., 2001). Rooke et al.
(2003) studied six transgenic wheat lines trans-
formed with HMW-GS and selection genes.
They found that the transgenes were present in 1
to about 15 copies at 1 to 3 separate loci. Some
copies of genes in the same locus were clustered
but separated by genomic DNA. Some were
present as tandem repeats and others were rear-
ranged or truncated. In two cases, the HMW-GS
gene could be segregated from the selection gene
(Rooke et al., 2003).
Complex integration structures in wheat trans-
formants are diffi cult to fully characterize because
of the size and repetitive nature of the wheat
genome. Since commercialization requires exten-
sive molecular characterization of the transgenic
plant, plants destined for commercial applica-
tions must have simple insertion structures. A
change in the structure of the DNA used for
bombardment, originally implemented to reduce
the occurrence of antibiotic resistance genes in
transgenic plants, shows promise for reducing
the complexity of integration sites in biolistic
transformants. In this method, plasmid backbone
sequences are physically removed from the DNA
before transformation. The resulting linear
DNAs comprise “minimum gene cassettes,” con-
sisting only of the sequences needed for expres-
sion of the genes in wheat: functionally linked
promoter, coding, and transcription termination
sequences. Bombardment with these so-called
clean genes has been shown in rice, wheat, and
other crops to result in lower transgene copy
numbers and less-complex insertions (Fu et al.,
2000; Agrawal et al., 2005; Yao et al., 2006;
Gadaleta et al., 2008b). Transformation effi cien-
cies are at least as high as those with intact
plasmid DNA (Yao et al., 2006), resulting in
integration of one to four copies of each trans-
gene. These results for both rice and wheat
suggest that, for biolistic transformation, the
sequences in plasmid backbones of circular DNAs
facilitate interactions among separate plasmids,
either by homologous recombination or as hot-
spots for illegitimate recombination, and that
such interactions can lead to formation of multi-
gene arrays before integration of the whole unit
into the chromosome (Kohli et al., 2003).
To the extent that co-integration of the selec-
tion gene and the gene(s) of interest occurs, it
becomes more diffi cult to fi nd progeny of the
transformants that contain only the gene of inter-
est. Linkage of the selection gene and gene of
interest can sometimes be advantageous in that
the phenotype of the selection gene can be used
to easily follow inheritance. This can be of value
when the gene of interest has a phenotype that is
costly or diffi cult to assay, such as grain protein
content or resistance to Fusarium head blight.
However, transgenic plants that retain selection
genes are becoming increasingly undesirable for
several reasons (Natarajan and Turna 2007). For
basic research, such plants cannot be retrans-
formed using the same selection gene. For plants
destined for commercial production, the pres-
ence of unnecessary new proteins in the food
supply could make transgenic wheat less accept-
able to the general public. For plants destined for
testing or release into fi eld environments, the
absence of herbicide resistance genes (in cases
where resistance is not the targeted trait) is desir-
able to prevent their widespread dispersal via
pollen or seed mixing with cultivated nontrans-
formed wheat and weedy relatives, to which they
could confer a selective advantage (Matus-Cadiz
et al., 2004; Brule-Babel et al., 2006). In the US,
jointed goatgrass (
Aegilops cylindrica
Host,
CCDD) is the only known wild wheat relative
that can produce, albeit at low frequencies, semi-
fertile hybrids with domesticated wheat (reviewed
in Hegde and Waines 2004; Hanson et al., 2005;
Schoenenberger et al., 2005). In parts of the
world where wild wheat relatives fl ourish, the
possibility of gene fl ow via pollen transmittal
would be more problematic (Weissmann et al.,
2005; Zaharieva and Monneveux 2006; Felber
et al., 2007).
