Introduction

To improve insect pest management, genetic modified crop plants such as cotton and corn based on insecticidal proteins from Bacillus thuringiensis (Bt) are now the most widely used strategies1. Since failures of synthetic chemicals, microbial insecticides derived from Bt have proven ample potential for pest management2. Insecticidal proteins are extensively used both in sprays formulations and transgenic crops to control lepidopteran pests having negligible effects on non-target pests3,4,5. These Bt insecticidal proteins are valuable as they can bring season long protection against lepidopteran pests6.

Since worldwide commercialization of transgenic Bt crops in 1996, cry genes that codify for insecticidal proteins such as Cry1Ab, Cry1Ac, Cry1F, Cry1Ah, and Cry3Bb1 have been expressed in corn, cotton and soybean and grown extensively in several countries7. However, the efficacy of transgenic crops can be reduced by the continuous use of single trait products. Insect resistance is the major hazard to the enduring success of Bt crops8. Bt resistance in field conditions has been reported in different insect pests including Ostrinia nubilalis 9, Plutella xylostella 10, Trichoplusia ni 11, Spodoptera frugiperda 12, Busseola fusca 13, Diabrotica virgifera virgifera 14 and Pectinophora gossypiella 15. In order to delay resistance evolution, implementation of suitable resistance management practices is necessary. The application of high-dose/refuge strategy and pyramiding of two or more toxins with different mode of action have been the foremost strategies used worldwide to delay evolution of resistance in different pests toward Bt crops16,17,18. The theory underlying the refuge strategy implies the planting of non-Bt host plants in close vicinity to Bt crops to assure that the rare resistant individuals selected on Bt crops will mate with susceptible individuals from nearby refuges of host plants without Bt toxins therefore generating heterozygous progenies that should be sensitive to the high dose concentration of the Bt toxin expressed in the transgenic plants, implying recessive inheritance19,20,21,22. Multiple components are predicted to favor the success of the refuge strategy: recessive inheritance of resistance, low initial resistance allele frequency, pest movements and mating patterns, fitness costs and incomplete resistance19,23. Moreover, the success of the strategy based on applying rotations of different toxins or toxin mixtures works best if the inheritance of resistance to each toxin is recessive.

Transgenic corn expressing the activated 65 kDa Cry1Ah was developed by Origen Seeds Ltd to target Asian corn borer (ACB). Knowledge of the genetic basis of resistance to Cry1Ah toxin is imperative for designing and refining managing tactics to minimize evolution of resistance in this pest. The present study offers insights into the mechanisms linked to resistance evolution. We describe here the results of the experiments that analyzed the mode of inheritance for Cry1Ah resistance in a laboratory selected strain resistant to Cry1Ah (ACB–AhR), we evaluated maternal effects, sex linkage and effective dominance. In addition, backcrosses were performed to estimate the number of loci affecting the inheritance. Also, the assumption of functional recessive resistance to Cry1Ah toxins was measured by determining the survival of Cry1Ah-resistant, susceptible parental strains and the F1 progenies from reciprocal crosses in experiments with Bt corn plants tissues and non-Bt plant in laboratory bioassays. Finally, cross resistance patterns among different Bt toxins were also analyzed. The implications of this study will be helpful for the future managing of ACB resistance to transgenic Cry1Ah-maize.

Results

Selection of ACB for Cry1Ah resistance

Bioassays using Cry1Ah toxin and Bt maize tissues revealed that ACB-AhR strain developed a very high level of resistance to Cry1Ah toxin after 48 generations of selection. The diet bioassays validated that larvae from the resistant strain had a higher level of resistance to Cry1Ah toxin compared with larvae from the susceptible strain. The LC50 values for the susceptible (ACB-BtS) and Cry1Ah-selected colonies (ACB-AhR) were significantly different showing LC50 values of 0.33 and 63.91 μg/g Cry1Ah toxin, respectively, (Table 1). The differences in LC50 values between ACB-BtS and ACB-AhR strains and the resulting resistance ratio of more than 193.67-fold validates that resistance to Cry1Ah toxin is attainable for this species.

