CABI - Biocontrol News and Information 19(2) June 1998 News:Biorational

June 1998, Volume 19 No. 2

Biorational

Integrated pest management (IPM) involves the use of many techniques, including biological control, to provide effective control of crop pests with minimum harmful side-effects. Those techniques which are compatible with the use of biological control or have little impact on natural enemies have been described as -biorational·.

Bt: What Future?

Does biotechnology herald another revolution in biological pest management? Or does it presage another venture in unsustainable technology like the -pesticide treadmills· of the 1970s and 1980s? The debate surrounding Bacillus thuringiensis (Bt) encapsulates many of the arguments.

The multinational agrochemical industry seems to have adopted a collective view that transgenic crops are the best way to deliver Bt toxins to pests. It has invested massively in the new technology, concentrating initially (because of the high development costs) on large-area/high-value crops. But although the -green· credentials of Bt crops have been trumpeted (better yields but less pesticide), the organic and IPM lobbies remain unconvinced. They fear that, unless properly managed, Bt crops may undermine rather than contribute to IPM and sustainable agriculture. To begin with, they argue, a fundamental tenet of pest management seems to have been ignored: too much reliance on one tool is a bad thing. The more exposure an insect has to Bt, the greater the selection pressure on it to develop resistance. Worryingly, the industry appears to accept that pests will eventually become resistant to Bt crops. Some are worried that the industry views Bt as a disposable technology, an expendable resource1. This is unwelcome news for those who have relied on Bt for decades.

Transgenic Crops

A joint World Bank/CGIAR (Consultative Group on International Agricultural Research) study recently concluded that agricultural biotechnology is an increasingly important tool for sustainable agriculture and may prove essential in the fight for global food security - and this in a world where some 85 million people (about 1 in 7 of the world·s population) do not have sufficient to eat2. The study also argues that transgenic crops are not in principle more ecologically harmful than conventionally bred crops. The many protagonists of transgenic crops point to the significant economic and environmental benefits of decreased pesticide use. For example, 0.7 million hectares (13%) of the US crop area was planted with Bt cotton in 1996, leading to an estimated reduction in insecticide use of a quarter of a million gallons (more than a million litres)2. An International Rice Research Institute (IRRI) report says that half of the insecticides used on rice in Asia are applied against caterpillars. It suggests that the introduction of Bt rice in Asia could help to reduce the estimated 25 million tonnes of rice lost annually through stemborer attack3. But sceptics argue that we may be stepping off one treadmill onto another.

Transgenic crops expressing Bt toxins have been widely adopted and accepted in the USA. More than a million hectares of Bt crops were planted there in 1996, and there was a five-fold increase in the area planted with transgenic varieties as a whole between 1996 and 1997 2. Yet elsewhere, for example in Europe, there is both government and public scepticism. There is also a strong international lobby opposed to their introduction. Last year Greenpeace International, in conjunction with other groups including IFOAM (the International Federation of Organic Agricultural Movements), petitioned the US government to halt the registration and use of genetically engineered plants expressing Bt toxins. Some of the agrochemical companies are now acknowledging that they misjudged the public mood, in Europe at least.

Yet while governments, pressure groups and researchers argue, the agrochemical industry is realigning itself. Companies such as Monsanto, DuPont and Dow AgroSciences have acquired or entered into agreements with seed companies. This move is seen as consolidating their positions in the transgenic crop market by integrating the crop protection and seed sectors. Acquisition of biotechnology expertise and technology is also widespread. This is occurring not only in the USA, but also in Europe and, with an eye to the future, in other areas of the world - from Asia to South America - where the introduction of Bt cotton in the first instance may be a bridgehead for establishing other transgenic crops. According to the Rural Advancement Foundation International (RAFI), which monitors alliances and mergers in the seed industry, commercial plant breeding and seed sales worldwide are now dominated by big agrochemical companies. For the agrochemical industry, the future is apparently clear.

