Protein Engineering of Vip3A in a Selected Bacillus thuringiensis Host for Consistent High Protein Production.

This study aimed to obtain reliable high Vip3A production from Bacillus thuringiensis (Bt) by modifying Vip3A to acquire higher thermostability in a suitable host. Bt117 is a great host for Vip3A production due to protein production consistency, low protease activity in culture media, and large amounts of mostly full-length protein, but it produces Vip3A with lower thermostability (Vip3Aa35). The C-terminal region of Bt117 Vip3A was replaced with that of a Vip3A with higher thermostability (Vip3Aa64 from Bt294) to generate the recombinant Bt117-Vip3Aa64-C. Like the parental strain Bt117, this strain expressed mostly full-length protein and exhibited low protease activity and similar protein expression profiles in culture media but retained greater larvicidal activity upon 37 °C storage like Bt294 Vip3Aa64. Importantly, every culture batch of Bt117-Vip3Aa64-C yielded over 200 mg/l Vip3A, which is a notable improvement over the original Vip3Aa64-producing strain Bt294 where 45% of culture batches failed to produce Vip3A at the same level. Successfully, we combined the superior qualities of two Bt strains, Bt294, which produces thermostable Vip3A but at low and inconsistent levels, and Bt117, which produces Vip3A with low thermostability but at consistently high levels. Protein engineering of Vip3A in Bt117 ultimately yielded an improved strain producing a thermostable Vip3A with reliably high protein production.

Screening and characterization of Bacillus thuringiensis isolates for high production of Vip3A and Cry proteins and high thermostability to control Spodoptera spp.

Bacillus thuringiensis (Bt) is an entomopathogenic bacterium that produces crystalline (Cry and Cyt) and soluble (vegetative insecticidal proteins or Vips) proteins during the sporulation and vegetative growth phases, respectively. Combining Cry and Vip proteins could delay insect resistance development and exhibit synergistic activity against various insect pests. This study aims to screen Bt isolates collected from Thailand for high Vip3A and Cry protein production levels and high thermostability to control Spodoptera spp. Among the selected Bt isolates with high target protein synthesis, Bt isolate 506 was found to be safe for further biopesticide formulation due to the absence of non-specific metabolite, as determined by the detection of thermo-stable ß-exotoxin I based on biological assays and PCR analysis. Bt isolate 506 showed the presence of Cry1A, Cry2A, and Vip3A-type proteins identified as Cry1Aa45, Cry2Aa22, and Vip3A87, respectively. The insecticidal activity of whole culture extracts containing Vip3A and Cry mixtures and culture supernatants containing secreted Vip3A protein was evaluated against the second-instar larvae of S. exigua and S. frugiperda. The Bt isolate 506 showed high toxicity against both insects, and the insecticidal proteins produced by this isolate retained their activity after heating at 50 °C. This Bt isolate is a promising candidate for further development as a biopesticide against lepidopteran pests.

Modulation of Cas9 level for efficient CRISPR-Cas9-mediated chromosomal and plasmid gene deletion in Bacillus thuringiensis.

OBJECTIVES: To set up an efficient gene editing system in Bacillus thuringiensis (Bt) using CRISPR-Cas9 by demonstrating deletion of chromosomal and plasmid genes. RESULTS: CRISPR-Cas9 from Streptococcus pyogenes was found to function in Bt cells, resulting in DNA cleavage that is lethal to the cells. The system was assessed for its ability to mediate gene editing by knock-out of the protease genes nprA (neutral protease A) and aprA (alkaline protease A). Gene editing was not detected when the Bacillus-derived pBCX was used to carry CRISPR-Cas9 elements and a DNA repair template. When the Cas9 promoter was replaced with the sporulation-specific promoter cyt2A, a Bt ∆nprA clone was obtained, but this plasmid construct did not give reproducible results. Bt ∆nprA ∆aprA and Bt ∆aprA deletion mutants were finally generated when the Lactobacillus plantarum-derived plasmid pLPPR9 was used, likely due to its lower copy number reducing Cas9 toxicity. Only three to four clones each needed to be screened to identify the desired gene-modified mutants. Conversely, efficient editing of the plasmid vip3A gene required the use of pBCX and longer homology sequences for the repair template. CONCLUSIONS: Capitalizing on the differential impact of plasmid copy number and homology arm length, we devised distinct yet simple and efficient approaches to chromosomal and plasmid gene deletion for Bt that condense the screening process, minimize screening, and facilitate multiple consecutive gene editing steps.