Table 1 Susceptibility of ACB-AhR and ACB-BtS of O. furnacalis to 5 Bt toxins.

Cross-resistance

The LC50 values for Cry1Ab and Cry1Ac were significantly higher in the ACB-AhR strain compared to the ACB-BtS strain, i.e. bioassays with Cry1Ac led to a 22.11-fold resistance, indicating that the selected ACB-AhR strain is slightly cross-resistance to Cry1Ac. Similarly, a 28.38-fold resistance was observed when exposed to Cry1Ab toxins (Table 1). The highest level of cross-resistance was detected with Cry1F with a resistance ratio of up to 464 fold, whilst no cross-resistance was observed with Cry1Ie (Table 1). These data indicate that there is no cross-resistance between Cry1Ie and Cry1Ah in ACB-AhR strain.

Survival on plant tissues

Neonate larvae of ACB-AhR, ACB-BtS strains and their F1 progeny displayed significantly lower rate of feeding on Bt maize tissues than on non-Bt maize tissues. The survival rates of ACB-AhR strain were significantly greater than ACB-BtS strain and F1 progenies when larvae were fed on Cry1Ah-expressing maize leaf tissue after 7 days (Table 2). In addition, ACB-AhR strain exhibited less survival on Bt kernel followed by Bt leaves, highest survival of resistant strain was perceived on silk tissues (Table 2). In contrast, in case of non-Bt leaves these strains did not differ significantly (F 3, 8 = 12.3; P = 0.0023). ACB-AhR strain showed more survival rate than susceptible strain when fed on non-Bt silk tissues (F 3, 8 = 3.52; P = 0.0689). The larvae of all strains significantly differed when they were fed on kernel tissue of non-Bt maize (F 3, 8 = 25.2, P = 0.0002) (Table 2).

Table 2 Survival of ACB-AhR, ACB-BtS strains of Ostrinia furnacalis and F1 progenies fed on Bt and non-Bt maize tissues.

Quantification of Cry1Ah toxin

Cry1Ah concentrations were determined in leaves, silk and kernels tissues of the Cry1Ah- maize. Kernel showed the highest concentration of Cry1Ah (Fig. 1). The concentration of Cry1Ah toxin in these plant tissues was 41.84, 57.10 and 78.66 ng/g dry weight, respectively (F 2, 29 = 28.3; P < 0.05).

Figure 1
figure 1

Quantification of Cry1Ah toxin (ng/g) in Cry1Ah maize leaves, silk and kernel tissue. Means ± SE followed by different letters are significantly different (P < 0.05) according to Fisher’s Protect LSD test. The concentration level of Cry1Ah toxin increased through the leaf to kernel tissue.

Maternal effects and sex linkage

The maternal influence and the sex-linked nature of the resistance were examined by comparing observed larval concentration responses of the F1 progeny. The LC50 for the F1 progeny from reciprocal crosses to Cry1Ah was significantly greater than the LC50 for the susceptible parental strain and significantly less than the LC50 for the resistant parental strain (Table 3). Likewise, the mean slope of the concentration-mortality line did not differ between the reciprocal crosses. Thus, inheritance was autosomal; neither maternal effects nor sex-linkage were evident.

Table 3 Responses of F1 progenies from reciprocal crosses between resistant and susceptible strains of Ostrinia furnacalis to Cry1Ah toxin.

Estimation of the degree of dominance

Dominance estimations at different concentrations of Cry1Ah toxin revealed that the resistance was dominant in low concentration treatment but decreased if the concentration increased. The level of dominance is obtained by the calculation of effective dominance at different toxin concentrations, h varied with concentration, from dominant inheritance at low concentrations to recessive inheritance at high concentrations. For example, results presented partially dominance at 0.5 μg/g (h = 0.79), and declined to incomplete-recessive by treatment concentrations of 50.0 μg/g (h = 0.4) and complete recessive 250 μg/g (h = 0) (Table 4).

