General introduction
Helicoverpa armigera, a significant polyphagous pest, inflicts considerable damage on various agricultural crops, notably cotton, maize, grain legumes, and tobacco The reliance on pesticides for its control has led to serious side effects, including the rapid development of resistance, making effective pesticide resistance management essential Since 1996, the adoption of Bt cotton, which expresses insecticidal toxins from Bacillus thuringiensis, has been implemented in Australia, now accounting for nearly 90% of the cotton crops This shift has resulted in a marked decrease in insecticide use against this pest.
Bt cotton effectively controls Helicoverpa spp in Australia; however, reports indicate that larvae of various sizes occasionally survive in all growing regions A survey from 2005-2008 revealed that approximately 15% of Bt cotton areas had larvae exceeding recommended control thresholds The survival of these larvae is not solely attributed to physiological resistance, but may also stem from factors such as inadequate gene expression in genetically modified plants, increased pest pressure due to climatic conditions, and behavioral mechanisms Behavioral resistance may allow larvae to evade Bt toxin by choosing specific oviposition sites or seeking refuge from plant defenses These behaviors suggest a potential genetic basis that could evolve in environments dominated by Bt crops.
This research investigates how Bt-susceptible H armigera larvae survive on Bt cotton plants, focusing on the concept of 'behavioural resistance.' The study explores two key areas: the influence of egg deposition and the movement of larvae, aiming to understand their roles in the survival of these susceptible larvae Additionally, several specific questions are examined within these two main areas of inquiry.
1 Do female moths choose oviposition sites within a crop randomly or do they place eggs on plants and/ or locations within plants that express relatively low levels of Bt toxin? If the latter, then I expect to find a relationship between oviposition site and the survival of first instar larvae.
2 Do first instar larvae move the same way or differently on cotton with and without Bt toxin? Can larvae detect Bt toxin levels in cotton plants and move to sites with relatively low expression levels or do they move independently of toxin levels? Do larger neonates (from bigger eggs) have greater reserves, which afford them a higher chance of finding sites that express relatively low levels of Bt toxin or finding un-hatched eggs as a food source and so survive the critical first instar stage?
This article examines the physiological resistance of a specific H armigera strain to Bt toxin (Cry2Ab) by comparing it with a susceptible strain to explore the relationship between behavior and resistance It reviews the mechanisms behind host-plant selection, highlighting the connections between oviposition sites, neonate movement, and overall survival Additionally, the article discusses the current state of Bt cotton production in Australia, focusing on resistance management for Helicoverpa spp The review emphasizes the oviposition behavior of female moths and the movement and cannibalism of first instars on both Bt and non-Bt substrates, including artificial diets and cotton plants.
Host-plant selection behaviour
Phytophagous insects, or herbivores, are classified based on their dietary specialization into three categories: monophagous (specialists) that feed on a single plant species, oligophagous species that consume multiple plants within the same botanical family, and polyphagous (generalists) insects that feed on a diverse range of plants across various families.
Identifying appropriate host plants poses a significant challenge for insects, which rely on a combination of olfactory and visual cues to locate them from afar This complex process involves various sensory inputs, including chemical signals and physical characteristics like color, shape, and texture Notably, volatile chemical cues are crucial for enabling insects to recognize and orient toward their host plants at a distance.
1998) Plant volatiles are often complex mixtures of several hundred compounds (Visser 1986;
The volatile compounds emitted by host plants vary over time due to physiological changes within the plants (Fraser et al 2003; Johnson et al 2004) Helicoverpa armigera, a highly polyphagous herbivore with over 100 host species, shows a preference for plants in the flowering growth stage (Cunningham & Zalucki 2014) Moths are attracted to the volatiles released by flowering hosts and can locate these plants by learning to associate specific volatile blends with their presence (Cunningham et al 2004; Cunningham 2012).
Specialist herbivores possess physiological adaptations that enable them to tolerate plant defensive compounds better than generalists, as noted by Cornell & Hawkins (2003) Research by Ali & Agrawal (2012) confirmed that while specialists exhibit greater tolerance to low levels of toxins, few can withstand the effects of higher concentrations In contrast, generalist herbivores have evolved mechanisms to mitigate the impact of plant defenses, allowing them to feed on a wider variety of plant species compared to specialists (Dussourd & Denno 1994; Eichenseer et al.).
Polyphagous holometabolous insects, particularly in their larval stages, exhibit remarkable tolerance to contact insecticides, likely due to evolutionary adaptations that enhance their endurance to diverse biochemical stresses from various food sources These generalist insects also tend to suppress plant defenses more effectively, which may improve their survival rates For instance, caterpillars of Helicoverpa zea have been observed to secrete salivary glucose oxidase (GOX), an enzyme that helps mitigate plant defense mechanisms Recent research has examined GOX levels across 85 species, highlighting the intricate relationship between these insects and their host plants.
Research on 23 families of Lepidoptera indicates that polyphagous species exhibit higher levels of glucose oxidase (GOX) compared to specialized species (Erb et al 2012) In contrast, specialist insects may tolerate plant defenses without needing to manipulate them, optimizing their fitness through strategies like feeding during less defended phenological stages or selecting less toxic feeding sites (Ali & Agrawal 2012) Numerous studies and the specialist–generalist paradigm suggest a consistent pattern in plant recognition and herbivore elicitation across various herbivorous insects (Dussourd & Eisner 1987; Karban & Agrawal 2002).
Cornelius and Bernays (1995) identified three strategies employed by herbivores in relation to plant toxins: sequestering specialists thrive on intermediate toxin levels, non-sequestering specialists tolerate low toxin levels, and generalists benefit from suppressing toxin induction While the first two strategies incur costs due to toxins, some generalists gain advantages from consuming toxic plants without sequestering the toxins themselves.
Ali & Agrawal (2012) found 20 studies to interpret this prediction by comparing the responses of a plant to both specialist and generalist herbivores using one feeding guild (Table 1.1).
Table 1.1 presents a comparison of plant defensive responses to both specialist and generalist insect herbivores within the same feeding guild, as outlined by Ali and Agrawal (2012) The ordinal numbers in the results column indicate the consistency with the hypothesis: (1) signifies no consistent pattern, (2) suggests that the level of specialization does not predict plant responses, and (3) denotes consistent findings, albeit based on a comparison of only two insect species.
Plant Generalist Specialist Measure of plant response
The generalist exhibited a greater impact on gene expression compared to the specialist sequesterer, with both types inducing similar levels of general stress-responsive genes and genes involved in octadecanoid and indole glucosinolate synthesis However, the specialist triggered a lower glucosinolate response than the generalist.
(1) Induction pattern by the two species depended on water status of the plant (Winz & Baldwin 2001).
(1) Expression of GS genes was similar for generalist and specialist, but
GS levels only showed an increase in response to S exigua Mean aliphatic
GS levels were equal Pieris rapae caused a higher increase in indolyl GS content (Mewis et al 2006).
(2) Transcription profiles were indistinguishable (Reymond et al 2004).
Parasitoid specificity for herbivore induced plant volatiles (HIPVs)
(1) Parasitoid attracted to damaged plants over controls for both generalists and specialists Parasitoids only discriminate between induction by insects in different guilds (van Poecke et al 2003)
(2) Transcriptional responses and GS were not consistently influenced by degree of insect specialization (Bidart-Bouzat & Kliebenstein 2011).
GS (2) Indole GS was significantly higher after feeding by P rapae and M brassicae than after P xylostella feeding (Loon van et al 2008).
