Highly Toxic and Broad-Spectrum Insecticidal Bacillus thuringiensis Engineered by Using the Transposon Tn917 and Protoplast Fusion

Highly Toxic and Broad-Spectrum InsecticidalBacillus thuringiensisEngineered by Using the Transposon Tn917 and Protoplast Fusion

Jianxiu Yu, Yi Pang, Mujin Tang, Ruiyu Xie, Le Tan, Shaoling Zeng, Meijin Yuan, Jingyu Liu

State Key Laboratory for Biocontrol, Zhongshan University, Guangzhou, People’s Republic of China 510275

Received: 30 November 2000 / Accepted: 10 January 2001

Abstract. The chromosome of theBacillus thuringiensisstrain S184 that was toxic against the thirdinstar larvae ofSpodoptera liturawith the LC50 of 9.74mg/ml was successfully integrated into two genesof cyt1Aa andcry11Aa using the transposon Tn917, yielding the primary engineered strain TnX. Thestrain TnX was highly toxic to the third instar larvae ofCulex pipiens fatiganswith the LC50 of 5.12ng/ml which was 1.82-fold higher than that ofB. thuringiensissubsp.israelensis, but lowly toxic tolepidopterous larvae. By the protoplast fusion of the strain TnX and the strain S184-Tetr (resistance totetracycline), the target engineered strain TnY was obtained. Against the third instar larvae ofS. litura,the strain TnY LC50 was of 4.68m g/ml and increased by 2.08-fold in comparison with the parent strainS184. Against the third instar larvae ofC. pipiens fatigans, the strain TnY LC50 was of 103.20 ng/ml. Thetwo target genes ofcyt1Aa andcry11Aa integrated into the chromosome were extremely stable and hadlittle possibility of a second transposition. It was unclear whether some factors existing in the parentstrain, S184, contributed to the high toxicity of the strains TnX and TnY.

Bacillus thuringiensisis a gram-positive soil bacteriumthat produces large amounts of insecticidal crystal pro-teins (ICPs) which accumulate in the cytoplasm to formcrystals with a highly specific insecticidal activity. TheseICPs fall into two unrelated groups, Cry proteins (Cry1;28) and Cyt proteins (Cyt 1;2) [8, 12]. The currentview on the mode of action is similar for all Cry proteins.The pathways firstly require ingestion, solubilization,and enzymatic activation by midgut proteases and highvalue of pH [13]. These activated toxins bind themselvesto specific glycoprotein receptors on the insect midgutbrush-border epithelium and form pores bringing aboutosmotic swelling, cell lysis and subsequent death bystarvation or septicemia [13, 15, 23]. Although the pri-mary affinity of Cyt proteins appears to be for lipids inthe microvillar membrane [14, 18, 24], they also causemidgut cell lysis. These proteins are not only toxic to thelarvae of lepidopteran, dipteran and coleopteran insects,but also to nematodes, protozoan pathogens, and animal-parasitic liver flukes and mites [10].B. thuringiensishasbeen used widely as the most promising insecticideamong alternatives.

However, the potential for development of resis-tance and cross-resistance in target insect populations toCry proteins used alone or in combination threatens themore widespread use of this novel pest control technol-ogy. High levels of resistance to one of Cry proteins canlead to substantial cross-resistance to other Cry protein[11]. Wirth et al. (1997) have reported that high levels ofresistance to Cry4 proteins (Cry4A, Cry4B, and Cry11A)in larvae of the mosquito,Culex pipens quinquefasciatus,can be suppressed or reduced markedly by combiningthese proteins with sub-lethal quantities of Cyt1Aa [26].Thus, they put forward the suggestion that the Cyt1Aa/Cry4 model provides a potential molecular genetic strat-egy for engineering resistance management for Cry pro-teins directly into bacterial insecticides and transgenicplants. Furthermore, it is reported that Cyt1Aa proteinsuppresses resistance to other Cry proteins, such asCry11B [25] and Cry3Aa [9]. In addition, Cyt1Aa syn-ergizes other three Cry proteins (Cry4A, Cry4B, andCry11A) in B. thuringiensissubsp.israelensisagainstmosquitoes [6, 7, 27, 30]. Cyt proteins are highly hydro-phobic d-endotoxin that share no sequence homologywith Cry proteins and appears to have a different mode ofaction. Cyt proteins lack significant cross-resistance toCorrespondence to:Yi Pang;email: [email protected]