Table 4 Effective of dominance (h) of resistance to Cry1Ah-selected Asian corn Borer.

Plant tissue bioassays showed that the h values were 0.13, 0.06 and 0 for Cry1Ah maize leaves, silks and kernel, respectively (Table 5), which indicated that the resistance was functionally recessive at leaves stage of maize plant development. At the silking stage of the plant, the dominance was virtually recessive but it was completely recessive at the kernel stage of the plant development.

Table 5 Effective dominance values (h) of Cry1Ah resistance in ACB-AhR strain of the Ostrinia furnacalis based on Cry1Ah maize plant tissues bioassays.

Number of loci influencing resistance

The F1 population was backcrossed with ACB-AhR and tested for goodness-of-fit to a monofactorial model. The direct tests use chi-square values calculated from the observed and expected mortalities of the backcross population. The pattern of response was not consistent with a monofactorial model (∑χ2 = 43.61 > (\(\sum {\chi }_{0.05}^{2}\) = 3.84 (df = 1) (Table 6). The same response was observed in all backcross interactions showing that inheritance of resistance to Cry1Ah in Asian corn borer ACB-AhR strain is polygenic.

Table 6 Direct test for deviation between observed and expected mortality for a monogenic model (df = 1).

Discussion

Bt maize, specifically the first generation of biotech Bt maize hybrids target both O. nubilalis and O. furnacalis. A few strains of O. nubilalis have developed resistance to the expressed protein in different events of Bt maize, Cry1Ac, Cry1Ab or Cry1F in laboratory-selection experiments24,25,26,27. In this work we showed that the selection of O. furnacalis with Cry1Ah resulted in a high level of resistance to this toxin (200-fold for ACB-AhR). We showed that ACB-AhR could feed and survive on leaf, silk and kernel tissues of Bt-maize, although survival rates were low on leaves and kernels (Table 2). The low survival on Cry1Ah-leaves of ACB-AhR was unexpected since this tissue showed the lower Cry1Ah concentration (Fig. 1). It could be possible that other plant molecules present in leaves and kernels of Cry1Ah-maize contribute to the low survival rate observed in these tissues. In agreement with this, both ACB-BtS and ACB-AhR showed higher survival rates on non-Bt silk than on non-Bt leaves or kernels (Table 2). In earlier findings, Cry1F-selected strain (ACB-FR) established a high level of resistance (1700-fod) and the larvae presented potential to survive and consume Cry1F toxin27. Four strains of O. nubilalis that were exposed to Cry1Ab toxin for 10 generations developed low level of the resistance (2.0- to 10-fold) and caused reduced susceptibility to other toxins to which these selected strains were not exposed26. However, high level of resistance to Cry1Ac with a peak of 162-fold resistance was observed in a Minnesota strain of O. nubilalis after only eight generations of selection24.

Several factors are involved in resistance levels among strains. However, resistance levels are linked to the number of generations selected, concentrations of toxins consumed, selective agents itself, bioassay procedures and susceptibility among unselected strains28,29,30. In the present study, the levels of resistance to Cry1Ah in laboratory-selected strains resulted in about 200-fold resistance to purified Cry1Ah toxin under moderate selection pressure. These are the highest levels of the resistance to a Cry1Ah toxin ever documented for O. furnacalis. Previously, different O. furnacalis resistant strains to different Cry1 toxins were selected and characterized, including 106-fold resistance to Cry1Ab28, 40-fold and 113-fold resistance to Cry1Ab and Cry1Ac respectively31, or 1700-fold resistance to Cry1F after 49 generations of selection32. Factors contributing in increased tolerance to Bt crops may be the time exposure to Bt crops, resistance genes and genetic background of different populations33.