P rapae Foliar trichomes, sinigrin, foliar nitrogen
(1) Differential induction by specialist versus generalist led to increased trichomes, but the effect reversed on different leaf positions (Traw &
Specialist herbivores trigger the expression of salicylic acid (SA) and ethylene-associated genes in plants, while generalist herbivores induce jasmonic acid (JA) and ethylene (ET) genes Although specialist herbivores may be better adapted to their host plants, these plants primarily defend themselves against generalist herbivores.
Variation in induction was observed, but it did not correlate with insect specialization The pests P xylostella and S exigua were found to induce resistance across all tested species, while P rapae induced resistance only to itself and S exigua In contrast, T ni did not induce any resistance.
(3) Specialist (sequesterer) and mechanical wounding induced GS and MYR threefold, whereas generalist induced only GS (twofold) (Travers- Martin & Mueller 2008) – generalist might be adaptively suppressing defense.
Peroxidase activity (POD), C/N ratio, protein content, insect bioassays
(3) POD activity was more strongly induced by generalist than specialist (no difference in bioassay) (Mooney et al 2009) – plant might be adaptively defending against generalist.
Iridoid GS (IrGS), protein, foliar nitrogen
(3) Higher IrGS induced by specialist
(sequesterer) compared with generalist (Stamp & Bowers 1994)– plant might be adaptively defending against generalist.
Parasitoid specificity for herbivore induced plant volatiles
(3) Natural enemies preferred roots attacked by specialist over roots damaged by generalist The specialist induced significantly more (E)-b- caryophyllene than the generalist. Solanaceae
Phytohormones (3) Specialist induced JA/ET burst, generalist induced SA (Diezel et al
2009) – might be adaptive for generalists to suppress resistance by activating SA.
The plant response to generalist herbivores showed greater similarity compared to the response to specialist herbivores, which was linked to the influence of fatty acid-amino acid conjugates (FACs) and oral secretions Both generalist herbivores, identified as noctuids, downregulated a significant number of similar genes, highlighting the distinct mechanisms of plant defense within the Solanaceae family (Voelckel & Baldwin 2004).
Phytohormones (1) M sexta induced a JA and SA response, whereas S littoralis and T ni induced stronger SA responses (Heidel & Baldwin 2004).
Lipoxygenase (LOX), proteinase inhibitors (PIs), nicotine, peroxidase (POD), polyphenol oxidase (PPO)
Both herbivores triggered a comparable defensive response in plants, yet the intensity of this response varied The specialist herbivore elicited a lower polyphenol oxidase (PPO) response while promoting a more pronounced nicotine and peroxidase (POD) response compared to the generalist herbivore Notably, levels of jasmonic acid (JA), lipoxygenase (LOX), and protein inhibitors (PIs) remained consistent across both herbivore types.
The success of lepidopteran larvae is largely dependent on their ability to establish themselves on a host plant, as they have limited sensory capabilities and must physically contact the plant to detect cues If a caterpillar is dislodged or leaves its hatching site, finding another suitable host can be challenging, particularly for less mobile species or in areas with low host density While highly mobile late-stage caterpillars may navigate these challenges more easily, the risks of exposure to environmental hazards, predators, and starvation significantly increase for walking insects in such conditions.
After hatching, caterpillars actively search for suitable plants to feed on, utilizing their senses to assess and interact with potential food sources They respond to both chemical and physical signals emitted by the plants, with volatile compounds in the leaf boundary layer significantly influencing their behavior This initial interaction often leads to a critical decision: caterpillars either continue to evaluate the plant for feeding or reject it based on its surface characteristics.
Oviposition preference
Oviposition behaviour, a crucial aspect of host finding, encompasses searching, orientating, encountering, landing, and acceptance behaviours, all influenced by various sensory cues Research has predominantly examined the initial stages of host finding, focusing on odours and visual elements like plant size, shape, and colour Upon landing, female moths assess the suitability of host plants through physical and chemical signals from leaves and surfaces Helicoverpa spp moths show a preference for laying eggs on rough or hairy surfaces, flowering plants, and fruiting bodies Recent studies on H armigera's nectar feeding behaviour further highlight moths' attraction to floral odours.
In addition, learning or experience can affect an insect’s preferences (Stanton 1984; Papaj 1986;
Insects, particularly Lepidoptera, demonstrate learning capabilities that allow them to adapt to changing environments, enhancing their activity efficiency (Papaj & Rausher 1987; Papaj & Prokopy 1989; Sadek 2010; Stephens & Krebs 1986) Research indicates that female Lepidoptera preferentially lay eggs on host plants they have previously encountered, with experienced females showing a higher acceptance rate of these hosts compared to those lacking prior experience (Stanton 1984; Traynier 1984, 1986; Papaj 1986; Cunningham et al 1998).
Learning plays a basic role in responses by individual H armigera to suitable hosts (Cunningham et al 1998; Cunningham et al 1999; Cunningham & West 2001).
The relationship between female oviposition preference and offspring survival is fundamental to insect-plant interaction theory, influencing insect dispersal among plant species over evolutionary time (Thompson 1988a) Females often prioritize host plants in a hierarchical manner, utilizing lower-ranked hosts when preferred options are unavailable (Jaenike 1978; Wiklund 1981; Singer et al 1983; Ward 1987; Courtney et al 1989) Preference is typically measured by the proportion of eggs laid on different plant species, though this method can be misleading due to variations in female presence and host availability (Wiklund 1975; Stanton 1979; Thompson 1988b; Stanton 1982; Ahman 1985) The performance of offspring—encompassing survival, growth rate, and reproduction—may not always align with this preference hierarchy (Thompson 1988a; Nylin & Janz 1993; Jallow & Zalucki 2003), as evidenced by Jallow et al (2001), who found that H armigera larvae thrived on a wider range of plant species than those selected by ovipositing females.
Helicoverpa armigera is a widely distributed polyphagous pest known for its varying oviposition preferences across different regions, which impacts offspring performance Research indicates that larval survival and growth are significantly higher on the reproductive structures of pigeon pea (Cajanus cajan) compared to other host plants like cotton, tomato, and linseed Specifically, within pigeon pea plants, larvae thrive more on flowers and pods than on leaves, highlighting the importance of these reproductive parts for the pest's development.
Helicoverpa armigera exhibits distinct preferences for oviposition and feeding based on geographic location In Japan, this pest favors okra and cotton over green pepper, tomato, and Chrysanthemum species Conversely, in Australia, H armigera shows a high oviposition rate on tobacco, sorghum, and maize, while cotton is less preferred Notably, the density of H armigera pupae is highest under pigeon pea crops compared to other plants like cotton, maize, and sorghum This species lays significantly more eggs on flowering pigeon pea, resulting in the development of robust larvae and pupae, which ultimately leads to the production of highly fecund moths.
Numerous studies have explored the relationship between individual preferences and performance, indicating that over generations, populations can exhibit changes in physiological or behavioral traits that enhance resource utilization However, there is a notable absence of clear genetic evidence linking preference and performance Research by Jallow and Zalucki (2003) highlights that the performance of offspring is significantly influenced by plant quality rather than the genetic lineage of the female parent.
Helicoverpa armigera selects oviposition sites based on the morphological characteristics and volatile compounds of leaves Female moths exhibit a preference for laying eggs on plant parts with dense trichomes and elevated levels of stimulatory chemicals Additionally, young larvae tend to feed on the reproductive parts of plants, which are typically more nutritious and rich in protein.
Research by Jallow et al (2001) indicates that oviposition preference has a minimal impact on the survival of larvae, with no significant correlation between the growth and survival of H armigera offspring and the host plant choices of adult moths (Jallow & Zalucki 2003) They emphasize the importance of conducting preference-performance experiments at the individual or family level, rather than relying on average population data Therefore, it is crucial to investigate whether the choice of oviposition site, which determines the initial environment for neonates, plays a significant role in their subsequent survival.