CURRENT MICROBIOLOGY Vol. 43 (2001), pp. 112–119DOI: 10.1007/s002840010271 Current

MicrobiologyAn International Journal© Springer-Verlag New York Inc. 2001

Cry proteins, indicating that Cyt proteins may play aneven more important long-term role in managing resis-tance to Cry proteins.

Most of the genes coding for these ICPs are plasmid-borne. A limitation in engineering the recombinantB.thuringiensisfor expanding the toxic spectrum and en-hancing their toxicity by plasmid conjugation transfer-ence or coexpression of ICPs has been shown in previouswork due to unstability and incompatibility of the plas-mids. In this study, two recombinant transposon-derivedvectors were constructed by cloningcyt1Aa gene andcry11Aa gene that are originally isolated from the plas-mids of B. thuringiensis subsp. israelensis into thepTV1TS (containing Tn917 fromStreptococcus faeca-lis), respectively. Cry11A (or named CryIVD) protein istoxic only to dipterous larvae [12], whereas Cyt1A pro-tein is highly cytolytic to various cells and toxic tomosquitoes (and related dipterans) [12, 13, 15, 19] andthe cotton leaf beetle,Chrysomela scripta[9]. The re-combinant transposon-derived vectors were transferredinto the wild-typeB. thuringiensisstrain S184 againstlepidopterous larvae by electroporation. The primary en-gineered strain TnX, which integrated successfully twogenes ofcyt1Aa andcry11Aa into the genomic DNA(chromosome) ofB. thuringiensisstrain S184 by trans-position frequency of approximately 5.93 1025, wasachieved. The integrated ICP-genes were expressed inthe strain TnX and normal parasposal crystals were ob-served under electron microscope. Owing to losing ofsome large plasmids from the parent strain S184, thestrain TnX showed no toxicity to lepidopterous larvae.The lost plasmids were re-obtained by the protoplastfusion of the strain TnX and the strain S184-Tetr (thestrain S184 transformed with theBt-E. colishuttle vectorpBR261 that harbors two genes responsible for resis-tance to ampicillin and tetracycline [32]), and as well asthe target engineered strain, designated as TnY. Bioassayshowed that the strain TnY was highly toxic not only tolepidopterous but also to dipterous larvae. These resultsprovide a new, important way for engineering novelgeneticB. thuringiensisinsecticides.

Materials and Methods

Bacterial strains and media. The wild-type B. thuringiensisstrainS184 that was toxic against lepidopterous larvae, such asSpodopteralitura, was isolated from the soil, and its serotype was H3aH3b (data notshown). The strain S184 contained two major crystal proteins of 130kDa and 60 kDa (Fig. 4). Two strains of TnX and TnY were engineeredin this study. The strain TnY has been conserved in the ConservationCenter of Typical Cultures of China (Wuhan University, China), thenumber of conservation: CCTCC M98017, the date of conservation: 26November 1998. Two commercial strains ofB. thuringiensissubsp.aizawai(Bta) andB. thuringiensissubsp.israelensis(Bti) were used inthis study. Grown in Luria-Bertani medium and G-Tris medium [1], as

required, an antibiotic such as tetracycline (Tet), ampicillin (Ap), orerythromycin (Em), was added. Protoplasts were regenerated on a MMculture plate which contained 0.5% yeast extract, 1% tryptone, 20 mMMgCl2, 0.5 M saccharose, and 1.5% agar [5].