Intracellular localization and cytotoxicity of Bacillus thuringiensis Vip3Aa against Spodoptera frugiperda (Sf9) cells.

Vip3Aa protein is produced by Bacillus thuringiensis during vegetative growth and displays high toxicity against a wide range of lepidopteran insect larvae such as Spodoptera exigua and Spodoptera frugiperda, both important insect pests worldwide. Vip3Aa protein is synthesized as a protoxin (proVip3Aa) and becomes activated by digestion with either trypsin or insect gut proteases. The activated Vip3Aa protein (actVip3Aa) binds to a specific receptor in the brush border epithelial midgut cells, causing cell death via apoptosis, possibly induced by its pore-forming activity. Here we investigated the actVip3Aa intracellular localization to explain the molecular mechanism leading to the cytotoxicity of Vip3Aa toxin. The Spodoptera frugiperda (Sf9) cell line was incubated with fluorescently labeled Vip3Aa, namely Alexa488-actVip3Aa, and the intracellular localization was analyzed through a laser scanning confocal microscope. The Alexa488-actVip3Aa was internalized into the Sf9 cells. Immunofluorescence detection demonstrated that Alexa488-actVip3Aa did not colocalize with early endosomes which is usually implicated in clathrin-mediated endocytosis, suggesting that the actVip3Aa does not use clathrin-dependent endocytosis to transport into the cytosol. Intracellular visualization also shows that actVip3Aa does not directly target to mitochondria upon entry into the cytosol. Following internalization, actVip3Aa causes cell division disruption that subsequently could trigger cell death via apoptosis.

Lipid phase influences the binding of Bacillus thuringiensis Cyt2Aa2 toxin on model lipid membranes.

Bacillus thuringiensis is a bacterium that produces many insecticidal proteins including cytolytic proteins or Cyt toxins. Although the Cyt toxin shows specific toxicity against Dipteran insect species, the toxin binds directly to the lipid membrane without a specific protein receptor requirement. In this work, we have investigated the interaction of Cyt2Aa2 toxin with lipid bilayers composed of different lipid phases. By means of atomic force microscopy (AFM), lipid phase separation was observed for 1:1 and 4:1 M mixtures of DPPC/POPC bilayers. The exposure of Cyt2Aa2 to these lipid bilayers revealed that the toxin selectively bound to Ld lipid bilayer (corresponding to POPC). In turn, it did not bind to the Lo and So phases (corresponding to DPPC). Interestingly, for the bilayer of 4:1 DPPC/POPC mixture, the binding of Cyt2Aa2 was localized at the lipid phase boundary instead of Ld domain as occurred for the 1:1 DPPC/POPC bilayer. In addition, quartz crystal microbalance with dissipation experiments confirmed AFM results. In particular, the measurements showed that amount of protein bound to 1:1 DPPC/POPC (with phase separation) was half of the binding quantified for the Ld phase lipid bilayer (pure POPC and 1:4 DPPC/POPC mixture). These results indicate that the lipid phase (lipid acyl chain) influences the Cyt2Aa2-lipid interaction.

Tyrosine-776 of Vip3Aa64 from Bacillus thuringiensis is Important for Retained Larvicidal Activity During High-Temperature Storage.

Vip3Aa (vegetative insecticidal protein) secreted by Bacillus thuringiensis (Bt) is highly toxic to lepidopteran insects. The Bt isolate M190 produces Vip3Aa35 at high concentrations, and Vip3Aa35 was found to be very effective against Spodoptera exigua. Unfortunately, the use of Vip3Aa35 in pest control is limited by its short shelf life when stored at high temperatures, retaining activity for only 1 month at 37 °C. To find a more stable alternative, we screened 500 isolates of Bt collected from various locations in Thailand and discovered Bt isolate 294 which produced large amounts of Vip3Aa64 that exhibited high toxicity against S. exigua but could be stored at 37 °C for up to 3 months. Vip3Aa35 and Vip3Aa64 have only nine amino acid differences between them, with six of those residues being located at the C terminus. Vip3Aa35 and Vip3Aa64 chimeras revealed that the C-terminal sequence is important for the retained larvicidal activity observed with Vip3Aa64. Various single amino acid substitutions were created to identify the key amino acids responsible for this stability. A single residue, Tyr776, was found to be solely responsible, with the Vip3Aa35:N776Y acquiring thermostability similar to Vip3Aa64 while the Vip3Aa64:Y776N exhibited Vip3Aa35-like thermostability.

