PEDS Advance Access originally published online on January 25, 2006
Protein Engineering Design and Selection 2006 19(3):107-111; doi:10.1093/protein/gzj009
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Redesigning Bacillus thuringiensis Cry1Aa toxin into a mosquito toxin
Department of Biochemistry, The Ohio State University, Columbus, OH 43210-1292, USA
1 To whom correspondence should be addressed. E-mail: dean.10{at}osu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionAcknowledgementsReferences |
|---|
The Bacillus thuringiensis crystal protein Cry1Aa is normally selectively active to caterpillar larvae. Through rational design, toxicity (µg/ml) to the mosquito Culex pipiens was introduced by selected deletions and substitutions of the loop residues of domain II. Toxicity to its natural target Manduca sexta was concomitantly abolished. The successful grafting of the alternate mosquito toxicity onto the original lepidopteran Cry1Aa toxin demonstrates the possibility of designing and engineering a desired toxicity into any toxin of a common scaffold by reshaping the receptor binding region with desired specificities.
Keywords: mosquito/protein engineering
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionAcknowledgementsReferences |
|---|
Bacillus thuringensis, an aerobic, Gram-positive spore-forming bacterium commonly found in soil, produces parasporal crystal (Cry) proteins with insecticidal activity against a wide range of pests. The structure and function of these toxins is well reviewed (Schnepf) (Schnepf et al., 1998
; Jenkins and Dean, 2000; Aronson and Shai, 2001; de Maagd et al., 2001). The N-terminal domain I is a bundle of eight 
-helices with the central, relatively hydrophobic helix surrounded by amphipathic helices. Domain I reportedly functions in the formation and operation of ion channels (Schnepf et al., 1998). Domain II, consisting of three anti-parallel ß-sheets connected by loops, has been linked to specific receptor binding (Schnepf et al., 1998). The C-terminal domain III adopts a lectin-like ß-sandwich topology. A number of functional roles have been suggested for this domain, including receptor binding (Burton et al., 1999; Lee et al., 1999; Jenkins et al., 2000) and ion channel formation (Chen et al., 1993; Wolfersberger et al., 1996; Schwartz et al., 1997).
Several examples of protein engineering of Bacillus thuringiensis Cry toxins have demonstrated enhancements of activity in toxins that already expressed some level of activity. In vivo domain substitutions of Cry1Ab resulted in a 4-fold enhancement of activity against Spodoptera (de Maagd et al., 1996). Site-directed mutations of individual residues in domain II loop regions of Cry3Aa led to a 10-fold increase of activity against Tenebrio molitor (Wu et al., 2000) and mutations in domain II loop regions of Cry1Ab resulted in a 34-fold increase in activity against the gypsy moth, Lymantria dispar (Rajamohan et al., 1996). More extensive deletions and substitutions of domain II loop regions of a mosquitocidal toxin, Cry4Ba, that has activity against Anopheles and Aedes, mosquitoes but no measurable activity against Culex mosquitoes, resulted in robust activity against Culex species (Abdullah et al., 2003). To date, no manipulation of Cry proteins has completely changed the specificity of a toxin to a different order of insect. This project was a test of the ability of rational design, based on current knowledge of receptor binding epitopes, to synthesize a completely new activity into a Cry protein.
In the present study, homology alignment and sequence comparison served as the basis for protein engineering design. Here, we show that by rational design, a mutant of Cry1Aa, after several rounds of deletions and substitutions made in loops of domain II, displays enhanced toxicity against Culex pipiens. These results further suggest that the introduction of short variable sequences of the loop regions from one Cry toxin into another might provide a general rational design approach to enhancing toxicity of B.thuringiensis Cry toxins.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionAcknowledgementsReferences |
|---|
Homology modeling
The following three programs were used to model the structure of Cry4Ba: (i) an internet-based free software Clustal W (http://www.ebi.ac.uk/clustalw/); (ii) SWISS-MODEL (http://www.expasy.org/swissmod/SWISS-MODEL.html); (iii) Swiss-Pdb Viewer Version 3.7b2 (Peitsch, 1995, 1996; Guex, and Peitsch, 1997).
