Protein-carbohydrate regulation in Helicoverpa amigera and H. punctigera and how diet protein-carbohydrate content affects insect susceptibility to Bt toxins

Journal of Insect Physiology xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Journal of Insect Physiology journal homepage: www.elsevier.com/locate/jinsphys

Protein-carbohydrate regulation in Helicoverpa amigera and H. punctigera and how diet protein-carbohydrate content affects insect susceptibility to Bt toxins ⁎

Ashley E. Tessnowa, , Spencer T. Behmera,c, Tom K. Walshb, Gregory A. Sworda,c a b c

Department of Entomology, Texas A & M University, College Station, TX 77843, USA CSIRO, Black Mountain Laboratories, Canberra, ACT 2601, Australia Ecology and Evolutionary Biology Interdisciplinary Program, Texas A & M University, College Station, TX 77843, USA

A R T I C L E I N F O

A B S T R A C T

Keywords: Nutritional ecology Helicoverpa Bt resistance Vip3Aa Cry1Ac Cry2Ab

Many animals, including insects, demonstrate a remarkable ability to regulate their intake of key macronutrients (e.g., soluble protein and digestible carbohydrates), which allows them to optimize fitness and performance. Additionally, regulating the intake of these two macronutrients enhances an animal’s ability to defend itself against pathogens, mitigate the effects of secondary plant metabolites, and decrease susceptibility to toxins. In this study, we first compared how Bt-resistant and -susceptible lines of Helicoverpa armigera and Helicoverpa punctigera regulate their intake of protein (p) and digestible carbohydrates (c). We found that there was no difference in the self-selected protein-carbohydrate intake target between resistant and susceptible genotypes of either species. We then explored the extent to which food protein-carbohydrate content altered the susceptibility of these species to three Bt toxins: Cry1Ac, Cry2Ab, and Vip3Aa. We found that H. armigera on diets that had protein-carbohydrate profiles that matched their self-selected protein-carbohydrate intake target were significantly less susceptible to Cry1Ac. In contrast, diet protein-carbohydrate content did not affect H. punctigera susceptibility to Cry1Ac. For both H. armigera and H. punctigera, susceptibility to Cry2Ab and Vip3Aa toxins did not change as a function of diet protein-carbohydrate profile. These results, when combined with earlier work on H. zea, suggest food protein-carbohydrate content can modify susceptibility to some Bt toxins, but not others. An increased understanding of how the nutritional environment can modify susceptibility to different Bt toxins could help improve pest management and resistance management practices.

1. Introduction Active regulation of protein-carbohydrate intake is widespread across organisms, including insects (Lee et al., 2002; Raubenheimer and Simpson, 1993; Behmer, 2009), mammals (Felton et al., 2009), and even slime molds (Dussutour et al., 2010). This ability to select and achieve a balanced protein-carbohydrate intake can affect fecundity (Lee et al., 2008; Maklakov et al., 2008; Roeder and Behmer, 2014), lifespan (Grandison et al., 2009; Fanson et al., 2009; Fanson and Taylor, 2012), immune responses (Siva-Jothy and Thompson, 2002; Lee et al., 2006), and influence the mode of action of plant secondary compounds (Simpson and Raubenheimer, 2001; Behmer et al., 2002; PascacioVillafán et al., 2016). Nutrient regulation in animals is best examined using the ‘Geometric Framework’ for nutrition (GFN), as originally outlined by Simpson and Raubenheimer (1993), and more recently updated (Simpson and Raubenheimer, 2012). The strength of the GFN

⁎

is that it models how organisms navigate through a heterogeneous nutritional landscape to achieve an optimal balance of multiple nutrients. This optimal balance – called an intake target (IT) – is usually species-specific (Behmer and Joern, 2008; Behmer, 2009). Caterpillars, or larvae from the order Lepidoptera, are a classic example of an insect herbivore that exhibits strong protein and carbohydrate regulation (Behmer, 2009; Deans et al., 2015). Many species of caterpillars are considered agricultural pests causing massive economic losses each year, often due to their targeting of specific plant tissues. Importantly, these tissues are highly variable in their macronutrient composition, even within a single plant (Machado et al., 2015; Deans et al., 2016b). This means that in the field, caterpillars can reach their protein-carbohydrate IT by feeding on different plant tissues that are individually sub-optimal, but collectively nutritionally complimentary. When caterpillars are able to reach their intake target, their performance is optimized (Roeder and Behmer, 2014), and their potential as a

Corresponding author at: 2475 TAMU Heep Building 412, College Station, TX 77843, USA. E-mail address: [email protected] (A.E. Tessnow).

http://dx.doi.org/10.1016/j.jinsphys.2017.07.004 Received 11 March 2017; Received in revised form 27 June 2017; Accepted 17 July 2017 0022-1910/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Tessnow, A.E., Journal of Insect Physiology (2017), http://dx.doi.org/10.1016/j.jinsphys.2017.07.004

Journal of Insect Physiology xxx (xxxx) xxx–xxx

A.E. Tessnow et al.

antimicrobials; this diet has protein and carbohydrate content of 42% and 30%, respectively (on a dry weight basis). After 6 days, 3rd-instar larvae were moved into individual, diet-coated wells of a 32-well rearing tray until pupation. In total, 40 pupae were transferred to a 10 L bucket to eclose and mate. Moths had constant access to a 10% honey solution, and eggs were collected daily. All insects were maintained at 25 °C with a 14:10 (L:D) cycle.

