SEARCH

SEARCH BY CITATION

Keywords:

  • Bacillus thuringiensis;
  • Bt cotton;
  • dominance;
  • genetically modified crops;
  • Heliothis virescens;
  • Pectinophora gossypiella;
  • refuges;
  • resistance management;
  • transgenic crops

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

The refuge strategy is used widely for delaying evolution of insect resistance to transgenic crops that produce Bacillus thuringiensis (Bt) toxins. Farmers grow refuges of host plants that do not produce Bt toxins to promote survival of susceptible pests. Many modelling studies predict that refuges will delay resistance longest if alleles conferring resistance are rare, most resistant adults mate with susceptible adults, and Bt plants have sufficiently high toxin concentration to kill heterozygous progeny from such matings. In contrast, based on their model of the cotton pest Heliothis virescens, Vacher et al. (Journal of Evolutionary Biology, 16, 2003, 378) concluded that low rather than high toxin doses would delay resistance most effectively. We demonstrate here that their conclusion arises from invalid assumptions about larval concentration-mortality responses and dominance of resistance. Incorporation of bioassay data from H. virescens and another key cotton pest (Pectinophora gossypiella) into a population genetic model shows that toxin concentrations high enough to kill all or nearly all heterozygotes should delay resistance longer than lower concentrations.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

‘If Darwin were alive today the insect world would delight and astound him with its impressive verification of his theories of survival of the fittest. Under the stress of intensive chemical spraying the weaker members of the insect populations are being weeded out’ (Carson, 1962). Evolution of insecticide resistance, which has been documented in more than 500 species of insects and mites (Georghiou & Lagunes-Tejeda, 1991), threatens agriculture and human health worldwide. The myriad examples of insecticide resistance offer rich opportunities for examining hypotheses about how response to selection is affected by various factors, including dominance, gene flow, fitness tradeoffs, genetic constraints, major and minor genes, haplodiploidy, founder events, life-history traits and multitrophic interactions (Roush & McKenzie, 1987; Rosenheim & Tabashnik, 1991, 1993; Carrière & Roff, 1995; Rosenheim et al., 1996; ffrench-Constant et al., 1998; Bourguet & Raymond, 1998; Groeters & Tabashnik, 2000; Carrière et al., 2001a; Tabashnik, 2002; Carrière, 2003). Whereas study of resistance can provide fundamental insights about evolution, efforts to manage resistance enable application and testing of evolutionary theory. One of the most urgent challenges is to determine if strategies based on evolutionary principles can delay pest resistance to transgenic crops.

Transgenic corn and cotton that produce insecticidal proteins derived from Bacillus thuringiensis (Bt) grew on 14 million ha during 2002, with a cumulative total of more than 62 million ha since 1996 (James, 2002; Tabashnik et al., 2003). Such Bt crops can reduce reliance on insecticide sprays, thereby providing economic, health, and environmental benefits (Shelton et al., 2002; Carrière et al., 2003). Evolution of resistance by pests, however, would cut short the efficacy of Bt crops and the associated benefits. Pest resistance to Bt crops in the field has not been documented yet (Tabashnik et al., 2003), but Plutella xylostella (diamondback moth) has evolved resistance to Bt sprays in the field, and many key pests have evolved resistance to Bt toxins in the laboratory (Tabashnik, 1994; Ferré & Van Rie, 2002).

To delay pest resistance to Bt crops, the refuge strategy has been widely adopted. This strategy is based on evolutionary theory developed in dozens of papers (e.g. Georghiou & Taylor, 1977; Curtis et al., 1978; Tabashnik & Croft, 1982; Roush, 1994; Gould, 1998; Peck et al., 1999; Caprio, 2001; Carrière & Tabashnik, 2001; Onstad et al., 2002) and on small-scale experiments with P. xylostella (Liu & Tabashnik, 1997a; Shelton et al., 2000; Tang et al., 2001). The theory underlying the strategy is to reduce heritability of resistance by providing refuges of non-Bt host plants that produce susceptible adults, by promoting mating between resistant and susceptible adults, and by decreasing the dominance of resistance. The strategy relies on the principle that dominance can be altered by the environment (Roush & Daly, 1990; Bourguet et al., 1996, 2000). In particular, bioassay results from some key pests show that the dominance of their resistance to Bt toxins decreases as toxin concentration increases (Tabashnik et al., 1992, 2002; Gould et al., 1995; Liu & Tabashnik, 1997b; Tang et al., 1997; Zhao et al., 2000; Liu et al., 2001b).

To implement the refuge strategy, farmers grow refuges of non-Bt host plants near Bt crops to promote survival of susceptible pests. This strategy is expected to work best if resistance is conferred by rare, recessive alleles and if most of the extremely rare resistant adults emerging from Bt crops mate with susceptible adults from refuges. The theory predicts that such conditions will greatly delay evolution of resistance.

The refuge strategy is sometimes called the ‘high-dose refuge’ (HDR) strategy because many modelling studies show that refuges are most effective if the dose of toxin ingested by insects eating Bt plants is high enough to kill all or nearly all individuals heterozygous for resistance (e.g. Curtis et al., 1978; Tabashnik & Croft, 1982; Roush, 1997; Gould, 1998). In other words, refuges are predicted to work best if the toxin concentration in Bt plants is high enough to make resistance functionally recessive.