Cholesterol Increases Lipid Binding Rate and Changes Binding Behavior of Bacillus thuringiensis Cytolytic Protein.

Cytolytic protein (Cyt) is a member of insecticidal proteins produced by Bacillus thuringiensis. Cyt protein has activity against insect cells and mammalian cells, which differ in lipid and cholesterol composition. This study presents the lipid binding behavior of Cyt2Aa2 protein on model membranes containing different levels of cholesterol content by combining Quartz Crystal Microbalance with Dissipation (QCM-D) and Atomic Force Microscopy (AFM). QCM-D results revealed that cholesterol enhances the binding rate of Cyt2Aa2 protein onto lipid bilayers. In addition, the thicker lipid bilayer was observed for the highest cholesterol content. These results were confirmed by AFM. The analysis of protein surface coverage as a function of time showed a slower process for 5:0 and 5:0.2 (POPC:Chol) ratios than for 5:1 and 5:2 (POPC:Chol) ratios. Significantly, the Cyt2Aa2-lipid binding behavior and the protein⁻lipid layer were different for the 5:3 (POPC:Chol) ratio. Furthermore, AFM images revealed a transformation of Cyt2Aa2/lipid layer structure from strip pattern to ring shape structures (which showed a strong repulsion with AFM tip). In summary, cholesterol increases the binding rate and alters the lipid binding behavior of Cyt2Aa2 protein, although it is not required for Cyt2Aa2 protein binding onto lipid bilayers.

pH regulates pore formation of a protease activated Vip3Aa from Bacillus thuringiensis.

Vip3Aa insecticidal protein is produced from Bacillus thuringiensis and exerts a broad spectrum of toxicity against lepidopteran insect species. Although Vip3Aa has been effectively used as part of integrated pest management strategies, the mechanism of the toxin remains unclear. Here, we investigated the effect of pH in a range from 5.0 to 10.0 on the pore-forming activity of the trypsin activated Vip3Aa (actVip3Aa) by in vitro pore-forming assays. Based on calcein release assay, actVip3Aa could permeabilize the artificial neutral liposomes under all the pH tested, except pH10.0. The maximum membrane permeability of actVip3Aa was detected at pH8.0 and the permeability decreased and abolished when exposing to acidic and alkaline conditions, respectively. The planar lipid bilayer experiment revealed that actVip3Aa formed ion channels at pH5.0-8.0 but no current signals were detected at pH10.0, consistent with the observation from calcein release assay. The toxin formed ion channels with a diameter of 1.4nm at pH8.0 and pore size was gradually decreased when reducing the pH. This study provided a view of the molecular mechanism of Vip3Aa by which the pore-forming activity is regulated by pH.

Bacillus thuringiensis Cyt2Aa2 binding on lipid/cholesterol bilayer depends on protein concentration and time.

Bacillus thuringiensis produces cytolytic proteins (Cyt) that show toxicity against dipteran insect larvae acting directly on the cell membrane. Up to now, two different models have been proposed to explain the interaction mechanism of the cytolytic protein Cyt2Aa2 on lipid membranes: pore-forming and detergent-like action. Here we report on the interaction of Cyt2Aa2 with lipid/cholesterol bilayers at early stage (far from equilibrium) as a function of protein concentration. Quartz crystal microbalance with dissipation (QCM-D) measurements showed that the rate of protein adsorption increased with concentration, although the mass of the final protein-lipid was similar after two hours. In addition, the dissipation (compliance of the hybrid lipid/protein layer) increased with decreasing protein concentration. Furthermore, atomic force microscopy (AFM) revealed that the structure of the protein-lipid layer was concentration and time dependent. A rigid hybrid homogeneous layer was observed at protein concentrations of 50 µg/ml and 100 µg/ml after 30 min. At lower concentrations, 10 µg/ml and 17.5 µg/ml, protein adsorption on the lipid layer led to the formation of small aggregates. Interestingly, at 25 µg/ml a transition of a hole-like structure into a homogeneous layer was observed. This suggests that 25 µg/ml is a threshold concentration for the binding mechanism of Cyt2Aa2 on to lipid membranes.

