www.checkbiotech.org ; www.raupp.info ; www.czu.cz

Guidelines to ensure the efficient coexistence of genetically modified (GM)
and non-GM crops are currently being considered across the European Union,
November 2005 by Ewen Mullins.

Curtailing pollen/seed-mediated gene flow between GM and non-GM crops is
central to effective coexistence. While models have been designed for
specific crops,1 traditional commentary associated with a crop's potential
for gene flow would typically rank the crop as a high, medium, or low risk.
By its qualitative nature, this approach does not provide the detail
required to highlight those aspects of a crop's biology that will serve to
challenge coexistence management. The substitution of this classification
system with a numerical gene flow index would permit a background level of
gene flow, specific for a particular crop, to be calculated. In turn, this
would underscore those crops that require additional measures when
genetically modified, in order to minimize gene flow in accordance with
anticipated coexistence guidelines.

The concept of a gene flow index or botanical file is not new2 and their
potential as tools to assist risk assessment strategies has already been
suggested. 3 However, present systems fall short by not encompassing all
modes of gene flow that are of relevance to coexistence. Here we present a
gene flow index (GFI) 4 model that we have applied to seven conventional
crops in Ireland. By combining four strands of gene flow?crop pollen-to-wild
relative (CPW); crop pollen-to-crop (CPC); crop seed-to-volunteer (CSV); and
crop seed-to-feral (CSF)?we have established a baseline data set that
describes the potential of Ireland's primary arable crops for both pollen-
and seed-mediated gene flow.

Approach

Information inputted into the model was collated from a broad literature
base, and only information that pertained to systems comparable to the Irish
agricultural and geographical environment was used. The calculated GFI value
pertains to the propensity of each crop to form viable
hybrid/volunteer/feral individuals. A clear distinction was made between the
volunteer and feral niches by differentiating between the ability of a plant
to grow within/outside a managed crop system, respectively. Responsive to
regional parameters, we applied the model to sugar beet, oilseed rape,
potato, perennial ryegrass, maize, wheat, and barley. Alternative pathways
for gene flow (e.g., wild, volunteer, or feral originating pollen to a
related crop) were not considered.

For all four strands the decisive factor for successful gene flow was deemed
to be the establishment of a viable, reproducing hybrid/volunteer/feral
individual, without which the introgression/gene spread exposure element of
any future GM crop risk assessment could not occur. By restricting the
analysis to just the dispersal and preliminary stage of establishing a
viable individual/population, it is accepted that the model excludes the
issue of hybrid/feral competitive ability. It does, however, provide an
initial data set that will quantify the propensity of a conventional crop to
spread its genetic material.

Retaining a simple format, each of the four strands (CPW, CPC, CSV, CSF)
contains several sequential questions, with each question designed to
provide a ?yes/no' answer, which in turn equates to a relevant score.4 By
following this linked progression, when a question incurs an answer with a
zero value, that strand automatically records a total value of zero, as no
gene flow can take place for the specified crop under the selected
criterion. The adoption of this worst-case scenario approach was intentional
and maintains the practicality of the model by encompassing real-life
factors that, while not desired, will occur all the same?for example the
occurrence of bolters in a sugar beet crop.

Outcome

A composite GFI value for each crop was calculated from which the gene flow
potential of both ryegrass and sugar beet was clear (GFI=25/27). The
justification for such a value is supported by the fact that both species
co-exist in Ireland with inter-fertile wild relatives, both can disperse
their pollen over large distances, and the initiation of feral populations
from each species is a reality. Importantly, the high GFI value for
conventional sugar beet does not necessarily advocate the non-cultivation of
GM sugar beet in Ireland. Conversely, it underlines the importance of bolter
control in the effective coexistence of GM and non-GM sugar beet. This point
is clear when a readjusted model utilizes the sugar beet data for a
management system that assumes stringent bolter control. In this scenario,
GFI = 6 where the potential for gene-flow is minimized to the establishment
of volunteer and feral populations from harvested tuber fragments. The
potential for pollen and seed-mediated gene flow in potato (GFI = 11/27) is
related to combined tuber and true potato seed (TPS) production. When
recalculated for districts where potato production is strictly for tuber
production, the GFI = 6/27. Both wheat and barley recorded low indices (GFI
= 8/27 for each), in contrast to oilseed rape, which confirmed its ability
to disperse its genetic material with a GFI = 19. Cultivation of maize in
Ireland is solely for animal forage. Coupled with an absence of wild
relatives, the gene flow potential (GFI = 9/27) for maize is limited to
pollen-mediated crop-to-crop and seed-mediated crop-to-volunteer.