In contrast to the aforementioned results, Vacher et al. (2003) concluded that the HDR strategy is suboptimal, in part, because sustainable control of pests is achievable with ‘transgenic plants producing lower doses of toxin’. They based this conclusion on simulations of a diallelic (R for resistance, S for susceptibility) single-locus model of the interaction between Bt cotton and the major pest Heliothis virescens (tobacco budworm). They noted, however, that their model relies on unverified assumptions. In particular, they state, ‘first and foremost, toxin dose-mortality curves must be accurately estimated for insect larvae’. Modelling results reported here, which incorporate mortality data from bioassays instead of hypothetical mortality assumed by Vacher et al. (2003), indicate that high rather than low toxin concentration should delay pest resistance longest. We focus on H. virescens, because it was modelled by Vacher et al. (2003), and on Pectinophora gossypiella (pink bollworm), because it is another key lepidopteran pest targeted by Bt cotton for which relevant bioassay data are available.

Sources of survival data

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

We used published data on the relationship between Bt toxin concentration, larval mortality, and functional dominance from bioassays with Bt toxin Cry1Ab against H. virescens (Gould et al., 1995) and Bt toxin Cry1Ac against P. gossypiella (Tabashnik et al., 2002). Cry1Ab is closely related to Cry1Ac (Tabashnik et al., 1996), the toxin in commercially grown Bt cotton. The data for H. virescens were obtained from the 14-day bioassay summarized in the upper left corner of Figure 7 of Gould et al. (1995). The data for P. gossypiella were obtained from the 21-day bioassay summarized in Figure 1 and Table 3 of Tabashnik et al. (2002).

image

Figure 1. Effects of refuge size and Bacillus thuringiensis (Bt) toxin concentration on increase in resistance (R) allele frequency in the first generation. Survival of Heliothis virescens larvae exposed to Bt toxin was based on bioassays (Gould et al., 1995, see Table 1). A recessive fitness cost was assumed; fitnesses in refuges were 1 for SS, 1 for RS and 0.85 for RR. We assumed that the initial R allele frequency was 0.0015 as in Vacher et al. (2003).

Download figure to PowerPoint

In the bioassays, larvae ate artificial diet into which various concentrations of Bt toxin had been incorporated. Each set of bioassays tested three groups of larvae representing putative RR, SS and RS genotypes. These three groups are labelled in the figures, respectively, as selected (YHD2), control (YDK) and F1 (YHD2 × YDK) for H. virescens; and as resistant (AZP-R), susceptible (APHIS-S), and F1 (AZP-R × APHIS-S) for P. gossypiella. In addition to the 10 concentration-mortality data points for H. virescens plotted in the upper left corner of Figure 7 (Gould et al., 1995), the bioassay results showed survival of 0% for SS at 40 μg toxin mL−1 diet; 0% for SS and RS at 200 μg toxin mL−1 diet; and 100 and 97%, respectively, for RR at 1.6 and 8.0 μg toxin mL−1 diet (F. Gould, unpublished data).

Both resistant strains (YHD2 and AZP-R) show ‘mode 1’ resistance, which is the most common type of resistance to Bt toxins in Lepidoptera (Tabashnik et al., 1998, 2003). Mode 1 is characterized by >500-fold resistance to at least one Cry1A toxin, recessive inheritance, little or no cross-resistance to Cry1C, and reduced binding of at least one Cry1A toxin. In both the YHD2 and AZP-R strains, resistance is tightly linked to a cadherin locus that encodes a receptor for Cry1A toxins (Gahan et al., 2001; Morin et al., 2003).

Population genetic model

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

We used a standard population genetic model (Hartl & Clark, 1989) to predict the change in resistance (R) allele frequency in the first generation:

  • image

where p is the R allele frequency, q is the S allele frequency, WM is mean fitness, and WRR, WRS and WSS are the fitnesses of RR, RS, and SS, respectively. Although this equation can be iterated to project changes over many generations, we focus here on relative rates of resistance evolution, which are indicated by the change in R allele frequency in the first generation. Fitness for each genotype was calculated as (Carrière & Tabashnik, 2001):

  • image
  • image
  • image

where PBt is the proportion of Bt plants and PRef is the proportion of non-Bt plants (refuge), and fitnesses for RR, RS, and SS, respectively, are BtWRR, BtWRS, and BtWSS on Bt plants, and RefWRR, RefWRS, and RefWSS in refuges. Mean fitness was calculated with the standard equation (Hartl & Clark, 1989):

  • image

As in Vacher et al. (2003), the initial R allele frequency was 0.0015, which is the empirical estimate for field populations of H. virescens from Gould et al. (1997). For simplicity, we also used this assumption for P. gossypiella, for which mean estimated R allele frequency ranged from 0 to 0.16 in Arizona during a 5-year period (Tabashnik et al., 2003). Fitnesses on Bt plants (BtWRR, BtWRS, and BtWSS) at a series of Bt toxin concentrations were based either on the assumptions of Vacher et al. (2003) for low, middle and high doses or on mortality data from bioassays as described above (Gould et al., 1995; Tabashnik et al., 2002). We evaluated the effects of hypothetical vs. observed mortality on resistance evolution by holding all other assumptions constant and changing assumptions about the effect of Bt toxin concentration on mortality.

