Transgenic Sunflower - Genetics, Genomics and Breeding of Crop Plants - page 305

Two strategies designed to minimize environmental risks proposed by
Gressel and Al-Ahmad (2005) have been suggested for sunflower. Transgenes
could be contained in male-sterile varieties, such as those studied for latex
production (Cornish et al. 2007) or inserted in chloroplasts. However, those
strategies would not ensure a complete barrier for transgene containment.
As in other species, a small part of sunflower paternal cytoplasm carrying
cp-DNA could be transmitted to the progeny (Haygood et al. 2004) while
volunteer plants would obviously carry them through maternal inheritance.
Bervillé et al. (2005) proposed to mitigate gene flow by using transgenic
varieties carrying chromosome translocations and inversions. For example,
Helianthus laciniatus differs from sunflower by at least eight reciprocal
translocations and one inversion, which render gene flow almost impossible
(Jan and Seiler 2007). Hybrids of wild or volunteer plants pollinated by
cultivated biotypes carrying these chromosome rearrangements would lack
a normal meiotic pairing having a very low fertility. Although these varieties
would show a high fecundity under normal crop conditions, chromosome
segments from wild species would carry agronomically undesirable traits,
as it occurred with the first generation of IMI-tolerant sunflower varieties.
An alternative would consist in transgene linkage to traits of low
persistence in the wild. Snow et al. (1998) reported a low fecundity in
domesticated-like genotypes, whereas wild types predominated following
introgression of crop genes within wild populations. In order to mitigate
gene flow, transgenes should be linked to domesticated traits, such as no
branching. One-headed plants would produce large seeds easily detected
by predators (Alexander et al. 2001), lessening in this way the soil seed
bank, which constitutes the main component of population dynamics
(Claessen et al. 2005).
In the present state of art, this option would be difficult to realize given
the low probability of a transgene being inserted in a specified locus within
the genome. Moreover, as the genus has one of the highest recombinational
rates among plants (Burke et al. 2004) this strategy would not be reliable
enough and linkage should be extremely tight not to be broken.
A fecundity increase due to Bt transgene acquisition could be expected
in wild plants (Snow et al. 2003) however other events would have the
opposite or neutral effects. Gene flow-acquired herbicide tolerance could
increase weedy sunflower interference, both of volunteers and wild-crop
hybrids, but even in the worst scenario this trait would not entail the creation
of “superweeds”. These variants would be controlled through herbicide
rotation, a technique traditionally recommended to avoid mutants tolerant
to the chemical groups utilized ( www.weedscience.org ).
If environmental constraints are to be overcome through regulation
flexibility or by obtaining varieties harboring containment or mitigation
mechanisms, transgenic sunflower diffusion would be strongly conditioned