Oviposition preference significantly influences larval survival on Bt-cotton plants, as moths that can detect toxins may choose to lay their eggs on the least toxic parts of the plant This behavior increases the likelihood of larval survival, a hypothesis supported by research conducted in China (Zhao et al.).
2016) They indicated that H armigera moths prefer non-Bt cotton for oviposition Such a preference would reduce selection for physiological resistance.
Research indicates that H armigera exhibits cannibalistic behavior, allowing them to dominate limited resources by consuming potential competitors, including their own egg shells and smaller neonates (Kakimoto 2003) This behavior may enhance the survival of neonates by enabling them to avoid Bt toxins while feeding on conspecifics Nutrition plays a crucial role in this cannibalism, as it helps larvae achieve an optimal protein-to-carbohydrate ratio, improving their survival and performance against Cry toxins in their diet (Deans et al 2016) Additionally, locating an appropriate feeding site is essential for the survival of first instars on any host plant, including Bt cotton (Perkins et al 2008).
Limited research has been conducted on how oviposition preference influences the movement, feeding, and cannibalistic behavior of first instar larvae of H armigera on conventional versus Bt cotton plants, and the role of physiological resistance in this context remains unknown This knowledge is crucial for understanding the evolution of insect behavior and could inform pest management strategies in Bt cotton-dominated landscapes Current resistance management approaches rely on mathematical models that presume moths are unable to differentiate between cotton plants containing Bt toxins and those that do not (Zalucki et al 2012).
The movement behaviour of first instar H armigera
The movement behavior of first instar larvae of Helicoverpa armigera is influenced by various factors, including egg placement, plant characteristics, and environmental cues such as light and gravity These larvae utilize ambient light, plant volatiles, and substrate angles to identify optimal feeding sites Once they find a suitable location, their responsiveness to these environmental stimuli decreases, leading to reduced movement.
Egg placement on pea plants significantly affects the movement of H armigera larvae, with those hatching from eggs in lower regions tending to move further away from their natal leaf compared to those from middle or upper placements After hatching, first instars quickly seek concealed spots, such as buds or flowers, and shift to active feeding mode within the first two hours Initial feeding is limited, but substantial feeding occurs on the second day, leading to molting into the second instar by day three, influenced by temperature The larvae typically exhibit upward movement on plants, displaying positive phototaxis and negative geotaxis, with a preference for steeper substrates over horizontal ones However, plant volatiles can alter their movement preferences Flowering plants attract more larvae on stipules and tendrils, while fewer are found on leaflets, due to the interaction between floral odors and substrate angle Additionally, larvae tend to seek the softest plant tissues, largely ignoring the hardness of plant components at a micro scale Leaf properties, such as texture, wax, and trichomes, play a crucial role in larval movement, with higher trichome density and thicker cuticles causing larvae to spend more time on these plants as they require additional time to process the wax from the leaf surface.
Mortality rates in the first instar stage of Helicoverpa spp can be alarmingly high, ranging from 88% to 97% (Zalucki et al 2002) and reaching 93% to 100% within 4-5 days post-oviposition (Kyi et al 1991) Various factors contribute to this mortality, including predation, adverse weather conditions, dispersal, and the inability to find host plants after dislodgment (Titmarsh 1992; Zalucki et al 2002) Notably, dropping off the plant significantly increases first-instar mortality in crops (Perovíc et al 2008) Understanding the causes of mortality is particularly crucial in the context of Bt crops, where studies have shown that larvae across all stages exhibit no physiological resistance on Bt cotton (Lu et al 2011) and Bt corn (Kurtz 2005) In Australia, research by Whitburn & Downes (2009) indicated no significant difference in the frequency of Bt-resistance alleles between surviving larvae and randomly collected conspecifics from the same populations Laboratory and field studies further reveal that surviving larvae are often found on structures such as squares and flowers (Yang et al 2008; Lu et al 2011).
Research from 2011 indicates that certain plants express low levels of Bt toxin, which may influence the foraging behavior and movement of newly hatched larvae A deeper understanding of these behaviors will help determine whether their presence on these plants is incidental or genetically determined This knowledge could significantly enhance pest management strategies.
Previous studies on H armigera larvae have primarily examined Bt-susceptible strains, but analyzing resistant strains allows for a direct comparison of behavioral shifts in response to Bt toxins in plants, as opposed to the natural chemicals like tannins present in cotton Additionally, assessing both susceptible and resistant larvae on non-Bt plants will shed light on how resistance genes influence key behavioral traits This research specifically focuses on the movement and feeding behavior of larvae shortly after hatching, a critical period when they are most susceptible to mortality from toxin exposure.
Experiments in Chapters 3 and 4 investigated the oviposition and movement patterns of larvae to determine if they preferentially fed on plant parts with lower concentrations of Bt-toxin, indicating potential behavioral resistance Alternatively, the study examined whether the larvae simply moved upward, a typical behavior for neonates of this species, ultimately consuming reproductive structures (Yang et al 2008).
Bacillus thuringiensis toxins
Bacillus thuringiensis is a gram-positive bacterium found in soil, known for producing insecticidal toxins that effectively eliminate various insect species The main types of Bt toxins include crystal (Cry) toxins, cytolytic (Cyt) toxins, and vegetative insecticidal proteins (Vip), which contribute to its pest control capabilities.
(Castagnola & Jurat-Fuentes 2012) Cry and Cyt toxins are synthesized during sporulation (Hannay
Bt toxins, including Cry and Vip, are produced during different growth phases of the bacterium, with Cry toxins generated in the late exponential growth phase and Vip toxins during vegetative growth Both types consist of single or multiple toxins and can be stored as parasporal crystalline bodies, making them valuable for commercial insecticides While Vip genes have been integrated into Bt crops, Cry toxin genes are predominantly used in transgenic plants due to their early discovery and favorable characteristics Over 400 Bt toxin genes have been cloned and sequenced, comprising 218 Cry and 28 Vip toxins.
Bt crops and Bt cotton
The first commercial Bt crop, NewLeaf potato, expressing the cry3A toxin gene, was introduced in 1995 to combat Colorado potato beetle larvae Following this, maize and cotton were rapidly commercialized as transgenic Bt crops In Australia, transgenic cotton varieties, such as Ingard, were developed to control Helicoverpa spp and the pink bollworm, resulting in a significant 56% reduction in insecticide applications within six years of introduction By 2012, Bt cotton adoption reached approximately 90% in Australia and the U.S., with similar trends observed in China and India, where the use of Bt cotton has also led to increased yields and decreased reliance on insecticides.
The primary concern with transgenic Bt crops is the potential development of resistance to Bt proteins in insect populations Laboratory studies have shown that certain insect strains can develop resistance to Bt toxins, indicating a genetic predisposition for this issue to arise in natural settings Notably, instances of resistance to Cry1Ac in H armigera have been recorded in Australia, China, and India, highlighting the urgency of addressing this challenge.
In laboratory studies, resistance is primarily attributed to alterations in toxin binding to midgut receptors, typically governed by a single autosomal recessive gene.
In 2012, the global distribution of Bt cotton cultivation varied by country, with China leading at 27% of world production, followed by India at 22%, the United States at 15%, and Australia at 3.4% All Bt cotton varieties produced Cry1Ac, while Australia and India exclusively cultivated two-toxin cotton, which included Cry2Ab From 2004 to 2012, 86% of two-toxin cotton in the United States produced both Cry1Ac and Cry2Ab, while 14% produced Cry1Ac and Cry1F.
