Characterization of the resistance to Vip3Aa in Helicoverpa armigera from Australia and the role of midgut processing and receptor binding

Crops expressing genes from Bacillus thuringiensis (Bt crops) are among the most successful technologies developed for the control of pests but the evolution of resistance to them remains a challenge. Insect resistant cotton and maize expressing the Bt Vip3Aa protein were recently commercialized, though not yet in Australia. We found that, although relatively high, the frequency of alleles for resistance to Vip3Aa in field populations of H. armigera in Australia did not increase over the past four seasons until 2014/15. Three new isofemale lines were determined to be allelic with previously isolated lines, suggesting that they belong to one common gene and this mechanism is relatively frequent. Vip3Aa-resistance does not confer cross-resistance to Cry1Ac or Cry2Ab. Vip3Aa was labeled with 125I and used to show specific binding to H. armigera brush-border membrane vesicles (BBMV). Binding was of high affinity (Kd = 25 and 19 nM for susceptible and resistant insects, respectively) and the concentration of binding sites was high (Rt = 140 pmol/mg for both). Despite the narrow-spectrum resistance, binding of 125I-labeled Vip3Aa to BBMV of resistant and susceptible insects was not significantly different. Proteolytic conversion of Vip3Aa protoxin into the activated toxin rendered the same products, though it was significantly slower in resistant insects.

confer higher protection and delay insect resistance (http://www.epa.gov/oppbppd1/biopesticides/pips/pip_list. htm). However, a significant threat to Bt-based insect control is the potential development of insect resistance that could jeopardize their-long term success. Resistance to Vip3Aa has already been selected in laboratory colonies of at least three heliothine species. Selection for Vip3Aa resistance in Heliothis virescens over 13 generations resulted in insects with a resistance ratio of 2040-fold relative to susceptible insects 18 . Of more concern, F 2 screens detected frequencies of alleles conferring resistance to Vip3Aa as high as 0.03 and 0.01, respectively, in field populations of H. armigera and H. punctigera from Australia that had not been exposed to plants expressing the toxin 19 .
Vip3A is an intestine specific virulence factor; after being ingested, the proteins are processed by the insect midgut proteases [20][21][22][23] . The major proteolytic products of Vip3Aa are approximately 62 and 20 kDa fragments that are inseparable in size-exclusion chromatography 24 . The processed protein binds to its specific receptors in the midgut epithelial brush border membrane and forms pores 20,21,24 . Although both Vip3 and Cry proteins need to be activated and bind to membrane receptors to exert their toxic action, the two proteins display different levels of stability and processing rates in the insect midgut 23 and bind to different specific receptors in the BBMV of the susceptible insects [24][25][26] . These differences in the mode of action are thought to be responsible for the absence of cross-resistance to Vip3 proteins observed in Cry-resistant insects from all insect species tested [27][28][29][30][31][32][33][34] . However, studies on cross-resistance to Cry proteins in Vip3-resistant insects are lacking.
The mechanisms underlying resistance to the Cry proteins have been studied in some field resistant populations and laboratory selected populations. In many cases the gene and mutation responsible has been identified. For Cry1Ac, a number of mutations in resistant individuals have been identified as responsible for phenotypic resistance 35 and recently genes involved in Cry1Ca resistance in Spodoptera exigua and in Cry2Ab resistance in H. armigera were also isolated 36,37 . In addition, gene expression alterations have also been characterized in some other cases 38,39 . Most cases of insect resistance to Cry proteins reported to date belong to one of the sequential steps proposed for their mode of action: impaired proteolysis activation 40,41 or decreased binding to midgut receptors 42,43 . However, there are a few cases of resistance to Cry proteins that could not be associated with either impaired activation or decreased binding [44][45][46] .
