APPENDIX 1
Memorandum submitted by the Scottish Crop
Research Institute (G 2)
SUMMARY
Gene flow, when considered on a regional scale
in realistic contexts, is more frequent and can occur over longer
distances than some previous studies suggest. Such gene flow generally
does not normally compromise the ability of seed companies to
meet purity standards for conventional crops. Some of the factors
which influence the rate of gene flow over longer distances are
known. Low levels of gene flow over very long distances are inevitable
for some crops. It should be noted that such gene flow has been
a feature of agriculture since man first attempted growing crops.
1. The Scottish Crop Research Institute
is a Non-Departmental Public Body funded by grant-in-aid from
The Scottish Executive Rural Affairs Department and by competitive
income from a variety of sources. A special strength of the institute
is the wide range of skills of its scientists and the integration
of these skills to tackle important issues in crop biology.
2. For about a decade, SCRI has maintained
a forward-thinking programme of research on gene flow and other
types of risk assessment studies, including participation in the
current Farm Scale Evaluations of GM Crops. These projects, primarily
won through competitive bids, are listed in Annex 1. During the
Committee's deliberations on the Segregation of Genetically Modified
Foods in December 1999, SCRI's MAFF-funded research on the quantification
of gene flow at the regional scale was discussed by Professor
Alan Gray of ITE and ACRE. Aspects of SCRI's investigations of
the persistence of feral oilseed rape plants in a project funded
by DETR were also discussed. These studies have, in part, looked
at the "long tail" mentioned by Professor Gray of cross-fertilisation
over distance or population persistence over time and bring together
expertise in genetics, pollination biology, seed-bank dynamics,
vegetation systems and mathematics.
3. A particular focus of the SCRI studies
has been to consider events at the regional scale. This regional
focus has brought with it higher estimates of gene flow through
pollen movement than in earlier studies, and raised controversy
on a number of occasions. We consider these studies to be relevant
to the understanding of the segregation issues arising from the
recent problem with GM-tainted seed sold by Advanta, and offer
this note to the Committee to aid its deliberations.
4. In 1992, SCRI reported that oilseed rape
plants, deliberately emasculated, could be pollinated 2.5 km from
fields of the crop1-4. In 1997, MAFF funded a three-year programme
at SCRI to quantify gene flow at the regional scale. At every
distance from oilseed rape fields (up to the 4 km tested), pollination
events were detected on genetically male-sterile recipient plants5-7.
Parentage was verified for some events by DNA fingerprinting8.
Genetically male-sterile plants (of a similar type to those used
by Advanta in F1 Hybrid seed production) were used in this study
to enable fertilisation events to be detected readily on a relatively
large scale. The lack of competition from self pollen does, of
course, also enable higher rates of cross-pollination. Experiments
to quantify this aspect are not yet complete, but are a main focus
of the remaining experimental work of the project.
5. Understanding the mechanism of pollen
transfer in any crop is important for the prediction of the decay
of cross-pollination with increasing distance from the source.
Honeybees were identified as important vectors in long-distance
pollination events in oilseed rape6. They may be expected to transfer
pollen up to 5 km and perhaps in very exceptional situations up
to 10 km in any direction from the hive. The foraging range of
other social pollinating insects such as bumble bees, or the dispersal
of other potential pollinators such as pollen beetles and flies
is less well known but will, along with airborne pollen, make
a contribution to gene dispersal in oilseed rape. Such gene flow
will likely extend well beyond the distances already observed.
6. F1 Hybrid seed production in oilseed
rape entails growing strips of male-sterile lines interspersed
with strips of pollen donors. Honeybees are normally introduced
into the area at higher densities than is optimal for honey production
to ensure the efficient transfer of pollen. In these circumstances,
foraging well beyond the confines of the seed production plots
is inevitable. The blocks of male-sterile plants in such seed
production systems will encourage a degree of cross-pollination,
not just to nearby intended male parents but also to other fields
in the region, although the majority of pollinations will still
be with the intended parents.
Figure 3
Predicted transmission of a contaminating GM
trait into a commercial F1 Hybrid field containing X per cent
male sterile GM contaminants and propagation into the succeeding
generation.