The widespread adoption of single gene transgenic Bt crops raised concerns about the potential evolution of insect resistance, leading to the development of second-generation Bt cotton and maize that express multiple Bt genes with various modes of action Regulatory agencies view multiple Bt toxins as having diverse modes of action, making them suitable candidates for pyramiding This strategy requires target insects to develop simultaneous mutations in different toxin receptors, thereby reducing the likelihood of resistance evolution Notable examples of this approach include the pyramids of cry1Ac and cry2Ab.
Next-generation Bt crops, featuring toxin genes such as those identified by Chitkowski et al (2003) and Cry and Vip from Estruch et al (1997), have demonstrated enhanced efficacy in controlling target pests Studies by Stewart et al (2001), Fitt (2003), and Jackson et al (2004) highlight the increased expression of these toxins, contributing to improved pest management strategies.
In 2004, Australia introduced Bollgard II®, featuring two Bt genes, Cry1Ac and Cry2Ab, which effectively controlled Helicoverpa spp throughout the season Despite this, some regions reported surviving larvae of Helicoverpa spp in Bollgard II® crops, with an average infestation rate of 15% from 2005 to 2008 The variations in survival and infestation patterns remained unclear, prompting numerous studies on the resistance levels of Helicoverpa spp to Bt toxins Notably, no significant changes in H armigera survival rates have been observed since the introduction of Bt cotton Possible explanations for the presence of surviving larvae include inadequate expression of Cry1Ac and Cry2Ab genes and the high density of Helicoverpa spp eggs laid on less toxic plant structures.
(3) behavioural resistance (Yang et al 2008; Lu et al 2011; Knight 2013) and / or (4) cannibalism behaviour (Kakimoto et al 2003).
The expression of Bt genes on Bt cotton
The stability of Bt toxin expression in cotton plants is influenced by various factors, including seasonal changes, differences among plant structures, and specific environmental conditions Research indicates that these variations can occur throughout the growing season and may differ between different parts of the plant, highlighting the complexity of Bt toxin dynamics in cotton cultivation.
The expression levels of Bt toxin can be affected by plant age Many studies have shown that
A reduction in Cry1Ac levels has been linked to decreased plant development in the field, as noted by Fitt et al (1998) and others The effectiveness of Cry1Ac against susceptible H armigera larvae appears to diminish and cease during the post-squaring stage, according to Olsen et al (2005b) This decline in Bt-toxin efficacy could potentially enhance the survival rates of Bt-resistant larvae.
The effectiveness of Cry proteins varies throughout the growing season, influenced by different plant structures and cultivars Research indicates that protein content remains stable in plants from the 19-node stage onward, with Cry2Ab expression peaking in large bolls later in the season Notably, Cry2Ab levels are higher in squares and small bolls compared to leaves and large bolls Conversely, Cry1Ac concentration is highest in leaves, although by the end of the season, its expression levels are consistent across all plant structures.
2013) In addition, Cry proteins were expressed at a lower level in white flowers (Gore et al 2001)
Lower toxin levels could provide an opportunity for larvae to survive on Bollgard II ® cotton
Higher percentages of larvae were found in the lower canopy of plants within flowers and bolls in
Bt cotton than in non-Bt cotton (Gore et al 2002).
Environmental factors, including site, temperature, and humidity, play a crucial role in the expression of Bt toxin While Adamczyk & Sumerford (2001) reported that site differences did not influence Cry1Ac expression, Sachs et al (1998) highlighted the significant impact of site, soil moisture, and fertility on Cry1Ac efficacy Additionally, Olsen (2005a) demonstrated that mild temperature variations can trigger a response in cotton, significantly affecting larval survival.
A study by Chen et al (2012) revealed that environmental stresses had minimal impact on the Bt toxin levels in cotton leaves during the square and flowering stages However, exposure to high temperatures for 24 hours during the boll period led to a notable reduction in the Cry1Ac protein content of Bt cotton plants.
Bt expression levels reduced by environmental factors provide H armigera with opportunities to be exposed to “good” (less toxic) food resources, and increased their survival on Bt cotton plants.
Behavioural resistance
Insect resistance mechanisms are categorized into biochemical, physiological, and behavioral types (Sparks et al 1989) Biochemical resistance involves changes in insecticide detoxification and sensitivity, while physiological resistance pertains to alterations in the penetration, transport, storage, and excretion of insecticides Behavioral resistance includes modifications in insect activities that affect their contact with toxic substances and host preferences, which can be further divided into stimulus-dependent and stimulus-independent mechanisms The former increases repellency and irritancy, reducing contact with insecticides, while the latter allows insects to avoid exposure due to innate behavioral traits (Sparks et al 1989) Both mechanisms enhance survival by minimizing exposure to toxicants However, behavioral resistance is often overlooked in research, as it is considered an experimental side effect and challenging to study (Gould 1984; Sparks et al 1989) Hoy et al (1991) highlighted that the interplay between physiological and behavioral responses significantly impacts the selection for physiological resistance.
Research on the behavioral responses of larvae has primarily focused on their movement, feeding, and oviposition behaviors, with studies indicating some evidence of behavioral resistance, except for Schwartz et al (1991) Sparks et al (1989) discovered pyrethroid resistance in horn flies and tobacco budworms, noting that resistant larvae exhibited reduced movement in the presence of pyrethroids Zhang et al (2004) found that H armigera larvae could detect and avoid transgenic Bt cotton in choice tests Knight (2013) suggested that increased movement of Helicoverpa spp larvae on Bt cotton may lead to better resource encounters, potentially explaining the higher larval populations in Bollgard II cotton fields Conversely, Schwartz et al (1991) found no evidence of behavioral resistance in diamondback moths (Plutella xylostella) against B thuringiensis on cabbage, highlighting the need for further experiments to explore this contentious resistance mechanism.
Cannibalism in natural population
Cannibalism, or intraspecific predation, is observed in various animal groups and is particularly common among herbivorous insects like butterfly larvae, leaf-eating beetles, and bark beetles This behavior can help regulate population density when resources are scarce and may enhance the fitness of insects in natural environments by reducing competition.
2006) or by providing access to essential nutrients (Polis 1981) Insect cannibals are usually juveniles (75% of reports across insect orders) and often consumed eggs (21% of reports)
Cannibalism in non-carnivorous insects may stem from neonates consuming eggs, as suggested by Richardson et al (2010) and earlier studies by Sotherton et al (1985) and Watanabe & Yangmaguchi (1993) These insects tend to lay eggs in clusters rather than individually, which increases the likelihood of encounters among them The prevalence of cannibalism is influenced by factors such as the availability of alternative food sources, the density and behavior of both cannibalistic and victimized individuals, and conditions of extreme food scarcity or overcrowding (Fox 1975).
Although starvation is not critical for initiating cannibalism it may increase this tendency (Fox
When food resources become scarce, animals may resort to cannibalism as a survival strategy This behavior is less frequent when food supply is sufficient Although starvation is primarily caused by food shortages, the effects of overcrowding often contribute to cannibalistic tendencies in animal populations High density situations can trigger this response, highlighting the complex relationship between resource availability and animal behavior.
(2006) indicated that protein and salt satiation reduced cannibalism in Mormon cricket (Anabrus simplex Haldeman (Orthoptera: Tettigoniidae)); they moved less in condition of protein satiation
Reduced mobility in damselfly larvae, specifically Lestes nympha, increases their risk of cannibalism, particularly under crowded conditions where survival rates are low despite ample food and nesting materials This high mortality rate is largely due to inadequate parental care, which includes cannibalistic behavior, although the specific factors contributing to mortality have not been quantified Additionally, cannibalism may significantly influence population density regulation in certain Lepidoptera species.