The first resistance alleles to Vip3A in field populations of Australian H. armigera were isolated in 2009. Pooling F 2 screen data across 2009/10 and 2010/11 yielded an r frequency for H. armigera of 0.027 (28 positive lines, 273 tested lines) with a 95% CI between 0.019 and 0.038. Complementation tests involving crosses of the first two isolates (SP85 and SP477) demonstrated that the F 1 progeny were also resistant to Vip3A, implying that the resistance in both isolates is due to alleles at a common locus. Characterization of these early isolations showed that the resistance to Vip3Aa is recessive and maps to a locus different from that conferring resistance to Cry2Ab 19,33 . Herein, in addition to providing up-to-date information on frequencies of resistance alleles in H. armigera field populations and further information on allelism among different resistant families, we provide one of the first demonstrations of a lack of cross-resistance to Cry1Ac and Cry2Ab in a Vip3Aa resistant colony. These studies provide important context to a further investigation reported herein which examines the possible role of Vip3Aa processing and binding to midgut receptors as mechanisms of resistance.

Results
Characteristics of Vip3Aa resistance. Complementation tests involve crossing a standard resistant colony with the new isolate and then testing the offspring by exposing them to the discriminating dose of the toxin. If the offspring survive the discriminating dose of Vip3Aa, then the resistance in each colony is due to the same mutation or variants (alleles) at the same gene which implies a common mechanism. If a complementation test performed on a new resistant isolate is negative (the offspring of a cross between it and the standard colony fail to survive), it is likely that different genes are involved in conferring resistance.
Previously we reported data for two isolations of H. armigera which were allelic with SP85 -here we test an additional four isolations from two seasons of monitoring (2011/2012 and 2012/2013) using F 2 tests. Three of these Vip3Aa isolations were found to be clearly allelic to the resistant laboratory line SP85 (11-1112, 12-2602, 12-2998). In one line  there was substantial mortality from Vip3Aa (~60%) in the offspring from the crosses to SP85 when compared to the control. This could be explained if another gene is involved in conferring the resistance or the tested individuals were heterozygous which would produce approximately 50% mortality. Unfortunately it was not possible to maintain the 11-2201 line to investigate this issue further. However, in all cases the survival rates are greater than would be expected if no resistance allele was present (p = < 0.001 χ 2 = 171432). This result supports the notion of a relatively common mechanism for Vip3Aa resistance in field populations of H. armigera in Australia, and justifies using F 1 screens to estimate SP85-like Vip3Aa resistance frequencies (Table 1). However, it would be prudent to continue to perform some F 2 screens to track whether resistance involving other potential mutations increases in frequency after the deployment of plants expressingVip3Aa.
Mahon et al. 19 reports frequencies of Vip3Aa resistance alleles for 2009/10 and 2010/11 based on F 2 screens. Here we report F 2 screen data from 2011/12 to 2012/13, and F 1 screen data from 2013/14 and 2014/15. This reflects a shift in the approach used for resistance monitoring. Since the allelism data show one common form of Vip3Aa resistance, in 2013/14 we shifted our focus to the common resistance using the more efficient F 1 screen (this shift is outlined in more detail in Walsh et al. 33 ).  and 0.016 with a 95% CI between 0.007 and 0.025 (10 positive lines, 313 tested lines) respectively; there is no statistically significant difference between these approximations (Fisher's Exact test, P < 0.05). Summed across both years the estimated r frequency for Vip3Aa in H. armigera based on F 1 screens is 0.013 with a 95% CI between 0.006 and 0.019 (16 positive lines, 634 tested lines).
As part of F 2 screens performed during the monitoring program from 2009/10 to 2012/13 we examined cross-resistance against Cry1Ac and Cry2Ab by screening isofemale families of H. armigera that scored positive for carrying a resistance allele for Vip3Aa. Table 2 summarizes these data and shows that none of the randomly selected 16 families examined showed a greater propensity for survival against Cry1Ac and Cry2Ab toxin than did a Vip3Aa susceptible laboratory colony. The sample is representative of the 54 families of H. armigera that scored positive for carrying a resistance allele for Vip3Aa. We therefore conclude that larvae resistant to Vip3Aa are not cross-resistant to Cry1Ac of Cry2Ab.
