Cutting edge: Nick Birch

While insect-resistant GM crops save on pesticides, one test plant also hit ladybirds down the food chain. Case-by-case assessment is the solution.

As an entomologist and ecologist, I am acutely aware of the delicate balance between insects, plants and humans. Insects, which comprise at least 56 per cent of all species on earth and number at least a quintillion individuals at any one time, consume enough crops to feed a billion people a year, despite an annual $9 billion spend on pesticides.

This means scientists need to improve crop production and protection through new and established technologies in order to feed the rapidly growing human population. We have been breeding improved crops for 30,000 years, taking advantage of wild plants' incredible genetic diversity. We now largely rely on a relatively small number of high-yielding elite varieties, grown mainly as monocultures (ie in low biodiversity systems). Some crop varieties have "built-in" pest and disease resistance, but most are bolstered heavily by fertilisers, pesticides and herbicides.

The question is: "Will genetically modified crops help overcome production problems or merely add to existing problems?" Pests and diseases quickly adapt and fight back against our scientific efforts, overcoming control strategies, while some pesticides are being withdrawn on environmental and health grounds. New crop varieties typically take ten to 20 years to reach the marketplace, while new pest and disease problems can evolve in less than half that time. This co-evolutionary warfare both fascinates and concerns agricultural scientists.

In my own research, I have studied how pest aphids have overcome a succession of introduced "natural" anti-aphid resistance genes in raspberry over the past 30 years. We have now run out of aphid resistance genes readily available through conventional breeding. Biotechnology offers the exciting prospect of introducing resistance genes from new sources, but are they safe and will they provide sustainable crop protection?

One possible way ahead is to use anti-aphid genes from other plants. We recently tested experimental GM potatoes that expressed a gene for a lectin from snowdrops. While the gene reduced the aphid problem by about 50 per cent (without pesticides), it also adversely affected beneficial ladybirds feeding on these insects, by reducing their life-span and egg fertility. This, together with similar studies in Switzerland on a different anti-pest gene, was the first demonstration that "food chain" effects on predators of pests are potentially problematic. Further studies on the snowdrop lectin at the Scottish Crop Research Institute and in Edinburgh have also indicated biological interactions with human blood cells, which require further investigation before lectins can be considered as toxicologically safe in edible crop plants.

Each combination of transgene/crop/pest/predator/environment is probably unique and requires case-by-case assessment in a contained laboratory, glasshouse and then controlled field trials. We did not ban all pharmaceutical drugs because of the thalidomide disaster; we test them more rigorously and warn of possible side effects. This sounds logical, but when spin doctors on both sides of the GM debate get hold of the basic science, the polarised arguments produce confusing and often alarming misinformation.

The tiered approach, which includes measuring "non-target" long-term effects on beneficial insects and other organisms, is now widely adopted by regulatory bodies around the world, so our initially controversial research has made a significant impact on the risk-benefit assessment of GM crops. I hope that independent scientists are funded to do essential long-term studies and publish freely on these important issues, in a way that is accessible to both policy-makers and the public.

My research is focused on modelling and predicting the potential risk-benefit of different insect-resistant GM crops, if they were released in the United Kingdom. This draws on data and expertise from more than 20 years' studying non-GM pest-resistant crops and how pests counter-adapt to them.

Because of limitations on field-based GM crop research in the UK, field data from large-scale releases must still be imported from the United States, Australia and other countries. They have released and monitored insect-resistant crops on a large-scale, commercial basis for the past four years. Interesting trends in their ecological and economic impacts are now emerging, from which we can learn. However, we still need to carry out carefully controlled and monitored field trials of candidate GM crops under UK conditions, once screened and passed in the laboratory and glasshouse tests, before we can really assess the true risk-benefit of each new transgenic crop line for British agriculture.

Nick Birch is a senior research scientist at the Scottish Crop Research Institute, Invergowrie, Scotland.