Discussion

Ecologically, the consequence of gene flow from a GM crop is wholly
dependent upon the physiological impact of the transgene and must be
addressed on a case-by-case basis. In contrast, the potential for gene flow
is primarily reliant upon the reproductive biology of the crop (be it GM or
non-GM) and this can be addressed by calculating a crop's GFI value. In this
research, several conventional crops (oilseed rape, ryegrass, and sugar
beet) attained a high GFI value. Importantly, it must not be implied from
this result that these crops are not suitable for GM development. Similarly,
for the crops that scored low GFI values, this does not imply gene flow will
not occur. Rather a high GFI score implies that a specific crop/variety
possesses a higher propensity for gene flow and thus requires greater
management precautions if efficient coexistence is to be attained.
Conversely, a low GFI value indicates a crop which should not pose a
significant challenge to the implementation of a coexistence strategy.

Notably, our work has highlighted several coexistence-based questions that
require further research and which should be addressed prior to the
commercialization of GM crops in Ireland. Specifically, the potential for
seed-mediated gene flow requires attention, for, due to a scarcity of
research data, we were limited in the number of questions we could ask in
regard to the efficacy of seed-mediated gene flow for each crop. This
contrasts with pollen-mediated gene flow (strand CPW and CPC) for which a
substantial research data set is available. The role of volunteers as
potential ?genetic bridges', facilitating the transfer of genetic material
from crop-to-wild/crop-to-crop/wild-to-crop, is also of particular concern,
especially as it would be naive to assume that total volunteer control will
be achieved in a coexistence-based management system.

From a non-scientific perspective, it is hoped that the GFI ranking scheme
will increase the public's understanding of ?gene flow', an issue central to
the GM debate. Within the scientific community, it is hoped that the
described index will revive discussion on the merits of gene flow indices;
specifically in regard to the feasibility of establishing a collective GM
crop risk index that encompasses not only a crop's propensity for gene flow
but also the elements that contribute to invasiveness, changes in genetic
diversity, and broader ecological disturbance.

Acknowledgements

This research is supported by the Irish National Development Fund.

References


Colbach N, Angevin F, Fargue A & Meynard JM. (2003) Using the GeneSys model
quantifying the effect of cropping systems on gene flow from GM rape
varieties to rape volunteers for designing and evaluating scenarios for
co-existence of GM, non-GM and organic crops. Proceedings of GMCC-03, Nov.
13th-14th, Denmark


de Vries AP, Meijden VD, & Brandenburg WA. (1992) Botanical files - a study
of the real chances for spontaneous gene flow from cultivated plants to the
wild flora of the Netherlands. Gorteria, Supplement 1


Ammann K, Jacot Y, & Al Mazyad RP. (2001) Safety of genetically engineered
plants: an ecological risk assessment of vertical gene flow. In Custers R,
ed, Safety of Genetically Engineered Crops. Zwijnaarde, Belgium, Flanders
Interuniversity, Institute for Biotechnology, pp 60-87


Flannery M-L, Meade C, & Mullins E. (2005) Employing a composite gene-flow
index to numerically quantify a crop's potential for gene flow: an Irish
perspective. Environmental Biosafety Research 4, p.29-43

[www.isb.vt.edu]

------------------------------------------
Posted to Phorum via PhorumMail