Conditions for homogeneous preparation of stable monomeric and oligomeric forms of activated Vip3A toxin from Bacillus thuringiensis.

Bacillus thuringiensis vegetative insecticidal proteins like Vip3A have been used for crop protection and to delay resistance to existing insecticidal Cry toxins. However, little is known about Vip3A's behavior or its mechanism of action, and a structural model is required. Herein, in an effort to facilitate future crystallization and functional studies, we have used the orthogonal biophysical techniques of light scattering and sedimentation to analyze the aggregation behavior and stability of trypsin-activated Vip3A toxin in solution. Both scattering and sedimentation data suggest that at pH 10 the toxin is monomeric and adopts an elongated shape, but after overnight incubation aggregation was observed at all pH values tested (5-12). The narrowest size distribution was observed at pH 7, but it was consistent with large oligomers of ~50 nm on average. The addition of ß-D-glucopyranoside (OG) helped in achieving preparations that were stable and with a narrower particle size distribution. In this case, scattering was consistent with a 4-nm monomeric globular Vip3A form. After OG dialysis, 40-nm particles were detected, with a molecular weight consistent with homotetramers. Therefore, OG is proposed as the detergent of choice to obtain a Vip3A crystal for structural studies, either before (monomers) or after dialysis (tetramers).

Effect of the Concentration of Cytolytic Protein Cyt2Aa2 on the Binding Mechanism on Lipid Bilayers Studied by QCM-D and AFM.

Bacillus thuringiensis is known by its insecticidal property. The insecticidal proteins are produced at different growth stages, including the cytolytic protein (Cyt2Aa2), which is a bioinsecticide and an antimicrobial protein. However, the binding mechanism (and the interaction) of Cyt2Aa2 on lipid bilayers is still unclear. In this work, we have used quartz crystal microbalance with dissipation (QCM-D) and atomic force microscopy (AFM) to investigate the interaction between Cyt2Aa2 protein and (cholesterol-)lipid bilayers. We have found that the binding mechanism is concentration dependent. While at 10 µg/mL, Cyt2Aa2 binds slowly on the lipid bilayer forming a compliance protein/lipid layer with aggregates, at higher protein concentrations (100 µg/mL), the binding is fast, and the protein/lipid layer is more rigid including holes (of about a lipid bilayer thickness) in its structure. Our study suggests that the protein/lipid bilayer binding mechanism seems to be carpet-like at low protein concentrations and pore forming-like at high protein concentrations.

Interaction of Lysinibacillus sphaericus binary toxin with mosquito larval gut cells: Binding and internalization.

The binary toxin produced by Lysinibacillus sphaericus is composed of BinA and BinB subunits. Together, but not separately, the two subunits are highly toxic to Culex quinquefasciatus larvae, but show no toxicity to Aedes aegypti. The molecular mechanism underlying intoxication has not been clearly elucidated. The present study compares the binding and the internalization of binary toxin into the midgut epithelial cells of susceptible C. quinquefasciatus mosquito larvae with those of Bin-refractory A. aegypti. The guts from larvae fed with fluorescently labeled toxin were dissected and analyzed using a confocal laser scanning microscope. When fed with a mixture of both components, co-localization of BinA and BinB was detected both on the cell surface and in the cytoplasm of Culex larval gut cells. However, administration of BinA alone resulted in localization only on the cell membrane, whereas BinB alone was detected both on the cell membrane and inside the cytoplasm. In contrast, when a mixture of both components, or each individual component, was fed to Aedes larvae, BinA and BinB were unable to reach the cytoplasm and were localized only on the cell membrane. These results are consistent with the suggestion that the internalization of BinA is essential for toxicity, and that BinB is required for this internalization into susceptible larval gut cells.

Lysinibacillus sphaericus binary toxin induces apoptosis in susceptible Culex quinquefasciatus larvae.