Fitness costs, refuge size and sublethal effects

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

We compared effects of hypothetical vs. observed mortality with two sets of assumptions about fitnesses in refuges. In the first set, we used the assumptions of Vacher et al. (2003), which incorporate an additive fitness cost associated with resistance: RefWRR = 0.85, RefWRS = 0.97, and RefWSS = 1. In the second set, we used the more realistic assumption that the fitness cost is recessive (Carrière et al., 2001a,b): RefWRR = 0.85, RefWRS = 1, and RefWSS = 1. Analysis of these two sets, which differ only in that RS have reduced fitness in refuges in set 1 (0.97) but not in set 2 (1), enabled direct examination of the effect of dominance of fitness cost.

To evaluate the effects of refuge size and toxin concentration on resistance evolution, we varied refuge size (5, 20 or 50%) for each of six concentrations tested against H. virescens larvae in bioassays (Gould et al., 1995). In these calculations, we used observed survival from the bioassays and assumed a recessive fitness cost (see above).

In all of the calculations described above, we assumed that larval survival corresponds to genotypic fitness, as in Vacher et al. (2003). However, for larvae exposed to Bt toxin, larval survival provides a good estimate of overall fitness when all larvae die, but not necessarily when some survive (Bourguet et al., 2000). Indeed, larval exposure to Bt toxin has sublethal effects on survivors of H. virescens and P. gossypiella, including reduced developmental rate and fecundity (Gould et al., 1995; Liu et al., 1999, 2001a). To evaluate the impact of sublethal effects and toxin concentration, we examined resistance evolution for each of six concentrations tested in bioassays against H. virescens larvae under three sets of assumptions about sublethal effects: (a) none, (b) fixed and (c) genotype-specific. Under (a), we assumed that sublethal effects were negligible, such that larval survival was a good estimate of overall fitness. Under (b), we assumed that the proportional reduction in fitness caused by sublethal effects of Bt toxin was the same for each genotype (0.50 × survival). Under (c), we assumed that the proportional reduction in fitness caused by sublethal effects of Bt toxin was greatest for SS (0.25 × survival), intermediate for RS (0.50 × survival), and weakest for RR (0.75 × survival). For example, survival of H. virescens larvae exposed to 0.32 μg toxin mL−1 diet was 0.47 for SS, 1 for RS, and 1 for RR. For individuals exposed to this concentration, fitness values for SS, RS, and RR were, respectively, 0.47, 1, and 1 under (a); 0.235, 0.5, and 0.5 under (b); and 0.1175, 0.5 and 0.75 under (c). The genotype-specific trend in (c) is consistent with data on developmental rate for H. virescens and P. gossypiella (Gould et al., 1995; Liu et al., 2001a). In these calculations, we assumed a recessive fitness cost (see above).

Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

Variation in larval survival as a function of Bt toxin concentration in bioassays with H. virescens and P. gossypiella differed markedly from the assumptions of Vacher et al. (2003) (Table 1). Bioassays with H. virescens larvae show the following: (1) survival at the lowest concentration tested (0.32 μg toxin mL−1 diet) was substantially lower for SS (47%) than for RS (100%), (2) survival of RS was 0% at high toxin concentrations (≥40 μg toxin mL−1 diet), and (3) survival of RR decreased from 100% at a low concentration (1.6 μg toxin mL−1 diet) to 59% at the highest concentration tested (500 μg toxin mL−1 diet).

Table 1.  Effects of hypothetical vs. observed survival of larvae exposed to Bacillus thuringiensis (Bt) toxin on functional dominance of resistance and change in resistance (R) allele frequency in the first generation*.
Bt toxin concentration (μg mL−1)Survival of larvae exposed to Bt (%)Dominance (h)Change in R allele frequency in the first generation
SSRSRRRS − SSAdditive cost†Recessive cost‡
  1. *In all cases, we assumed that the initial R allele frequency was 0.0015 as in Vacher et al. (2003) and that the habitat had 80% Bt cotton and 20% refuges.

  2. †With additive fitness costs, fitnesses in refuges were 1 for SS, 0.97 for RS and 0.85 for RR, as in Vacher et al. (2003).

  3. ‡With recessive fitness costs, fitnesses in refuges were 1 for SS, 1 for RS and 0.85 for RR.

  4. §ND, no data available; if concentration higher than concentration causing 0% survival, assumed to be 0% survival; if concentration lower than concentration causing 100% survival, assumed to be 100% survival for calculation of h and change in R allele frequency in the first generation.

Hypothetical: Heliothis virescens (Vacher et al., 2003)
 Low40408500.000−1.6 × 10−51.4 × 10−6
 Middle20218510.0151.1 × 10−53.6 × 10−5
 High028520.0248.2 × 10−51.3 × 10−4
Observed: Heliothis virescens (Gould et al., 1995)
 0.3247100ND§531.0001.1 × 10−31.1 × 10−3
 1.61675100590.7022.1 × 10−32.1 × 10−3
 8.003197310.3201.8 × 10−31.9 × 10−3
 40009700.000−3.6 × 10−58.4 × 10−6
 200007500.000−3.8 × 10−56.4 × 10−6
 500NDND59ND0.000−4.0 × 10−55.0 × 10−6
Observed: Pectinophora gossypiella (Tabashnik et al., 2002)
 0.323793ND560.8891.3 × 10−31.3 × 10−3
 1.0452ND480.5002.4 × 10−32.5 × 10−3
 3.208ND80.0804.4 × 10−44.9 × 10−4
 1000.51000.50.005−2.6 × 10−63.9 × 10−5

In contrast, Vacher et al. (2003) made the following three assumptions about H. virescens larvae exposed to Bt toxin: (1) survival at a low dose is equal for SS (40%) and RS (40%), (2) survival of RS is 2% at a high dose, and (3) survival of RR does not vary as a function of dose (85% survival at low, middle, and high doses). The bioassay data for H. virescens show that the assumptions of Vacher et al. (2003) underestimate survival of RS at low toxin concentration and overestimate survival of RS and RR at high toxin concentration.