The introduction and expansion of Bt crops in the USA over the last three years gives an interesting insight into their impact on the traditional pesticide market. The case of Bt cotton is particularly striking. In recent years, cotton has accounted for more than a quarter of insecticides used worldwide, and about half of this has been used on lepidopteran pests that could potentially be controlled by Bt crops. In the USA in 1997, as the area planted with Bt cotton increased to a million hectares, representing 18% of the total crop area2, some pesticide companies saw a fall in profits and overall the agrochemical market remained flat. This was widely attributed to a dramatic decrease in cotton pesticide sales following the expansion of Bollgard Bt cotton. In Australia, 12% of the total expenditure on pesticides is accounted for by the cotton sector, and 80% of that represents insecticides for Helicoverpa pests. Insecticide applications on Ingard Bt cotton in its first season were halved compared with conventional varieties1. The impact on the pesticide market is not confined to conventional pesticides: for example, Ecogen·s 1997 biopesticide sales remained static and pheromone sales declined sharply, and this was also attributed to the impact of Bt cotton.

However, the actions of the agrochemical industry need also to be seen in the context of other changes ahead. There is pressure to reduce pesticide use around the world. In the extreme, members of the Danish parliament have called for the country to be made entirely organic by 2010, and the government have initiated an assessment of the impact of a total pesticide ban. Elsewhere governments are being selective but nonetheless looking for reductions: for example, Australia wants to restrict endosulfan use, and China wants to phase out monocrotophos on cotton. In the USA, the Environmental Protection Agency (EPA) says that the re-assessment of tolerances required by the 1996 Food Quality Protection Act may lead to the banning of a large number of organophosphate and carbamate insecticides, and have suggested that if planned cumulative risk assessments were to go ahead, there is a possibility that all organophosphates may be deemed unacceptable.

There has therefore been a major refocusing, and in some cases down-sizing, of pesticide and biopesticide production. For example, Novartis are cutting their insecticide range and have sold their Bt business to Thermo Trilogy; and Mycogen recently closed a Bt manufacturing plant in Wyoming, USA. On the other hand, 95% of commercial funding for biopesticide research is being spent on transgenic crops1.

Bt has been engineered into some 40 crops, notably major food and cash crops, although only a small number are yet commercially available. In the USA, increasing numbers of transgenic varieties are being planted on increasing areas; Bt maize, cotton and potatoes are available (and tomatoes are being trialled). Enough seed has been produced to plant in excess of ten million hectares with Bt crops this year.

Elsewhere, areas of Bt crops are increasing as the worldwide expansion gathers momentum. Bt cotton will be planted on more than 70,000 ha in Australia and Mexico in 1998. Bt cotton seed is expected to be available commercially in China (the world·s largest cotton market) for the first time this spring, sufficient for some 200,000 ha. If approval is given in time, it will also be planted in Argentina and South Africa later this year, and it is expected to be available in Brazil within the next two years. Trials of Bt cotton are underway in India. However Thailand halted similar trials, begun three years before, on human health and environmental grounds. There, cotton is not just used for cloth: 16 members of the Malvaceae are used in the production of traditional medicines, and there were fears of adverse effects in these. It was argued that tests in Thailand should take account of local biodiversity and social conditions4.

Although the USA has far and away the largest area of Bt maize, Canada is likely to have some 300,000 ha in 1998, an 80-fold increase on last year, and the European Commission and the national government in France have finally cleared the way for planting Bt maize in Europe. The first 20,000 ha is expected to be planted in France this spring; both Italy and Spain may follow suit.

Significant advances are also being made in the development of Bt rice in Asia. IRRI has created varieties highly resistant to stemborers (the tropical Scirpophaga incertulas and temperate Chilo suppressalis species). IRRI researchers suggest that, although rice stemborers do not constrain yields in all rice-growing areas, the introduction of Bt rice could still have a massive impact. Stemborers are estimated to reduce average yields in Asia by some 5%, and this represents 25 million tonnes of rice - enough to feed 120 million people3.