Cannibalism among larvae is influenced by specific behavioral patterns, particularly in two closely related species of Helicoverpa Helicoverpa armigera larvae exhibit active movement and aggressive responses towards other larvae, resulting in higher cannibalism rates compared to the more passive Helicoverpa puntigera, which shows significantly lower cannibalistic behavior Additionally, cannibalism can be triggered by stress factors; for instance, Gastrimargus transversus nymphs engage in cannibalism even at low population densities when food is abundant, primarily due to high temperatures and low humidity Similarly, desert locust neonates enhance their drought survival by cannibalizing their conspecifics, highlighting the complex interplay between environmental stressors and cannibalistic behavior in these species.
Cannibalism in certain species can be influenced by the presence of vulnerable individuals, with the likelihood of such behavior increasing in response to these encounters (Fox 1975) Insects like coccinellid beetles exhibit specific cannibalism patterns based on egg group size and hatching timing, as they typically avoid cannibalizing eggs or newly hatched larvae if all eggs hatch before the older larvae begin their prey search (Kaddou 1960) Interestingly, herbivorous insects may resort to cannibalism even in the presence of abundant plant food and low population density (Brower 1961) For example, Helicoverpa armigera larvae will cannibalize others that arrive later to the corn husk, despite sufficient food for multiple larvae to grow (Kirkpatrick 1957) Furthermore, the rate of cannibalism in H armigera increases with higher population density when on an artificial diet (Twine 1971).
Cannibalism among newly hatched siblings may serve as a reliable nutritional resource for young animals before they disperse While egg cannibalism differs from that of older individuals, it restricts the sharing of resources as the animals grow However, this behavior does not lead to population extinction, as the rate of cannibalism declines when resources become more abundant and vulnerable individuals are less frequently encountered Consequently, cannibals must adapt by seeking alternative food sources.
Cannibalism can interact with toxins, such as allelochemicals from plants or pesticides, potentially magnifying their effects (Joyner & Gould 1985) The widespread adoption of Bt crops is seen as compatible with biological control due to their selectivity and effective delivery methods (Wearing & Hokkanen 1994, Romeis et al 2009) Understanding the relationship between cannibalism and Bt transgenic plants is crucial for population dynamics Research by Kakimoto et al (2003) indicates that cannibalism is more prevalent under poor nutritional conditions or high larval densities of H armigera Additionally, H zea larvae exhibit higher cannibalism rates on Bt corn compared to non-Bt corn, likely due to reduced defensive capabilities of the cannibalized larvae or lower food quality (Chilcutt 2006) This behavior appears to enhance larval survival on Bt plants, making it comparable to survival rates on non-Bt plants (Chilcutt).
2006) It is suggested that cannibalism indirectly enhances fitness by enabling larvae to avoid unsuitable environmental conditions (Kakimoto et al 2003; Chilcutt 2006) Benedict et al (1992,
A study conducted in 1993 revealed that larvae were more likely to abandon Bt cotton plants compared to non-Bt cotton, often by spinning down or dropping off The larvae on Bt cotton exhibited reduced feeding time (18%) and increased resting time (60%), in contrast to their behavior on non-Bt cotton, where they spent 50% of their time feeding and only 33% resting.
A study concluded that there was no significant discrimination between Bt and non-Bt cotton plants when third instar H zea and Heliothis virescens were tested However, after 24 hours of infestation, fewer larvae from both species were observed on Bt cotton, likely due to increased mortality or movement away from Bt plants These findings indicate that larvae feeding on Bt toxins may exhibit greater mobility, leading to more frequent contact and a higher likelihood of cannibalism Consequently, cannibalism may enhance larval survival on Bt plants, making their survival rates comparable to those of larvae on non-Bt plants.
Structure of the thesis
The thesis is structured into three sections that logically progress through the research findings Section 1 (Chapter 2) focuses on the oviposition behavior of female moths on both Bt cotton and non-Bt cotton This section evaluates the oviposition preferences of moths among and within cotton plants to identify any differences in host selection behavior between two strains of moths.
H armigera larvae, and whether there was a relationship between oviposition site and the survival of first instar larvae The second section accessed the effect of larval weight on the time to death by starvation, larval recovery if food is found earlier and the ability of larvae to detect and discriminate between Bt and non-Bt diet The third section included two experiment chapters I studied the drop- off behaviour of two H armigera strains (Bt-resistant and -susceptible larvae) to determine whether they moved in the same way or differently in response to Bt and non-Bt cotton plants, and if they moved independently of toxin levels (Chapter 4) In chapter 5, experiments were carried out to determine if neonates preferred young or old eggs, and whether there was a preference for alive or dead eggs (frozen eggs) In addition, cannibalism rate and the survival of Bt-susceptible larvae were examined when they were exposed to leaf discs before and after cannibalizing eggs
The thesis is structured in Journal Article Format, with each chapter representing a distinct article Chapter 2 has been submitted and accepted for publication in the Bulletin of Entomological Research, presented in the same format as the submitted manuscript, co-authored with my supervisors The remaining chapters are also prepared as papers intended for submission to academic journals.
Journals The use of the Journal Article Format has necessitated some duplication of material between Chapters, especially in the Introduction and Discussion sections, since each
Chapter/Article will eventually be published separately.
Chapter 2: Oviposition site selection and the survival of Bt-resistant and Bt-susceptible larvae of
Helicoverpa armigera on Bt and non-Bt cotton
This study investigates the oviposition preferences of Helicoverpa armigera from both Bt-resistant and susceptible colonies when given a choice between Bt and non-Bt cotton plants We aim to determine if female moths exhibit random oviposition or preferentially lay eggs on potentially less toxic plants, such as non-Bt cotton Additionally, we explore whether different strains of moths select less toxic parts of plants based on Bt expression Furthermore, we analyze the relationship between oviposition sites and the survival rates of first instar larvae from both H armigera strains on various plant structures, including young leaves, mature leaves, stems, squares, and flowers, across both Bt and non-Bt cotton Survival rates are assessed through laboratory assays.
Chapter 3: Feeding behaviour and the survival of Bt-resistant and Bt-susceptible neonates of
Helicoverpa armigera when exposed to a diet with Bt-toxin.
Some larvae exhibit increased mobility before feeding, which may lead them to more suitable feeding locations Larger and more mobile larvae are likely to survive longer, enhancing their ability to find optimal feeding sites This raises important questions: (1) How long can H armigera larvae survive without food? (2) Are there differences in these behaviors between Bt-resistant and Bt-susceptible larvae?
I conducted choice tests to assess the ability of H armigera larvae to detect Bt toxin in an artificial diet Additionally, I examined the influence of initial larval size on starvation duration and their capacity to recover after food deprivation The larvae's responses to food scarcity could impact their survival rates and mobility in search of suitable feeding sites To evaluate these factors, I performed a comprehensive assay that tracked the larvae's movement choices and monitored their subsequent survival rates.
Chapter 4: The drop-off behaviour of Bt-resistant and -susceptible Helicoverpa armigera larvae on
Bt-cotton and non-Bt cotton plants
This chapter investigates the movement patterns of two strains of H armigera larvae, specifically focusing on Bt-resistant and Bt-susceptible individuals, to assess whether their drop-off behavior differs in response to environmental factors.
Bt and non-Bt cotton varieties play a crucial role in understanding the behavior of H armigera larvae Research aims to determine whether these larvae can detect Bt toxin levels in cotton plants or if their movement is independent of toxin concentrations This knowledge is essential for assessing how H armigera, both with and without physiological resistance, can persist on Bt cotton plants.