The differences between observed and hypothetical survival for H. virescens larvae yield opposite patterns of dominance (h) as a function of toxin concentration (Table 1). In bioassays, resistance was completely dominant (h = 1) at the lowest concentration tested and completely recessive (h = 0) at the highest concentrations. In contrast, the assumptions of Vacher et al. (2003) yield complete recessiveness (h = 0) at the low dose and the highest degree of dominance (h = 0.024) at the high dose. Thus, bioassays showed that dominance decreased as toxin concentration increased, but Vacher et al. (2003) assumed the reverse.

Observed patterns of survival and functional dominance for P. gossypiella were similar to those observed for H. virescens (Table 1). In bioassays of P. gossypiella larvae, survival at the lowest concentration tested (0.32 μg toxin mL−1 diet) was substantially lower for SS (37%) than for RS (93%). At the highest concentration at which RS were tested (10 μg toxin mL−1 diet), their survival was 0.5%. Functional dominance (h) was 0.889 at the low concentration and 0.005 at the high concentration. Thus, similar to results with H. virescens, dominance decreased as toxin concentration increased.

Effects of observed vs. hypothetical survival on resistance evolution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

As expected from the differences described above, the predicted effects of toxin concentration on resistance evolution differ markedly between observed and hypothetical values for survival of H. virescens larvae (Table 1). When survival data for H. virescens were incorporated in the model, resistance evolved faster at the three lowest concentrations than at the three highest concentrations. Resistance evolved fastest at a concentration of 1.6 μg toxin mL−1 diet, which allowed survival of 16% for SS, 75% for RS and 100% for RR. Resistance evolved slowest at the highest concentration tested (500 μg toxin mL−1 diet), which allowed survival of 0% for SS and RS, and 59% for RR. In contrast to these findings, the hypothetical survival values for H. virescens larvae assumed by Vacher et al. (2003) make resistance evolve faster as dose increases.

With bioassay data for P. gossypiella larvae incorporated in the model, the predicted effects of toxin concentration on resistance evolution were similar to those based on bioassay data for H. virescens larvae (Table 1). For P. gossypiella, resistance evolved faster at the three lowest concentrations than at the highest concentration. Resistance evolved fastest at 1.0 μg toxin mL−1 diet, which allowed survival of 4% for SS and 52% for RS. Resistance evolved slowest at the highest concentration tested against RS (10 μg toxin mL−1 diet), which allowed 0% survival for SS, 0.5% for RS and 100% for RR.

Effects of recessive vs. additive fitness costs on resistance evolution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

Consistent with previous modelling results (Carrière & Tabashnik, 2001; Carrière et al., 2002), the additive fitness cost assumed by Vacher et al. (2003) delayed resistance more than a recessive fitness cost (Table 1). When resistance was completely recessive (h = 0) or almost completely recessive (h = 0.005), an additive fitness cost caused decreases in R allele frequency in five cases, but no such decreases occurred when the fitness cost was recessive. As the dominance of resistance increased, the effect of the dominance of the fitness cost diminished. For example, the difference between recessive and additive fitness cost was negligible when h was 1 or 0.889 (observed data for 0.32 μg toxin mL−1 diet for H. virescens and P. gossypiella, respectively).

Effects of Bt toxin concentration and sublethal effects on resistance evolution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

With no sublethal effects, sublethal effects fixed across genotypes, or genotype-specific sublethal effects (see Methods) and survival of H. virescens larvae exposed to Bt toxins based on bioassay data, resistance evolved faster at the three lowest concentrations than at the three highest concentrations (Fig. 2). Sublethal effects that reduced the fitness of individuals surviving exposure to Bt toxin generally slowed resistance evolution (Fig. 2). This is not surprising, as such sublethal effects reduce the fitness of larvae exposed to Bt toxin relative to larvae in refuges. An exception to this general trend occurred at the lowest concentration tested in bioassays (0.32 μg toxin mL−1 diet). At this concentration, resistance evolved faster with genotype-specific sublethal effects than without sublethal effects (Fig. 2). This occurred because the acceleration in resistance caused by the substantial reduction in fitness of SS associated with strong sublethal effects (from 0.47 to 0.1175) outweighed the delaying effect of proportionately smaller reductions in fitness of RS (from 1 to 0.5) and RR (from 1 to 0.75).

image

Figure 2. Effects of Bacillus thuringiensis (Bt) toxin concentration and sublethal effects on increase in resistance (R) allele frequency in the first generation. Survival of Heliothis virescens larvae exposed to Bt toxin was based on bioassays (Gould et al., 1995, see Table 1). We assumed that the initial R allele frequency was 0.0015 as in Vacher et al. (2003) and that the habitat had 80% Bt cotton and 20% refuges. Fitness of larvae exposed to Bt toxin depended on survival and sublethal effects. For no sublethal effects, larval survival values were used as fitness values. For fixed sublethal effects, fitness of each genotype was larval survival × 0.50. For genotype-specific sublethal effects, fitness of SS was survival × 0.25, fitness of RS was survival × 0.50 and fitness of RR was survival × 0.75.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References

The results here illustrate the principle that dominance can be altered by the environment. In particular, the dominance of resistance to Bt toxins of some key pests can be reduced by increasing toxin concentration. If most of the rare resistant adults emerging from Bt crops mate with susceptible adults from refuges, survival of their heterozygous progeny largely determines the response to selection. Under such conditions, reduced dominance of resistance decreases the heritability of resistance, thereby delaying evolution of resistance to Bt crops. When heritability is constant, increasing selection intensity increases the response to selection (Falconer, 1989). However, incorporation of bioassay data into a simple population genetic model indicates that increasing toxin concentration from moderate to high levels can decrease the heritability of resistance sufficiently to decrease the response to selection, despite the higher intensity of selection.