Biopesticides

Microbial Bt in foliar pesticides has been used for decades in a wide variety of crops around the world. It is a crucial component of many IPM programmes, where it serves as an emergency control measure. First identified as toxic to Lepidoptera in the early 1900s, it has been registered with the US Environmental Protection Agency (EPA) since 1961. Sales of Bt in the USA currently amount to some US$60 million annually. For some time now, evidence has been mounting that resistance to Bt is developing in the field as a consequence of its over-use. It has been reported widely in diamondback moth (Plutella xylostella) in China, Malaysia, Japan and the USA. It is argued that the development of resistance to these sprays is the result of an increased reliance by farmers on Bt, as the diamondback moth has developed resistance to more and more synthetic pesticides5. Bt resistance has also been identified in tobacco budworm (Heliothis virescens) in the USA: a study of individuals collected from four cotton-growing states, in pre-transgenic 1993, found that one in 350 individuals carried an allele for Bt resistance, considerably higher than estimates had assumed and indicative of a potential for the swift evolution of resistant populations6.

IPM practitioners in the field are attempting to manage resistance by stopping the use of Bt sprays where resistance has begun to appear, and thus removing the selection pressure. However, the built-in toxin in Bt crops makes such rapid response to selection pressure very hard: planting them to manage pests is analogous to using a very persistent pesticide.

There is more bad news for those who have argued that Bt produces a vast arsenal of different toxins that can be exploited. Laboratory studies have provided evidence of a simple mechanism of multiple resistance to different Bt toxins. This sheds serious doubt on the received wisdom that the development of resistance to Bt could be managed or even prevented by the judicious use of the different toxins, each of which was thought to need a separate resistance mechanism7.

Set against these revelations, the cornerstone of the transgenic lobbys defence against the development of resistance to Bt, the resistant management plan, has come in for much criticism.

The Resistance Management Plan

The argument runs that Bt sprays give a directed dose at a given time and the toxin is quickly inactivated by UV light. The selection pressure is therefore less than for Bt crops, which deliver a high dose of a single toxin throughout the season, which is protected from degrading radiation. Resistance is therefore more likely to develop, and more quickly, with Bt crops than with Bt sprays.

The main strategy so far proposed to limit the development of resistance in transgenic plants is the provision of refuges of conventional crops. Controversies abound as to the optimal dose of Bt, the size of such refuges, their proximity to the Bt crop, and the extent to which such unstructured refuges can slow down the development of resistance8. Existing resistance management plans rely on the -high dose/refuge· strategy. A high dose is the amount of Bt toxin that a plant must deliver to kill virtually all insects feeding on the plant, leaving only rare, highly resistant pests. Refuges are plantings of non-Bt crops to provide a habitat for Bt-susceptible insects that can mate with the rare, highly resistant ones, thereby diluting resistance. A recent report from the US Union of Concerned Scientists (UCS) says that the current required US Environmental Protection Agency (EPA) Bt resistance management plans for Bt cotton and maize are too weak, and that there is currently no requirement for such a plan for Bt potato (although Monsanto have developed a voluntary management plan, a move commended by the report).

The report argues that while the strategy underlying current resistance management plans is theoretically sound, it was adopted before major assumptions were confirmed, and new data have cast doubt on some of these. To begin with, the Bt dose may be neither as high nor as consistent as the theory assumed. There is evidence from the field to support these concerns. It was clear from cotton bollworm (Helicoverpa zea) damage on Bt cotton in USA in 1996 that Bt cotton did not always produce a high dose for this pest: 40% of growers, notably in Texas, had to spray Bt cotton against the bollworm. This species, while a minor pest of cotton in the USA, is a serious pest of maize, so the spectre of cross-resistance raised its head. Yet the fact that not all pests are equally susceptible to Bt crops was used, at the time, as an argument against the alleged failure. It was argued that Bollgard cotton was successful against its principal target, tobacco budworm. Farmers were only spraying outbreaks of the secondary pest, cotton bollworm, which in conventional crops is rarely present in sufficient numbers to cause damage. But it can equally be argued that the outbreaks were the result of the bollworm·s higher tolerance to Bt, together with inadequate expression of Bt, particularly late in the season9. Farmers in Australia also had to spray against cotton bollworm, and although the resistance management plan there was designed to allow for this eventuality, there are now fears that the measures taken, including increasing the size of the refuges, may not have been sufficient1. Other studies have suggested that some, maybe most, Bt-maize cultivars do not deliver a season-long high dose for the European cornborer (Ostrinia nubilalis). This argument has been taken up by concerned groups in Europe who suggest that Bt maize may be ineffective against the late-season second generation of borers that occurs in southern Europe. Studies on tobacco budworm in the USA and Australia have concluded that the use of transgenic crops with the present management system could lead to the development of resistance to Bt in as little as three years, or at most ten. Resistance is likely to develop more quickly in less susceptible pests such as the cotton bollworm and the European cornborer6,9.