7. In F1 seed production systems, male-sterility
is normally transmitted through the female line, from mother to
daughter. When the pollen arriving on the mail-sterile plants
carries a restorer gene, the plants grown from these seeds have
normal fertility. Such restorer genes are unlikely to be present
in neighbouring fields and hence the seeds produced by unintended
crossing give plants which are in most cases male-sterile. Advanta
use this system to generate restored F1 hybrid seed under the
cultivar name Hyola. Seeds produced in Canada have contained a
small proportion of GM and male-sterile off-types, apparently
from crossing the fields 800m or more distant. A normally male-fertile
oilseed rape flower is shown in figure 1 [not printed] and a male-sterile
flower with small anthers from a RoundUp Ready contaminant in
Hyola 38 is shown in figure 2 [not printed]. Some predictions
can be made on the fate of male-sterile contaminants in restored
F1 hybrid cultivars. They will be adequately pollinated by neighbouring
plants and will contribute fully to the seed harvest. If these
seeds are grown again about half will be male-fertile as about
half of the pollen grains produced in the field carry a restorer
gene. Equally, the male-sterile contaminants will transmit the
GM trait to half of their offspring (figure 3) and thereby the
level of GM contamination in the stock in this second generation
will be halved.
8. It should be re-iterated that ACRE have
been aware of the gene flow research at SCRI through scientific
papers, meetings and personal contacts, and have taken account
of our results in their deliberations. It has not been their intention
to ensure zero gene flow from GM field releases, but to accept
that some gene flow will occur and to focus on its implications.
One contribution of our research to the debate has been to point
out that levels of gene flow depend totally on context. In the
presence of efficient insect pollinators such as bees, in realistic
situations where a patchwork of large pollen sources can be expected,
and where the ability of small patches of recipient plants to
receive pollen is maximised, surprisingly high levels of gene
flow can be detected over very long distances. Changing one of
these criteria, for example where recipient plants are fully male-fertile
or are present in larger blocks, will reduce the height of the
"long tail" but probably not its length. In other words,
seed production, even if not using a male sterility system to
generate hybrid seed, will always be liable to contamination from
distant sources at a low level. "Distant" in this context
could mean a few hundred metres, a few km or even a few hundred
km. The seed industry already has experience of meeting the requirements
of purity thresholds laid down for non-GM seed crops. Increasingly
stringent thresholds would, on the basis of our results, become
increasingly impractical for seed producers to meet as technological
advances in detection accrue and as some insist on the right of
absolute freedom from all traces of detectable GMOs. It may be
argued that in an agriculture where there is still the possibility
of meeting the rights of farmers and consumers who wish to realise
the benefits of the technology, sensible threshold levels should
be adopted. In this context, the move towards setting thresholds
for GM admixture in non-GM seed stocks is welcome and will provide
certainty and security for both the seed producer and the consumer.
9. In addition to issues of GM admixture
through cross-pollination, there are other routes through which
OSR genes may become dispersed in space and time. Feral colonies
and volunteer plants occur through local spillage at the production
site, during bulk transport, on farm machinery, by the use of
agricultural soil containing seeds in landscaping, and possibly
by birds. Although most such colonies do not appear to persist,
some are capable of surviving for 10 years or more9.
10. Many of the notes above apply to seed
crops of oilseed rape. Oilseed rape has all the features required
for interesting pollination biology: a high attractiveness for
bees and indeed beekeepers intent on a honey crop; open flowers
also visited by pollen beetles, seed weevils, lepidoptera, hover
and other flies; pollen which can become airborne and travel far
in large quantities in the right conditions; and a somewhat unpredictable
ability to freely accept self pollen or promote out-crossing.
Other UK crops can be self-pollinated, vegetatively propagated,
wind-pollinated or pollinated by different types of insects. Realistically,
no UK crop will fall completely into one category: various combinations
of some of these elements will contribute to gene flow. Many are,
however, predominantly self-pollinated and are unlikely to arouse
such intense interest in gene flow as oilseed rape.