Plasmids.Five plamids were used in this study, including pHT3101,pWF26, pWF45, pBR261 and pTV1Ts. pHT3101 is aBt-E. colishuttlevector that was described by Lereclus et al. [17]. pWF45 [28] andpWF26 [29] are individually sources of thecry11A gene and thecyt1Aa gene, kindly provided by Dr. Dong Wu. pBR261, aBt-E. colishuttle vector, contains two resistance genes resposible for Ap and Tet[32]. pTV1Ts carries Em resistance transposon Tn917 fromStrepto-coccus faecalis[22].

Construction of the vectors pTYP45, pTYP26 and pHtnpA.pWF45contains thecyt1Aa gene and the gene for the 20 kDa protein undercontrol of thecry1Ac promoter [28]. pWF26 contains thecry11Aa geneand the gene for the 20 kDa protein [29]. These two plasmids weretreated byHindIII to have about 8-kb inserted fragments. The latterwere cloned into the large fragment of pTV1Ts, which was also treatedby HindIII, losing the 1.3-kb small fragment located intnpA region ofTn917, to produce two recombinant plasmids of pTYP45 (Fig. 1) andpTYP26 (Fig. 2), respectively. To construct pHtnpA, a fragment con-taining the complete genetnpA and its promote was cloned frompTV1Ts into pHT3101 at the sites ofEcoRI andKpn I (Fig. 3).

Electroporation transformation. B. thuringiensiscells were trans-formed by electroporation, as described by references [4, 33].

Construction of the primary engineered strain TnX. The primaryengineered strain TnX was constructed as described by reference [31].In brief, three plasmids of pTYP26, pTYP45 and pHtnpA (1:1:1) weresimultaneously transformed into the strain S184 by electroporation, andspread on the LB-agar plate containing antibiotic Em. By the identifi-cations of Dot blot, Southern blot and PCR, the primary engineeredstrain TnX, of which two target genes integrated into the chromosome,was obtained.

Fig. 1. Construction of the vector pTYP45.

J. Yu et al: Engineered Strain ofB. thuringiensis 113

Construction of the strain S184-Tetr. The wild-type strain S184 wastransformed with theBt-E.coli shuttle vector, pBR261, which harborstwo genes responsible for resistance to ampicillin and tetracycline toget the strain S184-Tetr.

Protoplast fusion of B. thuringiensis. The engineered strain TnY wasconstructed by the protoplast fusion of the strain TnX and the strainS184-Tetr. Overnight cultures of the strain S184-Tetr and the strainTnX were diluted 1:100 in 20 ml of Luria–Bertani medium, containinga moderate antibiotic, with shaking at 30°C for 3 h. The cell pelletsresulting from the centrifugation were washed once with SMM (20 mMMgCl2, 0.5 M sucrose, 20 mM maleate buffer, and pH6.5) and sus-pended in 4 ml SMM buffer at a final concentration of 13 108 cells/ml.The cells were digested at 37°C for 1 h in thepresence of 2.5 mg/mllysozyme (Sigma Co.) in SMM buffer until most ofbacilli turned intoprotoplasts, and then resuspended in 1 ml 1% bovine serum albumin(BSA) in SMM buffer after they were centrifuged at 3500 rpm for 15min. Then, 0.5 ml protoplast suspension of each parent strain wasmixed and centrifuged at 3500 rpm for 15 min. Immediately after 3 minexposure to 0.9 ml PEG and 0.1 ml neonatal calcium phosphatecontaining 0.01 M KH2PO4 and 0.5 M CaCl2, the mixture was centri-fuged and appropriately diluted. The primary fusants were selected byincubating them on a MM culture plate at 28°C, then by replica filterplating onto a LB medium plate, containing 10m g/ml Tet and Em. Thefusants were analyzed by sodium dodecyl sulfate-polyacylamide gelelectrophoresis (SDS-PAGE), observed under electron microscopy,and used in bioassay.