pest increases (Deans et al., 2016a). Since their commercial introduction in 1996, Bt transgenic crops expressing insecticidal proteins have been widely adopted as a control measure for many caterpillars (Tabashnik et al., 2013). However, the evolution of resistance continues to be a major concern for the sustainability of these transgenic technologies (Gould, 1998; Downes et al., 2007; Dively et al., 2016). Typically, resistance is defined as a genetically-determined change in susceptibility (Tabashnik et al., 2014), but this definition ignores all aspects of an insect’s ecology that could reduce an insect’s susceptibility. Although genetic mechanisms underlie many forms of resistance, insect nutritional ecology can also play a role in modulating the susceptibility to a toxin. For example, Deans et al. (2016a) proposed a phenotypic plasticity model that suggests the nutritional environment could play a role in mediating the insects’ susceptibility to Bt. Recent data on Helicoverpa zea supported this hypothesis (Deans et al., 2017). Compared to caterpillars on nutritionally suboptimal protein-carbohydrate diets, those on diets that had proteincarbohydrate ratios that matched their self-selected protein-carbohydrate intake target showed a 100-fold reduction in susceptibility. In the current study, we examined how nutritional intake alters the susceptibility of two other Helicoverpa species, H. armigera and H. punctigera, that both have a history of quickly evolving resistance to synthetic insecticides (McCaffery, 1998; Kranthi et al., 2002; Downes et al., 2007). First, we empirically determined the self-selected intake target for susceptible and resistant lines of H. armigera and H. punctigera to assess whether genetic resistance was associated with a change in nutritional intake, as has been suggested elsewhere (Shikano and Cory, 2014a). Next, we conducted neonate dose response assays with three Bt toxins (Cry1Ac, Cry2Ab, and Vip3Aa). Here the insects were allowed to feed on either a diet containing a p:c ratio that matched each species’ self-selected intake target, or on a suboptimal diet with a nutritional profile that mimicked a common carbohydrate-biased commercial rearing diet. Understanding how food protein-carbohydrate content impacts the responses of a range of species to Bt toxins has implications for monitoring and managing the evolution of resistance (Shikano and Cory, 2014b; Orpet et al., 2015; Deans et al., 2017).

2.2. Artificial diets All experiments were conducted using an artificial diet described by Ritter and Nes (1981), with modifications by Jing et al. (2013). A total of 9 diets were used that differed in their ratio of soluble proteins (p) and digestible carbohydrates (c) for both experiments. These diets were made by altering the amounts of casein and sucrose in the diet, while maintaining the same concentrations for all other ingredients. All diets had a total macronutrient concentration (p + c) of 42%. 2.3. Intake target choice test Choice tests were used to determine the self-selected protein-carbohydrate intake target (p:c IT) for 5 strains of Helicoverpa armigera, and 2 strains of Helicoverpa punctigera. These strains differed in their resistance to three common Bt toxins (Cry1Ac, Cry2Ab, and Vip3Aa; Table 1). In this paper, the H. armigera susceptible, Cry1Ac resistant, Cry2Ab resistant, Vip3Aa resistant and Cry2Ab/Vip3Aa dual resistant strains are henceforth referred to as Ha-Susceptible, Ha-Cry1Ac, HaCry2Ab, Ha-Vip3Aa and Ha-Dual, respectively. The H. punctigera strains are similarly referred to as Hp-Susceptible and Hp-Cry1Ac. Upon molting to their final instar, larvae were weighed and placed into a 10cm petri dish containing two diet cubes that differed in their composition of protein and digestible carbohydrates. Each insect was assigned to one of three diet pairings: (1) 35% protein, 7% carbohydrate (p35:c7) with p7:c35, (2) p28:c14 with p14:c28, (3) p35:c7 with p14:c28. The choice arenas were kept at 25 °C with a 14:10 (L:D) and the diets were checked twice daily to ensure that both diets were always available to the insects; diets were changed as needed with a minimum of every three days until the larvae reached the prepupal stage and ceased feeding. One day after the insects pupated, the pupae were weighed and sexed. For H. armigera, 10 insects per strain were placed onto each diet treatment, and two trials were run for a total of 20 larvae per strain per diet pairing. For H. punctigera, all insects were evaluated in a single trial and 20 insects per strain were placed on each diet treatment. All strains within a species were assessed simultaneously. Larvae that died during the trial, did not pupate within 20 days, or failed to move between the two diets were removed from the analyses. As a result the number of insects per diet pairing per strain for H. armigera was 14–20. A higher mortality rate was observed for H. punctigera, so the number of insects per diet pairing per strain ranged from 11–16. The total amount of protein and carbohydrates consumed was calculated as the difference between the initial and final dry mass of the diet blocks (Lee et al., 2006). The initial wet mass of the diet was converted to dry mass using linear regression.

2. Methods 2.1. Insects Bt susceptible and resistant colonies of H. armigera and H. punctigera maintained at the CSIRO laboratory in Canberra, Australia were used for these experiments (Table 1). For routine rearing, approximately 50 neonates were transferred to a 500 ml deli container coated in artificial diet, as described by Akhurst et al. (2003). This rearing diet contains soyflour, wheat germ, brewers yeast, agar, vegetable oil and Table 1 H. armigera and H. puctigera strains evaluated. We used susceptible strains of H. armigera and H. punctigera, along with multiple strains that were resistant to Bt toxins found in transgenic plants. The key characteristics of each strain are described in the cited references. Strain

CSIRO Name

Species

Resistance

Reference

Ha-Susceptible Ha-Cry1Ac

Gr ISOC

H. armigera H. armigera

Susceptible Cry1Ac

Ha-Cry2Ab Ha-Vip3Aa

SP15 HA85

H. armigera H. armigera

Cry2Ab Vip3A

Ha-Dual Hp-Susceptible Hp-Cry1Ac

DRES HPM 9.3784

H. armigera H. punctigera H. punctigera

Cry2Ab & Vip3A Susceptible Cry1Ac

Mahon et al. (2007) Bird and Akhurst (2004) Mahon et al. (2007) Chakroun et al. (2016b) Walsh et al. (2014) Walsh et al. (2014) Walsh et al. (personal communication)

2.4. Toxin production Cry1Ac protein was produced from a Bt strain HD73 as described in Akhurst et al. (2003), and Cry2Ab protein was produced from a clone of the Cry2Ab gene from B. thuringiensis variety kurstaki HD-1 in B. thuringiensis; the original clone was provided by L. Masson (National Research Council, Montreal, QC, Canada). The concentration of toxin was estimated by analyzing the density of the toxin band relative to a bovine serum albumin standard using Scion Image 1.62 software (Scion Corporation, Frederick, MD). The source of the Vip3A toxin was an E. coli 2

Journal of Insect Physiology xxx (xxxx) xxx–xxx

A.E. Tessnow et al.

