[To contribute to this conference, send your message to biotech-room3@mailserv.fao.org. For further information on the Electronic Forum on Biotechnology in Food and Agriculture see Forum website.
NB - participants are assumed to be speaking on their own behalf, unless they state otherwise.]

-----Original Message-----
From: Biotech-Mod3
Sent: 21 June 2002 09:12 To: 'biotech-room3@mailserv.fao.org'
Subject: 64: GM versus conventionally bred plants

I agree with the Moderator's latest comments (message 63, June 20) that it may seem off the main point of the debate, but if Professor Cummins' view of the difference between transgenic plants or those produced by traditional breeding (message 56, June 18) is accepted as correct, then it alters the nature of the debate about gene flow, since the hazard from GM plants then differs fundamentally from that from conventional crops, in both developed and developing countries.

I would therefore like to briefly respond to Professor Cummins' reply (message 56, June 18) to my earlier question (message 17, June 6). In framing my question -- which was a response to Suzanne Wuerthele (message 1, 31 May) who said:"Because GMOs are fundamentally different from conventionally-bred organisms, they raise novel concerns about their effects on ecosystems at the genetic level and about their behaviour in ecosystems at the agricultural level." -- I asked "What is the evidence for this broad statement?"

I am puzzled by the responses which assume that the nature of the recombinant product is so different that the GM plant cannot be compared with those derived by 'conventional' plant breeding, (which itself may well involve such unnatural procedures as chemical and X ray mutagenesis). The products of the initial transformation first go through a searching and prolonged selection process which starts in the growth chamber, proceeds to the greenhouse and finally to field plots. The great majority of the initial transformants are discarded en route. But before they can be considered for commercial use they then, after the completion of this screen, have to go though the same approval process as other potential cultivars which have been derived from 'conventional' plant breeding, and in the United Kingdom that means satisfying the DUS criteria (Distinctiveness, Uniformity and Stability) before they can be approved for Plant Varieties Rights regulations. This path has been followed by all transformed plants which came for regulatory approval.

So there is a double selection procedure and yet Professor Cummins maintains that the product is so fundamentally different that it cannot be compared. Is there any scientific evidence for this? Or has it come from a belief that insertion of a transgene is such an 'unnatural' process that the product must be different in kind, not even in degree. Are we not then sliding into a pseudo-philosophical use of the word 'natural' here? So that, in such a view, 'natural' is intrinsically a good and 'contamination' an evil.

My general point is made in a much more scholarly way in the review by Dale, Clarke and Fontes (Nature Biotechnology, June 2002, Vol 20, pp 567-574).

Professor Derek Burke
13,Pretoria Road
Cambridge CB4 1HD
United Kingdom
Tel/Fax 01223 301159
e-mail: dcb27 (at) cam.ac.uk

-----Original Message-----
From: Biotech-Mod3
Sent: 21 June 2002 14:11
To: 'biotech-room3@mailserv.fao.org'
Subject: 65: Re: GM versus conventionally bred plants

Prof. Joe Cummins, Prof. Emeritus, University of Western Ontario, Canada.

Prof. Derek Burke (message 64, June 21) says "Is there any scientific evidence for this? Or has it come from a belief that insertion of a transgene is such an 'unnatural' process that the product must be different in kind, not even in degree. Are we not then sliding into a pseudo-philosophical use of the word 'natural' here? So that, in such a view, 'natural' is intrinsically a good and 'contamination' an evil."

In reply, the discussion was about the fact that conventional plant breeding depends on homologous recombination, the normal process by which plant sexual generate recombinant forms, while plant nuclear transgenes are always inserted using illegitimate recombination, a process by which genes are inserted into chromosomes at DNA double strand breaks that are most frequently essentially random.

Professor Burke promotes the view that homologous and illegitimate recombination are equivalent, and the products of the two types of recombination are equivalent. That opinion is simply wrong and one that is so elementary that it will cause even beginning students of genetics to titter. Most seriously, Burke's views seem to dominate policy in plant genetic engineering and tend to discourage development of a system using meiotic recombination to produce high quality transgenic crops.