Protein analysis and quantification. The ICPs ofB. thuringiensiswere analyzed and quantified by SDS-PAGE [21].B. thuringiensisstrains were grown separately in 1 ml PWYE medium (5% peptone,0.1% yeast extract, 0.5% NaCl, and pH7.5) overnight at 30°C, and thendiluted 1:100 with 100 ml G-Tris medium [1] with moderate antibioticor without. Cultures were grown at 37°C with shaking for 3-4 days, bywhich time more than 95% of the cells had sporulated and lysed.Spores, crystals, and cell debris were sedimented by centrifugation. Thepellet was suspended in a 50m 1 Laemmli sample buffer and boiled for5 min until completely dissolved. Protein content was determined by

subjecting 50m1 samples to SDS-PAGE as described by Laemmli [16].The gel was stained with 0.125% Coomassie Blue R-250, destained,dried, and the protein bands were scanned using a ElectrophoresisDocumentation and Analysis 120 System (Eastman Kodak Co.) Theamount of protein in each band was quantified using ImageMaster VDSsoftware (Amersham Pharmacia Biotech).

Transmission and scanning electron microscopy.For electron mi-croscopy, inclusions were purified and treated as described previously[21, 26, 29].

Bioassay.Standardized bioassay procedures were used as described [2,20]. For lepidopteran insects,Argyrogramma agnatalarvae, ICPs weremixed into the semi-solid artificial diet to give the final concentrationas needed. The mixture was made with the diet temperature below55°C. After the diet cooled and solidified, the test insects were placedon the treated diet and the containers were covered and incubated in28°C for a suitable period of time. Sterile, distilled water was added tothe control. The 50% lethal concentrations (LC50s) were calculated48 h after treatment, using a minimum of 30 larvae per concentration,and six dilutions per sample. This process was replicated three times.For mosquito,Culex pipens fatigans, twenty third-instar larvae wereplaced in 50-ml plastic cups containing 20 ml of distilled water. Theappropriate concentration of ICPs was added, and mortality was deter-mined after 24 h. Twelve different concentrations were used, whichyielded mortality rates ranging from 0–100%. Tests were replicated atleast five times on 4 or 5 different days. For each bioassay, LC50s wereestimated by using the probit analysis.

Results and Discussion

Analysis of the primary engineered strain TnX. Onaccount of few available insertion sites in the nonessen-tial region of Tn917, the target fragment containingcyt1Aa andcry11Aa was inserted into thetnpA gene inthe essential region in order to produce the recombinantplasmids. As a result, the recombinant transposon-de-rived vectors pTYP45 and pTYP26 didn’t express thetransposase, TnpA, and the transposition frequency wasextremely low. By using thetnpA gene on the pHtnpAvector that can express TnpA, the transposition fre-quency of the strain TnX largely increased by approxi-mately 5.93 1025 and was similar to that of pTY1Ts.

Commonly, the ICP-genes ofB. thuringiensisareusually located on large (30- to 200-MDa) plasmids [3].Determination by SDS-PAGE (Fig. 4) showed the strainTnX lost the 60 kDa protein of the strain S184. One ofthe probable explanations is that the strain TnX lost mostof its large plasmids from the strain S184 in the processof molecular genetic manipulation. The results of bioas-say against lepidopterous larvae showed that the strainTnX had almost no toxicity.

Screening of stable fusants.For easy screening, thestrain S184-Tetr, containing the Tetr gene as a selectedmarker, was constructed. Resistance to Tet was a veryimportant characteristic of the strain S184-Tetr sincemost of B. thuringiensiswere sensitive to Tet. On theother hand, the strain TnX had resistance to Em while the

Fig. 2. Construction of the vector pTYP26.

114 CURRENT MICROBIOLOGY Vol. 43 (2001)

strain S184-Tetr had not. So the target fusant wasscreened on the LB plate with Tet and Em.