The effects of Bt toxin concentration on mortality of H. virescens larvae observed in bioassays (Gould et al., 1995) differ markedly from the effects assumed by Vacher et al. (2003). Results from a population genetic model reported here demonstrate that the differences between the observed and assumed patterns of mortality yield opposite predictions about the effect of toxin concentration on resistance evolution. Confirming the results of many earlier studies and refuting Vacher et al. (2003), incorporation of bioassay data for H. virescens and P. gossypiella larvae into a model led to the prediction that concentrations high enough to kill all or nearly all RS larvae will delay resistance most effectively. Also confirming patterns described from earlier models (e. g. Tabashnik & Croft, 1982), results reported here based on incorporation of bioassay data from H. virescens and P. gossypiella show that resistance evolved fastest at intermediate concentrations causing <20% survival of SS and >50% survival of RS. These conclusions were robust across various assumptions about fitness costs (Table 1), refuge size (Fig. 1) and sublethal effects (Fig. 2).

We emphasized one bioassay data set for each pest species (Table 1), but additional data for these two pests and data for other pests suggest that typical concentration-mortality responses favour high doses for delaying pest resistance to Bt toxins. Although we focused on data for H. virescens from one set of bioassays with the resistant YHD2 strain, the susceptible YDK strain, and their F1 progeny (Gould et al., 1995, Fig. 7 upper left), data from a second set of bioassays with these three groups of larvae show the same trends (Gould et al., 1995, Fig. 7 lower left). At a low concentration (1.6 μg toxin mL−1 diet) survival was much lower for SS (45%) than for RS (93%), whereas at a high concentration (200 μg toxin mL−1 diet) survival was 0% for SS and RS. So, as with the data for H. virescens summarized in Table 1, resistance was more dominant at a low concentration (h = 0.87) than at a high concentration (h = 0). Other bioassay data with Bt toxins showing a similar pattern include the APHIS-98R strain of P. gossypiella vs. Cry1Ac (Liu et al., 2001b), P. xylostella vs. Cry1C (Liu & Tabashnik, 1997b; Zhao et al., 2000), and P. xylostella vs. formulated Bt toxins (Tabashnik et al., 1992; Tang et al., 1997).

In some published bioassay studies with Bt toxins, the potential impact of high toxin concentrations cannot be assessed readily because survival of RS larvae is not reported for concentrations higher than those killing 95% of RS larvae (e. g. McGaughey, 1985; McGaughey & Beeman, 1988; Gould et al., 1992; Huang et al., 1999). In some cases, only concentrations killing 50% (LC50s) and slopes of concentration-mortality lines are reported, or estimated concentration-mortality lines are plotted in figures without individual data points. Further, some studies report estimates of dominance (D, Stone, 1968) based only on LC50s. These practices reflect an emphasis on probit analysis, in which concentrations causing 0% survival are problematic and estimation of LC50s is a primary focus. However, to facilitate evaluation of the refuge strategy, we encourage researchers to test SS, RS and RR larvae at concentrations killing <25 to >99% of RS larvae and to report estimates of dominance (h) as a function of concentration across this range.

We are not aware of published bioassay results with Bt toxins showing the type of dose-mortality relationships assumed by Vacher et al. (2003). For example, we know of no bioassays where survival of 40% occurs for both SS and RS at the same Bt toxin concentration. Further, bioassay results show that if Bt toxin concentration is increased sufficiently, survival of RR declines. Because the assumptions of Vacher et al. (2003) do not correspond with published data for H. virescens or other pests, the relevance of their conclusions is unclear.

The results reported here confirm numerous other studies showing that, in theory, high concentrations of Bt toxins in transgenic plants will be most effective for delaying resistance. Experimental evidence shows that commercial cultivars of Bt cotton are sufficiently toxic to kill all or nearly all F1 larvae from crosses of resistant and susceptible adults (putative RS) of P. gossypiella (Liu et al., 1999, 2001b; Tabashnik et al., 2000) and H. virescens (F. Gould, L. Carter and L. Seltmann unpublished data). Thus, for these two pests that are extremely susceptible to the toxin in Bt cotton, it appears that the theoretically envisioned high dose is achieved. This also appears to be true for P. xylostella tested on noncommercial cultivars of transgenic crucifers that produce high concentrations of Bt toxins (Metz et al., 1995; Tang et al., 1997; Zhao et al., 2000). On the contrary, as noted by Vacher et al. (2003) and others, some pests of cotton such as Helicoverpa armigera and Helicoverpa zea are much less susceptible to the Cry1Ac toxin in Bt cotton. Their resistance to commercial cultivars of Bt cotton might not be functionally recessive (Akhurst et al., 2003; Burd et al., 2003).