Professor Joe Cummins,
University of Western Ontario.
Canada
jcummins (at) uwo.ca

[1. Further messages on this subject will only be posted if dealing with the impact on, or relevance for, gene flow from GM to non-GM populations.
2. Definitions of homologous and illegitimate recombination were provided in message 59...Moderator]

-----Original Message-----
From: Biotech-Mod3
Sent: 21 June 2002 15:21
To: 'biotech-room3@mailserv.fao.org'
Subject: 66: Legitimacy of a GM cultivar

This is David Heaf, UK co-ordinator of Ifgene -- International Forum for Genetic Engineering

Professor Joe Cummins (message 56, June 18) portrays GM plants as genetic 'bastards'. His post is part of a thread, also addressed by others, which essentially concerns the naturalness of genetic engineering and any resulting gene flow. The philosophical core of this thread was portrayed nearly 400 years ago in Shakespeare's 'A Winter's Tale' (act IV Scene 4):[glossary appended]

"Perdita: .. the fairest flowers o' the season
Are our carnations, and streaked gillyflowers
Which some call nature's bastards: of that kind
Our rustic garden's barren; and I care not
To get slips of them
Polixenes: Wherefore gentle maiden, Do you neglect them?
Perdita: For I have heard it said
That is an art which in their piedness shares
With great creating nature.
Polixenes: Say there is;
Yet nature is made better by no means
But nature makes that mean: so over that art
Which you say adds to nature, is an art
That nature makes. You see, sweet maid, we marry
A gentler scion to the wildest stock
And make conceive a bark of baser kind
By bud of nobler race. This is an art
Which does mend nature - change it, rather - but
The art itself is nature
Perdita: So it is
Polixenes: Then make your garden rich in gillyflowers
And do not call them bastards
Perdita: I'll not put
The dibble in earth to set one slip of 'em!"

Although Professor Derek Burke's post (message 64, June 21) comes close to it, what the discussion seems to overlook so far is that the human being is part of evolution and what he does has natural consequences for evolution. The alternative is a somewhat unscientific form of dualism. Generalising Polixenes' sense then, we could say that any gene flow arising from GM cultivars in developing countries would be natural. This would mean that an appeal to the legitimacy (bastard status) of such cultivars is not valid and we would therefore need to find quite other moral intuitions to take as our motives in determining the future of agriculture in the developing countries. This issue of 'naturalness' will be addressed at a public workshop later this year on 'Genetic Engineering and the Intrinsic Value and Integrity of Animals and Plants' (www.anth.org/ifgene/2002.htm).

Glossary: gillyflower (gillivor) - wallflower
slip - cutting of plant
piedness - part one colour, part another
scion - cutting of plant
dibble - dibber, pointed wooden stick to make hole in ground for young plant

-----Original Message-----
From: Biotech-Mod3
Sent: 21 June 2002 17:20
To: 'biotech-room3@mailserv.fao.org'
Subject: 67: Risks of genetic engineering

I am Roberto Verzola, former member of the National Biosafety Committee of the Philippines.

I find it hard to accept the claims that the risks of genetic engineering are not very different from the risks of conventional breeding for the following reason:

In conventional breeding, I would expect that the frequency of damaging mutation is much lower than 1 in a thousand, hence more than 99.9% of the results will be viable. The scientists here can probably supply the more accurate figure.

I've asked scientists the frequency of damaging mutations in a genetic engineering (GE) transformation and my impression is that more than 99% of transformed cells are considered damaged or unviable. Less than 1% actually survive "mega-mutation" as somebody here put it, and then, as the transformants grow into adults, more are even eliminated because of less obvious, and more subtle, problems that later show up. Again, those who actually do transformations can probably supply the more accurate figures. Have these genetically damaged results of transformation been analyzed statistically, as a whole? Apparently not. They represent the *real* risks of genetic engineering.

If conventional breeding results in a higher than 99.9% viability rate, while genetic engineering has a less than 1% viability rate, that's a world of difference in terms of risk, isn't it? That genetic engineers then have to resort again to conventional breeding to eliminate the genetically-damaged and reduce the risks created by GE clearly shows that conventional breeding reduces the risk, while GE increases it.

My own field is software. Making changes in the genome is very much like making changes in a complex software system. Such changes would obviously be tested if they accomplished the intended effect. However, they also introduce unintended side-effects, what we call "bugs". If, after testing, we discover a bug, then the possibility rises that a few more hidden bugs exist, which may surface only on rare occasions or under certain conditions. Even simple bugs can result in catastrophes. There's enough literature about this. Since the unintended side-effects of soya and corn transformation are now showing up in commercial products (higher lignin, allergenicities, etc.), the probability has also gone up that more hidden side-effects exist.

The difference between complex software systems and genomes is that we already have, theoretically, 100% knowledge of the entire software system, because it is, after all, human designed and constructed. Very often, the documentation is available, where the intentions of the designers are explicitly stated. Still, software changes invariably introduce bugs. How much more in a genome where we don't really know the design or the intentions of the designer, have no documentation and do not even know how the whole thing works in detail?

Roberto Verzola
Philippines
rverzola (at) gn.apc.org