Incubated on LB medium plates containing antibi-otic, 186 Emr Tetr colonies were selected. They werepropagated on LB medium plates, and, at the same time,on LB plates with Tet and Em to detect the phenotype.The results of the resistance marker segregation of fus-ants are shown in Table 1. During eight rounds, the

resistance markers of 172 fusants were segregated.About 74.2% of fusants preferred to the resistance ofTnX-Emr, and about 18.3% of them preferred to theresistance of the strain S184-Tetr. Only 7.5% of them hadthe characteristic of two parent strains. After culturingfor more than 20 rounds, the 14 fusants were stable.

Determination of inclusion protein composition bySDS-PAGE.The inclusions isolated from the 16 fusants,consisting of 2 Emr fusants and 14 Emr Tetr fusants, wereanalyzed by SDS-PAGE. Among them, 11 Emr Tetr

fusants and 2 Emr fusants contained four major proteinswhose migration corresponded to molecular mass posi-tions of 130 kDa, 67 kDa, 60 kDa, and 27 kDa, respec-tively. So, the 13 fusants were target fusants, designed asTnY1-13. Among the other 3 fusants, 1 fusant containedtwo major proteins of 130 kDa and 60 kDa, coincidingwith the protein profiles of the strain S184, while 2fusants contained three major proteins of 130 kDa, 67kDa and 27 kDa, being consistent with the protein pro-files of the strain TnX. The results show that there is nolinkage between ICP genes and resistant genes.

ICPs from the strain TnY contained not only pro-teins of 27 kDa (Cyt1Aa) and 67 kDa (Cry11Aa) but alsocrystal proteins of 130 kDa and 60 kDa derived from the

Fig. 3. Construction of the vectorpHtnpA.

Fig. 4. Analysis of ICPs from the wild-typeB. thuringiensisstrain S184and the two engineered strains of TnX, TnY by SDS-PAGE. Lane 1:blank. Lane 2: TnX. Lane 3: TnY. Lane 4: S184.

J. Yu et al: Engineered Strain ofB. thuringiensis 115

strain S184. The results suggested that most of largeplasmids lost from the original strain have been fullyretrieved. The protein quantification revealed that theyield of 60 kDa, 67 kDa and 130 kDa proteins from thestrain TnY was no less than that of corresponding pro-teins of the two parent strains, S184 and TnX. However,the yield of 27 kDa protein from the strain TnY wasdistinctly lower than that of corresponding protein of thestrain TnX, as illustrated in Fig. 4. It was perhaps that theresources (such as amino acids) required for proteinsynthesis were limited in a cell.

Determination of ICPs by transmission and scanningmicroscopy. The crystals from three strains of S184,TnX and TnY were prepared for transmission and scan-ning microscopy, photographed, and measured. Thecrystals from the wild-type strain S184 were bipyramidalin shape, measuring 2.01mm long and 0.86mm in widthon average (Fig. 5). The crystals from the strain TnXwere bipyramidal and oval (or irregular) in shape, cor-responding to Cyt1Aa and Cry11Aa (Fig. 6), respec-tively, as described previously [24, 28, 29]. The crystalsin bipyramidal shape measured 0.92mm long and 0.71mm in width on average. The crystals from the strainTnY were also bipyramidal and oval (or irregular) inshape (Fig. 7), similar in shape to those from the strainS184 and the strain TnX. This results further showed that

the engineered strain TnY derived from the fusion of thestrain S184-Tetr and the strain TnX because it containedthe crystal proteins of the two strains, as shown in Table2. However, the crystals from the strain TnY weresmaller in size compared to those from the two parentstrains. For example, crystals in bipyramidal shapewhich were similar to those from the strain S184 mea-sured 1.42mm long and 0.63mm in width, on average,and the others in bipyramidal shape similar to those fromthe strain TnX measured 0.52mm long and 0.4mm inwidth on average.