In conclusion, models incorporating data from bioassays with Bt toxins in diet show that high rather than low toxin concentrations are expected to delay pest adaptation longest. Further, bioassays testing H. virescens and P. gossypiella larvae on transgenic plants indicate that the toxin concentration in Bt cotton is high enough to kill all or nearly all F1 progeny of matings between resistant and susceptible adults. Thus, for these pests, the assumption of functionally recessive inheritance of resistance to Bt cotton appears valid. Nonetheless, the hypothesis that high doses are best for delaying pest resistance remains to be tested experimentally.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Sources of survival data
  6. Estimation of dominance
  7. Population genetic model
  8. Fitness costs, refuge size and sublethal effects
  9. Results
  10. Observed vs. hypothetical effects of Bt toxin concentration on survival and functional dominance
  11. Effects of observed vs. hypothetical survival on resistance evolution
  12. Effects of recessive vs. additive fitness costs on resistance evolution
  13. Effects of Bt toxin concentration and refuge size on resistance evolution
  14. Effects of Bt toxin concentration and sublethal effects on resistance evolution
  15. Discussion
  16. Acknowledgments
  17. References
  • Ffrench-Constant, R.H., Pittendrigh, B. & Vaughan, A. 1998. Why are there so few resistance-associated mutations in insecticide target genes? Philos. Trans. R. Soc. Lond. B 0: 16851693.
  • Akhurst, R.J., James, W. & Bird, L. 2003. Resistance to the Cry1Ac δ-endotoxin of Bacillus thuringiensis in the cotton bollworm, Helicoverpa armigera (Lepidoptera: Noctuidae). J. Econ. Entomol. 96: 12901299.
  • Bourguet, D. & Raymond, M. 1998. The molecular basis of dominance relationships: the case of some recent adaptive genes. J. Evol. Biol. 11: 103122.
  • Bourguet, D., Prout, M. & Raymond, M. 1996. Dominance of insecticide resistance presents a plastic response. Genetics 143: 407416.
  • Bourguet, D., Genissel, A. & Raymond, M. 2000. Insecticide resistance and dominance levels. J. Econ. Entomol. 93: 15881595.
  • Burd, A.D., Gould, F., Bradley, J.R., Van Duyn, J.W. & Moar, W.J. 2003. Estimated frequency of nonrecessive Bt resistance genes in bollworm, Helicoverpa zea (Boddie) (Lepidoptera: Noctuidae) in eastern North Carolina. J. Econ. Entomol. 96: 137142.
  • Caprio, M.A. 2001. Source-sink dynamics between transgenic and nontransgenic habitats and their role in the evolution of resistance. J. Econ. Entomol. 94: 698705.
  • Carrière, Y. 2003. Haplodiploidy, sex, and the evolution of pesticide resistance. J. Econ. Entomol. 6: 16261640.
  • Carrière, Y. & Roff, D.A. 1995. Change in genetic architecture resulting from the evolution of insecticide resistance: a theoretical and empirical analysis. Heredity 74: 357368.
  • Carrière, Y. & Tabashnik, B.E. 2001. Reversing insect adaptation to transgenic insecticidal plants. Proc. Roy. Soc. Lond. B 268: 14751480.
  • Carrière, Y., Ellers-Kirk, C., Patin, A.L., Sims, M.A., Meyer, S., Liu, Y.B., Dennehy, T.J. & Tabashnik, B.E. 2001a. Overwintering costs associated with resistance to transgenic cotton in the pink bollworm. J. Econ. Entomol. 94: 935941.
  • Carrière, Y., Ellers-Kirk, C., Liu, Y.B., Sims, M.A., Patin, A.L., Dennehy, T.J. & Tabashnik, B.E. 2001b. Fitness costs and maternal effects associated with resistance to transgenic cotton in the pink bollworm. J. Econ. Entomol. 94: 15711576.
  • Carrière, Y., Dennehy, T.J., Ellers-Kirk, C., Holley, D., Liu, Y.B., Simms, M. & Tabashnik, B.E. 2002. Fitness costs, incomplete resistance, and management of resistance to Bt crops. In: Biotechnology of Bacillus thuringiensis and its Environmental Impact: Proceedings of the 4th Pacific Rim Conference (R. J.Akhurst, C. E.Beard & P.Hughes, eds), pp. 8291. CSIRO Canberra.
  • Carrière, Y., Ellers-Kirk, C., Sisterson, M., Antilla, L., Whitlow, M., Dennehy, T.J. & Tabashnik, B.E. 2003. Long-term regional suppression of pink bollworm by Bacillus thuringiensis cotton. Proc. Natl. Acad. Sci. U S A 100: 15191523.
  • Carson, R. 1962. Silent Spring. Houghon Mifflin, New York, USA.
  • Curtis, C.F., Cook, L.M. & Wood, R.J. 1978. Selection for and against insecticide resistance and possible methods of inhibiting the evolution of resistance in mosquitoes. Ecol. Entomol. 3: 273287.
  • Falconer, D.S. 1989. Introduction to Quantitative Genetics, 3rd edn. John Wiley & Sons, New York, USA.