Vip3Aa processing with midgut juice of susceptible and resistant H. armigera. Since Vip3Aa is found in the protoxin form in cotton leaves 47 , we searched for differences in its conversion to the activate form between the susceptible and resistant insects. When midgut juice of GR and SP85 was incubated with Vip3Aa protoxin, many proteolytic products were obtained but no difference in the band profile between the two colonies was observed (Fig. 1a,b). The major proteolysis products were the 62 and the 20 kDa fragments in both cases. The kinetic analysis of the 89 kDa activation and the 62 kDa fragment formation showed a difference in the processing rate between the susceptible and the resistant H. armigera colonies (Fig. 1c). The processing of the 89 kDa protoxin was faster in the susceptible colony. After 15 min the protoxin completely disappeared with the midgut juice from the susceptible insects, however, with SP85 there was 31% residual protoxin which was completely activated after 60 min incubation.  Table 2. A sample of isofemale lines generated from F 2 screens that were confirmed to be homozygous resistant for Vip3Aa resistance, and their responses to Cry1Ac and Cry2Ab toxin in the F 3 generation. Assays were performed on neonates. After 7 days they were scored as being alive and at least 3 rd instar, or dead or not at 3 rd instar.

Discussion
In Australia monitoring for resistance to Vip3Aa in field populations of H. armigera has been ongoing since 2009 which enabled isolation of resistant alleles and development of colonies with these genes in the laboratory. These colonies will help with the understanding of Vip3Aa resistance in this global pest. This is timely because despite our ability to detect resistance alleles, until recently, H. armigera was not exposed to significant selection pressure by Vip3Aa. However, with the recent incursion of H. armigera into the New World 2 there is enormous potential selection for resistance primarily due to the large areas of corn expressing Vip3Aa proteins to control the closely related Helicoverpa zea and other lepidopteran corn pests. The bioassays performed herein to characterize Vip3Aa resistance support previous research which demonstrates that resistance alleles can readily be detected in Australian field populations despite no obvious selection (for more detail see Mahon et al. 19 ). This relatively high baseline level of resistance may reflect selection at a low level from naturally occurring Vip3 toxins and/or direct or indirect (e.g. linkage) selection to something other than Vip3Aa. Our results suggest that natural variation    exists in insect populations that could drive resistance once crops expressing Vip3Aa are introduced unless appropriate resistance management measures are implemented (mean expression levels of Vip3Aa in cotton plants is 25 μg/g dry weight, not substantially different from the discriminant dose used in our study) 47 . The frequency of Vip3Aa resistance alleles in Australian populations of H. armigera has not increased significantly in the six seasons that monitoring has taken place, the last four of which are presented herein. F 2 screens were performed from 2009/10 until 2012/13 and F 1 screens were performed in 2013/14 and 2014/15. This is not surprising given that products expressing Vip3A have yet to be commercialized. Interestingly, the estimates obtained using F 2 screens, which can yield false negatives at least for Cry2Ab (S. Downes, unpublished data), are substantially higher than those obtained using F 1 screens. It is possible that the different r frequencies from these methods reflect actual changes in frequencies over time but unlikely given the absence of Bt-crops that could impact on selection for Vip3A resistance (although see above for alternative explanations). Regardless of the reason(s), our data verify that Vip3A resistance alleles exist at relatively high frequencies, and are not rising. This is in contrast to Cry2Ab which has been more variable. In 2010-2012 in particular, Cry2Ab resistance in H. armigera doubled compared to the baseline which could have signaled the beginning of a significant resistance problem in response to cotton expressing this toxin 48 (see also Downes et al. 11 for a similar response in the closely related H. punctigera). In the 2014/15 season, the frequency for Cry2Ab declined to baseline levels, which reflects the variability of the presence of this resistance allele, and Cry1Ac resistance remains rare (Downes, unpublished data). For the Vip3Aa alleles detected to date using F 2 screens, cross-resistance to Cry2Ab and Cry1Ac was assessed and not identified. This is one of the first demonstrations of a lack of cross-resistance to Cry proteins in a Vip3Aa resistant colony, in contrast to previous studies in which Cry-resistant colonies were tested for cross-resistance to Vip3Aa.