Table 2 Assessment for non-random feeding in the five H. armigera strains. A positive t-stat value indicates larvae fed more on the p-biased diet. A negative t statistic value indicates larvae fed more on the c-biased diet. If there was a significant difference in at least one treatment, it was concluded that the insects were attempting to regulate their protein-carbohydrate intake. p35:c7 w/ p7:c35

Ha-Susceptible Ha-Cry1Ac Ha-Cry2Ab Ha-Vip3Aa Ha-Dual

p28:c14 w/ p14:c28

p35:c7 w/ p14:c28

t

df

P-value

t

df

P-value

t

df

P-value

3.442 1.033 0.383 3.642 2.463

18 19 18 18 13

0.003 NS NS 0.002 0.029

0.828 −0.981 −0.367 1.601 3.212

17 18 17 18 16

NS NS NS NS 0.005

−0.645 2.442 −2.475 −0.653 0.442

18 18 17 19 14

NS 0.025 0.024 NS NS

analyzed by Simple Probit Analysis in JMP®Pro 12.0.1 (SAS Institute Inc., Cary, NC).

clone; production and calibration is described in Mahon et al. (2012). 2.5. Diet incorporation dose response assay

3. Results To test whether a diet’s protein-carbohydrate profile affected the susceptibility of H. armigera and H. punctigera to the endotoxins found in Bt crops, a dose response assay using two diets with differing p:c ratios was conducted. The first diet matched each species’ intake target (IT), that was calculated in the choice test (p24:c18 for H. armigera; p23:c19 for H. punctigera). These diets are referred to as the species IT diets. The second diet matched the carbohydrate-biased p:c ratio found in common commercial rearing diets (p12:c30) for caterpillars (Deans et al., 2016a). This carbohydrate-biased diet will be referred to as the CB diet. Both the IT and CB diets contained 42% total macronutrients. Stock solutions of Cry1Ac, Cy2Ab, and Vip3Aa were removed from a −20 °C freezer and thawed; dilutions of the stock solution were prepared for incorporation into the artificial diet. For every 1 g of diet, 100 μl of the appropriate solution was thoroughly mixed/mashed into the artificial diet to achieve the desired concentration of the toxin. For the control (0 ppm) treatment, 100 μl of distilled water was added per gram of diet. For Cry1Ac there were a total of 7 toxin concentrations (0, 0.01, 0.1, 1, 10, 100, 1000 ppm) and for Cry2Ab and Vip3A there were 6 total toxin concentrations (0, 0.01, 0.1, 1, 10, 100 ppm). Approximately 0.2 g of treated diet was placed in each well of a 96 well plate. Each well was randomized with respect to diet. One newly hatched neonate was added to each well and the well plate was sealed with a heat activated adhesive lid. The plates were kept at 25 °C, 14:10 h (L:D), and larval mortality was recorded after seven days. Dose response assays for all three toxins were conducted for HaSusceptible (4 trials) and Hp-Susceptible (2 trials). An additional dose response assay using Cry1Ac was conducted on Ha-Cry1Ac (3 trials). For each trial, 16 individuals were placed on each diet treatment for each concentration of each toxin. For analysis, data for each strain was pooled. However, in some instances, neonates escaped from their well. Thus, the total sample size for each concentration varied from 50–61 larvae per diet per concentration for Ha-Susceptible, 27–32 larvae for Hp-Susceptible, and 43–48 larvae for Ha-Cry1Ac.

3.1. Choice test 3.1.1. H. armigera To assess the self-selected intake target of each H. armigera strain, insects were provided a choice between two diets that varied in their makeup of proteins and carbohydrates. Final instar larvae were placed on one of three diet pairings: (1) p35:c7c paired with p7:c35, (2) p28:c14 paired with p14:c28 and (3) p35:c7, paired with p14:c28) and allowed to feed freely between diets. All H. armigera strains fed nonrandomly on at least one diet paring indicating that the insects actively selected between the two diets. Within the p35:c7 and p7:c35 diet pairing, 3 strains (Ha- susceptible, Ha- Vip3Aa, and Ha-Dual) showed a significant preference for the p35:c7 diet choice. Ha-Dual also preferentially fed on p28:c14 over p14:c28 and both Ha-Cry1Ac and HaCry2Ab actively selected between diets within the p35:c7 and p14:c28 diet pairing with Ha-Cry1Ac showing a preference for p35:c7 and HaCry2Ab preferring p14:c28 (Table 2). Furthermore, when protein-carbohydrate regulation was compared across the three diet pairings within each H. armigera strain, no differences were observed for the HaSusceptible, Ha-Cry2Ab, Ha-Vip3Aa, or Ha-Dual strains (Table 3). However, a significant difference in protein-carbohydrate regulation was observed between the three treatment pairs for the Ha-Cry1Ac strain (Table 3). However, when protein-carbohydrate regulation was compared across the five H. armigera strains, no significant differences were observed (P = 0.204, Fig. 1); the selected p:c ratio, averaged over the 5 H. armigera strains was p1.3:c1. 3.1.2. H. punctigera Choice tests were also used to assess the self-selected intake target of two H. punctigera strains (Hp- susceptible and Hp-Cry1Ac). Here too the insects were provided a choice between two diets that differed in their protein:carbohydrate composition: (1) p35:c7 paired with p7:c35, (2) p28p:c14 paired with p14:c28 and (3) p35:c7 paired with p14:c28. The Hp-Cry1Ac strain fed non-randomly on one diet pairing, showing a

2.6. Data analysis A t-test was initially used in the choice experiments to determine if insects were feeding non-randomly. If there was a significant difference in at least one treatment, it was concluded that the insects were attempting, at a minimal level, to regulate their protein-carbohydrate intake. An ANOVA model was used to determine the effects of diet pairing (treatment) and trial on the p:c ratio consumed. This model included the ln(p:c) as the dependent variable, and either treatment or trial as the fixed factor. An ANOVA was also used to determine significant differences between strains of the same species, with the ln(p:c) as the dependent variable and the strain as a fixed factor. All t-tests and ANOVA analyses were performed using SPSStatistics, version 22.0 (SPSS Inc., Chicago, IL, USA). All dose response data were

Table 3 ANOVA results comparing the independent effects of trial and treatment on the ln(p:c) intake target for each strain of H. armigera. For this model, ln(p:c) was the dependent variable and either trial or treatment served as the fixed factor. Strain

Ha-Susceptible Ha-Cry1Ac Ha-Cry2Ab Ha-Vip3Aa Ha-Dual

3

Trial df = 1

Treatment df = 2

F-Ratio

P-value

F-Ratio

P- value

0.234 1.108 0.626 0.349 3.415

NS NS NS NS NS

2.036 8.785 0.434 3.077 2.644

NS 0.001 NS NS NS

Journal of Insect Physiology xxx (xxxx) xxx–xxx

A.E. Tessnow et al.

