In this study we analyzed the different steps in the mechanism of action of de Cry toxins to determine if the formulation itself may affect performance of Bt Cry toxins. Specifically, we compared commercial non-encapsulated products with microencapsulated products named MP-Bta and MP- Btk. The same Bt strains were used as the basic material to make Xentari and MP-Bta, or to make Dipel and MP-Btk. According to the manufacturer, Dipel® is formulated with Bt kurstaki, containing the proteins Cry1Aa (15%), Cry1Ab (39%), Cry1Ac (23%), Cry2Aa (22%), while Xentari® is formulated with from Bt aizawai, containing Cry1Aa (21%), Cry1Ab (53%), Cry1Ca (20%) and Cry1Da (6%) (Valent 2022).
Our first result showed that non-encapsulated commercial products showed a higher concentration of total protein than the microencapsulated formulations after suspension of similar water volume. Similar protein concentrations of these formulations were then subjected to solubilization analysis. We did not observe changes in the profile of the solubilized protoxin bands of the SDS-PAGE analysis, indicating that the matrices used in the microencapsulation process did not influence the solubilization of Cry proteins. However, the results presented in Table 1 show that none of the tested formulations have 100% solubilization efficiency. It important to mention that these formulations (conventional and microencapsulated) present a mixture of Cry toxins, which may influence their solubilization process. Aronson et al. (1991), tested different solubilization buffers for Bta and Btk strains, obtaining solubilization efficiencies that varied from 8 to 70%, and these authors concluded that the solubilization differences between these strains may be related to the protoxin composition of the crystals. Our data confirmed that Btk strain showed higher solubilization efficiency in relation to the Bta strain. In relation to conventional and microencapsulated formulations, variations in solubilization values were observed, with microencapsulated formulations showing lower solubilization values in in vitro conditions. However, when midgut juice was used to solubilize these samples, it was evident that microencapsulated formulations showed a better solubilization than the conventional formulations.
The solubilization of Bt parasporal crystals is a fundamental step to initiate the intoxication process. It was described that the solubilization process is facilitated by the physicochemical conditions (mainly pH and reducing conditions) that are found in the host's digestive fluids (Gill et al. 1992; Bravo and Soberón 2008; Deist et al. 2014). In the work by Du et al. (1994), the authors compared two strains (Bta and Btk) in relation to the solubility of their parasporal crystals. The authors showed that for both strains, the solubilization started with pH values of 9.5, reaching a complete solubilization only when pH reaches values above pH 11, and it was proposed that similar high pH conditions could be found inside the midgut lumen of the lepidopteran larvae. The authors describe that the different insecticidal crystals produced present distorted and destabilized disulfide bonds, which can also influence the processes of solubilization and toxicity. In another study, Naimov et al. (2008), evaluated the solubilization of crystals of Bt thompsoni HD542 composed of Cry15Aa toxin. The authors tested different buffers and pH values (ranging from 6.0 to 11.0) with or without the reducing agent DTT. The authors obtained complete protein solubilization only at pH 11.0 in sodium hydrogen carbonate buffer and CAPS buffer, respectively. In the presence of the reducing agent, the reduction of disulfide bonds allowed solubilization at pH 10. Other buffers (ethanolamine, Tris, borate-buffered saline) did not solubilize a significant amount of protein, with or without the addition of DTT. Here we show that the midgut lumen on S. frugiperda has a pH of 9.5, and solubilization is extremely low under these conditions. Only after increasing the pH up to 10.5 we were able to observe complete solubilization and activation of Cry proteins. However, our results also demonstrate that incubation time is also important for solubilization. We reinforce the importance of studying additional factors, which can contribute to altering the pH values of the midgut, consequently altering the solubility and toxicity of these proteins, such as, for example, the different instars and also the geographic region where the pest may be found (Bravo and Soberón 2008). It could be worth to isolate in the future some Cry mutants with improved solubilization, specially at lower pH values that may correlate with improved toxicity.