11. It is important to retain a sense of
perspective in this debate. Almost all of our crop and horticultural
species are effectively aliens to UK ecosystems. Indeed many crops,
including Brassica napus, oilseed rape, are hybrids which
do not occur in nature and are assumed to have arisen in man's
fields. Despite extensive use of GMOs in countries less nervous
about the technology than the UK, no confirmed adverse effects
on the environment have been reported. According to the report
of the recent OECD conference in Edinburgh earlier this year "many
consumers eat GM foods and no significant effects have yet been
detected on human health". At some time in the future, GM
approaches may be more readily accepted by the UK public as a
supplement to more traditional breeding practices including the
movement of genes between species by hybridisation and artificial
mutagenesis. The challenge facing us now is to ensure that by
reducing the climate of distrust by every available means, including
careful regulation and the underpinning science which informs
it, this acceptance is not delayed far into the future.
REFERENCES
1. Mackay, GR, McNicol, RJ, Wilkinson, MJ,
Timmons, AM & Dubbels, S (1992), Monitoring the flow of pollen
in crop plants. OTTAWA '92: The OECD workshop on methods for monitoring
organisms in the environment. Ottawa, Canada, September 1992:
pp 43-49.
2. Timmons, AM, O'Brien, ET, Charters, YM
and Wilkinson MJ (1995). Aspects of environmental risk assessment
for genetically modified plants with special reference to oilseed
rape. SCRI Annual Report 1994, pp 43-45.
3. Timmons, AM, O'Brien, ET, Charters, YM,
Dubbels, SJ, Wilkinson, MJ (1995) Assessing the risks of wind
pollination from fields of genetically modified Brassica napus
ssp oleifera. Euphytica 85, 417-423.
4. Timmons, AM, Charters YM, Crawford, JW,
Burn, D, Scott, SE , Dubbels, SJ, Wilson, NJ, Robertson, A, O'Brien,
ET, Squire, G and Wilkinson MJ, (1996) Risks from transgenic crops.
Nature 380, 487.
5. Thompson, CE, Squire, G, Mackay, GR,
Bradshaw, Crawford, J, and Ramsay G (1999) Regional patterns of
gene flow and its consequence for GM oilseed rape. 1999 BCPC Symposium
Proceedings No 72. Gene Flow and Agriculture: Relevance for Transgenic
Crops pp 95-100.
6. Ramsay, G, Thompson, CE, Neilson, SJ
and Mackay, G (1999) Honeybees as vectors of GM oilseed rape pollen.
1999 BCPC Symposium Proceedings No 72. Gene Flow and Agriculture:
Relevance for Transgenic Crops, pp 209-214.
7. Squire, G, Crawford, JW, Ramsay, G, Thompson,
CE and Bown, J (1999) Gene flow at the landscape level. 1999 BCPC
Symposium Proceedings No 72. Gene Flow and Agriculture: Relevance
for Transgenic Crops. pp 57-64.
8. Charters YM, Robertson, A, Wilkinson,
MJ and Ramsay G 1996. PCR analysis of oilseed rape cultivars (Brassica
napus L, ssp oleifera) using 5 ft. anchored simple
sequence repeat (SSR) primers. Theoretical and Applied Genetics
92, 442-447.
9. Squire, GR, Augustin, N, Bown, J, Crawford,
JC, Dunlop, G, Graham, J, Hillman, JR, Marshall, B, Marshall,
D, Ramsay, G, Robinson, DJ, Russell, J, Thompson, C, and Wright,
G (1999) Gene flow in the environment: genetic pollution? SCRI
Annual Report 1998-99 pp 45-54.
11 July 2000
Annex 1
RELEVANT PROJECTS
UNDERTAKEN AT
SCRI
Industry/AFRC/DTI PROSAMO initiative (participant),
1989-91.
SERAD FF340 Computation of safe isolation distances
for field-grown genetically modified crops. 1991-95.
DETR PECD 7/8/237 Investigations of feral oilseed
rape. 1993-96.
MAFF RG0208 (Desk study) Risk assessment of
the release of genetically modified plants: a review 1994.
SERAD Modelling impact of herbicide resistant
transgenic oilseed rape. 1995-96.
MAFF CSA4202 An experimental and mathematical
study of the local and regional scale movement of an oilseed rape
transgene. 1997-2000.
MAFF CTB9802 (subcontractor to CSL) Consequences
for agriculture of the introduction of genetically modified crops.
1999-2001.
DETR Farm Scale Evaluations of GM Crops. (participant)
1999-2002.
|