Sequence alignment of Cry1Aa and Cry4Ba was done by Clustal W, a general purpose multiple sequence alignment program for DNA or proteins. It produces biologically meaningful multiple sequence alignments of divergent sequences. It calculates the best match for the selected sequences, and aligns them to reveal the identities, similarities and differences.
Model structure of Cry4Ba was constructed using the ‘Optimize (project) mode’ in SWISS-MODEL, in conjunction with Swiss-Pdb Viewer. The sequence of the target protein (Cry4Ba) was aligned with the template sequences (Cry4Aa modeled based on Cry1Aa and 3Aa) in Swiss-Pdb Viewer according to the alignment produced by CLUSTAL W (Figure 2). Unaligned residues at the N- and C-terminal of the target protein were removed before submitting the project to the SWISS-MODEL program.
Mutating Cry1Aa toxins by site-directed mutagenesis
cry1Aa gene cloned in pBluescript vector was expressed under the control of the lacuv5 promoter in DH5 Escherichia coli cells for DNA isolation and protein expression. Several rounds of site-directed mutagenesis (primer sequences are listed in Table I) were performed using the modified QuickChange (Stratagene) method. The protocol for the PCR is available upon request from the authors. Mutations were confirmed by automated DNA sequencing (Plant–Microbe Genomics Facility, The Ohio State University).
|
Expression and purification of Cry toxin
The method used was essentially as described elsewhere (Abdullah et al., 2003). Toxins were expressed in E.coli cells. Crystal inclusion protein was solubilized in high salt sodium carbonate buffer (30 mM Na2CO3, 20 mM NaHCO3, pH 9.5 and 0.5 M NaCl) at 37°C for 1 h. Wild-type and mutant Cry1Aa 130 kDa protoxin produced a 
65 kDa toxin fragment when digested with trypsin. Protein concentration was measured using the Coomassie protein assay reagent (Pierce) with BSA as standard.
Toxin structure analysis
To detect changes in secondary structure, CD spectra of wild-type and mutant toxins were measured with an Aviv Circular Dichroism spectrometer-model 62A DS (Lakewood, NJ) in a 32-Q-10 quartz cuvette at 25°C using Star Stationary 3.0 software. Toxins were at 2 µM in the high salt sodium carbonate–bicarbonate buffer. Readings were taken from 200 to 300 nm with 100 sampling numbers and averaged from five replicate measurements.
Determining toxicity of Cry toxins by Manduca sexta larvae bioassay
Manduca sexta eggs (Carolina Biological Supply Company) were hatched under room temperature for 3–4 days. Larvae were reared on artificial diet (Bioserv, Inc.) to first instar. Activity of toxins was determined by surface contamination method. Briefly, artificial diet was poured into 24-well tissue culture plates (Corning Costar, Corning, NY). Each well had a surface area of 2 cm2. Toxins were serially diluted in high salt sodium carbonate–bicarbonate buffer, pH 9.5 and 50 µl of toxin dilution was applied to each well, with four wells for each concentration. After liquid was absorbed and dried two larvae were placed in each well. Mortality was recorded after 5 days. LC50 (50% lethal concentration) values was calculated by Probit analysis in softtox 1.1 program (WindowChem, Fairfield, CA).
Determining toxicity of Cry toxins by Culex mosquito larvae bioassay
Colonies of the mosquito C.pipiens (egg rafts from Carolina Biological Supply Company) was reared in an environment-controlled room at 28°C and 85% humidity, with a photoperiod of 14 h light/10 h dark. Larvae were maintained on fish food pellets (Koi Floating Blend, AquaricareTM), as suggested by Dr Mark Q. Benedict (Centers for Disease Control and Prevention, Chamblee, GA). Bioassays were performed as described previously (Abdullah et al., 2003). The LC50 was calculated by a Probit method using softtox version 1.1 (WindowChem).