During sporulation, a Gram-positive bacterium Lysinibacillus sphaericus (Ls) produces the mosquito larvicidal binary toxin composed of 2 subunits, BinA and BinB. Full toxicity against Culex and Anopheles mosquito larvae is achieved when both subunits are administered together at equimolar amounts. Although cellular responses to Bin toxin have been reported in previous studies, it remains essential to extensively examine the cytopathic effects in vivo to define the underlying mechanism of larval death. In this study, 4th instar Culex quinquefasciatus larvae fed with different doses of Bin toxin were analyzed both for ultrastructural as well as biochemical effects. Typical morphological changes consistent with apoptosis were observed in mosquito larvae exposed to Bin toxin, including mitochondrial swelling, chromatin condensation, cytoplasmic vacuolization and apoptotic cell formation. Bin toxin also induced the activation of caspase-9 and caspase-3 in larval midgut cells. Our current observations thus suggest that Bin toxin triggers apoptosis via an intrinsic or mitochondrial pathway in vivo, possibly contributing to larval death.

Crystal structure of BinB: a receptor binding component of the binary toxin from Lysinibacillus sphaericus.

The binary toxin (Bin), produced by Lysinibacillus sphaericus, is composed of BinA (42 kDa) and BinB (51 kDa) proteins, which are both required for full toxicity against Culex and Anopheles mosquito larvae. Specificity of Bin toxin is determined by the binding of BinB component to a receptor present on the midgut epithelial membranes, while BinA is proposed to be a toxic component. Here, we determined the first crystal structure of the active form of BinB at a resolution of 1.75 Å. BinB possesses two distinct structural domains in its N- and C-termini. The globular N-terminal domain has a ß-trefoil scaffold which is a highly conserved architecture of some sugar binding proteins or lectins, suggesting a role of this domain in receptor-binding. The BinB ß-rich C-terminal domain shares similar three-dimensional folding with aerolysin type ß-pore forming toxins, despite a low sequence identity. The BinB structure, therefore, is a new member of the aerolysin-like toxin family, with probably similarities in the cytolytic mechanism that takes place via pore formation.

Biosynthesis of Cry5B-Loaded Sulfur Nanoparticles using Arthrobotrys oligospora Filtrate: Effects on Nematicidal Activity, Thermal Stability, and Pathogenicity against Caenorhabditis elegans.

Cry5B, a crystal protein produced by Bacillus thuringiensis (Bt), is a bionematicide with potent nematicidal activity against various plant-parasitic and free-living nematodes. This protein, however, is susceptible to destruction by ultraviolet light, proteolytic enzymes, and high temperatures. This study aims to produce Cry5B protein for bionematicidal use and improve its stability and nematicidal efficacy by loading it intoArthrobotrys oligospora-mediated sulfur nanoparticles (AO-SNPs). Based on the mortality assay, the Cry5B protein exhibited dose-dependent nematicidal activity against the model organismCaenorhabditis elegans. The nematicidal activity, thermal stability, and pathogenic effects of Cry5B-loaded AO-SNPs (Cry5B-SNPs) were compared to those of free Cry5B. After 3 h of exposure to heat at 60 °C, Cry5B-SNPs had greater nematicidal activity than free Cry5B protein, indicating the effective formulation of Cry5B-SNPs that could be used as an alternative to current nematicide delivery strategies.

Bacillus thuringiensis Cry4Ba toxin employs two receptor-binding loops for synergistic interactions with Cyt2Aa2.

We previously demonstrated that co-expression in Escherichia coli of Bacillus thuringiensis (Bt) subsp. israelensis Cry4Ba and Bt subsp. darmstadiensis Cyt2Aa2 shows high synergistic toxicity against target mosquito larvae. Here, further insights into synergistic interactions between these two toxins were revealed through bioactivity restoration of particular inactive Cry4Ba-mutant toxins altered within the receptor-binding domain. Specific mutations at ß2-ß3 (Y332A) or ß4-ß5 (F364A) loops, but neither at three other ß-hairpin loops (ß6-ß7, ß8-ß9 and ß10-ß11) of Cry4Ba, adversely affect toxicity restoration by Cyt2Aa2. Binding analysis using quartz crystal microbalance verified a decrease in binding of these two bioinactive-mutant toxins (Y332A and F364A) to the immobilized Cyt2Aa2. This suggests that Cry4Ba utilizes these two critical aromatic loop-residues, Tyr(332) and Phe(364), for synergistic toxicity with its alternative receptor-Cyt2Aa2.