Table 5 ANOVA results comparing the independent effects of treatment on the ln(p:c) intake target for both strains of H. punctigera. For this model, ln(p:c) was the dependent variable and treatment served as the fixed factor. Statistics were calculated in SPSS. Strain

Treatment df = 2

Hp-Susceptible Hp-Cry1Ac

F-Ratio

P-value

0.035 0.926

NS NS

Fig. 1. Self-selected intake ratio of protein to carbohydrates for susceptible (white), Cry resistant (blue), and Vip resistant (red) strains of H. armigera. Purple indicates resistance to both Cry and Vip toxins. There was no significant difference in nutritional regulation between strains (F4 = 1.49, P = 0.204). H. armigera selected a protein to carbohydrate ratio of p1.3:c1. This intake target was calculated as the average protein-carbohydrate intake across all diet pairings (n = 46–58). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

preference for p28:c14 over p14:c28 (Table 4); in contrast, we failed to reject the null of random feeding for the Hp-Susceptible strain. In both the Hp-Susceptible and Hp-Cry1Ac strains, there was no difference in protein to carbohydrate intake across diet pairings. In other words, the average intake of proteins and carbohydrates on all three treatments (p35:c7 and p7:c35, p28:c14 and p14:c28, p35:c7 and p14:c28) was the same within each strain (Table 5). When the two strains were compared no significant difference in protein-carbohydrate intake was observed between the two strains (P = 0.103); the observed p:c ratio, averaged over the two H. punctigera strains, was p1.1:c1 (Fig. 2). 3.2. Dose response data Fig. 2. Self-selected intake ratio of protein to carbohydrates for susceptible (white) and Cry1Ac resistant (blue) strains of H. punctigera. There was no significant difference between the two strains (F1 = 2.72, P = 0.103). H. punctigera selected a protein to carbohydrate ratio of p1.1:c1. This intake target was calculated as the average protein-carbohydrate intake across all diet pairings (n = 38–43). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The LC50 concentrations to Cry1Ac, Cry2Ab and Vip3Aa for susceptible H. armigera and H. punctigera were determined against two different nutritional backgrounds: 1) a carbohydrate biased diet that matched the nutritional profile of commonly used commercial diets (CB), and 2) a diet that had a macronutrient profile that matched the self-selected protein-carbohydrate intake target (IT) for H. armigera and H. punctigera (as determined in the experiment above). For Ha-Susceptible caterpillars, the Cry1Ac LC50 on the IT diet was 1.957 ppm compared to 0.456 ppm on the CB diet. The Cry2Ab LC50 was 0.674 ppm on the IT diet compared to 0.192 ppm on the CB diet. Overall, when Ha-Susceptible neonates were exposed to these Cry proteins, LC50 values were about 4x higher on the intake target diet compared to the carbohydrate biased diet (Fig. 3a and c). For Hp-Susceptible caterpillars, the Cry1Ac LC50 was 2.311 ppm on the IT diet, but 1.418 ppm on the CB diet. The Cry2Ab LC50 was 0.860 ppm on the IT and 0.392 ppm on the CB diet. When Hp-Susceptible insects were allowed to feed at their self-selected intake target, the concentration of Cry proteins required to kill 50% of the population (LC50) was approximately 2× higher than when insects were forced to feed on a carbohydrate biased diet (Fig. 3b and d). Overall, for caterpillars reared

on diets containing Cry1Ac and Cry2Ab, LC50 values were consistently higher on the diets with protein-carbohydrate profiles mimicking the IT for each species (Fig. 3a, b, d and e). However, differences in the LC50 values were only statistically significant for H. armigera caterpillars reared on diets containing Cry1Ac (Fig. 3a; Table 2). LC50 values for Vip3Aa on the CB and IT diets were also measured for Ha-Susceptible and Hp-Susceptible caterpillars. Here, however, the lethal concentration of Vip3Aa did not differ as a function of food protein-carbohydrate profile (Table 6; Fig. 3c and f). For the Ha-Susceptible strain, the Vip3Aa LC50 was calculated as 10.012 ppm on the IT diet and 9.317 ppm on the CB diet. The Vip3Aa LC50 for Hp-Susceptible strain was 31.577 ppm on the IT diet and 23.623 on the CB diet. When delivered at the maximum dose of 100 ppm, Vip3Aa did not kill all of the susceptible individuals of either Helicoverpa species tested; between

Table 4 Assessment for non-random feeding in the two H. punctigera strains. A positive t-stat value indicates larvae fed more on the p-biased diet. A negative t statistic value indicates larvae fed more on the c-biased diet. If there was a significant difference in at least one treatment, it was concluded that the insects were attempting to regulate their protein-carbohydrate intake. p35:c7 w/ p7:c35

Hp-Susceptible Hp-Cry1Ac

p28:c14 w/ p14:c28

p35:c7 w/ p14:c28

t

df

P-value

t

df

P-value

t

df

P-value

1.483 −1.525

13 12

NS NS

−1.037 3.339

10 13

NS 0.005

−1.133 −0.709

12 15

NS NS

4

Journal of Insect Physiology xxx (xxxx) xxx–xxx

A.E. Tessnow et al.

