I am contributing a short piece on the likelihood of gene flow from plants to soil bacteria. I am a full-time scientist with 20 years of postdoctoral research experience in soil microbiology and ecology, including bacterial gene transfer in soil and plant-microbe-soil interactions at Rothamsted Research, an Institute sponsored by the BBSRC. I am a member of ACRE and I chair its subgroup on Soil Ecology.

Penny Hirsch
Dr Penny R. Hirsch
Agriculture and the Environment Division
Rothamsted Research
Harpenden, Herts AL5 2JQ, UK

The likelihood of gene flow from plants to soil bacteria

Crops related to native plants and common agricultural weeds and can cross-hybridise with them (e.g. rape), others with no close relatives in Europe (e.g. maize; potatoes) cannot. Some plants produce large quantities of pollen, others are sterile or have very low out-crossing rates. These properties are well documented; the separation distances required to maintain "pure lines" were set empirically in the past by plant breeders. The advent of molecular genetics allows detection of low frequency hybridisation, not possible using conventional markers. GM plants with unique marker genes enable detection of very rare events, especially when combined with PCR detection. Using PCR requires stringent controls and exacting experimental conditions, otherwise the technique is highly prone to errors and misinterpretation.

Where gene transfer is extremely rare, it is unlikely that hybrids will ever be detected unless they have a survival advantage. In theory, resistance to a damaging pest or a frequently used herbicide could provide such selective pressure and must be considered in risk assessments. If a very low level of cross-hybridisation occurs but is never detected and hybrids are rapidly lost from the population, it is important only when perceived to be a problem for regulatory reasons. All plant-derived produce contains traces of other organisms (weeds, bacteria, fungi, insects) and unless highly processed, also contain their genetic material.

Gene transfer from plants to bacteria is extremely unlikely. It can be demonstrated in laboratory conditions if DNA introduced into plants is designed to be able to transfer to bacteria and be maintained and expressed in them ^{1, 2}. Many bacteria can take up DNA from any source but only incorporate, express and maintain it when it offers a selective advantage. Although theoretically possible, the probability of this occurring in the field is exceedingly low. Many of the genes used in GM plants at present (herbicide tolerance; Bt; antibiotic resistance markers) are derived from soil bacteria in the first place. The few soil bacteria genomes sequenced to date have not revealed homology to plant DNA although when bacterial and other genomes are compared there is some evidence for inter-kingdom gene transfer over evolutionary time. Problems in treating certain human infections due to the spread of antibiotic resistance genes is due mainly to inappropriate medical use of antibiotics in the past and it highly unlikely that GM plants could ever contribute to this problem ³.

Summary of key points

Genes can move from GM crops to nearby, sexually compatible plants; new techniques improve detection but it is predictable for risk assessments. Transfer to soil bacteria is very unlikely.

1 Gebhard F, Smalla K 1999 Monitoring field releases of genetically modified sugar beets for persistence of transgenic plant DNA and horizontal gene transfer. FEMS Microbiology Ecology 28: 261-272

2 Kay E, Vogel TM, Bertolla F, Nalin R, Simonet P 2002 In situ transfer of antibiotic resistance genes from transgenic (transplastomic) tobacco plants to bacteria Applied & Environmental Microbiology 68: 3345-3351.

3 Seveno NA, Kallifidas D, Smalla K, van Elsas JD, Collard JM, Karagouni AD, Wellington EMH 2002 Occurrence and reservoirs of antibiotic resistance genes in the environment. Reviews in Medical Microbiology 13: 15-27.