  • Ferré, J. & Van Rie, J. 2002. Biochemistry and genetics of insect resistance to Bacillus thuringiensis. Annu. Rev. Entomol. 47: 501533.
  • Gahan, L.J., Gould, F. & Heckel, D.G. 2001. Identification of a gene associated with Bt resistance in Heliothis virescens. Science 293: 857860.
  • Georghiou, G.P. & Lagunes-Tejeda, A. 1991. The Occurrence of Resistance to Pesticides in Arthropods. Food and Agriculture Organization of the United Nations, Rome.
  • Georghiou, G.P. & Taylor, C.E. 1977. Operational influences in the evolution of insecticide resistance. J. Econ. Entomol. 70: 653658.
  • Gould, F. 1998. Sustainability of transgenic insecticidal cultivars: integrating pest genetics and ecology. Annu. Rev. Entomol. 43: 701726.
  • Gould, F., Martínez-Ramírez, A., Anderson, A., Ferré, J., Silva, F.J. & Moar, W.J. 1992. Broad-spectrum resistance to Bacillus thuringiensis in Heliothis virescens. Proc. Natl. Acad. Sci. USA 89: 79867990.
  • Gould, F., Anderson, A., Reynolds, A., Bumgarner, L. & Moar, W. 1995. Selection and genetic analysis of a Heliothis virescens (Lepidoptera: Noctuidae) strain with high levels of resistance to Bacillus thuringiensis toxins. J. Econ. Entomol. 88: 15451559.
  • Gould, F., Anderson, A., Jones, A., Sumerford, D., Heckel, D.G., Lopez, J., Micinski, S., Leonard, R. & Laster, M. 1997. Initial frequency of alleles for resistance to Bacillus thuringiensis toxins in field populations of Heliothis virescens. Proc. Natl. Acad. Sci. USA 94: 35193523.
  • Groeters, F.R. & Tabashnik, B.E. 2000. Roles of selection intensity, major genes, and minor genes in evolution of insecticide resistance. J. Econ. Entomol. 93: 15801587.
  • Hartl, D. & Clark, A.G. 1989. Principles of Population Genetics, 2nd edn. Sinauer Associates, Inc., Sunderland, Massachusetts, USA.
  • Huang, F., Buschman, L.L., Higgins, R.A. & McGaughey, W.H. 1999. Inheritance of resistance to Bacillus thuringiensis toxin (Dipel ES) in the European corn borer. Science 284: 965967.
  • James, C. 2002. Global Status of Commercialized Transgenic Crops: 2002. ISAAA Briefs No. 27: Preview. International Service for the Acquisition of Agri-biotech Applications, Ithaca, NY.
  • Liu, Y.B. & Tabashnik, B.E. 1997a. Experimental evidence that refuges delay insect adaptation to Bacillus thuringiensis. Proc. Roy. Soc. Lond. B 264: 605610.
  • Liu, Y.B. & Tabashnik, B.E. 1997b. Inheritance of resistance to Bacillus thuringiensis toxin Cry1C in diamondback moth. Appl. Environ. Microbiol. 63: 22182223.
  • Liu, Y.B., Tabashnik, B.E., Dennehy, T.J., Patin, A.L. & Bartlett, A.C. 1999. Development time and resistance to Bt crops. Nature 400: 519.
  • Liu, Y.B., Tabashnik, B.E., Dennehy, T.J., Patin, A.L., Sims, M.A., Meyer, S.K. & Carrière, Y. 2001a. Effects of Bt cotton and Cry1Ac toxin on survival and development of pink bollworm (Lepidoptera: Gelechiidae). J. Econ. Entomol. 94: 12371242.
  • Liu, Y.B., Tabashnik, B.E., Meyer, S.K., Carrière, Y. & Bartlett, A.C. 2001b. Genetics of pink bollworm resistance to Bacillus thuringiensis toxin Cry1Ac. J. Econ. Entomol. 94: 248252.
  • McGaughey, W.H. 1985. Insect resistance to the biological insecticide Bacillus thuringiensis. Science 229: 193195.
  • McGaughey, W.H. & Beeman, R.W. 1988. Resistance to Bacillus thuringiensis in colonies of Indianmeal moth and almond moth (Lepidoptera: Pyralidae). J. Econ. Entomol. 81: 2833.
  • Metz, T.D., Roush, R.T., Tang, J.D., Shelton, A.M. & Earle, E.D. 1995. Transgenic broccoli expressing a Bacillus thuringiensis insecticidal crystal protein: implications for pest resistance management strategies. Mol. Breed. 1: 309317.
  • Morin, S., Biggs, R.W., Sisterson, M.S., Shriver, L., Ellers-Kirk, C., Higginson, D., Holley, D., Gahan, L.J., Heckel, D.G., Carrière, Y., Dennehy, T.J., Brown, J.K. & Tabashnik, B.E. 2003. Three cadherin alleles associated with resistance to Bacillus thuringiensis in pink bollworm. Proc. Natl. Acad. Sci. USA 100: 50045009.
  • Onstad, D.W., Guse, C.A., Porter, P., Buschman, L.L., Higgins, R.A., Sloderbeck, P.E., Peairs, F.B. & Cronholm, G.B. 2002. Modeling the development of resistance by stalk-boring Lepidopteran insects (Crambidae) in areas with transgenic corn and frequent insecticide use. J. Econ. Entomol. 95: 10331043.
  • Peck, S., Gould, F. & Ellner, S.P. 1999. Spread of resistance in spatially extended regions of transgenic cotton: implications for management of Heliothis virescens (Lepidoptera: Noctuidae). J. Econ. Entomol. 92: 116.