A number of the different H. armigera Vip3Aa lines were further characterized by performing complementation tests with the first isolated Vip3Aa resistant line (SP85). In the complementation tests, three of them were found to be allelic with the lines previously identified and isolated while the results for another line were less convincing. This suggests that the mechanism present in SP85 is relatively frequent but raises the possibility of other mechanisms/genes being involved in the other F 2 isolated Vip3A resistant detections. In the case of Cry1Ac and Cry2Ab resistance in H. armigera, multiple alleles were identified in the same gene which had the same phenotypic effect 37,49 . The fact that the majority of resistant lines tested were allelic allows us to characterize the mechanism of Vip3Aa resistance in a single line and gives us more confidence in extrapolating the findings to the whole population.
The Vip3Aa protein is produced in planta as a full length protein of ca. 89 kDa and its purification from cotton leaves indicates that the protein is stored in its protoxin form 47 . Upon ingested by the insect larva, the Vip3Aa protoxin is cleaved by serine proteases to several fragments, with two main products of around 62 kDa and 20 kDa when the incubation is performed under mild conditions (reviewed by Chakroun et al. 50 ). We chose conditions that yielded the 62 kDa and 20 kDa bands as the main products of the Vip3Aa incubation with H. armigera midgut juice to search for differences between the two colonies. Although no differences were observed in the band pattern, the conversion of protoxin (89 kDa) into active toxin (the 62 kDa fragment) was faster, under the same experimental conditions, with midgut juice from the susceptible insects than from their resistant counterparts (Fig. 1). It is difficult to evaluate how this difference in the activation rate may contribute to the resistance to Vip3Aa. In some cases, the kinetics of the protoxin processing to the active toxin has been proposed to be one of the factors determining the potency of Vip3A proteins 22,23,51 . However, given the narrow spectrum of resistance observed, it is unlikely that a protease-based mechanism is the only factor contributing to the resistance to Vip3Aa 42,43 . On the contrary, binding site alteration is a well-documented mechanism of resistance to Cry1A and Cry2A toxins 42,43,52 . This type of alteration is very specific and cross-resistance is found only in those toxins that bind to the altered binding site.
Specific binding of Vip3A proteins to lepidopteran BBMV has been shown in several insect species using biotin-labeled Vip3Aa competed by unlabeled toxin (reviewed in Chakroun et al. 50 ), in particular, in H. armigera 53 . The use of 125 I-labeled ligands allows to increase sensitivity and to obtain quantitative results out of the binding assays. Conditions to successfully label Vip3Aa with 125 I, to perform binding analyses, were set up with S. frugiperda BBMV 24 . We have used these conditions and shown specific binding of 125 I-Vip3Aa to H. armigera BBMV (Fig. 2). Equilibrium binding parameters did not show any significant difference between insects from the two colonies (Table 3), indicating that alteration of the binding to the epithelial membrane does not seem to be the reason for the difference in susceptibility of the two insect colonies to Vip3Aa. This result is somewhat unexpected, since binding site alteration confers high levels of resistance to a very small set of structurally related toxins. Several studies have shown that Cry1A and Cry2A toxins do not share binding sites with Vip3A toxins [24][25][26][53][54][55] . The fact that Vip3Aa-resistant SP85 insects are not cross-resistant to Cry1Ac or Cry2Ab, suggests a highly specific change in the resistant insects but, similarly to other cases of resistance to Cry toxins, this change does not seem to affect binding to the epithelial membrane of the midgut 29,56,57 .
In conclusion, alleles for Vip3Aa resistance occur at a relatively high frequency in the field in Australian populations of H. armigera, despite the fact that Bt crops expressing this toxin are not yet prevalent in the agroecosystem. Complementation tests with the various alleles isolated by the F 2 test in isofemale lines indicate that all alleles identified so far are alleles of the same gene. Biochemical analyses of resistant and susceptible insects have shown no differences at the level of binding and minor differences in the activation rate of the Vip3Aa protoxin, which may or may not contribute to resistance. Since the mode of action of Vip3 proteins is not yet well understood, further study with resistant insects may shed light on specific targets of Vip3A proteins which so far are not known.