The results from the analysis of F1 progenies from reciprocal crosses performed between resistant and susceptible strains suggest that resistance to Cry1Ah was autosomally inherited with no maternal effects. The values of the present study are consistent with nearly all previous findings with Bt resistance including resistance to Cry1F in O. furnacalis 32, or to Cry1Ab and Cry1Ac31, Greenhouse-Derived Strain of Trichoplusia ni to Cry1Ac and B. thuringiensis subsp. kurstaki,11,34, Helicoverpa armigera to Cry1Ac35 and resistance to Cry1F maize in a strain of S. frugiperda 36. Usually, resistance to Bt toxins behave in agreement with autosomal mode of inheritance. In contrast, inheritance of Cry1Ab resistance in a field-derived strain of O. nubilalis and resistance to Cry1Ac in P. xylostella in leaf dips bioassay had proven maternal influence on the survival of F1 progenies9,37. Also, sex linkage phenomenon was described to inheritance pattern of resistance to Cry3Bb1 in Diabrotica virgifera virgifera 38. These dissimilarities in results are possibly related to use of different insects species.

The spectrum of cross-resistance occurs when the Bt toxins demonstrate similarities in their mechanisms of actions39,40. In the present study, selection for Cry1Ah resistance has resulted in slight cross-resistance to Cry1Ab and Cry1Ac with resistance ratios of 28.38 and 22.11-fold, respectively. Cross-resistance of Cry1Ah was not unexpected for Cry1Ab and Cry1Ac because the toxic fragment of Cry1Ah has 72 and 82% similarity to those of Cry1Ab and Cry1Ac, respectively41. Similarly, in O. furnacalis, a Cry1Ab-selected strain has indicated cross-resistance to Cry1Ah28. The highest level of cross-resistance was with Cry1F toxin (464-fold). This high level of cross-resistance between Cry1Ah and Cry1F indicates that these toxins shared similar mechanisms of action and sequence similarities. These results varied from the previous study in resistance of O. furnacalis to Cry1F that showed no cross-resistance to Cry1Ah32, suggesting that in these two populations a different mechanism of resistance was selected. The different cross-resistance patterns of these two populations could be due to the different protein used for the selection procedure (Cry1Fa vs. Cry1Ah). Our results suggest that in ACB-AhR resistant population a step shared between Cry1Ah and Cry1Fa toxins was affected, while in the ACB-FR resistant population a step involved in Cry1Fa action but not in Cry1Ah was affected. The presence of multiple binding sites shared by different Cry toxins have been documented before, for example the binding competition assays previously performed in Diatraea saccharalis that showed that Cry1Ac and Cry1Ab share binding sites, while Cry1Fa shared binding sites with Cry1Ab but not with Cry1Ac toxin42. These results indicate that in D. saccharalis Cry1Ab toxin may have two binding sites, one that is shared with Cry1Ac and the other shared with Cry1Fa42. Our data suggest that in the two O. furnacalis populations, ACB-AhR and ACB-FR, different Cry1Ah binding sites were affected, one shared with Cry1F and other not shared with Cry1F, explaining the differences observed in cross-resistance in these populations. Our results indicate that the lack of cross-resistance in a Cry resistant population does not necessary means that cross-resistance to other Cry toxins could not be selected when the insects were exposed to a different toxins. In some lepidopteran insect cross-resistance pattern of Cry1F to other Bt toxins has been defined. In O. nubilalis a strain selected for Cry1F resistance (3000-fold resistance), showed no cross-resistance to Cry1Ab or Cry9C, and a low cross-resistance to Cry1Ac (7-fold)40. Interestingly, no cross-resistance was observed to Cry1Ie. The lack of cross-resistance to Cry1Ie to other Cry toxins agree with the previous binding analysis in O. furnacalis 28,31,32. This lack of cross-resistance may be due to binding of the secondary toxins on different binding receptors on brush border membrane vesicles (BBMV). When cross-resistance is not observed, binding studies have shown that the two toxins interact with different binding sites43.