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionAcknowledgementsReferences |
|---|
Owing to the enormous selective pressure imposed by widespread use of B.thuringiensis Cry proteins in agriculture worldwide, the development of better Cry toxins is of ever increasing importance. The ultimate goal of protein engineering of the insecticidal Cry proteins is to be able to design any Cry toxin to possess toxic activity against any insect. A more immediate goal is to introduce a specific activity into a toxin that does not possess it.
Cry1Aa and Cry4Ba are presumed to share a similar mode of action, but target distinct insect species. With its known tertiary structure and relatively well characterized receptor binding regions, Cry1Aa is an ideal candidate for the design of alternate specificity. Cry1Aa is a lepidopteran toxin with no natural activity toward mosquito. In this study, we have introduced alterations in domain II loops of Cry1Aa to introduce mosquito toxicity.
Loop regions are excellent targets for genetic re-designing of novel toxins with diverse specificity by exchanging residues or chain lengths of the active sites without major disruption of the overall integrity of the toxin. Previously, we predicted the loop sequences of Cry4Ba (Abdullah et al., 2003). This prediction was recently verified with the solution of the 3D structure (Boonserm et al., 2005). Toxicities of Cry4Ba toward three species of mosquitoes were successfully modified by manipulating the loop regions (Abdullah et al., 2003). A Cry4Ba mutant called 4BL3PAT with its original first two loops and a loop 3 mimicking Cry4Aa gained toxicity to Culex, whereas its toxicity to the natural target species, Anophelles and Aedes, was not negatively affected.
Based on secondary sequence alignment of lepidoptera-specific Cry1Aa and diptera-specific Cry4Ba, done by Clustal W (shown in Figure 1) and structural analysis, significant differences in length and composition were found in the first two of three loops in Domain II of Cry1Aa and Cry4Ba (4BL3PAT). Interestingly, it was reported by Abdullah et al. (2003) that when loop 3 (PAT) of 4BL3PAT was replaced by GAV, a Cry1Aa homologous loop sequence, its toxicity toward mosquito Culex was further enhanced. For this reason the third loop of Cry1Aa was left unchanged in the subsequent protein engineering work.
|
The first two loop regions of Cry1Aa were changed to acquire toxicity toward diptera, using 4BL3PAT as a template. Loop 1 (residues 311RG312) in Cry1Aa was replaced by YQDL, the loop 1 sequence in Cry4Ba, to extend its length. This mutant was named L1. Cry1Aa loop2, LY367RRIILGSGPNNQ378, was altered in two separate steps. LYRRIIL was first deleted to produce an intermediate mutant called D3. The loop 1 mutation in L1 was then introduced to D3 creating a reference mutant called l1d3. When changed individually or in combination none of L1, D3 and L1D3 gave rise to a mosquitocidal toxin. Under the guidance of molecular modeling, a third mutant named 1AaMosq with an additional substitution of NNQ by G was built into L1D3 to mimic the shorter second loop in Cry4Ba but maintain the turn between two ß-sheets.
Shown in Figure 2 are the solved structure of Cry1Aa wild-type toxin and the modeled structure of 1AaMosq mutant. The loops at the bottom of the molecules are loops 1, 2 and 3 from the right to left. Loop 1 was elongated when RG was mutated to YQDL, the relatively long loop 2 was shortened to merely a turn by two rounds of deletion, whereas loop 3 was left unchanged in 1AaMosq.
|
The near-UV spectral region (Figure 3) of wt and mutant Cry1Aa showed no significant variation, indicating that the defined tertiary structure was not disturbed. The gradual differences in far UV region agree with the changing ratio of loop components.