Chapter 5: Egg cannibalism in Helicoverpa armigera larvae: overcoming the plant establishment hurdle
Helicoverpa armigera larvae can enhance their survival by cannibalizing the eggs or siblings when food sources are scarce, particularly on Bt cotton plants This behavior of egg cannibalism may significantly improve larval survival, especially for neonates during the critical first instar stage Given the limited research on this phenomenon and its relationship with Bt toxins, it is essential to investigate how cannibalism may contribute to survival on Bt cotton, which has important implications for the development of resistance to this technology Our study focuses on the egg cannibalism behavior of H armigera from both resistant and susceptible colonies to identify their egg preferences and assess the effects of cannibalism on the survival of early-stage larvae on Bt cotton.
In the final chapter, I evaluate the research findings in relation to the initial questions and highlight possible avenues for future research This analysis explores various mechanisms that could contribute to the survival of larvae in Bt cotton fields.
Introduction
Helicoverpa armigera is a significant polyphagous pest that severely impacts agricultural crops, particularly cotton, maize, legumes, and tomatoes The extensive use of chemical pesticides to manage this pest has led to adverse effects, including the development of resistant insect populations Therefore, managing pesticide resistance is crucial, especially with the use of chemical sprays and genetically engineered crops In Australia, the introduction of genetically modified cotton expressing Bacillus thuringiensis (Bt) toxins since the mid-1990s has effectively controlled H armigera and reduced insecticide usage significantly, from 30% with single toxin varieties to 85% with dual toxin varieties Consequently, Bt cotton has been widely adopted by growers, now accounting for nearly 90% of cotton crops in the region.
Although it has greatly improved the control of H armigera, the threat of an increasing frequency of insects that are physiologically resistant to the toxins remains (Downes et al 2010)
The widespread planting of Bt cotton significantly impacts the biology and ecology of H armigera, yet these effects remain largely unknown This major crop has emerged as an effective population sink (Rochester et al 2002; Baker et al 2015) Research by Zalucki et al (2012) indicates that there has been no change in host selection by H armigera among tobacco, conventional cotton, and cabbage following the extensive adoption of genetically modified cotton in Australia.
Research conducted in 2012 involved experiments with Bt-susceptible moths, notably excluding Bt cotton from oviposition assays In the USA, Torres & Ruberson (2006) observed that H virescens and H zea exhibited no oviposition preference between non-Bt and Bt cotton, with a similar distribution of eggs on both types of plants This indicated that moths did not alter their oviposition behavior despite varying toxin concentrations in Bt cotton Additionally, a field study by Kumar & Stanley (2010) in India confirmed that H armigera moths also showed no preference for Bt over non-Bt cotton.
Research indicates that H armigera exhibits behavioral avoidance in selecting oviposition sites, effectively minimizing exposure to toxins in Bt cotton plants (Men et al 2005; Liu et al 2010) In China, studies show a preference for non-Bt cotton for oviposition, which may help reduce the selection pressure for physiological resistance (Zhao et al 2016) Additionally, in mixed plantings, moths tend to favor non-Bt cotton for laying eggs (Liu et al 2010) In Gujarat, India, larger patches of Bt cotton corresponded with a decrease in egg density (Lodaya & Borad 2014) Notably, the number of eggs laid on conventional cotton was approximately 95% higher than on Bt cotton during both the bud-flower and flower-boll stages (Lodaya & Borad 2014).
Research on oviposition behavior in Australian populations of H armigera has primarily focused on Bt-susceptible moths To understand the implications for resistance evolution, it is essential to conduct experiments comparing the oviposition behavior of both Bt-resistant and Bt-susceptible H armigera moths on Bt cotton This comparison will help clarify whether differences in adult and larval behavior contribute to the differential survival of these genotypes on Bt cotton plants.
This study investigates the oviposition preferences of Helicoverpa armigera from both Bt-resistant and susceptible colonies when presented with Bt and non-Bt cotton plants We aim to determine whether female moths exhibit random oviposition choices or prefer potentially less toxic plants, such as non-Bt cotton Additionally, we explore if moths from different strains select less toxic plant parts based on Bt expression, using published toxicity data as a reference Finally, we analyze the correlation between oviposition sites and the survival rates of first instar larvae from both strains on various plant structures, including young leaves, mature leaves, stems, squares, and flowers, across both Bt and non-Bt cotton, through laboratory survival assays.
Materials and methods
The H armigera Bt-resistant strain SP15 was established from a single mating pair collected as eggs on corn in Griffith, NSW, in December 2002 Progeny from this pair underwent an F2 screen to identify resistance to Cry toxins, resulting in the formation of the SP15 colony from F2 offspring that survived a discriminating dose of Cry2Ab Initially, SP15 had a limited gene pool due to its origin from a single isofemale line, which can lead to inbreeding depression and reduced vigor in Lepidopteran colonies To counteract this, SP15 has been outcrossed with the Bt-susceptible strain GR multiple times to enhance fitness and develop a strain that is nearly isogenic with the susceptible counterpart.
In 2007, after each outcross, the colony was maintained without selection for one generation before being re-selected using a diet surface treatment of 1-2 µg/cm² Cry2Ab toxin The source of the Cry2Ab toxin was dried and ground corn (Zea mays L.) leaf material, provided by Monsanto as lyophilized leaf powder containing 6 mg/g of the transgenically expressed B thuringiensis crystal protein Cry2Ab The concentration of the toxin in the leaf was calibrated using an enzyme-linked immunosorbent assay (ELISA) after freeze-drying and homogenization, with detailed methods found in Holt et al (2002) All subsequent generations were selected at this toxin dose Moths used to establish a Bt-susceptible H armigera colony were collected from various crops, including chickpea, pigeon pea, and cotton, and were bulk mated to form a colony, which was maintained at the Australian Cotton Research Institute in Narrabri, New South Wales.
Bt-resistant (SP15) and Bt-susceptible (GR) eggs of H armigera were transferred to the School of Biological Science at The University of Queensland, Australia, to establish experimental colonies Neonates were reared individually on a modified artificial diet until the 3rd instar stage in 45-well plastic trays, after which they were moved to 32-well trays for pupation The rearing trays were sealed with a perforated lid, and male and female pupae were separated and housed in vermiculite within an incubator at 25±1°C and 80% humidity for synchronous adult emergence Fifteen male and fifteen female moths were then placed together in plastic holding containers, which were securely covered and provided with a substrate for egg laying, along with a 10% honey/sugar solution as a food source.
Monthly testing of Bt-resistant and -susceptible colonies was conducted to monitor their responses to Bt toxin and assess resistance levels Only those families confirmed to be non-resistant through screening were included in the Bt-susceptible colony.
In experiments designed to test oviposition preference, a transgenic cotton cultivar containing Cry genes (Bollgard II® cotton, Sicot 71 BRF) was compared to a conventional cotton variety (Sicot 71 RRF, referred to as "non-Bt") that shares the same genetic background but lacks the Cry genes To facilitate the study, a pool of plants was established by sowing three seeds per pot in a UC soil mix, using pots that are 30 cm in height and 25 cm in diameter.
In a controlled greenhouse environment at The University of Queensland, a mixture of sand, bark, and peat moss was used to cultivate plants After two weeks, the healthiest seedling was selected, while the others were discarded The plants were kept at a temperature of 24 ± 6 °C and a relative humidity of 56 ± 10% RH, receiving water three times a week and a general-purpose soluble fertilizer (Thrive 16: 9: 12: 2 N: K: P: MgO) every four weeks For oviposition choice tests, plants of similar size (50–60 cm in height) and development stage, featuring open flowers, squares, and bolls, were utilized.
