Insecticidal Effect of Cry Toxins Produced by Bacillus thuringiensis on Diceraeus melacanthus (Dallas, 1851) and Euschistus heros (Fabricius, 1798) (Hemiptera: Pentatomidae)

Some species of the Heteroptera (Hemiptera) suborder are of great agricultural importance, mainly as pests, and their control is often necessary. The use of Bacillus thuringiensis , an entomopathogenic bacterium normally extracted from the soil and used in biological control, is an alternative to the chemical control of these insects. Mortality tests must be carried out in order to select and determine a viable toxic strain, but currently there is no validated methodology for conducting those tests. In this context, this research aimed to develop and improve a selective bioassay methodology to assess the toxic effect of B. thuringiensis Cry toxins on Diceraeus melacanthus (green-belly stink bug) and Euschistus heros (neotropical brown stink bug) nymphs. A bioassay methodology consisting of tubes and artificial diet was proposed. Bioassays with D. melacanthus and E. heros nymphs were performed incorporating Cry toxins (Cry1Aa, Cry1Ab, Cry1Ac, Cry1B, Cry1C, Cry1F, Cry1G, Cry1Ia, Cry2Ab, Cry2A, Cry2Ae, Cry4A, Cry4B, Cry10, and Cry11Aa) into their liquid diet. The artificial feeding system developed in order to carry out the stink bug mortality tests was conducted. Among the toxins tested, we can highlight 2 causing 80-85% nymphal mortality on D. melacanthus , and 4 toxins causing 90-100% nymphal mortality for E. heros after 7 days of incubation. Both species are susceptible to different Cry toxins, with emphasis on Cry2Ab and Cry4B for D. melacanthus and Cry1B, Cry1G, Cry1Ia and Cry2Ab for E. heros .


Introduction
Species in the Heteroptera suborder (Insecta: Hemiptera) are commonly known as true or typical bugs. Most species are phytophagous, but there are predators and hematophagous species as well. They are distributed in practically all regions of the world, and have great agricultural importance as pests of many different crops (Possebom et al., 2020). The main difference between Heteroptera and other insects of the Hemiptera order is the presence of scent glands located in the abdomen as nymphs, and in the abdomen and thorax as adults, with different functions depending on the species (Panizzi et al., 2000;Possebom et al., 2020). Within the Heteroptera suborder, Pentatomidae is the most abundant family of insects found in many crop areas in Brazil, causing significant losses and damage (Panizzi et al., 2012;Sosa-Gómez et al., 2020;Steinhaus et al., 2022).
Bacillus thuringiensis is a Gram-positive aerobic spore-forming bacterium of the Bacillaceae family, of cosmopolitan coverage (Krywunczyk & Fast, 1980;Bravo et al., 2011), and can be found in several substrates such as soil, water, plant surfaces, dead insects, spider webs, and stored grains (Bravo et al., 1998;Valicente, 2019). The entomopathogenic activity of this bacterium is due to proteinaceous inclusions produced during the sporulation phase, which are crystals composed of proteins called endoproteins or crystal proteins (Monnerat & Bravo, 2000;Bravo et al., 2017;Chen et al., 2021). Also known as Cry toxins or Cry proteins, delta-endotoxins participate in the formation of protein crystals, linked to bacterial sporulation, formed from the phase following sporulation, and released when cells are lysed (Bravo et al., 2013). These proteins are the most used for insect biocontrol, they have an action spectrum usually restricted to a specific order of insects (Palma et al., 2014) and have an extremely toxic action (Bravo et al., 2013;Adang et al., 2014;Jurat-Fuentes & Crickmore, 2017).
The use of B. thuringiensis in insect pest biocontrol has many advantages such as specificity to target insects, a non-polluting effect on the environment, innocuousness to mammals and vertebrates, and non-toxicity to plants, which allows its direct application (Whiteley & Schnepf, 1986;Schünemann et al., 2014;Jurat-Fuentes & Crickmore, 2017). Bacillus thuringiensis has merited researchers' attention for controlling insect larvae/worms/nymphs of the Lepidoptera, Diptera, Coleoptera, Orthoptera, and Hemiptera orders, in addition to other organisms such as mites, nematodes, and protozoa (Palma et al., 2014).
Few studies have been carried out to evaluate the toxicity of Cry toxins with regard to insect mortality in the Heteroptera suborder. Therefore, this research aimed to develop and improve on a selective bioassay methodology to assess the toxic effect of individual Cry toxins on D. melacanthus and E. heros nymphs, under laboratory conditions.

Insects Used in the Bioassays
This study was performed using D. melacanthus and E. heros second-instar nymphs obtained at the Insect Rearing Platform of Embrapa Genetic Resources and Biotechnology, Brasilia, Brazil. The insect-rearing conditions were as described by Blassioli-Moraes et al. (2014).

