Vol.12_No.2 - Pesticide Alternatives Lab - Michigan State University

Spring 2003 Resistant Pest Management Newsletter Vol. 12, No.2enforcing a similar system for small holding farmerswill not be possible in most parts of Asia. In a typicalvillage in Asia, it is unlikely that all farmers will plantBt crops on all their land, and farmers grow severaldiverse crops, which serve as hosts for H. armigera. Insuch a scenario, it may not be necessary to enforce thecultivation of refuge crops (Sharma and Ortiz 2001). Btgenes will be one of many factors that the farmers willconsider when choosing which varieties to grow. Thegovernments can promote the maintenance of refugesby restricting the number and diversity of Bt cultivarsthat can be released. For example, in the Indian state ofPunjab, rice farmers grow traditional Basmati varietiesand modern semi-dwarf varieties. Stem borer damageis higher in basmati varieties, and thus the governmentcould authorize the release of Bt-transformed basmativarieties, but not Bt-transformed semi-dwarf varieties(Cohen 2000).Although Bt cotton that produces Cry1Ac toxinhas been effective against pink bollworm (Patin et al.1999, Tabashnik et al. 2000b), the slower developmentof resistant larvae on Bt cotton as compared tosusceptible larvae on non-Bt cotton could reduce theprobability of mating between susceptible and resistantinsects, and this asynchrony could reduce the expectedbenefits of the refuge strategy (Liu et al. 1999, Liu etal. 2001, Storer et al. 2001). Though there was slowlarval growth, the corn borer larvae were successful incompleting development on transgenic corn plants,causing similar amounts of damage as on non-Bt plants(Storer et al. 2001). Each insect/Bt crop system mayhave unique management requirements because of thebiology of the insect, but the studies have validated theneed for a refuge (Shelton et al. 2000). Therefore, caremust be taken to ensure that refuges, particularly thosesprayed with insecticides, produce adequate numbersof susceptible insects. Models and experimental datashowed that separate but adjacent refuges might besuperior to other strategies for insects that can movebetween plants in their larval stage (Shelton et al.2002).A concern is often raised that insect damage innon-Bt fields will increase after introducing Bt crops.The implication is that farmers will be even less likelyto grow non-Bt crops because of the increased damage,and therefore there will be even fewer refuge fields.However, Cohen (2000) suggested that withdiamondback moth on Bt collards and the Europeancorn borer on Bt maize, many of the moths that emergefrom fields of non-Bt crops would disperse and laytheir eggs in Bt fields. In contrast, very few moths willemerge from Bt fields and move from Bt fields to non-Bt fields. As a consequence, insect damage in non-Btfields may decrease if most fields are planted with Btcrops.There is also a debate regarding the spatial designof the refuge system (separate/seed-mixture) to beadapted. Roush (1997a) pointed out that seed mixescan actually promote resistance development forinsects that move from plant to plant. There is noevidence to show that moths can detect whether or nota plant contains Bt toxin. In some studies, it has beenfound that after the feeding begins, caterpillars moveaway from Bt plants faster than from non-Bt plants, butvery few larvae crawl far enough to move from onefield to another. Ramachandran et al. (1998) found thatP. xylostella larvae move away from transgenic canolaplants within 24 hours. Similarly, H. virescens and H.zea larvae are known to move between plants, so seedmixtures might not work. For endophytic insects suchas P. gossypiella and other stem and root-feedingspecies with limited larval and adult movement,within-field refuge would be best (Gould 1998). Mallet& Porter (1992) pointed out that in seed mixture refugesystem, if the pest's feeding stages could move betweenplants, instead of ingesting a high dose of toxin or notoxin at all, they would often consume intermediatedoses nullifying the advantages of high-dose refuge.The same argument can be extended to transgenicplants with tissue specific expression of toxins.Because of the importance of maintaining appropriaterefuges, insect biology and behavior should also to beconsidered for implementing a refuge system that ispractical and economic.Increasing the size of the refuge delays thedevelopment of resistance. Some workers have calledfor refuges as large as 50%, if farmers are allowed tospray them, which may present a dilemma and reducefarm profitability (Gould & Tabashnik 1998). On theother hand, farmers may be reluctant to sacrifice alarge number of refuge plants to insects just to maintainsusceptible alleles. In China, H. armigera naturallypossesses a vast refuge as it can feed on corn, soybean,peanut, and many other crops. Studies that havemonitored the sensitivity of H. armigera fieldpopulations to Bt insecticidal protein Cry1Ac from1998 to 2000 indicated that H. armigera is stillsusceptible to Cry1Ac protein (Wu et al. 2002b).Although development of H. armigera on Bt cottonwas much slower than on common cotton, there was ahigh probability of mating between populations fromBt cotton and other sources due to scattered emergencepattern of H. armigera adults and overlap of secondand third generations. Thus, in a cotton, soybean, andpeanut mix system, non-cotton crops provided a naturalrefuge (Wu et al. 2002a). As indicated earlier in thediverse cropping systems of the tropics (Sharma et al.2001), where the insects have several alternative andwild hosts, there may not be any need to grow therefuge crops.FUSION GENE STRATEGY Theoretical models suggestthat pyramiding two dissimilar toxin genes in the sameplant has the potential to delay the onset of resistance20