Figure 2.1 Materials used for oviposition experiments in glasshouse (A) large cage; (B) cotton plants arranged in large cage; (C) Bt cotton plant; (D) non-Bt cotton plant
Between December 2012 and April 2014, four experiments were conducted using eight cages for two different larval strains A total of 108 cotton plants were utilized for each strain, along with over 240 female moths The experiments were organized into distinct stages: the first two replicates took place in December 2012, the third replicate in July 2013, and the fourth, fifth, and sixth replicates in November 2013, concluding with the final stage in April 2014.
In 2014, replicates 7 and 8 involved releasing groups of 30 moths (15 males and 15 females) into a large cage (180cm x 180cm x 180cm) containing cotton plants within a glasshouse Each cage housed twelve to sixteen plants, consisting of either 6 or 8 Bt cotton and 6 or 8 non-Bt cotton plants, arranged randomly To provide additional nutrients, pots with a 10% sucrose solution were placed in a plastic box with water inside the cage The moths were acclimated to their new environment and permitted to lay eggs for two nights.
On the second day of the experiment, moth survival was assessed, and additional moths from the laboratory culture were added to maintain a total of 30 moths (15 males and 15 females) in each cage All eggs laid within the first two nights were removed from the plants to ensure cleanliness for subsequent experiments To minimize positional effects, the plants were re-randomized daily Egg counts were conducted the following day, focusing only on the eggs present on the plants, as it was not feasible to track the number of eggs placed in the cage The eggs were categorized based on their location on the plant structures, including immature leaves, mature leaves, stems, squares, or flowers, and counted over a single day.
In a study conducted from December 2012 to April, the average number of eggs laid per plant by female H armigera moths, both Bt-resistant and susceptible, was recorded on Bt cotton and non-Bt cotton plants The results, presented in Table 2.1, indicate a comparison of egg-laying behavior in controlled cage conditions, with a sample size of eight.
2014 Means within a column followed by same letter are not significantly different (α = 0.05, Tukey’s Multiple Range Test) The replicate with plants at different growth stages is highlighted
In the third replicate conducted in July 2013, Bt cotton plants exhibited 3-4 more nodes compared to non-Bt cotton, despite both varieties being the same height The number of eggs laid on Bt cotton was significantly higher, with 355 eggs per plant, compared to only 143 eggs per plant on non-Bt cotton Consequently, this replicate was excluded from further statistical analysis, while all other experiments maintained similar height and developmental stages among the plants.
2.2.4 Survival of newly hatched larvae
Experiments were conducted in a temperature-controlled environment at 25 ± 1°C with a natural light cycle to evaluate the survival of newly hatched larvae Neonates were placed in round plastic containers, each containing a small pot filled with water to maintain the freshness of plant parts for two days The experimental setup included large containers covered with polypropylene nappy liners for adequate air circulation, with four treatments (young leaf, mature leaf, square, and flower) for each cotton strain (SP15 and GR), including both Bt and non-Bt varieties Ten neonates were introduced into each treatment for two days, with ten replicate experiments performed for each plant part and larval strain After two days, the surviving larvae were counted and subsequently transferred to an artificial diet to assess their survival over an additional four days, reaching a total age of six days.
Statistical analyses were performed using the Statgraphic Century procedure, version 15.1
Statpoint Technologies, Inc conducted a study in Washington D.C to analyze the variation in total egg numbers across different cages, expressing the data as a percentage of total egg lay per plant To address heterogeneity of variances, all percentage data were arcsin-transformed before undergoing analysis of variance (ANOVA), focusing on strains and plant lines as primary factors The distribution of eggs between two moth strains on parts of Bt versus non-Bt cotton plants was assessed using Wilcoxon Signed-Rank Tests Additionally, oviposition preference and larval survival rates on Bt and conventional cotton were evaluated with a two-way ANOVA Mean percentages of oviposition preference on various plant structures and larval survival across different plant positions were further analyzed using Tukey’s Multiple Range Tests.
Figure 2.2 Helicoverpa armigera eggs were laid on different cotton plant structures (A) young leaf; (B) mature leaf; (C) stem; (D) square; (E) flower; (F) boll
Results
In the analysis of preference, data from 7 out of the 8 replicates were utilized due to one replicate containing plants at varying phonological stages The results indicated no significant differences in oviposition percentages between Bt-resistant and Bt-susceptible moths, with statistical values of F1, 191 = 1.89 and P > 0.05.
Both Bt-resistant and Bt-susceptible moths showed no preference for Bt cotton over non-Bt cotton, with statistical analyses indicating no significant differences (Bt-resistant: F1, 95 = 0, P = 0.97; Bt-susceptible: F1, 95 = 0.29, P = 0.59) The proportion of eggs laid by Bt-resistant moths on Bt cotton was identical to that on non-Bt cotton, with each receiving 50% of the eggs Similarly, Bt-susceptible moths laid 48% of their eggs on Bt cotton and 52% on non-Bt cotton.
The mean percentage of eggs laid by Bt-resistant and Bt-susceptible female H armigera moths on various plant structures was analyzed, comparing Bt cotton (represented by white bars) and non-Bt cotton (represented by black bars).
The study found no significant differences in the egg distribution of Bt-resistant moths across various parts of both Bt and non-Bt cotton plants The analysis revealed similar patterns in young leaves (t = 0.84, P = 0.40), mature leaves (t = 0.09, P = 0.92), squares (t = -0.52, P = 0.60), stems (t = -1.34, P = 0.18), and flowers (t = 1.17, P = 0.24).
The oviposition patterns of Bt-susceptible moths showed no significant differences between Bt and non-Bt cotton across various plant structures, including young leaves (0.84, P = 0.40), mature leaves (t = -1.31, P = 0.19), stems (t = 0.18, P = 0.85), and flowers (t = -1.41, P = 0.16) However, a notable exception was observed in the squares of Bt cotton, where susceptible moths laid significantly more eggs compared to non-Bt cotton, with rates of 8% and 5%, respectively (t = 2.21, P = 0.03).
The study found no significant effect of cotton plant line (Bt vs non-Bt) on the survival of Bt-resistant larvae across various plant structures, with survival rates of 75% on Bt cotton and 81% on non-Bt cotton In contrast, Bt-susceptible larvae showed a significant difference in survival rates, with only 59% surviving on Bt cotton compared to 81% on non-Bt cotton Notably, survival rates for Bt-susceptible larvae varied significantly on young leaves, mature leaves, and squares between the two cotton lines.
A study comparing the survival rates of Bt-resistant and susceptible H armigera larvae on various parts of Bt and non-Bt cotton revealed significant differences After two days, the mean percentages of larvae that survived on young leaves, mature leaves, squares, and flowers were recorded Notably, Bt-susceptible neonates showed a significant difference in survival on flowers between Bt cotton and non-Bt cotton, highlighting the effectiveness of Bt cotton in pest management.
A study revealed no significant difference in the survival rates of Bt-susceptible larvae on Bt cotton flowers compared to non-Bt cotton (F1, 1, 95 = 3.38, P = 0.08) However, larval survival was notably higher on flowers than on young leaves, mature leaves, and squares of Bt cotton (F3, 95 = 3.62, P = 0.02) Overall, both Bt-resistant and -susceptible larvae showed similar survival rates on non-Bt cotton Interestingly, Bt-resistant larvae had a better survival rate on Bt cotton than their susceptible counterparts, with no significant differences among various plant parts or cotton lines While Bt-susceptible larvae generally fared worse on Bt cotton, they exhibited improved survival on squares and flowers compared to other structures of Bt cotton.