|
Bioassay results shown in Table II indicated that Cry1Aa wild-type and intermediate mutants (L1, D3, L1D3) have no apparent toxicity to C.pipiens, whereas 1AaMosq with triple mutations in both loops 1 and 2 has enhanced C.pipiens activity at µg/ml levels. Concomitant with the gain in mosquito toxicity, toxicity toward M.sexta was abolished during several rounds of changes in loop residues, confirming the importance of the domain II loops in specificity and activity.
|
The idea of using a protein of known three-dimensional structure to present motifs of various functions or specificity has been a primary goal of the protein engineering (Dunn and Bungert, 2003
). The use of so-called protein scaffolds for generation of novel binding proteins via combinatorial engineering has emerged as a powerful alternative to natural or recombinant antibodies (Nygren and Skerra, 2004). The results of this study present an example of enhancing Cry toxicity through an approach that integrates sequence comparison, computational prediction and rational design of mutagenesis. Table III shows the toxicity of known mosquitocidal toxins from B.thuringiensis and Bacillus sphaericus. Toxicity of engineered Cry1AaMosq is greater than several natural toxins (Cry1Ca, Cry2Aa, Cry4Ba and Cry20Aa). The successful grafting of the alternate mosquito toxicity onto the original lepidopteran Cry1Aa toxin demonstrates the possibility to design and engineer desired toxicity into any toxin of a common scaffold by reshaping the receptor binding region with desired specificities. By varying the specificity elements in loop regions on a general scaffold, a customized toxin can be selectively tuned to target different insect species.
|
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionAcknowledgementsReferences |
|---|
We thank Dr Mohd A.F.Abdullah for valuable discussion at the outset of the project and Bin Ni for technical assistance. This work was supported by NIH grant to D.H.D. and M.J.Adang (Grant # R01 AI 29092).
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionAcknowledgementsReferences |
|---|
Abdullah,M.A. and Dean,D.H. (2004) Appl. Environ. Microbiol., 70, 3769–3771.
Abdullah,M.A., Alzate,O., Mohammad,M., McNall,R.J., Adang,M.J. and Dean,D.H. (2003) Appl. Environ. Microbiol., 69, 5343–5353.
Angsuthanasombat,C., Crickmore,N. and Ellar,D.J. (1992) FEMS Microbiol. Lett., 94, 63–68.
Aronson,A.I. and Shai,Y. (2001) FEMS Microbiol. Lett., 195, 1–8.[CrossRef][Web of Science][Medline]
Audtho,M. (2001) Mode of Action of Bacillus thuringiensis Cry2Aa. Ohio State University, Columbus.
Boonserm,P., Davis,P., Ellar,D.J. and Li,J. (2005) J. Mol. Biol., 348, 363–382.[CrossRef][Web of Science][Medline]
Burton,S.L., Ellar,D.J., Li,J. and Derbyshire,D.J. (1999) J. Mol. Biol., 287, 1011–1022.[CrossRef][Web of Science][Medline]
Chen,X.J., Lee,M.K. and Dean,D.H. (1993) Proc. Natl Acad. Sci. USA, 90,9041–9045.
Crickmore,N., Bone,E.J., Williams,J.A. and Ellar,D.J. (1995) FEMS Microbiol. Lett., 131, 249–254.[CrossRef]
Dankocsik,C., Donovan,W.P. and Jany,C.S. (1990) Mol. Microbiol., 4, 2087–2094.[CrossRef][Web of Science][Medline]
de Maagd,R.A., Bravo,A. and Crickmore,N. (2001) Trends Genet., 17, 193–199.[CrossRef][Web of Science][Medline]
de Maagd,R.A., Kwa,M.S.G., van der Klei,H., Yamamoto,T., Schipper,B., Vlak,J.M., Stiekema,W.J. and Bosch,D. (1996) Appl. Environ. Microbiol., 62, 1537–1543.[Abstract]
Delécluse,A., Poncet,S., Klier,A. and Rapoport,G. (1993) Appl. Environ. Microbiol., 59, 3922–3927.