Artificial Feeding System Initial Setting
The artificial feeding system consisted of a sterile centrifuge tube (50-mL Falcon tube) containing seven nymphs ( Figures 1A and 1B) in triplicate. Each tube with nymphs was considered an experimental unit. The tubes were covered with Parafilm® (3 × 3 cm), sterilized under ultraviolet light (UV) and elongated to twice its size. About 100 µL of liquid aphid diet (Dadd & Mitter, 1966) were placed over Parafilm® ( Figure 1C) and then 50 µL of the treatment (first aniline blue, then Cry toxins). Soon after, another Parafilm® of the same size was used to cover the droplet in order to produce a sachet containing the mixture ( Figure 1D). The liquid diet was previously filtered through a Millipore® membrane (0.22 µm) with the aid of a sterile syringe and stored at -20 °C until the moment of use. The entire procedure took place in a laminar flow cabinet with UV-sterilized materials. The tubes with the feeding system and insects were placed on shelves with the feeding system facing upwards (Figures 1E and 1F)  Purified pr buffer (Na polyacryla polyacryla Coomassie concentrat mL) of the trials.

Cry To
The Dicer toxins (Cr Cry4B, Cr sachets in treatment.

Statisti
To test the Where LIC Treatment

Cry To
Between t mortality o 23.8% (Ta  Note. (-) data not available, the number of live nymphs in the Cry treatment was higher than in the control treatment. * Cry toxins efficiency on nymphs > 75%. Table 3 shows the trial's findings regarding the average number of dead insects, absolute mortality, and Cry toxins' efficiency on D. melacanthus nymphs at the end of 7 days of incubation. For the average number of dead insects, the Cry4B toxin had a 6.0 average compared to 1.6 in the control, with a total of 18 and 5 dead nymphs, respectively, and a sample size of 21 nymphs (n = 21) for each treatment. The Cry1F toxin was not efficient on nymphs. Lowest efficiency indices were observed for Cry1B and Cry1G toxins (6.2%). Note. * E. h   Note. (-) data not available, the number of live nymphs in the Cry treatment was higher than in the control treatment. * Cry toxins' efficiency on nymphs observed > 80%.
For the average number of dead insects, the Cry1Ia and Cry2Ab toxins had an average of 7.0 dead nymphs compared to 2.6 in the control ( Table 6). The absolute mortality of nymphs exposed to these same toxins was 21 compared to 8 in the control. Therefore, the Cry1Ia and Cry2Ab toxins' efficiency was 100% on E. heros nymphs. After seven days of incubation, the Cry1Aa and Cry10 toxins were not efficient, Cry1C and Cry11Aa toxin treatments had more live insects at the end of the trial than the control treatment. Note. a average number of dead E. heros nymphs from three replicates with 7 nymphs each. b total number of dead insects in treatments with 21 nymphs for each tested toxin.