Stability of the engineered strain.The two target genesof cyt1Aa andcry11Aa integrated into the chromosomewere extremely stable and there was little possibility of asecond transposition because thetnpA gene on the pHt-npA vector was lost in the process of heat-treatment.After the two engineered strains of TnY and TnX werepropagated for more than 100 rounds without pressure ofantibiotics, the determinations of resistance and proteinsdemonstrated that the lost frequency of the two targetgenes was less than 1025.

Toxicity of the engineered strain. The toxicity of allinclusions from four strains of S184, TnX, TnY and Btawas determined against the third instar larvae ofS. litura.The commercial strain Bta that is toxic toSpodopteraspecies was adapted as the positive control of the bioas-

Fig. 5. Morphology and size of ICPs produced in the wild-typeB.thuringiensisstrain S184. Transmission electron micrographs of puri-fied ICPs illustrating bipyramidal shape.

Fig. 6. Morphology and size of ICPs produced in the engineeed strainTnX. Transmission electron micrographs of purified ICPs illustratingbipyramidal and ovoid (or irregular) shape.

Table 1. Resistance marker segregation of fusants

Propagated rounds 1 2 3 4 5 6 7 8 9 10 11;20

Phenotype of strains TnX– – – –Emr 7 60 31 6 18 14 2 0 0 0 0S184– – –Tetr 3 15 10 5 1 0 0 0 0 0 0Fusants-Emr Tetr 176 101 60 49 30 16 14 14 14 14 14

116 CURRENT MICROBIOLOGY Vol. 43 (2001)

say. Results presented in Table 3 showed that the strainTnX had little toxicity to the larvae with a LC50 of morethan 2mg/ml, while the toxicity LC50 of the strain S184and the strain TnY was 3.04-fold and 6.33-fold higherthan that of Bta, respectively. It was noteworthy that thetoxicity of the strain TnY increased by 2.08-fold com-pared to that of the wild-type strain S184. The strain TnYcontained Cyt1Aa and Cry11Aa proteins, in addition tothe crystal proteins of the strain S184, and it was certainthat the Cry11Aa did not contribute to the toxicityagainstSpodopteraspecies. It has been reported that Cytproteins suppress resistance to Cry proteins, such asCry11B, Cry3Aa, Cry4A, Cry4B and Cry11A, and alsointeract synergistically with these proteins [9, 25, 26, 27,30]. However, the toxicity of the mixtures of Cyt1Aaprotein and crystal proteins from the strain S184 againstthe third instar larvae ofS. litura, was not significantlydifferent from that only from the strain S184. TheCyt1Aa crystal mentioned above was purified from thepWF45 transfomed 4Q7 that is an acrystalliferous strain

of B. thuringiensissubsp.israelensis. The results dem-onstrated that Cyt1Aa did not interact synergisticallywith the crystal proteins of the strain S184 to enhance thetoxicity againstS. litura (data not shown). It was notknown if some other factors contributed to the hightoxicity of the strain TnY.

The toxicity of all inclusions from four strains ofS184, TnX, TnY and Bti was determined against thethird instar larvae ofCulex pipiens fatigans. The inclu-sion of the commercial strain Bti was used as the positivecontrol for the bioassay. Results presented in Table 4showed that the strain TnX had very high toxicity to thelarvae with LC50 of 5.12 ng/ml, which was 1.82-foldhigher than that of the strain Bti and significantly higherthan that of the strain TnY with the LC50 of 103.20ng/ml, whereas the wild-type strain S184 had little tox-icity to the larvae with LC50 of more than 6mg/ml. Thestrain TnX contained Cyt1Aa and Cry11Aa proteins inaddition to 130-kDa proteins from the strain S184. It wasproven that the 130-kDa protein did not contribute to thetoxicity against dipterous larvae. In our previous re-search, the purified Cyt1Aa and Cry11Aa crystals, pro-duced by strains whichcyt1Aa andcry11Aa genes hadbeen individually integrated into the chromosome, wereused for bioassay. We obtained the results that the LC50

againstCulex pipiens fatiganslarvae of Cyt1Aa andCry11Aa were 64ng/ml and 83ng/ml, respectively, whilethe mixtures of Cyt1Aa and Cry11Aa (1:1) enhanced thetoxicity by 4–5 fold with the LC50 of 17 ng/ml [31]. Thatthere was significant synergism between Cyt1Aa and

Fig. 7. Morphology and size of ICPs produced in the engineeed strain TnY. Transmission (A) and Scanning (B) electron micrographs of purifiedICPs illustrating bipyramidal and ovoid (or irregular) shape.