Scientific RepoRts | 6:24311 | DOI: 10.1038/srep24311 Methods Insect colonies. The H. armigera Vip3Aa resistant colony used in this experiment (SP85) is described in detail elsewhere 19 . Briefly, it is a laboratory colony which was isolated using an F 2 screen during the summer of 2009-10 from individuals collected as eggs on non-Bt cotton from St. George, Queensland, Australia. This resistant colony was outcrossed five times to the susceptible laboratory colony (GR) to maintain fitness and to produce a colony that was 96.2% isogenic with the susceptible colony. Following each outcross, the colony was reselected with levels of toxin that killed all genotypes except those that were homozygous resistant to Vip3A. All subsequent generations were selected at this dose. The assays reported here were performed with individuals from the 3rd to the 5th generation. However, most of the analyses were conducted with assays on the near-isogenic 5 th outcross in order to reduce the potentially misleading effects of hybrid vigor that may be evident when crossing colonies of H. armigera.
Individuals from the SP85 colonies survive the maximum concentration of Vip3Aa toxin that can be practically delivered in a surface treatment assay (220 μg/cm 2 ) and larvae develop at the same rate as siblings reared on non-treated diet. Resistance for this colony is essentially recessive, with heterozygotes exhibiting concentration-response characteristics that are similar to those of susceptible insects 19 . Reciprocal backcrosses of heterozygotes to resistant colonies produced results for concentration-response assays which confirmed that resistance is essentially recessive -that is, 50% of offspring are homozygous resistant while the remainder are heterozygous and thus phenotypically susceptible 19 . These data are also consistent with the hypothesis that resistance is conferred by a single gene.
The GR colony used in our assays is susceptible to Vip3Aa, Cry1Ac and Cry2Ab toxins. This susceptibility is monitored regularly. The susceptible colony was employed during every screen to verify that a correctly administered discriminating concentration of toxin-containing material was applied. It has been in culture since the mid-1980s and is derived from material collected from cotton fields in the Namoi Valley, northern NSW Australia. On occasions it has been supplemented with additional collections from the same area that were screened for resistance and found to be susceptible.

Source of toxins.
Bioassays to characterize Vip3Aa resistance. A Vip3Aa clone in E. coli was used as a source of toxin. Production and calibration of the Vip3Aa toxin was described elsewhere 19 .
Cry1Ac toxin was produced by the HD-73 strain of B. thuringiensis var. kurstaki (producing only the Cry1Ac toxin and spores). Mass production via fermentation of HD-73 was performed by Genesearch (Brisbane, Australia) with a resulting spore/crystal mix. The pellets produced were resuspended and washed three times before use. The extract was used without activating the toxin by trypsin treatment.
Dried and ground corn leaf material was used as a source of Cry2Ab toxin. This corn powder was provided by Monsanto (US) as a lyophilized Zea mays leaf powder containing transgenically expressed B. thuringiensis crystal protein, Cry2Ab2 at a concentration of 6 mg/g of powder.
Biochemical tests. The E. coli BL21 expressing Vip3Aa16 58 used for the biochemical tests was kindly supplied by Dr. Slim Tounsi, CBS (Sfax, Tunisia).
Bioassays to characterize Vip3Aa resistance. Whole organism bioassays were conducted in 45 well (2.7 cm 2 ) trays which contained approximately 2 ml of rearing diet that was overlaid with an aqueous solution of toxin and allowed to air dry. Concentrations were calculated as μg of toxin per cm 2 of diet surface. After the addition of one neonate larvae per well, trays were heat sealed and maintained at 25 °C and 45-55% RH. Each bioassay consisted of a control (diet with no toxin), plus one toxin concentration. The concentration used was 10 μg of toxin per cm 2 . After 7 days, the larvae were scored as "alive" (exhibiting normal movement) or "dead" (dead, moribund, uncoordinated movement). The mortality of neonates in controls was minimal for all assays (mean mortality 4.1 ± 5%, range 0-11%, n = 242 neonate larvae in 6 control assays).
Allelism of different isolations of Vip3Aa resistance. Complementation tests were performed after spending 2 to 5 generations (include the two-generation F 2 tests) in the laboratory. They involved setting up reciprocal crosses between new Vip3Aa-resistant colonies and the SP85 colony. To determine if the characteristics of the captured alleles were similar to those of SP85, the response to a discriminating concentration of toxin in bioassays was determined for the progeny from the above cross and from the parental colonies (SP85, the new resistant colony, and GR). Forty five insects were normally tested in the control per test and the same number exposed to 10 μg/cm 2 Vip3Aa in 45 well trays (control = 242, tested = 270 in 6 toxin bioassays).