Fig. 3. Dose response curves for susceptible H. armigera (a,b,c, n = 50–61 insects per toxin concentration per diet) and susceptible H. punctigera (d,e,f, n = 27–32 per toxin concentration per diet) on three Bt toxins; Cry1Ac (a,d,g), Cry2Ab (b,e), and Vip3Aa (c,f). The black line represents the insect responses when fed on a carbohydrate biased diet (p12:c30). The red line represents the insect’s response when fed at their intake target (p24:c18 for H. armigera and p23:c19 for H. punctigera). All dose response curves were created using simple probit analysis in JMP. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

17 and 22% of neonates survived the 7-day assay on both the IT and CB diets. Finally, we also conducted a Cry1Ac dose response assay for the H. armigera strain that was resistant to Cry1Ac (Ha-Cry1Ac). The LC50 for this strain on the IT diet was 372.806 ppm compared to 229.269 ppm on the CB diet. Here a dose of 1000 ppm of Cry1Ac did not kill all of the Ha-Cry1Ac tested – 6.25% of the Ha-Cry1Ac caterpillars exposed to 1000 ppm Cry1Ac on the CB survived, whereas 12.77% of the HaCry1Ac caterpillars exposed to 1000 ppm Cry1Ac on the IT survived (Fig. 4). However, this difference was not statistically significant (proportions test, Z = 1.09). 4. Discussion All animals forage to acquire a suite of nutrients that support growth, development and reproduction (Simpson and Raubenheimer, 2012). Protein and digestible carbohydrates are two nutrients that greatly impact insect performance (Le Gall and Behmer, 2014; Roeder and Behmer, 2014), but the balance needed is often species-specific (Behmer and Joern, 2008). In the current study, we showed that H. armigera and H. punctigera self-select a protein-carbohydrate ratio of p1.3:c1 and p1.1:c1, respectively. These values fall within the range reported for a number of other noctuid and Helicoverpa species (Fig. 5). Most caterpillar species self-select a slightly protein-biased diet. In the field, caterpillars can reach their species-specific protein-carbohydrate

Fig. 4. Dose response curve for Ha-Cry1Ac when exposed to Cry1Ac on two diets with varying macronutrient profiles (n = 43–48 per Cry1Ac concentration per diet). The black line represents the insect responses when fed on a carbohydrate biased diet (p0.4:c1) and the red line represents the insect’s response when fed at their self-selected intake target (p1.3:c1). Dose response curve was created using simple probit analysis in JMP. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 6 LC50 values calculated on the two protein-carbohydrate diets for each Bt toxin. The 95% confidence intervals are listed in parentheses below each LC50 value. The p:c ratio of the CB diet was p0.4:c1. The p:c ratio of the IT diet for H. armigera and H. punctigera was p1.3:c1 and p1.1:c1, respectively. IT diets that were significantly different from the CB diet are marked with an asterisk (α = 0.05). Toxin

Cry1Ac Cry2Ab Vip3Aa

Diet

CB IT CB IT CB IT

Ha-Susceptible

Ha-Cry1Ac

Hp-Susceptible

LC50

95% CI

LC50

95% CI

LC50

95% CI

0.456 1.957* 0.192 0.674 9.317 10.012

0.226–0.920 0.999–3.832 0.103–0.359 0.280–1.619 1.236–81.126 2.266–38.303

229.269 372.806 – – – –

83.97–3583.5 78.23–671.96 – – – –

1.418 2.311 0.392 0.860 23.623 31.577

0.418–4.816 0.727–7.340 0.134–1.143 0.217–3.408 2.863–194.94 2.778–358.89

5

Journal of Insect Physiology xxx (xxxx) xxx–xxx

A.E. Tessnow et al.

Fig. 5. Self-selected protein-carbohydrate intake targets (IT) for 12 lepidopteran species from three families; Noctuidae (circles), Plutellidae (square) and Sphingidae (triangle). The ITs for H. armigera and H. punctigera come from the current study. The ITs for the remaining 10 species were taken from the literature: S. exempta (Lee et al., 2004), H. subflexa (Lee et al., 2006), M. sexta (Thompson and Redak, 2005), S. exigua (Merkx-Jacques et al., 2008), P. xyostella (Warbrick-Smith et al., 2009), T. ni (Shikano and Corey, 2014a,b), S. littoralis (Lee et al., 2002), H. virescens (Lee et al., 2006), H. zea (Deans et al., 2015), and S. litura (Lee, 2010).

content; susceptibility was greatly reduced when they were fed diets with protein-carbohydrate content that matched their protein-carbohydrate IT. We found a similar response to Cry1Ac in H. armigera, although diet effects were much stronger in H. zea (100-fold) compared to H. armigera (4-fold). There was no diet effect for H. armigera when challenged with Cry2Ab, or for H. punctigera challenged with Cry1Ac or Cry2Ab. However, in these three cases there was a trend to perform better when on the diet that had the optimal protein-carbohydrate ratio. Thus, the effect that individual nutritional state has on the insect’s susceptibility to Cry toxins indicates that nutrition should be taken into account when monitoring for resistance to Cry toxins. Differences in toxin production between our study and Deans et al. (2017) may have altered the degree of the diet effect (Chakroun et al., 2012) and help to explain difference in results across the two studies. In our study, we used a purified protoxin of Cry1Ac, whereas the H. zea study used a trypsin-activated form of Cry1Ac. Because we do not understand the mechanism of nutritionally mediated variation in susceptibility, we cannot rule out the possibility that this difference in toxin production altered the degree of the diet effect on insect susceptibility. Interestingly, when the Cry1Ac resistant line of H. armigera was challenged with Cry1Ac on diets that either had p:c ratios matching the IT, or a suboptimal carbohydrate-biased protein-carbohydrate ratio, there was a unique pattern of susceptibility. At very low doses, the individuals on the carbohydrate-biased rearing diet appeared to outperform those feeding at the intake target. However, as the dose increased, the individuals feeding on the intake target exhibited higher survival rates. At the highest dose tested (1000 ppm), nearly double the number of insects survived on the intake target as opposed to the suboptimal rearing diet (Fig. 4). Food protein-carbohydrate content altering an insect’s susceptibility to the Cry1Ac endotoxin has become a consistent trend across the sister species, H. zea and H. armigera (Deans et al., 2017). As a result, this effect should be considered when developing resistance-monitoring programs. Without accounting for the insect’s nutritional ecology, we risk overestimating the insects’ susceptibility to these toxins, which in turn can compromise the sustainability of Bt crops. In contrast, nutritionally mediated variation in susceptibility to the Vip3Aa toxin was not observed. The lethal concentration of Vip3Aa on