Discussion
The high damage potential of Euschistus heros and Diceraeus melacanthus had been previously reported in the literature (Torres et al., 2013;Gomes et al., 2020;Cordeiro & Bueno, 2021;da Silva et al., 2021). Among the options that seek to control insect pests without the use of chemical groups and with high specific toxicity, the use of microorganisms has taken an important position (Schünemann et al., 2014;Sosa-Gómez et al., 2020). Bacillus thuringiensis and their toxins are the most manipulated microbial pesticides worldwide and have been accepted by the public as a safe bioinsecticide (Bravo et al., 2011;Ruan et al., 2015).
While studies show that the hemipteran insects are susceptible to Cry protein toxicity (Porcar et al., 2009;Dorta et al., 2010;Melatti et al., 2010;Salazar-Magallon et al., 2015;Torres-Quintero et al., 2016;Schünemann et al., 2018;Torres Cabra et al., 2019;da Costa et al., 2021), brief is the knowledge about the effect of Cry toxins on pentatomids (Schünemann et al., 2014). Research using Cry proteins directly on insects in the Heteroptera suborder are scarce and focused on the effects that genetically-modified (GM) plants (plants expressing insecticidal crystalline proteins derived from B. thuringiensis) can also have on non-target organisms and in the third trophic level of the food chain (Wellman-Desbiens & Côté, 2005;da Cunha et al., 2012;Silva et al., 2014).
According to our results, the diet proved to be satisfactory and well accepted by the stink bugs at the tested artificial feeding system, conceivably, the adequate one for bioassays using Cry toxins. This is important to prove that insects were fed with the provided diet and, consequently, ingested bacterial proteins offered in each treatment. An artificial feeding system, like the one used in this research, was efficient for the sucking pest Myzus persicae (Sulzer, 1776) (Hemiptera: Aphididae) mortality testing (Paula et al., 2015). Some methodologies for stink bugs have already been established for chemical testing, mainly with 20-mL glass vials (Snodgrass, 1996;Snodgrass et al., 2005;Lopez Jr et al., 2012a, 2012b. Treatments with Cry toxins against Diaphorina citri Kuwayama (Hemiptera: Liviidae) nymphs in Citrus sinensis (L.) Osbeck systemically colonized seedlings (Dorta et al., 2010) showed different toxicities regarding the individual proteins. The highlights in that study were Cry4B, Cry10, Cry11, and Cyt1A, which caused around 65% mortality in D. citri nymphs. Interestingly, the results found in the current study for the Cry4B toxin showed a mortality rate above 85% in D. melacanthus nymphs, while the Cry10 toxin obtained the lowest rates (38.1%) on E. heros. The Cry2Ab protein demonstrated 100% mortality on E. heros assured within the first 48 hours (90.4% mortality). In our study, the Cry2Ab protein resulted in 100% mortality results when ingested by E. heros, and 80.9% on D. melacanthus. It could be seen as a possible interaction between this protein and these insects' intestine. In the interaction assessment of Cry toxins with the intestine of the hemipteran Lygus hesperus Knight (Heteroptera: Miridae), it was observed that there was no interaction with Cry1Ac, but Cry2Ab presented a strong extracellular interaction (Brandt et al., 2004).
The mechanism of action of different proteins has been identified as receptors for Cry toxins in the lepidopteran midgut (de Maagd et al., 2012;Bravo et al., 2017), but not for pentatomids, and, when reported, it is related to the third trophic level. Therefore, the insect gut pH, the presence of specific receptors on microvilli of midgut cells, as well as the proteolytic activation of Cry toxins ingested by insects are essential and necessary factors in the interactions for the occurrence of toxicity (Li et al., 2011;Bravo et al., 2013;Javed et al., 2019;Chen et al., 2021). Consequently, those factors may be associated with the differences in mortality obtained between both tested stink bug demonstrates by individual Cry toxins. However, there is little information regarding the digestive physiology of hemipterans associated with Cry toxins, and the literature is insufficient to elucidate about the mechanism of resistance or susceptibility of these insects (Schünemann et al., 2014).
Subsequently, spores and vegetative cells of B. thuringiensis were detected in the midgut of the predator Podisus nigrispinus (Dallas, 1861) (Hemiptera: Pentatomidae) fed on Bombyx mori (Lepidoptera: Bombycidae) treated with B. thuringiensis kurstaki (HD1) (Nascimento et al., 1998). Histological sections of the P. nigrispinus midgut shows the effects of proteins Cry1F, Cry1A.105, and Cry2Ab2, with histopathological changes, after predating Spodoptera frugiperda (J.E. Smith, 1797) (Lepidoptera: Noctuidae) fed on GM Bt-maize but not being lethal to 3 rd trophic level (Souza et al., 2021). In additional, other studies showed that the Cry1Ac toxin caused ultrastructural changes in the digestive cells of the predatory stink bug P. nigrispinus when it fed on S. frugiperda that had consumed GM Bt-cotton expressing the toxin (da Cunha et al., 2012). Nevertheless, mortality or impacts were not observed for E. heros fed on GM Bt-soybean plants expressing the Cry1Ac protein Schünemann et al., 2018). In the present study, the Cry1Ac protein caused low mortality (33.3%) in D. melacanthus and reasonable mortality (61.9%) for E. heros in a direct exposure to the toxin. The susceptibility to Cry toxins was evidenced for E. heros nymphs exposed in the absence of substrate, but only the combination of toxins (contained Cry2, Cry1, and Cry9 proteins) resulted in the highest mortality and efficiency (> 98%) jas.ccsenet.org Journal of Agricultural Science Vol. 14, No. 9; 2022 (Schünemann et al., 2018). B. thuringiensis' spectrum of action may depend on the combination of individual Cry toxins revealing synergisms between them (Estruch et al., 1997;Schünemann et al., 2018).
Euschistus heros and D. melacanthus are susceptible to Cry toxins correlating effect directly with the average mortality of these insects' pests. Individual toxins produced by B. thuringiensis have insecticidal activity and different degrees of lethality in both phytophagous-sucking insects of the Heteroptera suborder. Nonetheless, assays of protein interaction with the insect gut are suggested and can help with more information regarding the digestive physiology of hemipterans associated with Cry toxins, as well as the verification of interaction and possible synergism between the toxins against these pests. In this way, we can expand knowledge about the pathosystem so that toxins can be most efficiently used, with other strategies to control, against the stink bug pests.

Conclusions
The Cry toxin consumption system through artificial diet in Falcon tubes is efficient and can be used for phytophagous stink bug species. The two stink bug species evaluated in this study are susceptible to different Cry toxins, with emphasis on Cry2Ab and Cry4B for D. melacanthus and Cry1B, Cry1G, Cry1Ia, and Cry2Ab for E. heros.