The LC50 values of F1 progenies were intermediate between the LC50 values of resistant and susceptible parental strains. The h varied from 0 to 0.79 depending on the concentrations of the toxin, with resistance fully recessive at high concentration and almost partially dominant as concentration decreased. In previous studies similar findings have been reported in other insects9,37. Analysis of dominance values on plant tissues of Bt-maize showed that the Cry1Ah-resistance was partially recessive on leaves, nearly recessive on silk stage, but completely recessive on Cry1Ah maize kernel. These results show that high dose/refuge strategy is likely to be successful assuring high dose of Cry1Ah in transgenic crop. Dominance is dependent on several factors including the resistance level, toxins and specific strains30. High variability in the degree of dominance has been observed in different population9,44 and resistance is incompletely recessive in S. frugiperda even with low dose of Cry1F maize45. In contrast, some studies indicated that dominance decreased with higher levels of resistance in the selected strains29.

Toxicity assays of backcrosses (RS × RR) showed that the inheritance of resistance to Cry1Ah toxin in ACB-AhR was controlled by more than one locus. The Lande’s method to estimate number of loci involved in resistance is based on the expected mortality of offspring from the RS × RR backcross at each dose of toxin. This test of monogenic model analyzing the response of backcross progenies provided confirmation that more than one locus contributes for Cry1Ah resistance in ACB-AhR. The null hypothesis tested in the standard backcross method is that resistance is controlled by one locus with two alleles46. If the resistance had been controlled entirely by one locus with two alleles, the sole RR allele would have been fixed in several generations of selection and further increase in resistance would not have occurred47. The genetic basis for resistance to Bt toxins appears to be different in different insect species and different selection regimes. Resistance to Cry1Ac was controlled primarily by one or a few major loci in a field-derived strain of pink bollworm47. Furthermore, resistance to Cry1Ab resulted in polygenic control in Cry1Ab-selected Asian corn borer (ACB-AbR), but in Cry1Ac- selected Asian corn borer (ACB-AcR) resistance was controlled by a single locus31. A field-derived strain of diamondback moth fit best to polygenic inheritance to Cry1Ac48. Resistance to Cry1F in a population of O. nubilalis, the best fit happened with a single locus or a set of tightly linked loci49, and resistance to Cry1Ab in a population from the Philippines revealed contributions from two different genes50. Besides, resistance to Cry1Ac toxin in H. armigera initially was monogenic, but showed better fit to polygenic control as resistance increased29.

ACB-AhR laboratory-selected strain showed increased survival on Cry1Ah expressing leaf and silk tissues, although the surviving level was low in kernel tissue of transgenic maize. The Cry1Ah resistance was primarily autosomal, no maternal effects and under polygenic control. Dominance varies depending on toxin concentrations but resistance to Cry1Ah in ACB-AhR was functionally recessive in Bt-maize. These results should be useful in dealing the development of resistance management strategies, and the mechanism of inheritance and cross-resistance will be valuable to minimize the negative effects of cross-resistance on the durability of Bt toxins. Pyramided Bt crops that express two or more toxins that protect against same insect pest is considered a key tactic deployed to delay evolution of pest resistance to succeed in reducing refuges. One of the important components favoring the durability of pyramided Bt crops is that these plants express Cry proteins that lack cross-resistance. Our results show that ACB–AhR has high cross resistance with Cry1Fa toxin, indicating that careful studies should be done to determine cross-resistance since the absence of cross-resistance between two toxins derived from a resistant strain selected against only one-toxin could be not enough to make a decision to develop pyramided Bt crops. This should be taken in account for future development of genetically modified pyramided Bt crops.

Materials and Methods

Ethics statement

For insect collection in maize field, no specific permits were required by authorities.

None of the species used in this study are endangered or protected.