intake target by foraging between different plants, or different plant tissues. For example, even in a monoculture like cotton, Helicoverpa spp. larvae have the opportunity to feed selectively on different vegetative and reproductive structures to regulate their protein-carbohydrate intake (Deans et al., 2016a). Interestingly, genetic resistance to toxins can impact protein-carbohydrate regulation, as demonstrated in Trichoplusia ni (Fig. 5, Shikano and Cory, 2014b). However, we did not observe different protein-carbohydrate ITs between Bt resistant and susceptible lines of H. armigera or H. punctigera. This might indicate a lower fitness cost of resistance in these Helicoverpa species, because when the cost of resistance is low, there is less pressure to compensate by altering macronutrient intake (Shikano and Cory, 2014a). This idea is further supported by Mahon and Young (2010), who showed that the frequency of Cry2Ab resistant alleles in a H. armigera population could be maintained for multiple generations without selection, indicating that the cost of resistance was negligible. Although there is some evidence that genetic resistance to Cry1Ac can reduce the fitness of H. armigera (Bird & Akhurst 2007), this cost does not appear to be sufficient to alter protein-carbohydrate intake. However, the Cry1Ac resistant line of H. punctigera showed a tighter pattern of nutrient regulation, with slightly less individual variation compared to the susceptible line (Fig. 2). This might suggest there is stronger stabilizing selection pressure on proteincarbohydrate intake that penalizes deviations from the intake target. After empirically determining self-selected intake targets for each species and strain, we then tested whether deviation from the optimal intake target could alter the insects’ susceptibility to three Bt toxins found in transgenic crops (Cry1Ac, Cry2Ab, and Vip3Aa). Many resistance-monitoring assays in the United States use carbohydrate-biased diets that do not match the protein-carbohydrate profile that the insects select (Deans et al., 2015). As outlined in Deans et al. (2016a), an unintentional consequence of testing for resistance on diets that are suboptimal is that it may increase susceptibility to Bt toxins. More importantly, this undermines the ability of researchers to detect resistance. Therefore, it is crucial to understand how variation in food protein-carbohydrate content can affect insect’s susceptibility to a particular toxin (Deans et al., 2016a, 2017). Previously, Deans et al. (2017) found a significant difference in susceptibility in H. zea to Cry1Ac when reared on diets with different protein-carbohydrate 6

Journal of Insect Physiology xxx (xxxx) xxx–xxx

A.E. Tessnow et al.

Characterization of the resistance to Vip3Aa in Helicoverpa armigera from Australia and the role of midgut processing and receptor binding. Sci. Rep. 6, 24311. Chakroun, M., Bel, Y., Caccia, S., Abdelkefi-Mesrati, L., Escriche, B., Ferré, J., 2012. Susceptibility of Spodoptera frugiperda and S. exigua to Bacillus thuringiensis Vip3Aa insecticidal protein. J. Invertebr. Pathol. 110, 334–339. http://dx.doi.org/10. 1016/j.jip.2012.03.021. Deans, C.A., Behmer, S.T., Tessnow, A.E., Tamez-Guerra, P., Pusztai-Carey, M., Sword, G.A., 2017. Nutrition affects insect susceptibility to Bt toxins. Sci. Rep. 7. http://dx. doi.org/10.1038/srep39705. Deans, C., Sword, G., Behmer, S., 2016a. Nutrition as a neglected factor in insect herbivore susceptibility to Bt toxins. Curr. Opin. Insect Sci. 15, 97–103. http://dx.doi.org/ 10.1016/j.cois.2016.04.005. Deans, C.A., Behmer, S.T., Fiene, J., Sword, G.A., 2016b. Spatio-temporal, genotypic, and environmental effects on plant soluble protein and digestible carbohydrate content: implications for insect herbivores with cotton as an exemplar. J. Chem. Ecol. http:// dx.doi.org/10.1007/s10886-016-0772-1. Deans, C.A., Sword, G.A., Behmer, S.T., 2015. Revisiting macronutrient regulation in the polyphagous herbivore Helicoverpa zea (Lepidoptera: Noctuidae): new insights via nutritional geometry. J. Insect Physiol. 81, 21–27. http://dx.doi.org/10.1016/j. jinsphys.2015.06.015. Dively, G.P., Venugopal, P.D., Finkenbinder, C., 2016. Field-evolved resistance in corn earworm to cry proteins expressed by transgenic sweet corn. PLoS One 11, e0169115. http://dx.doi.org/10.1371/journal.pone.0169115. Downes, S., Mahon, R., Olsen, K., 2007. Monitoring and adaptive resistance management in Australia for Bt-cotton: current status and future challenges. J. Invertebr. Pathol. 95, 208–213. Dussutour, A., Latty, T., Beekman, M., Simpson, S.J., 2010. Amoeboid organism solves complex nutritional challenges. Proc. Natl. Acad. Sci. U.S.A. 107, 4607–4611. Fanson, B.G., Taylor, P.W., 2012. Protein:carbohydrate ratios explain life span patterns found in Queensland fruit fly on diets varying in yeast:sugar ratios. Age 34, 1361–1368. Fanson, B.G., Weldon, C.W., Pérez-Staples, D., Simpson, S.J., Taylor, P.W., 2009. Nutrients, not caloric restriction, extend lifespan in Queensland fruit flies (Bactrocera tryoni). Aging Cell 8, 514–523. Felton, A.M., Felton, A., Wood, J.T., Foley, W.J., Raubenheimer, D., Wallis, I.R., Lindenmayer, D.B., 2009. Nutritional ecology of Ateles chamek in lowland Bolivia: how macronutrient balancing influences food choices. Int. J. Primatol. 30, 675–696. Gouffon, C., Van Vliet, A., Van Rie, J., Jansens, S., Jurat-Fuentes, J.L., 2011. Binding sites for Bacillus thuringiensis Cry2Ae toxin on heliothine brush border membrane vesicles are not shared with Cry1A, Cry1F, or Vip3A toxin. Appl. Environ. Microbiol. 77, 3182–3188. Gould, F., 1998. Sustainability of transgenic insecticidal cultivars: integrating pest genetics and ecology. Annu. Rev. Entomol. 43, 701–726. Grandison, R.C., Piper, M.D.W., Partridge, L., 2009. Amino-acid imbalance explains extension of lifespan by dietary restriction in Drosophila. Nature 462, 1061–1064. Jing, X., Grebenok, R.J., Behmer, S.T., 2013. Sterol/steroid metabolism and absorption in a generalist and specialist caterpillar: effects of dietary sterol/steroid structure, mixture and ratio. Insect Biochem. Mol. Biol. 43, 580–587. Kranthi, K.R., Jadhav, D.R., Kranthi, S., Wanjari, R.R., Ali, S.S., Russell, D.A., 2002. Insecticide resistance in five major insect pests of cotton in India. Crop Protect. 21, 449–460. Kurtz, R.W., 2010. A Review of Vip3A Mode of Action and Effects on Bt Cry ProteinResistant Colonies of Lepidopteran Larvae. Southwestern Entomologist September. Le Gall, M., Behmer, S.T., 2014. Effects of protein and carbohydrate on an insect herbivore: the vista from a fitness landscape. Integr. Comp. Biol. http://dx.doi.org/10. 1093/icb/icu102. Lee, K.P., 2010. Sex-specific differences in nutrient regulation in a capital breeding caterpillar, Spodoptera litura (Fabricius). J. Insect Physiol. 56, 1685–1695. Lee, K.P., Behmer, S.T., Simpson, S.J., Raubenheimer, D., 2002. A geometric analysis of nutrient regulation in the generalist caterpillar Spodoptera littoralis (Boisduval). J. Insect Physiol. 48, 655–665. Lee, K.P., Simpson, S.J., Raubenheimer, D., 2004. A comparison of nutrient regulation between solitarious and gregarious phases of the specialist caterpillar, Spodoptera exempta (Walker). J. Insect Physiol. 50, 1171–1180. http://dx.doi.org/10.1016/j. jinsphys.2004.10.009. Lee, K.P., Cory, J.S., Wilson, K., Raubenheimer, D., Simpson, S.J., 2006. Flexible diet choice offsets protein costs of pathogen resistance in a caterpillar. Proc. R. Soc. BBiol. Sci. 273, 823–829. http://dx.doi.org/10.1098/rspb.2005.3385. Lee, K.P., Simpson, S.J., Clissold, F.J., Brooks, R., Ballard, J.W.O., Taylor, P.W., Soran, N., Raubenheimer, D., 2008. Lifespan and reproduction in Drosophila: new insights from nutritional geometry. Proc. Natl. Acad. Sci. U.S.A. 105, 2498–2503. Machado, R.A.R., Arce, C.C.M., Ferrieri, A.P., Baldwin, I.T., Erb, M., 2015. Jasmonatedependent depletion of soluble sugars compromises plant resistance to Manduca sexta. New Phytol. 207, 91–105. Mahon, R.J., Young, S., 2010. Selection experiments to assess fitness costs associated with Cry2Ab resistance in Helicoverpa armigera (Lepidoptera: Noctuidae). J. Econ. Entomol. 103, 835–842. Mahon, R.J., Olsen, K.M., Downes, S., Addison, S., 2007. Frequency of alleles conferring resistance to the Bt toxins Cry1Ac and Cry2Ab in Australian populations of Helicoverpa armigera (Lepidoptera: Noctuidae). J. Econ. Entomol. http://dx.doi.org/ 10.1093/jee/100.6.1844. Mahon, R.J., Downes, S.J., James, B., 2012. Vip3A resistance alleles exist at high levels in Australian targets before release of cotton expressing this toxin. PLoS One 7. Maklakov, A.A., Simpson, S.J., Zajitschek, F., Hall, M.D., Dessmann, J., Clissold, F., Raubenheimer, D., Bonduriansky, R., Brooks, R.C., 2008. Sex-specific fitness effects of nutrient intake on reproduction and lifespan. Curr. Biol. 18, 1062–1066.