There are also doubts about synchronization of Bt and refuge crops and their insect populations (insects grow more slowly on Bt crops), about distances moths travel before mating, and just how readily insects will move between crops and refuges. It may be almost impossible to ensure that resistant insects mate with susceptible ones. Even integrating the crops and refuges more closely may not help: partially resistant pests may move from the crop into the refuge and survive, thereby adding to the pool of resistant pests1.

Current EPA plans were evaluated in the UCS report. It found that even in crops with mandatory resistance management plans, the required refuges were not large enough to implement successfully the refuge/

high-dose strategy. The report calls for non-Bt refuges that comprise 20-50% of crop area, depending on the crop and pest (compared, for example, to the current requirement of 4% for unsprayed cotton), with distances separating refuges from Bt-crops specified for each crop. In Australia there are suggestions that Bt cotton should be re-engineered to increase the toxin dose, and calls for a massive effort to establish the large refuges required1,9. At IRRI, research on resistance management for Bt rice is underway3.

Other Concerns

More dogma has also come under fire following studies on non-target species in Bt crops. The belief that Bt crops have no adverse effects on these species may be optimistic at best, although most researchers are careful to point out that any impact of the transgenic toxins has to be judged in relation to the impact of pesticides used in conventional crops.

The biopesticides and the transgenic crops contain different toxin entities requiring different activation steps, and there are fears that the smaller, more-easily activated entities expressed by Bt crops may be less selective. One of the toxins used in transgenic maize has been shown to have an adverse impact on soil-dwelling organisms such as springtails (Collembola) when mixed with the soil. An added concern here is that decomposing plants at the end of the season could contaminate the soil with transgenic Bt toxins, which would not only harm beneficial species but also increase the selective pressure on target species over-wintering in the soil10. New evidence suggests that Bt toxins bound to soil particles may retain activity for weeks or even months.

There is also evidence that Bt plants may have a serious and direct negative impact on beneficial species feeding on prey that has fed on Bt varieties. Researchers at the Swiss Federal Research Station in Zurich have shown in laboratory experiments that predatory lacewing (Chrysoperla carnea) larvae were adversely affected when reared on prey that had fed on Bt maize. Total immature mortality was 66% when reared on Bt maize-fed European cornborer larvae (the target pest). Even more alarmingly, similar mortality was observed for lacewing larvae fed on live African cottonworms (Spodoptera littoralis) that had survived exposure to Bt maize11. Some circumstantial evidence of adverse effects on pollinators comes from Bt cotton trials in Thailand in which some 30% of bees were found to have died, although the cause of this was not investigated4.

There are also concerns that crops such as Bt cotton could trigger secondary pest outbreaks, as was observed with cotton bollworm in the USA and Australia in 1996/7 (see above). The secondary outbreaks may need to be controlled by conventional insecticides, which could harm natural enemies. Just such a threat has been identified for mirids in Bt cotton. In Australian trials unsprayed Bt cotton suffered heavy mirid infestations12, and studies in the USA indicated an enhanced performance of Lygus lineolaris in Bt cotton13. The only insecticides available for mirids on cotton are broad-spectrum. It appears that all lepidopterous herbivores in the Philippines may be susceptible to the toxin expressed in Bt rice being developed by IRRI. This could conceivably mean that Bt rice would be essentially a Lepidoptera-free crop, leaving a large resource to be exploited by other herbivores - for example, planthoppers. Although currently constrained by not being able to plant Bt rice in the field, IRRI already have experiments underway to assess the impact of the transgenic crop on insect communities3.

