Summary: Horizontal gene transfer from GM-crop cells to bacteria and fungi in the soil may not be an impossibility. If it does occur, then the effects will be difficult to predict and could, in certain situations, result in changes to the soil ecology, or even broader changes at the ecosystem level.

There are many ways in which horizontal gene transfer could theoretically occur from a GM-crop. (Transfer by pollen to a non-GM plant, or incorporation into the genome of an infective virus or bacterium spring to mind). Whilst researchers may toy with the possibility of controlling, or at least restricting, the risk of horizontal gene transfer above ground---(for example, transgene containment in chloroplasts to reduce the risk of pollen transfer of transgenes in species where the plastids are maternally inherited)---there seems to be NO possibility of 'control' when it comes to the rhizosphere.

Root cells, particularly root cap cells, are constantly being sloughed off and broken down in the surrounding soil. The DNA present in these cells could become 'available' for uptake by soil microorganisms. Indeed, the uptake of plant DNA from the soil by microorganisms may be a widespread natural phenomenon, but one in which the transferred nucleic acids are generally inactivated in one way or another, and then broken down to provide material for the growth of the microorganism. (1),(2).

But what happens to the hybrid genetic constructs released by cells from GM-crops? These are 'designed' to facilitate horizontal gene transfer, and to overcome the natural barriers to the incorporation of exotic genes into a host genome. Some of these genetic elements may not be destroyed, or even silenced. They may become incorporated into, and expressed by, the new 'host'.

The chance of this happening may be very low, nevertheless there is evidence from laboratory experiments that such transfers CAN occur, (3),(4),(5),(6), and some indications (although not proof) that they MAY occur in nature, (7),(8),(9). Clearly, much more research needs to be done before sound conclusions can be drawn. Consequently, open-minded scientists should for the moment remain unconvinced that horizontal gene transfer does not occur from GM-crop plants to soil microorganisms.

For myself, I believe that such gene transfers to bacteria and fungi in the soil could prove to be a most important consequence of the long-term cultivation of GM-crops. There is presently no possibility of control of this, and assurances from researchers that such transfers are unlikely or insignificant may prove to be unfounded.

'Problems' will not be noticed immediately, or even in a year or two. Indeed, it may be many years, or tens of years, before there are noticeable changes in the soil microflora. But changes may well occur,--and these will be both genetic in origin and evolutionary in nature. In short, the genetic makeup of the soil bacteria and fungi could well change in the longer term as a result of the cultivation of GM-crops. The nature of these changes, and their consequences, cannot be predicted,----and it is this 'unpredictability' which should cause concern. I can envisage many such potential 'problems', three of which I present below.

1. The accelerated spread of antibiotic resistance genes in bacterial populations, with the inevitable consequences for human (and animal) medical treatments.
2. Changes in the relationships between root nodule bacteria and their host plants. Any genetic variants of these bacteria (caused by horizontal gene transfer from GM-crops) which result in a less symbiotic, and more parasitic, association with the host plant could lead to reduced levels of nitrogen-fixation. For crop plants this could be a real problem,--with lower yields and possibly lower soil fertility. For wild species it could cause changes in local vegetation as the competitive balance between plant species is altered. And of course changes in flora will lead to changes in fauna. In other words, the longer term potential effect of such genetic changes in root nodule bacteria could be so fundamental that ecosystems themselves could change at every level (from species content and diversity to food chains and food webs).
3. Changes in the relationships between mycorrhizal fungi and their host plants. In recent years the importance of mycorrhizae in the growth of plants has only just been realised. It now seems that the majority of higher plants, including some crop species, form associations with these soil fungi. Any horizontal gene transfer to mycorrhizal fungi from GM-crops which changed the characteristics of their interactions with higher plants could potentially have far-reaching effects on ecosystem structure and dynamics (in much the same way as I have described under 'problem' 2).

Now, I am NOT saying that the cultivation of GM-crops WILL cause problems of this nature, but I am saying that I see cause for concern. I am far from being convinced that horizontal gene transfer will NEVER occur in the rhizosphere (from GM-crop cells to bacteria and fungi). In fact if the probability of such gene transfer is non-zero then, in the longer term, we can probably expect to see genetic changes in the soil microflora (which otherwise would not have occurred), and even changes in vegetation patterns and crop yields.

To conclude. the key factor here is 'unpredictability'. We cannot predict that horizontal gene transfer will never occur in the soil. If it does occur, we cannot predict what the consequences will be. The only thing we can predict is that we are unsure of the long-term effects on the biosphere of the cultivation of GM-crops.

References.
(1). Gene Technology and Gene Ecology of Infectious Diseases. Ho, et al. (1998). Microbial Ecology in Health and Disease 10. 33-59. (and references therein).
(2). Horizontal transfer from transgenic plants to terrestrial bacteria--a rare event? Nielsen, K.M., Bones, A.M., Smalla, K. and van Elsas, J.D. (1998). FEMS Microbiology Reviews 22. 79-103.
(3). Foreign DNA sequences are received by a wild-type strain of Aspergillus niger after co-culture with transgenic higher plants. Hoffman, T., Golz, C. and Schieder, O. (1994). Current Genetics 27. 70-76.
(4). Horizontal gene transfer from a transgenic potato line to a bacterial pathogen (Erwinia chrysanthem) occurs, if at all, at an extremely low frequency. Schluter, K., Futterer, J. and Potrykus, I. (1995). Bio/Technology 13. 1094-1098.
(5). Transformation of Acinetobacter sp. strain BD413 by transgenic sugar beet DNA. Gebhard, F. and Smalla, K. (1998). Appl. Environ. Microbiol. 64. 1550-1554.
(6). Detection of nptII (kanamycin resistance) genes in genomes of transgenic plants by marker-rescue transformation. De Vries, J. and Wackernagel, W. (1998). Mol. Gen. Genet. 257. 606-613.
(7). Bacterial gene transfer by natural genetic transformation in the environment. Lorenz, M.G. and Wackernagel, W. (1994). Microbiol. Rev. 58. 563-602.
(8). Development of engineered genomic DNA to monitor the natural transformation of Pseudomonas stutzeri in soil-like microcosms. Paget, E. and Simonet, P. (1997). Can. J. Microbiol. 43. 78-84.
(9). Monitoring field releases of genetically modified sugar beets for persistence of transgenic plant DNA and horizontal gene transfer. Gebhard, F. and Smalla, K. (1999). FEMS Microbiology Ecology 28. 261-272.

This is from Brian Stratton, Nottingham. Although I have degrees in Plant Science, I have not worked in Biology for 20 years. I must therefore be regarded as a 'non-scientist'. I teach Mathematics.