the intake target diet was the same as that for the carbohydrate-biased diet in both species of Helicoverpa. This could be due to a difference in the mode of action between the Cry and Vip proteins. Genes coding for the Cry and Vip toxins are commonly pyramided in Bt crops to delay the onset of resistance, due to differences in the mode of action and lack of cross resistance between these two classes of proteins (Kurtz, 2010; Chakroun et al., 2016a). Still, following ingestion, Cry and Vip proteins undergo a similar series of steps that ultimately result in the insect’s death. The toxins are activated by proteases in the midgut, then travel through the peritropic membrane to bind to receptors on the epithilial membrane resulting in the formation of pores (Whalon and Wingerd, 2003; Bravo et al., 2007; Chakroun et al., 2016a). The main differences between the Cry and Vip mode of action are the specific proteases involved in the activation of the toxin and the specific receptors that the toxin binds to on the epithelial cells (Chakroun et al., 2016a; Caccia et al., 2014; Gouffon et al., 2011). Still, there are steps within the Vip pathway that remain unclear. Our study suggests that nutrition interacts with the Cry and Vip pathways and their modes of action in different ways. Despite evidence indicating that food protein-carbohydrate content can alter the susceptibility of Helicoverpa pests to Cry endotoxins, notably Cry1Ac (Orpet et al., 2015; Deans et al., 2017), the molecular mechanisms involved in this change remain unknown. This difference in nutritionally mediated susceptibility between Cry and Vip toxins observed in this study may allow us to better pinpoint the mechanism involved in nutritionally mediated resistance while also revealing more specific differences in the mode of action between these two classes of Bt proteins. Further studies should address the effects of nutrition on physiological changes that occur within the insect to better understand the role of phenotypic plasticity in modulating insect susceptibility to Bt toxins (e.g. Deans et al., 2016a, 2017). A more comprehensive understanding of insect susceptibility could allow for better Bt crop management practices, increasing the sustainability of these transgenic technologies. Acknowledgments This work was made possible through a collaboration between Cotton Incorporated and the Cotton Research and Development Corporation fostered by Ryan Kurtz and Susan Mass. Funding was provided by Cotton Inc. CORE Project #16-413, Cotton Research and Development Corporation travel grant CLW1602, and the USDA-NIFA Biotechnology Risk Assessment Grant (BRAG) program grant #201533522-24099. We would also like to thank Bill James and Patrick Gleeson for their help with rearing and toxin production. References Akhurst, R.J., James, W., Bird, L.J., Beard, C., 2003. Resistance to the Cry1Ac -endotoxin of Bacillus thuringiensis in the cotton bollworm, Helicoverpa armigera (Lepidoptera: Noctuidae). J. Econ. Entomol. 96, 1290–1299. http://dx.doi.org/10.1093/jee/96.4. 1290. Behmer, S.T., 2009. Insect herbivore nutrient regulation. Annu. Rev. Entomol. 54, 165–187. Behmer, S.T., Joern, A., 2008. Coexisting generalist herbivores occupy unique nutritional feeding niches. Proc. Natl. Acad. Sci. U.S.A. 105, 1977–1982. http://dx.doi.org/10. 1073/pnas.0711870105. Behmer, S.T., Simpson, S.J., Raubenheimer, D., 2002. Herbivore foraging in chemically heterogeneous environments: nutrients and secondary metabolites. Ecology 83, 2489–2501. Bird, L.J., Akhurst, R.J., 2004. Relative fitness of Cry1A-resistant and -susceptible Helicoverpa armigera (Lepidoptera: Noctuidae) on conventional and transgenic cotton. J. Econ. Entomol. 97, 1699–1709. Bravo, A., Gill, S.S., Soberón, M., 2007. Mode of action of Bacillus thuringiensis Cry and Cyt toxins and their potential for insect control. Toxicon. Caccia, S., Chakroun, M., Vinokurov, K., Ferré, J., 2014. Proteolytic processing of Bacillus thuringiensis Vip3A proteins by two Spodoptera species. J. Insect Physiol. 67, 76–84. Chakroun, M., Banyuls, N., Bel, Y., Escriche, B., Ferré, J., 2016. Bacterial vegetative insecticidal proteins (Vip) from entomopathogenic bacteria 329–350. doi:10.1128/ MMBR.00060-15.Address. Chakroun, M., Banyuls, N., Walsh, T., Downes, S., James, B., Ferré, J., 2016b.