  • Rosenheim, J.A. & Tabashnik, B.E. 1991. Influence of generation time on the rate of response to selection. Am. Naturalist 137: 527541.
  • Rosenheim, J.A. & Tabashnik, B.E. 1993. Generation time and evolution. Nature 365: 791792.
  • Rosenheim, J.A., Johnson, M.W., Mau, R.F.L., Welter, S.C. & Tabashnik, B.E. 1996. Biochemical preadaptations, founder events, and the evolution of resistance in arthropods. J. Econ. Entomol. 89: 263273.
  • Roush, R.T. 1994. Managing pests and their resistance to Bacillus thuringiensis: can transgenics be better than sprays? Biocontrol Sci. Technol. 4: 501516.
  • Roush, R.T. 1997. Managing resistance to transgenic crops. In: Advances in Insect Control: The Role of Transgenic Plants (N.Carozzi & M.Koziel, eds), pp. 271294. Taylor & Francis, London.
  • Roush, R.T. & Daly J.C. 1990. The role of population genetics in resistance research and management. In: Pesticide Resistance in Arthropods (R.T.Roush & B.E.Tabashnik, eds), pp. 97152. Chapman & Hall, New York.
  • Roush, R.T. & McKenzie, J.A. 1987. Ecological genetics of insecticide and acaricide resistance. Annu. Rev. Entomol. 32: 361380.
  • Shelton, A.M., Tang, J.D., Roush, R.T., Metz, T.D. & Earle, E.D. 2000. Field tests on managing resistance to Bt-engineered plants. Nat. Biotechnol. 18: 339342.
  • Shelton, A.M., Zhao, J.-Z. & Roush, R.T. 2002. Economic, ecological, food safety, and social consequences of the deployment of Bt transgenic plants. Annu. Rev. Entomol. 47: 845881.
  • Stone, B.F. 1968. A formula for determining degree of dominance in cases of monofactorial inheritance of resistance to chemicals. WHO Bull. 38: 325326.
  • Tabashnik, B.E. 1994. Evolution of resistance to Bacillus thuringiensis. Annu. Rev. Entomol. 39: 4779.
  • Tabashnik, B.E. 2002. Pesticide resistance. In: Encyclopedia of Evolution (M.Pagel, ed.), pp. 880883. Oxford University Press, New York, USA.
  • Tabashnik, B.E. & Croft, B.A. 1982. Managing pesticide resistance in crop-arthropod complexes: interactions between biological and operational factors. Environ. Entomol. 11: 11371144.
  • Tabashnik, B.E., Schwartz, J.M., Finson, N. & Johnson, M.W. 1992. Inheritance of resistance to Bacillus thuringiensis in diamondback moth (Lepidoptera: Plutellidae). J. Econ. Entomol. 85: 10461055.
  • Tabashnik, B.E., Malvar, T., Liu, Y.B., Finson, N., Borthakur, D., Shin, B.S., Park, S.H., Masson, L., de Maagd, R.A. & Bosch, D. 1996. Cross-resistance of the diamondback moth indicates altered interactions with domain II of Bacillus thuringiensis toxins. Appl. Environ. Microbiol. 62: 28392844.
  • Tabashnik, B.E., Liu, Y.B., Malvar, T., Heckel, D.G., Masson, L. & Ferré, J. 1998. Insect resistance to Bacillus thuringiensis: uniform or diverse?. Phil. Trans. Roy. Soc. Lond. B 353: 17511756.
  • Tabashnik, B.E., Patin, A.L., Dennehy, T.J., Liu, Y.B., Carrière, Y., Sims, M.A. & Antilla, L. 2000. Frequency of resistance to Bacillus thuringiensis in field populations of pink bollworm. Proc. Natl. Acad. Sci. U S A 97: 1298012984.
  • Tabashnik, B.E., Liu, Y.B., Dennehy, T.J., Sims, M.A., Sisterson, M.S., Biggs, R.W. & Carrière, Y. 2002. Inheritance of resistance to Bt toxin Cry1Ac in a field-derived strain of pink bollworm (Lepidoptera: Gelechiidae). J. Econ. Entomol. 95: 10181026.
  • Tabashnik, B.E., Carrière, Y., Dennehy, T.J., Morin, S., Sisterson, M.S., Roush, R.T., Shelton, A.M. & Zhao, J.Z. 2003. Insect resistance to transgenic Bt crops: lessons from the laboratory and field. J. Econ. Entomol. 96: 10311038.
  • Tang, J.D., Gilboa, S., Roush, R.T. & Shelton, A.M. 1997. Inheritance, stability, and lack-of-fitness costs of field-selected resistance to Bacillus thuringiensis in diamondback moth (Lepidoptera: Plutellidae) from Florida. J. Econ. Entomol. 90: 732741.
  • Tang, J.D., Collins, H.L., Metz, T.D., Earle, E.D., Zhao, J.Z., Roush, R.T. & Shelton, A.M. 2001. Greenhouse tests on resistance management of Bt transgenic plants using refuge strategies. J. Econ. Entomol. 94: 240247.
  • Vacher, C., Bourguet, D., Rousset, F., Chevillon, C. & Hochberg, M.E. 2003. Modelling the spatial configuration of refuges for a sustainable control of pests: a case study of Bt cotton. J. Evol. Biol. 16: 378387.
  • Zhao, J.Z., Collins, H.L., Tang, J.D., Cao, J., Earle, E.D., Roush, R.T., Herrero, S., Escriche, B., Ferré, J. & Shelton, A.M. 2000. Development and characterization of diamondback moth resistance to transgenic broccoli expressing high levels of Cry1C. Appl. Environ. Microbiol. 66: 37843789.