Crystallization and preliminary X-ray crystallographic analysis of the functional form of BinB binary toxin from Bacillus sphaericus.

The binary toxin from Bacillus sphaericus consists of two proteins, BinA and BinB, which work together to exert toxicity against mosquito larvae. BinB is proposed to be a receptor-binding domain and internalizes BinA into the midgut cells, resulting in toxicity via an unknown mechanism. The functional form of BinB has been successfully crystallized. The crystals of BinB diffracted to a resolution of 1.75 Å and belong to space group P6(2)22, with unit-cell parameters a = b = 95.2, c = 154.9 Å. Selenomethionine-substituted BinB (SeMetBinB) was prepared and crystallized for experimental phasing. The SeMetBinB crystal data were collected at a wavelength of 0.979 Å and diffracted to a resolution of 1.85 Å.

Expression of mosquito-larvicidal toxin genes under the control of a native promoter in Enterobacter amnigenus An11.

Enterobacter amnigenus An11, that can colonize the gut of mosquito larva, is an alternative toxin-producing host to be used as a mosquito control since it is able to float in the feeding zone of mosquito larvae. To produce mosquito-larvicidal toxins in this bacterium, a native promoter has been identified from its genomic DNA. The promoter exhibited consensus sequences for -35 and -10 regions of bacterial promoters and constitutively drove the expression of gfp. This promoter was inserted into recombinant plasmids upstream of promoter-free cyt2Aa2 from Bacillus thuringiensis and mtx2 from Bacillus sphaericus. Results demonstrated that Cyt2Aa2 and Mtx2 are constitutively produced without induction. The recombinant E. amnigenus showed toxicity against mosquito larvae, demonstrating a potential to be applied in a mosquito control program.

Amino acid residues in the N-terminal region of the BinB subunit of Lysinibacillus sphaericus binary toxin play a critical role during receptor binding and membrane insertion.

The binary toxin produced by Lysinibacillus sphaericus is composed of BinA and BinB subunits that work together in governing toxicity against mosquito larvae. BinA is proposed to be important for toxicity, whereas BinB has been shown to act as a specific receptor-binding component. The precise function of both subunits, however, is not well established. Here, we investigated the function of the N-terminal region of BinB subunit initially by introducing triple alanine substitutions at positions 35PEI37 and 41FYN43. Both block mutations abolished the larvicidal activity. Single point mutations (P35A, E36A, I37A, F41A, Y42A, N43A) were generated in order to identify amino acids that are critical for the toxin activity. Mosquito-larvicidal activity was significantly reduced in P35A, E36A, F41A and Y42A mutants. However, these mutants retained ability to form in vitro interaction with the BinA counterpart. Immunohistochemistry analysis revealed that P35A, F41A and N43A bind to the larval midgut membrane at comparable levels to that of the wild type BinB. In contrast, greatly reduced binding activity was observed in the Y42A, suggesting an important role of this residue in receptor binding. Alanine substitution at P35 resulted in a marked decrease in membrane penetration, indicating its functional importance for the membrane insertion. These results suggest the important roles of the N-terminal region of BinB in both the receptor recognition and the membrane interaction.

Mutation of a Threonine Residue in αD-ß4 Loop of Cyt2Aa2 Protein Influences Binding on Fluid Lipid Membranes.

Cyt proteins are insecticidal proteins originally from Bacillus thuringiensis. The lipid binding of the Cyt2Aa2 protein depends on the phase of the lipid bilayer. In this work, the importance of the conserved T144 residue in the αD-ß4 loop for lipid binding on fluid lipid membranes was investigated via atomic force microscopy (AFM). Lipid membrane fluidity could be monitored for the following lipid mixture systems: POPC/DPPC, POPC/SM, and DOPC/SM. AFM results revealed that the T144A mutant was unable to bind to pure POPC bilayers. Similar topography between the wildtype and T144A mutant was seen for the POPC/Chol system. Small aggregates of T144A mutant were observed in the POPC and DOPC domains of the lipid mixture systems. In addition, the T144A mutant had no cytotoxic effect against human colon cancer cells. These results suggest that alanine replacement into threonine 144 hinders the binding of Cyt2Aa2 on liquid lipid membranes. These observations provide a possibility to modify the Cyt2Aa2 protein to specific cells via lipid phase selection.