One last but very substantial worry concerns the possibility of gene flow from the transgenic cultivars into non-transgenic traditional varieties and indigenous wild relatives. Recent evidence suggests that the chances of this happening have been underestimated. Recombination of viral transgenic traits and hybrid inheritance of transgenic traits have both now been demonstrated to occur with unexpected ease in plants. This is of particular concern in the areas of origin of the currently available commercial Bt crops (Mesoamerica for maize and cotton, and South America for potato) where the indigenous floras contain wild relatives that would be at particular risk of contamination. Greatest concern is being expressed about biodiversity contamination. Maize, for example, probably originated from teosinte, which is still cultivated in Mesoamerica. Countries such as Mexico see it as a natural heritage and justifiably want to preserve its integrity. However, there are also fears that rare and therefore protected insects (e.g. butterflies) living on wild relatives of Bt crops could become locally extinct. It could be disastrous if transgenic crops were to be cultivated close to conservation areas with high biodiversity value. There is a related (but so far unresearched and therefore theoretical) concern that gene transfer from Bt crops could lead to a decreased herbivore load on related wild species, and thus potentially turn them into weedy species.

So while a widening stream of evidence is accumulating to raise concern over the safety and sustainability of transgenic crops in general, and Bt crops in particular, the juggernaut of the international agrochemical industry rolls on, fuelled at least in part by the need to recoup its massive investment in transgenic crops. It remains to be seen whether the expansion already underway in the USA is reflected on a world scale, or whether scepticism and opposition elsewhere put the brakes on. Many remember that the pest management technology in the last green revolution foundered on the rock of pest resistance - but corporate memories are short. While biotechnology offers exciting new methods for pest management, and there is great excitement that it may be able to make a real contribution to IPM, there are also fears that Bt crops are being oversold by an agrochemical industry needing to recoup its investment. Whatever the financial imperatives, it is crucial that careful ecological assessments of the impact of transgenic crops are made, and that steps are taken to ensure that the invaluable resource of Bt is preserved.

1 Salleh, A. (in press) Wearing out our genes? The case of transgenic cotton. In: Hindmarsh, R.; Lawrence, G.; Norton, J. (eds) Altered genes: reconstructing nature. Sydney, Australia; Allen & Unwin.

2 James, C. (1997) International Service of the Acquisition of Agri-biotech Applications (ISAAA) Briefs No. 5. New York, USA; ISAAA, 31 pp.

3 IRRI (1997) Bt rice: research and policy issues. IRRI Information Series No. 5, 20 pp.

4 Anon (1997) Thai government forgoes pest-free cotton for now. Pestcide Monitor 6(4), 19.

5 Verkerk, R.H.J. Wright, D.J. (1997) Field-based studies with the diamondback moth tritrophic system in Cameron Highlands of Malaysia: implications for pest management. International Journal of Pest Management 43, 27-33.

6 Gould, F. et al. (1997) Initial frequency of alleles for resistance to Bacillus thuringiensis toxins in field populations of Heliothis virescens. Proceedings of the National Academy of Sciences, USA, 94 3519-3523.

7 Tabashnik, B.E. et al. (1997) One gene in diamondback moth confers resistance to four Bacillus thuringiensis toxins. Proceedings of the National Academy of Science, USA 94, 1640-44.

8 Alstad, D.N.; Andow, D.A. (1996) Implementing management of insect resistance to transgenic crops. AgBiotech News and Information 8, 177-181.

9 Roush, R.T.; Shelton, A.M. (1997) Assessing the odds: the emergence of resistance to Bt transgenic plants. Nature-Biotechnology 15, 816-817.

10 Greenpeace et al. (1997) Petition for rulemaking and collateral relief concerning the registration and use of genetically engineered plants expressing Bacillus thuringiensis endotoxins.

11 Hilbeck, A. et al. (1998) Effects of transgenic Bacillus thuringiensis corn-fed prey on mortality and development time of immature Chrysoperla carnea (Neuroptera: Chrysopidae). Environmental Entomology 27(2), 480-487.

12 Fitt, G.P. et al. (1994) Field evaluation and potential ecological impact of transgenic cottons (Gossypium hirsutum) in Australia. Biocontrol Science and Technology 4, 535-548.

13 Hardee, D.D.; Bryan, W.W. (1997) Influence of Bacillus thuringiensis-transgenic and nectariless cotton on insect populations with emphasis on the tarnished plant bug (Heteroptera: Miridae). Journal of Economic Entomology 90, 663-668.