During the activation step, approximately 40–60 amino acids from the N-terminus are removed by proteases for 70 kDa and 130 kDa protoxins. For the 130 kDa protoxins, in addition to N-terminal processing, about 500–600 amino acids are also cleaved out from the C-terminus, in both cases the active Cry toxins resulted in activate toxin proteins of ~ 55 to 65 kDa (Bravo et al. 2002; Bergamasco et al. 2013; Gómez et al. 2014). Our results corroborate the proteins found in Bta and Btk were correctly activated, since there was no great difference in the proteolytic patterns of the different formulations. It was possible to observe a double band of ~ 70 and ~ 65 kDa for the formulations containing mixture of Cry1 and for the formulations containing Cry2 it was possible to observe a band of ~ 55 kDa. Previously, Liu et al. (2020), obtained similar proteolytic profiles when they evaluated activation of Cry1Ac and Cry2Ab protoxins by proteases from the midgut juice of Helicoverpa armigera larvae. We also noticed that the profile observed for the cleavage with chymotrypsin differs from the other treatments, being possible to observe with more precision the bands of ~ 55kDa for the formulations containing Cry2 protoxin. However, the protein profiles obtained after incubation with trypsin and midgut juice were very similar, suggesting that it is possible that midgut juice contains high levels of trypsin and low levels of chymotrypsin. Actually, Saadaoui, Rouis and Jaoua (2009) showed that trypsin-like activity is predominant in the midgut homogenate of Ephestia kuehniella. However, it is worth noting that the activation step has been related to resistance mechanisms. Improper activation such as insufficient processing or over digestion can result in insect resistance to Cry protoxin action (Domínguez-Arrizabalaga et al. 2020).
After the activation process, the active toxins bypass the peritrophic matrix to interact with the brush border membrane (BBM) of the midgut tissue. This membrane is considered the main target for toxins. The toxin then undergoes a complex sequential binding steps with the different receptors present in BBM, which results in its insertion into the membrane, with the consequent formation of pores, and osmotic lysis, leading to insect death (Lu et al. 2013). In the literature, we found several works that report that Cry toxin receptors are located in BBM, including cadherin-like (Aimanova et al. 2006; Zhang et al. 2020; Jin et al. 2021), aminopeptidase N (APN) (Wei et al. 2016; Shao et al. 2018) and alkaline phosphatase (ALP) (Likitvivatanavong et al. 2011; Stalinski et al. 2016). Our studies concluded that activated toxin samples obtained from the different formulated products showed similar binding to BBMV from two lepidopteran insects. Bel et al. (2017) studied the binding of different Cry toxins (Cry1Ab, Cry1Ac, Cry1B, Cry1C and Cry2Ab) with BBMVs from different insects (S. exigua, S. litura, A. ipsilon and H. armigera) reporting different binding curves for these toxins, supporting that these toxins display different affinity values to the BBMVs from the different larval species. In our studies we are analyzing a mixture of activated toxins and we were not able to provide specific affinity values for each protein.
Bioassay data against M. sexta and S. frugiperda larvae supported that the different procedures that were used in the formulation of these products did not affect the mechanism of action of Cry toxins, since no differences in LC50 values between commercial and microencapsulated products were observed. It is worth mentioning that these tests were carried out on a diet, under the same condition such as been reported by Eski et al. (2019) where microencapsulated formulations of an indigenous strain of B. thuringiensis (Se13) was evaluated against S. exigua. showing similar LC50 values compared to the commercial formulations on laboratory conditions.
However, we show here that when Dipel® formulation was treated with UV and analyzed against S. frugiperda larvae, the toxicity was lower than MP-Btk formulation. The microencapsulation helped to maintain the insecticidal effect of Btk, most likely due to an improved protection of crystals and spores. Another point that is worth to mention is that the results showed that the combination of conventional and microencapsulated formulations can be an important management strategy.
Khorramvatan et al. (2014), compared the effect of three polymers (starch, gelatin and sodium alginate) as wall materials for the production of a microencapsulated Bt formulation. The authors observed that for the alginate polymer, the viability of the spores was 90% after exposure to long−term UV radiation (UVB 385 nm), while the viability of non−microencapsulated spores under this condition was only 40%. In another study,) Jalali et al. (2020) used the Pickering−emulsion technique to perform the microencapsulation of Bt. The authors tested different materials such as latex particles, graphene oxide nanosheets and olive oil as protective materials. The authors evaluated toxicity of these formulations against E. kuehniella larvae after UV−A radiation. The results showed that the combination of matrices at a concentration of 0.045% allowed an effective control and greater protection against radiation, supporting the effectiveness of microencapsulation.
Overall our results provide important information for the development of future more efficient biopesticide formulations.