Insect strains

Susceptible and resistant strains of O. furnacalis used in the present study were obtained from laboratory colonies. A susceptible strain of Asian corn borer (ACB-BtS) was originally collected from corn fields in Liaoning Province. This population was established in the laboratory by rearing on artificial diet without exposure to insecticides for 23 generations, at the Institute of Plant Protection (IPP), Chinese Academy of Agricultural Sciences (CAAS), Beijing. A Cry1Ah-resistant strain (ACB-AhR) was established by 90 pairs of male and female larvae, collected from the fields in the Shaanxi province located in summer maize region of central China. The resulting offspring were used to establish a laboratory strain that was maintained using standard rearing techniques under selection pressure with increasing concentrations of Cry1Ah protein incorporated into artificial diet for several generations. The concentrations initially was 0.05 μg/g (Cry1Ah toxin/diet), and was increased to 0.1 μg/g in the 2nd generation, 0.2 μg/g in the 3rd generation, 0.4 μg/g in the 4th generation, 0.8 μg/g in the 5th–7th generation, 2.5 μg/g in the 8th–10th generation, 4.0 μg/g in the 11th–13th generation, and 6.0 μg/g in the 14th–23th generation. The resistance characteristics were tested at the 48th generation. Larvae were reared at 27 ± 1 °C, 70–80% relative humidity (RH), with a photoperiod of 16:8 h light: dark (L:D).

Bt toxins

Cry1Ah expressed in Bt subsp. kurstaki strain Cry1Ah (HD-73) was trypsin-activated and used for selection and bioassays. Trypsin-activated Cry1Ab, Cry1Ac and Cry1F (98% pure protein), were produced by Marianne P. Carey, Case Western Reserve University, USA. Cry1Ie expressed as a recombinant protein in E. coli 51, was provided by the Biotechnology Group of Institute of Plant Protection, Chinese Academy of Agricultural Sciences. Cry1Ie protoxin was purified via NiNTA affinity chromatography and Superdex- 75 size-exclusion chromatography. These samples were analyzed by SDS-PAGE (8%), and the protein concentrations were determined (Universal Hood II, Bio-Rad, USA) using bovine serum albumin (BSA) as a standard. Test solutions of toxins were freshly prepared in distilled water with sodium carbonate in 50 mM sodium carbonate buffer pH10.

Quantification of Cry1Ah toxin expressed in Bt maize tissues

To determine the Bt proteins concentration in the corn leaves, silk and kernel tissues, 10 samples from each tissue were taken randomly. The leaf, silk and kernel tissues were diluted at a rate of 1:10 (milligram sample: microliter PBST buffer) and fully ground by mortar and pestle. The concentration of Cry1Ah in maize leaves, silks, and kernels was confirmed by enzyme-linked immunosorbent assay (ELISA) with Cry1Ah detection kits as per the protocol of manufacturer (Shanghai MLBIO Biotechnology Co. Ltd.).

Plant tissue bioassays

ACB neonate larvae were tested on the leaves, silks and kernels of transgenic maize (Bt-21) expressing Cry1Ah protein and a non-Bt control. Resistance of ACB-AhR to Bt transgenic maize was evaluated with leaves, silks and kernel to homozygous parental strains (ACB-BtS & ACB-AhR and F1 progenies). Hybrid F1 progenies were derived from reciprocal crosses between susceptible and resistant strains. To generate the F1 progeny, corn borer pupae were separated by gender, and 150 Cry1Ah-selected females were pooled with 150 control males, and 150 control females were pooled with 150 Cry1Ah-selected males in mating cages. The leaves were cut into small pieces with a pair of scissors and placed in individual wells of a 24-well rearing plates. One neonate larva was infested in each well, and then covered with a piece of moistened filter paper and lid. Plates were then incubated at 27 ± 1 °C, 70% RH and a photoperiod of 16:8 (L:D). The number of surviving larvae was recorded either daily or every 2 days, and fresh tissue was provided when required. The bioassays were evaluated in terms of insect survival. Each strain was assayed with 120 larvae and replicated three times. Silk tissue from 10 plants of Bt and non-Bt were collected and were placed into a plastic container with a disc of wet filter paper at bottom. Ten larvae (<12 h after hatching) were placed inside each container using a fine brush, and then covered with a piece of moistened filter paper and lid. Kernels were collected from field and were fast frozen with liquid nitrogen. 15–20 grains of kernel from Bt and non-Bt plants were placed into a container and were infested with 10 larvae. All rearing trays and containers were kept in an incubator 27 ± 1 °C with a photoperiod of 16:8 h (L:D) and 80% RH. The number of surviving larvae were recorded either daily or every 2 days, and fresh tissues were provided when necessary.