Current frequencies of Vip3Aa resistance. Assays to identify resistant insects included F 2 and F 1 screens that were conducted using published protocols 12 . We aimed to expose 90 neonate larvae to Vip3Aa toxin for each line.
(i) F 2 method. Eggs collected from field hosts of H. armigera were reared to pupae. On emergence, single male and female moths were placed in individual 850 ml plastic containers with a dilute honey solution. Eggs laid on the gauze opening of the container were collected every 1-2 days. If they were fertile, around 135 hatchings were reared to establish isofemale lines. On pupation, individuals were sexed and equivalent numbers of males and females were placed in a 5 litre container and allowed to mate.
Scientific RepoRts | 6:24311 | DOI: 10.1038/srep24311 F 2 offspring generated from these parents were challenged with a discriminating dose. If either field-collected insect carried a 'resistant allele' , we would expect at least 6.25% of the toxin-exposed larvae to be homozygous for that allele and thus survive and grow to at least 3 rd instar by day 7 59 .
(ii) F 1 method. This technique makes use of colonies of resistant insects in a similar fashion to that used by Gould et al. 60 to determine the frequency of resistance in H. virescens. Field-collected eggs were reared to pupae and male and female pupae were placed in groups in separate cages. As moths emerged, a male was placed in an 850 ml container with two virgin SP85 females. Similarly, a female was placed in an 850 ml container with two SP85 males.
If fertile eggs were obtained from such crosses, F 1 offspring were exposed to a discriminating dose. If the field-derived individual tested in this process was heterozygous for resistance, we would expect approximately 50% of the larvae to be homozygous for resistance and therefore to thrive. In the unlikely event that we collected and tested homozygotes from the field, the frequency of survivors would be close to 100%.
Cross-resistance to Cry1Ac and Cry2Ab in Vip3A resistant colonies. We present data for a sub-set of isofemale lines that were confirmed to be homozygous for alleles conferring resistance to Vip3A toxin and were challenged in the F 3 generation as neonates against Cry1Ac and Cry2Ab. The discriminating concentration for Cry1Ac was 0.25 μg/cm 2 of Cry1Ac delivered in a 50 μl/well solution. After 7 days this concentration killed 95.7 ± 1.8% of a susceptible general rearing colony (n = 628 larvae in 10 assays conducted over 7 days) and no surviving larvae grew beyond 2 nd instar. The discriminating concentration for Cry2Ab was 1 μg/cm 2 of Cry2Ab delivered in a 50 μl/ well solution. After 7 days this concentration killed 99.6 ± 0.4% of a susceptible general rearing colony (n = 286 larvae in 6 assays conducted over 7 days) and no surviving larvae grew beyond 3 rd instar.
Vip3Aa purification for biochemical analyses. Conditions for bacterial culture and expression of the Vip3Aa16 protein was described previously 22 . For proteolysis assays the expressed Vip3Aa was purified from an E. coli cell lysate using a HisTrap FF affinity purification column (GE Healthcare) following the manufacturer instructions. Fractions of 1 ml were eluted from the column and collected in tubes containing 50 μl of 0.1 M EDTA. The most concentrated fractions were pooled and dialyzed against 20 mM Tris, 150 mM NaCl, pH 9, before storage at −20 °C.
For binding assays, Vip3Aa protein was purified as described previously 24 . In brief, the Vip3Aa in the E. coli cell lysate was precipitated adjusting the pH to its isoelectric point using acetic acid. The precipitated protein was recovered in the pellet after centrifugation, dissolved in 20 mM Tris-HCl, 150 mM NaCl, pH 9, and treated with 1% trypsin for 2 h at 37 °C. The protein that was used for labeling was further purified by anion-exchange chromatography in an AKTA explorer 100 system (GE Healthcare, UK).

Midgut juice preparation.