Delécluse,A., Rosso,M.-L. and Ragni,A. (1995) Appl. Environ. Microbiol., 61, 4230–4235.[Abstract]
Donovan,W.P., Dankocsik,C. and Gilbert,M.P. (1988) J. Bacteriol., 170, 4732–4738.
Dunn,B.D. and Bungert,J. (2003) Nat. Biotechnol., 21, 1019–1021.[CrossRef][Web of Science][Medline]
Guex,N. and Peitsch,M.C. (1997) Electrophoresis, 18, 2714–2723.[CrossRef][Web of Science][Medline]
Jenkins,J.L. and Dean,D.H. (2000) Genet. Eng., 22, 33–54.
Jenkins,J.L., Lee,M.K., Curtiss,A. and Dean,D.H. (2000) J. Biol. Chem., 275, 14423–14431.
Lee,H.K. and Gill,S.S. (1997) Appl. Environ. Microbiol., 63, 4664–4670.[Abstract]
Lee,M.K., You,T.H., Gould,F.L. and Dean,D.H. (1999) Appl. Environ. Microbiol., 65, 4513–4520.
Moar,W.J., Trumble,J.T., Hice,R.H. and Backman,P.A. (1994) Appl. Environ. Microbiol., 60, 896–902.
Nygren,P.A. and Skerra,A. (2004) J. Immunol. Methods, 290, 3–28.[CrossRef][Web of Science][Medline]
Orduz,S. (1998) Biochim. Biophys. Acta., 1388, 267–272.[CrossRef][Medline]
Peitsch,M.C. (1996) Biochem. Soc. Trans., 24, 274–279.[Web of Science][Medline]
Peitsch,M.C. (1995) PDB Quart. Newslett., 72, 4.
Poncet,S., Delécluse,A., Klier,A. and Rapoport,G. (1995) J. Invert. Pathol.,66, 131–135.
Rajamohan,F., Alzate,O., Cotrill,J.A., Curtiss,A. and Dean,D.H. (1996) Proc. Natl Acad. Sci. USA, 93, 14338–14343.
Rosso,M.L. and Delecluse,A. (1997) Appl. Environ. Microbiol., 63, 4449–4455.[Abstract]
Schnepf,E., Crickmore,N., VanRie,J., Lereclus,D., Baum,J., Feitelson,J., Zeigler,D.R. and Dean,D.H. (1998) Microbiol. Mol. Biol. Rev., 62, 775–806.
Schwartz,J.L., Potvin,L., Chen,X.J., Brousseau,R., Laprade,R. and Dean,D.H. (1997) Appl. Environ. Microbiol., 63, 3978–3984.[Abstract]
Smith,G.P., Merrick,J.D., Bone,E.J. and Ellar,D.J. (1996) Appl. Environ. Microbiol., 62, 680–684.[Abstract]
Thorne,L., Garduno,F., Thompson,T., Decker,D., Zounes,M.A., Wild,M., Walfield,A.M. and Pollock,T.J. (1986) J. Bacteriol., 166, 801–811.
Widner,W.R. and Whiteley,H.R. (1989) J. Bacteriol., 171, 965–974.
Wolfersberger,M.G., Chen,X.J. and Dean,D.H. (1996) Appl. Environ. Microbiol., 62, 279–282.[Abstract]
Wu,S.J., Koller,C.N., Miller,D.L., Bauer,L.S. and Dean,D.H. (2000) FEBS Lett., 473, 227–232.[CrossRef][Web of Science][Medline]
Received July 15, 2005; revised November 15, 2005; accepted December 13, 2005.
Edited by Mark Zoller
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. R. Pigott, M. S. King, and D. J. Ellar Investigating the Properties of Bacillus thuringiensis Cry Proteins with Novel Loop Replacements Created Using Combinatorial Molecular Biology Appl. Envir. Microbiol., June 1, 2008; 74(11): 3497 - 3511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Gatehouse Biotechnological Prospects for Engineering Insect-Resistant Plants Plant Physiology, March 1, 2008; 146(3): 881 - 887. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||