Diet bioassay

We used survival bioassays to evaluate susceptibility to Cry1Ah of the susceptible strain, the resistant strain, and progeny of mass crosses. Bt test solutions were serial diluted in water; we tested 6 to 9 concentrations from 0.2–12.5 μg/g (toxin/diet) for ACB-BtS and 5–800 μg/g (toxin/diet) for ACB-AhR. Dilutions were added to an agar-free semi-artificial diet to form a testing medium52, which was then distributed into individual cells of 48-well trays. One neonate ACB larvae (<12 h after hatching) was placed in each well and plates were concealed with adhesive plate sleeves. The larval trays were placed in a climate controlled room 27 ± 1 °C with a photoperiod of 16:8 h (L:D) and 80% RH for 7 days and checked for survival. Those larvae that were visibly moving or that move when poked with a pin were considered surviving larvae. Bioassays were replicated three times in each experiment including a negative control that did not contain any Bt toxin. Cross-resistance bioassays with Cry1Ab, Cry1Ac, Cry1F and Cry1Ie were conducted using the methods defined above.

Inheritance experiments

To evaluate sex linkage and dominance the influence of sex on Cry1Ah inheritance was tested by bioassays of F1 progeny from reciprocal mass crosses between resistant and susceptible strains. The power of indirect tests for modes of inheritance is higher when the backcross progeny are originated from crosses between F1 progeny and the parental strain which is more dissimilar in susceptibility to the toxin46. In the reciprocal mass cross between resistant and susceptible strains, 100 resistant female pupae were pooled with 100 susceptible male pupae in one cross. In another reciprocal cross, we pooled 100 resistant male pupae with 100 susceptible female pupae. The estimation of number of genes involved in inheritance of Cry1Ah resistance was carried in bioassays of backcross progenies. Males and females of each F1 were backcrossed to resistant parental strains to produce four backcross populations and tested for susceptibility as described above.

Statistical analysis

For bioassays with multiple concentrations, probit regression using POLO-PC (LeOra Software 1987) was used to estimate the LC50 with 95% fiducial limits (FL), as well as the slopes of the concentration-mortality lines and their standard errors, Chi-square (χ2) values and resistance ratio (RR). Resistance ratios were calculated by dividing the LC50 for a particular strain by the LC50 for the susceptible strain. The data of F1 reciprocal crosses between ACB-AhR and ACB-BtS were also analyzed with the equality and parallel tests using Polo Plus.

To calculate dominance, the responses of F1 progeny were compared with parental susceptible and resistant strains. Dominance of Cry1Ah resistance at the different toxin concentrations was measured as reported by Bourguet, et al.53 and methods adapted from Liu and Tabashnik 199754. The values of degree of dominance (h) range from 0 that indicates completely recessive to 1 that indicates a fully completely dominant resistance and h value of 0.5 defines co-dominant or additive trait. The characters used in the calculation of dominance were larval survival after 7 days of infestation and the combination of the two traits.

The backcross generation obtained from mating F1 progenies with parental resistant strains was verified to estimate the number of genes affecting resistance by using the approach described by Tabashnik, et al.10 and using the method reported by Lande55. The monogenic inheritance model was tested to compare observed and expected mortality of the backcross progeny at different Cry1Ah concentrations by using the Chi-square test10,46. Expected mortality in the backcross can be calculated directly from experimental data following the method as described previously32,46. The survival percentage of larvae fed on Cry1Ah-corn and non-Bt corn was arcsine transformed to assure the normality of data prior to statistical analysis. The means difference in each corn tissue of Bt and non-Bt were tested using Fisher’s protected least significant difference (SAS Institute, Cary, NC. 2009).