Table 2. Crystal shapes and major molecular masses of ICPs fromthreeB. thuringiensisstrains of S184, TnX and TnY

Strains Crystal shapeMajor molecular mass

(kDa)

S184 bipyramid 130, 60TnX Bipyramid and ovoid (or irregular) 130, 67, 27TnY Bipyramid and ovoid (or irregular) 130, 60, 67, 27

J. Yu et al: Engineered Strain ofB. thuringiensis 117

Cry11Aa was in agreement with the result in reference[30]. Therefore, the lower yield of Cyt1Aa protein fromthe strain TnY than from the strain TnX, as indicated bySDS-PAGE and electron microscopy, accounted for thehigher LC50 value of the strain TnY. However, the syn-ergism between them may not be the most importantreason. Because TnX was extremely more toxic with theLC50 of 5.12 ng/ml than the mixtures of Cyt1Aa andCry11Aa in vitro, and also more toxic than the strain Btiwhich contains four proteins, Cyt1Aa (27kDa), Cry11Aa(67kDa), Cry4Aa (128kDa) and Cry4Ba (134kDa) [12].What causes the high level toxicity of the strain TnXagainstCulex pipieus fatiganshas not been made clear.

ACKNOWLEDGMENTS

This project was supported by 863 Plan of China (No. 863-101-03-01-01), National Natural Sciences Foundation of China (No. 39900098)and Natural Sciences Foundation of Guangdong Province (No. 980296and No. 963056).

Literature Cited1. Arvidson H, Dunn PE, Strnad S, Aronson AI (1989) Specificity of

Bacillus thuringiensisfor lepidopteran larvae: Factors involved invivo and in the structure of a purified protoxin. Mol Microbiol3:1533–1543

2. Beegle CC, Rose RI (1995)Bacillus thuringiensisbioassay meth-odology. In: Feng TY, (ed),Bacillus thuringiensisbiotechnologyand environmental benefit, Vol. 1. Taipei: Hua Shiang Yuan Pub-lishing Co., pp 349–362

3. Carlton BC, Gonzalez Jr JM (1985) Plasmids andd-endotoxinproduction in different subspecies ofBacillus thuringiensis. In:Hoch JA, Setlow P, (eds), Molecular biology of microbial differ-entiation. Washington, DC: American Society for Microbiology pp246–252

4. Chang C, Dai SM, Frutos R, Federici BA, Gill SS (1992) Proper-ties of a 72-kilodalton mosquital protein fromBacillus thuringien-sis subsp.morrisoni PG-14 expressed inB. thuringiensissubspkurstakiby using the shuttle vector pHT3101. Appl Environ Mi-crobiol 58:507–512

5. Chang S, Cohen SN (1979) High frequency transformation ofBacillus subtilis protoplasts by plasmid DNA. Mol Gen Genet168:111–115

6. Chilcott CN, Ellar DJ (1988) Comparative toxicity ofBacillusthuringiensisvar. israelensiscrystal proteins in vivo and in vitro.J Gen Microbiol 134:2551–2558

7. Crickmore N, Bone E, Williams JA and Ellar DJ (1995) Contri-bution of the individual components of thed-endotoxin crystal tothe mosquitocidal activity ofBacillus thuringiensissubsp. is-raelensis. FEMS Microbiol Lett 131:249–254

8. EMBL nucleotide database: www.susx.ac.uk./users/bafnb/bt/in-dex.html

9. Federici BA, Bauer LS (1998) