Agriculture Reference
In-Depth Information
of genetic variability available to plant breeders.
Current successful examples of transgenic traits
include herbicide tolerance, pathogen resistance,
insect resistance, abiotic stress tolerance (though
this work may need further validation), and
improved nutritional quality (reviewed by Vasil
2007). Though grain yield has been reported to
have been increased by transgenes in greenhouses
or growth chambers, fi eld evaluations are much
less promising (Meyer et al., 2007). However, this
result should not be surprising as only with recent
molecular marker technology have quantitative
trait loci (QTLs) affecting grain yield been identi-
fi ed (Campbell et al., 2003; Cuthbert et al., 2008).
Hence with our limited knowledge of grain yield
(the exception being Ashikari et al. 2005), it is not
surprising that a benefi cial transgene for grain
yield will be a challenge to identify initially and
be used by a plant breeder.
Aside from the diffi culty of identifi cation, a
related matter is whether the insertion of trans-
genes affect wheat grain yield. A possible reduc-
tion in grain yield could be due to the transgene
insertion; it could be due to somaclonal variation
from tissue culture during the recovery of the
transformed cell or plant; or, as a result of cultivar
heterogeneity, a single transformant could by
chance represent a lower yielding biotype within
the original cultivar. However, most transforma-
tions are done in highly transformable lines, and
the resultant transgene is backcrossed into elite
lines. Hence somaclonal variation or a lower
yielding selection would be lost in the backcross.
Most of the available data indicate that transgenic
wheat performs very similarly to the parental line
that does not carry the transgene (Bregitzer et al.,
2006; Vasil 2007).
Though often overlooked when considering
the importance of transgenic wheat, one area
of transformation that can greatly assist wheat
breeders is the ability to chemically or insertion-
ally link useful genes so that they segregate as a
unit or as a multiple-gene “locus” (Halpin 2005).
As described in a subsequent section (“Methods
of selecting while inbreeding to develop a culti-
var”), the ultimate goal of plant breeding is to
accumulate favorable alleles into one cultivar. As
more desirable independently segregating genes
are needed in a cultivar, the population size needs
to be proportionately larger, because the chance
of fi nding the line with the most desirable genes
is lower (Tables 13.1 and 13.2). However, the
DGQ is greater, and hence the predicted popula-
tion size is lower, if the genes are linked in cou-
pling phase.
In transgenic crops, the need to stack genes, or
pyramid desirable genes, has become important
as more benefi cial transgenic traits are identifi ed
(Halpin 2005). While various methods are used to
stack genes (e.g., iterative transformations, cross-
ing independently transformed lines, cotransfor-
mation with different transgenes, or chemically
linking the transgenes for a single insertion),
methods which lead to co-insertion of the benefi -
cial transgenes are preferred by plant breeders for
their ease of manipulation. An example of chemi-
cally linked multiple genes would be a herbicide-
selectable marker linked to another trait of interest
(e.g., virus resistance). The resultant transgenic
lines would be herbicide tolerant and virus resis-
tant. Whenever the breeder selected for herbicide
tolerance (often with a very easy selection assay),
the selected lines would be virus resistant accord-
ing to a simple F
2
segregation ratio of 3 : 1, herbi-
cide-tolerant and virus-resistant-to-herbicide- and
virus-susceptible.
A major limitation with using linked transgenes
is that the needed stacked traits will differ by
adaptation or geographic region. For example,
while herbicide tolerance could be widely used,
resistance to a specifi c pathogen may only be
needed in a localized region; hence a different
gene stack may be needed for different areas of
adaptation. A second limitation is that while it
might be best for wheat breeders to have a series
of different inserted stacked traits, the practicality
and expense of transgene regulation preclude this
possibility. Currently every commercial transgene
is heavily regulated and must undergo extensive
testing. This testing presents a major economic
impediment for having multiple and similar
transgenic clusters undergo regulation. The high
cost of regulatory approval, though it may not
hinder initial transgenic discoveries by the public
sector, requires cooperation with the private
sector for commercialization.