Midguts from ten 5 th instar larvae of Vip3Aa resistant (SP85) and susceptible (GR) H. armigera colonies reared on standard diet were dissected and the peritrophic membrane extracted with its bolus content, which was then homogenized and centrifuged for 10 min at 16000 g. The supernatant was collected and distributed in small aliquots, flash frozen in liquid nitrogen and stored in −80 °C. Total protein concentration in the midgut juice was quantified with Bradford reagent using BSA as standard.
Proteolytic processing of Vip3Aa. Proteolytic processing of Vip3A protoxin by the midgut juice of the susceptible (GR colony) and resistant (SP85 colony) H. armigera was first performed with different midgut juice dilutions to select the optimal dilution to perform the kinetic study. To compare the kinetics of Vip3Aa activation by the midgut juice of the susceptible and resistant H. armigera, 50 μg of affinity-purified protoxin was incubated with midgut juice at 1/250 (w/w, midgut juice: Vip3Aa) in 70 μl final volume of 20 mM Tris, 150 mM NaCl, pH 9, and incubated for 5, 10, 15, 20, 25, 30 and 60 min at 30 °C. The reaction was stopped by adding the SDS-PAGE loading buffer and heating for 5 min at 99 °C, after which the samples were loaded in 12% polyacrylamide gel. For a quantitative comparison of the processing rate, the amount of Vip3Aa protoxin (89 kDa) and activated toxin (62 kDa) at the different incubation times was quantified densitometrically using the TotalLab 1D v 13.01 software. The densitometry values from the 89 kDa and 62 kDa bands were relativized to the input values in each gel, and the background was corrected. Graphical representation was performed using the software GraphPad Prism v 5.00. Vip3Aa radiolabeling. Trypsin-activated Vip3Aa was labeled using the chloramine-T method as previously described 61,62 . The labeled protein was separated from the excess of iodine by size-exclusion chromatography in a PD10 (GE Healthcare) column. The purity of the labeled protein was checked by analyzing the elution fractions by SDS-PAGE with further exposure of the dried gel to an X-Ray film at −20 °C. The calculated specific activity of the protein was 0.38 mCi/mg. BBMV preparation. Fifth-instar larvae of H. armigera from both the susceptible (GR) and the resistant (SP85) colony reared on standard diet were dissected and the midguts (without the bolus content) were washed in MET buffer (300 mM mannitol, 5 mM EGTA, 17 mM Tris, pH 7.5) and frozen in liquid nitrogen and preserved at −80 °C until required. Alternatively, midguts in MET buffer were lyophilized and kept at 4 °C 63 . Brush border membrane vesicles (BBMV) were prepared from the frozen or the lyophilized midguts by the differential magnesium precipitation method 63,64 , and then frozen in liquid nitrogen, and stored at −80 °C until use. The protein Scientific RepoRts | 6:24311 | DOI: 10.1038/srep24311 concentration in the BBMV preparations was determined by Bradford 65 using bovine serum albumin (BSA) as standard.
Binding assays with 125 I-labeled Vip3Aa. Prior to use, the buffer of the BBMV was changed to binding buffer (20 mM Tris, 150 mM NaCl, 1 mM MnCl 2 , pH 7.4) supplemented with 0.1% BSA. To determine the appropriate concentration of BBMV to be used for the binding assays, 125 I-Vip3Aa (1.2 nM) was incubated with increasing amounts of BBMV. An excess of unlabeled Vip3Aa was used to calculate the non-specific binding. The reaction was stopped by centrifuging the tubes at 16,000 g for 10 min at 4 °C and the pellet was washed once with 500 μl of cold binding buffer. The radioactivity retained in the pellet was measured in a model 2480 WIZARD 2 gamma counter.
Competition experiments were performed by incubating 20 μg/ml of BBMV, from both the susceptible and resistant colonies, with 1.2 nM 125 I-Vip3Aa in 0.1 ml final volume of binding buffer for 90 min at 25 °C in the presence of an increasing amount of unlabeled Vip3Aa protein. The reaction was stopped and the remaining radioactivity measured as described above. The dissociation constant (K d ) and the concentration of binding sites (R t ) were calculated using the LIGAND program 66 .