Intraspecific Competition of an Insect-Resistant Transgenic Canola in Seed Mixtures

genic crops is considered a major threat to the durability of these crops (van Rie, 1991; Gould, 1994; McGaughey, Seed mixtures are recommended as a strategy to minimize or avoid 1994). Refugia in the form of susceptible (nontranshaving insects develop resistance to insect-resistant transgenic crops. genic) plants to prevent exposure of some portion of The objective of this study was to evaluate a canola, Brassica napus L., transgenic for a Bacillus thuringiensis cry1Ac gene for its resistance target insect population to the toxin are currently conagainst diamondback moth, Plutella xylostella L., and for its competisidered the best strategy to minimize or avoid insect tive ability with nontransgenic canola in seed mixtures. Transgenic adaptation to transgenic crops (Gould, 1988; McGaughey and nontransgenic canola were planted either as pure stands or in and Whalon, 1992; Tabashnik, 1994a). Also, some exmixtures of 75:25, 50:50, and 25:75 in plastic trays in greenhouse perimental evidence is available suggesting that refugia experiments or in field experiments at three locations during the 1996 would delay insect adaptation (Liu and Tabashnik, through 1998 field seasons. The trays and plots were either infested 1997). Refugia could be provided either by growing with diamondback moth neonates or left without any insect infestasusceptible plants in separate sections of the field or tion. In diamondback moth-infested treatments, transgenic plants had adjacent to the resistant (transgenic) plants as seed low levels of damage both as a pure stand and in mixtures. Nontransmixtures. genic plants in diamondback moth-infested trays and plots suffered high levels of defoliation and produced less biomass and seed yield Seed mixtures are easy to adopt (Hokkanen and compared with transgenic plants. Relative crowding coefficient Wearing, 1995); however, in a seed mixture there is the (RCC), a measure of competition between the two plant types, ranged possibility that the target insect would initially develop from 0.6 to 1.1 in plots where there was no diamondback moth infestaon nontransgenic plants and later move to transgenic tion and 1.1 to 12.8 in plots where there was diamondback moth plants. Because later growth stages of the target insect infestation. No competitive advantage was observed for either plant may better tolerate the toxin of a transgenic plant, insect type in seed mixtures when there was no diamondback moth infestamovement from nontransgenic to transgenic plants tion. Transgenic canola because of its high level of resistance was might cause economic damage to the transgenic plants competitively superior in seed mixtures when there was diamondback (Mallet and Porter, 1992; Tabashnik, 1994b). Ideally, moth infestation. transgenic plants should not suffer any economic damage from attack by any stage of the target insect. For a seed mixture to be economically successful, transgenic I transgenic crops containing differplants in seed mixtures must have levels of insect resisent toxin-producing genes from the bacteria, Bacillus tance equivalent to those in pure stands of transgenic thuringiensis Berliner, are being cultivated on a large plants. The first objective of this study was to evaluate scale. The possibility of insects adapting to these transthe resistance of transgenic canola, Brassica napus L. containing a cry1Ac gene of B. thuringiensis in seed S. Ramachandran, J.N. All, Dep. of Entomology, Univ. of Georgia, mixtures against the target insect diamondback moth, Athens, GA 30602; G.D. Buntin, Dep. of Entomology, and P.L. Plutella xylostella L. Raymer, Dep. of Crop and Soil Sciences, Georgia Experiment Station, Griffin, GA 30223; C.N. Stewart, Jr., Dep. of Biology, Univ. of North It is often argued that crop plants with improved Carolina, Greensboro, NC 27402. Received 9 Nov. 1998. *Corresponding author ([email protected]). Abbreviations: LSD, least significant difference; RCC, relative crowding coefficient. Published in Agron. J. 92:368–374 (2000). RAMACHANDRAN ET AL.: INSECT-RESISTANT CANOLA IN SEED MIXTURES 369 in 50:50 mixtures. In 25:75 and 75:25 mixtures 25% plant types resistance would be agronomically less fit because of the were planted at one per row. The experiment replicated four pleiotropic effects associated with the resistance genes times was arranged in a split-plot design with diamondback (Coley et al., 1985; Bergelson and Purrington, 1996). moth and no diamondback moth infestation as main plots and For example, transgenic Arabidopis thaliana plants with different ratios of transgenic and nontransgenic plants as the resistance to the herbicide, cholorosulfuron, were less subplots. In insect infestation treatments, 20 d after plant productive compared with the nontransgenic plants (Bergermination each tray was infested with approximately 350 gelson et al., 1996; Purrington and Bergelson, 1997). diamondback moth neonates mixed with corn grits with a This reduced fitness of transgenic plants was attributed bazooka applicator (Wiseman et al. 1980). Ten days after to the improved resistance to herbicides rather than insect infestation the percentage defoliation of transgenic and nontransgenic plants was recorded and plants were cut at the to changes associated with plant transformation and base from both insect-infested and uninfested trays. Plants regeneration. Furthermore, genetic transformation could were dried in an oven at 758C, and final dry weight was affect other characters of the transgenic plant in addition measured. to the targeted