After two days of exposure to plant parts, the larvae were transferred to an artificial diet and evaluated again at six days old The initial cotton variety did not influence the survival rates of Bt-resistant larvae after an additional four days on the diet While the survival percentage of Bt-resistant larvae on non-Bt cotton structures was higher (63%) compared to Bt cotton parts (59%), this difference was not statistically significant (F1, 19 = 0.40, P = 0.59) Furthermore, no significant differences were observed in the survival rates of Bt-resistant larvae on the diet after exposure to various cotton structures, including young leaves (F1, 19 = 2.59, P = 0.25), mature leaves (F1, 19 = 5.49, P = 0.94), squares (F1, 19 = 0.01, P = 0.94), and flowers (F1, 19 = 0.54, P = 0.54).
After four days on a diet, the survival rates of Bt-susceptible larvae significantly decreased following initial exposure to various structures of Bt cotton plants Specifically, survival dropped from 50% to 25% on young leaves, from 53% to 25% on mature leaves, and from 63% to 38% on squares, while the decline on flowers was less pronounced, from 70% to 47% These variations among the different plant structures were statistically significant (F 3, 39 = 3.62, P = 0.02).
A study analyzed the survival rates of Bt-resistant and susceptible H armigera larvae on an artificial diet after a 6-day period, which included 2 days feeding on various plant parts of Bt and non-Bt cotton, followed by 4 days on the artificial diet The results showed distinct mean percentages of survival for both resistant and susceptible larvae, highlighting the impact of Bt cotton on pest resistance.
Asterisks identified the significant difference in survival of Bt-susceptible larvae on flowers between Bt cotton and non-Bt cotton.
The survival rates of Bt-susceptible larvae showed no significant difference on diet after four days of exposure to non-Bt cotton structures, with an average survival decrease of approximately 14%, similar to that of Bt-resistant larvae However, the number of surviving Bt-susceptible larvae varied significantly when exposed to Bt versus non-Bt cotton plant parts, including young leaves (F1, 3 = 835, P = 0.00), mature leaves (F1, 3 = 65.11, P = 0.02), and squares (F1, 3 = 25.14, P = 0.04), while the difference in flowers approached significance (F1, 3 = 15.38, P = 0.06).
After four days on a diet, Bt-susceptible larvae exposed to Bt cotton structures showed significantly lower survival rates compared to those on non-Bt cotton, with survival rates being higher on the squares and flowers of Bt cotton Specifically, the average survival of Bt-susceptible larvae declined by 25% to 28% on young leaves, mature leaves, and squares of Bt cotton In contrast, the survival rates of Bt-resistant larvae on both Bt and non-Bt cotton plants remained similar after 2 and 6 days.
Discussion
Over a two-year period, oviposition experiments conducted under varying weather conditions revealed significant variability in the number of eggs laid per plant The findings indicated that both Bt-resistant and -susceptible moths exhibited similar behaviors when selecting sites for egg-laying.
Both Bt and conventional cotton, sharing the same genetic background, equally attracted eggs from moths The egg-laying percentages of both Bt-resistant and susceptible moths on Bt and non-Bt cotton showed no significant differences, indicating that the moth strains did not exhibit a preference for either type of cotton Additionally, the stage of the plant plays a crucial role in making these comparisons.
In our experiment, we enhanced Bt cotton to produce more flowers and fruiting structures Despite both Bt and non-Bt cotton plants being similar in height, the Bt plants attracted a greater number of eggs This outcome underscores the importance of carefully controlling assays that evaluate host preference between Bt and non-Bt cotton.
In studies of cotton plants, it was observed that female moths, particularly Helicoverpa spp and Pectinophora gossypiella, preferred to lay their eggs on young, hairy leaves, which are more toxic than other parts like squares, regardless of the plant's resistance to Bt toxins Research indicated that oviposition was largely independent of the toxicity levels associated with Bt and non-Bt cotton, as both types received similar egg counts, with 80-95% of eggs laid on the top nodes However, Bt-susceptible moths exhibited a notable preference for laying more eggs on squares of Bt cotton compared to non-Bt cotton, potentially allowing larvae to survive as these squares develop into flowers that express lower levels of toxins.
Data from the larval survival assay indicate that the mortality rate of Bt-susceptible H armigera neonates is significantly higher on various structures of Bt cotton plants, including young leaves, mature leaves, and squares, compared to non-Bt cotton This aligns with earlier research findings (Gore et al 2001; Kranthi et al 2005; Arshad et al 2009; Lu 2010) Additionally, H zea larvae exhibit higher survival rates on squares and flower anthers than on other floral structures in both non-Bt and Bollgard II® cotton (Gore et al 2001; Kranthi et al 2005) Furthermore, Arshad et al (2009) reported significantly greater mortality in larvae that consumed Bt cotton leaves compared to those that fed on Bt flower-bolls.
Early-stage Bt-susceptible larvae were predominantly found alive on the flowers of Bollgard II® cotton, indicating that these plant parts may have lower levels of Bt toxin.
Bt-susceptible larvae show a notable preference for flowers, which may be linked to the expression of Cry proteins in different plant structures, particularly the levels of Cry1Ac expression.
Bollgard ® cotton plants are typically higher on vegetative tissues compared to floral structures, such as pollen and flower petals (Greenplate 1999; Adamczyk et al 2001b; Gore et al 2001) In
In Australian growing conditions, Cry1Ac expression levels were found to be highest in leaves, with lower levels in squares and flowers, as confirmed by Lu (2010) Within the floral structures, bracts exhibited the highest expression of Cry1Ac, followed by petals, while anthers showed the lowest levels Our study corroborated the finding that flowers generally express low levels of Bt toxin, as there was no significant difference in the survival rates of Bt-susceptible larvae on flowers from both Bt and non-Bt cotton This low expression in Bollgard II® cotton squares and flowers provides an opportunity for neonate larvae to survive by seeking out plant parts with reduced Bt toxin levels Additionally, the oviposition behavior of moths, which tends to favor squares for egg-laying, may enhance the survival of hatching neonates at the flowering stage if they feed on these less toxic plant parts.
Adult moths did not show a preference for Bt-expressing plant structures, but Bt-susceptible moths laid more eggs on squares, potentially increasing larval survival on flowers About 15% of eggs were deposited on Bt squares, with 70% of those larvae surviving on Bt flowers after two days, and 43% continuing to survive after four days on a non-toxic diet Overall, 6-7% of larvae could survive on Bt cotton if they found less-toxic food early on Current thresholds for chemical control are two small larvae or one medium larva per meter If H armigera females laid 50 eggs per plant, it is estimated that three larvae could survive, reaching the threshold for spraying However, since the introduction of Bt cotton in Australia, egg counts per plant have significantly decreased, making 50 eggs per plant unlikely High local egg loads during the squaring stage and larval movement may explain occasional high levels of non-resistant larvae Research suggests that H armigera larvae might increase their survival by moving from high-toxin areas, like leaves, to less toxic food sources such as flowers.
In summary, research indicates that both Bt-resistant and Bt-susceptible female H armigera moths in Australia exhibit similar oviposition behaviors on Bt and non-Bt cotton plants, with only minor variations Notably, approximately 20% of Bt-susceptible larvae survived two days on Bt cotton, suggesting potential strategies such as feeding avoidance or minimal feeding that prevented them from ingesting a lethal dose of the toxin Further studies are needed to explore whether H armigera larvae can detect Bt toxin, endure starvation, or recover after periods without food.