7

Journal of Insect Physiology xxx (xxxx) xxx–xxx

A.E. Tessnow et al.

geometry of nutritional decisions. Philos. Trans. R. Soc. B: Biol. Sci. 342, 381–402. Simpson, S.J., Raubenheimer, D., 2012. The Nature of Nutrition: A Unifying Framework From Animal Adaptation to Human Obesity. Princeton University Press. Siva-Jothy, M.T., Thompson, J.J.W., 2002. Short-term nutrient deprivation affects immune function. Physiological Entomology 27, 206–212. Tabashnik, B.E., Brévault, T., Carrière, Y., 2013. Insect resistance to Bt crops: lessons from the first billion acres. Nat. Biotechnol. http://dx.doi.org/10.1038/nbt.2597. Tabashnik, B.E., Mota-Sanchez, D., Whalon, M.E., Hollingworth, R.M., Carrière, Y., 2014. Defining terms for proactive management of resistance to Bt crops and pesticides. J. Econ. Entomol. http://dx.doi.org/10.1603/EC13458. Thompson, S.N., Redak, R.A., 2005. Feeding behaviour and nutrient selection in an insect Manduca sexta L. and alterations induced by parasitism. J. Comp. Physiol. A: Neuroethol. Sens. Neural Behav. Physiol. 191, 909–923. http://dx.doi.org/10.1007/ s00359-005-0013-5. Walsh, T.K., Downes, S.J., Gascoyne, J., James, W., Parker, T., Armstrong, J., Mahon, R.J., 2014. Dual Cry2Ab and Vip3A Resistant Strains of Helicoverpa armigera and Helicoverpa punctigera (Lepidoptera: Noctuidae); Testing Linkage Between Loci and Monitoring of Allele Frequencies. J. Econ. Entomol. 107, 1610–1617. Warbrick-Smith, J., Raubenheimer, D., Simpson, S.J., Behmer, S.T., 2009. Three hundred and fifty generations of extreme food specialisation: testing predictions of nutritional ecology. Entomol. Exp. Appl. 132, 65–75. http://dx.doi.org/10.1111/j.1570-7458. 2009.00870.x. Whalon, M.E., Wingerd, B.A., 2003. Bt: mode of action and use. Arch. Insect Biochem. Physiol. 200–211.

McCaffery, A.R., 1998. Resistance to insecticides in heliothine Lepidoptera: a global view. Philos. Trans. R. Soc. Lond B 353, 1735–1750. Merkx-Jacques, M., Despland, E., Bede, J.C., 2008. Nutrient utilization by caterpillars of the generalist beet armyworm, Spodoptera exigua. Physiol. Entomol. 33, 51–61. http://dx.doi.org/10.1111/j.1365-3032.2007.00604.x. Orpet, R.J., Degain, B.A., Tabashnik, B.E., Carrière, Y., 2015. Balancing Bt toxin avoidance and nutrient intake by Helicoverpa zea (Lepidoptera: Noctuidae) Larvae. J. Econ. Entomol. 108, 2581–2588. Pascacio-Villafán, C., Williams, T., Birke, A., Aluja, M., 2016. Nutritional and non-nutritional food components modulate phenotypic variation but not physiological tradeoffs in an insect. Sci. Rep. 6, 29413. http://dx.doi.org/10.1038/srep29413. Raubenheimer, D., Simpson, S.J., 1993. The geometry of compensatory feeding in the locust. Anim. Behav. 45, 953–964. Ritter, K.S., Nes, W.R., 1981. The effects of the structure of sterols on the development of Heliothis zea. J. Insect Physiol. 27, 419–424. Roeder, K.A., Behmer, S.T., 2014. Lifetime consequences of food protein-carbohydrate content for an insect herbivore. Funct. Ecol. 28, 1135–1143. http://dx.doi.org/10. 1111/1365-2435.12262. Shikano, I., Cory, J.S., 2014a. Genetic resistance to Bacillus thuringiensis alters feeding behaviour in the cabbage looper, Trichoplusia ni. PLoS One 9. Shikano, I., Cory, J.S., 2014b. Dietary mechanism behind the costs associated with resistance to Bacillus thuringiensis in the cabbage looper, Trichoplusia ni. PLoS One 9. Simpson, S.J., Raubenheimer, D., 2001. The geometric analysis of nutrient - allelochemical interactions: a case study using locusts. Ecology 82, 422–439. Simpson, S.J., Raubenheimer, D., 1993. A multi-level analysis of feeding behaviour: the

8