character of the transformation process. Negative impacts of the transformation process have Field Experiments been associated with the poor performance of a few transgenic cotton lines (Jenkins et al., 1997). Thus, there Field experiments were conducted at the Univ. of Georgia, Horticultural Farm, Watkinsville, during the 1996–1997 season are two possibilities that could lower the performance and at the Bledsoe Research Farm, Griffin, GA, and Gibbs of insect-resistant transgenic plants: (i) metabolic costs Research Farm, Tifton, GA, during the 1997–1998 season. associated with the improved resistance, and (ii) negaFields were prepared by chisel plowing followed by disking. tive impact of the transformation process. Lime and fertilizers were added to achieve a moderate level Seed mixtures add a new dimension to the use of of fertility. Preemergence herbicide trifluralin (Treflan 4EC) transgenic plants; will the transgenic plants outcompete was applied at the rate of 0.56 kg a.i. ha2. Weeds were renontransgenic plants when grown in seed mixtures? In moved by hand as they emerged. All other normal agronomic a recent study, no differences in competitive interactions practices for the crop were performed. were noticed when transgenic canola with a herbicide Sixteen metal nails (6 cm long) were fixed on to a wooden board (145 by 20 cm) at 15-cm intervals in two rows (15 cm resistance gene was grown in mixtures with nontransapart). The wooden board laden with nails was pressed firmly genic counterpart or with barley (Fredshavn et al., 1995). in the soil to make 64 holes plot2 for planting seeds. This However, intraspecific competition has not been studied method ensured that the plants within each plot were equally for transgenic crops with insect-resistant gene(s), which spaced. Each hole was hand-planted with two seeds of the is of prime importance because of the possibility of using required genotype on 10 Oct. 1996 at Watkinsville, on 7 Oct. seed mixtures for insect-resistant transgenic crops. Thus, 1997 at Griffin, and on 18 Nov. 1997 at Tifton. In 50:50 mixture, the second objective of this study was to examine intratransgenic and nontransgenic plants were planted alternately specific competition between transgenic and nontranswithin a row. In 25:75 and 75:25 mixtures, 25% plant types genic canola plants in seed mixtures with and without were uniformly arranged as every fourth plant. A spacing of insect pressure. 1 m was provided on all sides of each plot to minimize interaction among the plots. Ten days after planting, the plants were thinned to one per hole and gaps were filled with new seeds. MATERIALS AND METHODS Treatments in the field experiments were similar to greenPlants and Insects house experiments. In insect-infested plots, each plot was infested 30 d after plant germination with approximately 3000 Canola cultivar Oscar transformed with a synthetic cry1Ac gene (line O52–6; referred to as transgenic or Bt canola) (Stewdiamondback moth neonates mixed with corn grits using a bazooka applicator (Wiseman et al., 1980) three times at art et al., 1996) and nontransgenic Oscar (referred to as nontransgenic or NBt canola) were tested for diamondback moth weekly intervals for a total of 9000 neonates. Two weeks after final infestation, the percent defoliation of transgenic resistance and intraspecific competition in seed mixtures. Transgenic Oscar synthesized 238 6 29 ng of Cry1Ac protein and nontransgenic plants was visually estimated. When the plants attained physiological maturity (pods were golden per gram of total extractable protein (Ramachandran et al., 1998a) and conferred high levels of resistance against diabrown in color), stems were cut at the base and dried in an oven at 758C. After recording the final biomass of transgenic mondback moth under laboratory conditions (Stewart et al., 1996) and as a pure stand under field conditions (Ramachanand nontransgenic plants from each plot, the seed was threshed and cleaned using an air blast seed cleaner (Allan Machine dran et al., 1998b). Diamondback moth eggs (obtained from Abbott Laboratories, Chicago, IL) were hatched at room temCo., IA), and weight was recorded. perature, and freshly hatched neonates were used in artificial infestations. Data Analysis A t-test was performed within each seed mixture to identify Greenhouse Experiments differences in the percent defoliation between transgenic and nontransgenic plants. Percent defoliation data from WatkinsA pure stand of transgenic and nontransgenic canola and mixtures of both at 75:25, 50:50, and 25:75 were planted in ville and Griffin trials were combined and a single analysis was performed. An analysis of variance using general linear plastic trays (40 by 25 by 15 cm) filled with Craven’s potting soil mix (Commerce, GA). Each tray contained a total of 16 models procedure was performed to identify significant differences (P , 0.05) in biomass and seed production between plants in four rows (each row contained four plants) with equal spacing between the plants. Transgenic plants were main plot treatments and also to identify interaction between main plot and subplots (SAS Inst., 1985). When there was a marked with white stakes for identification purposes. Transgenic and nontransgenic plants were alternated within a row significant interaction between main plot and subplots, the 370 AGRONOMY JOURNAL, VOL. 92, MARCH–APRIL 2000 Table 1. Mean (6 SE) percent defliation of transgenic (Bt) and nontransgenic (NBt) canola plants grown in different mixtures by diamondback moth infestations in the greenhouse, Watkinsville, and Griffin field experiments.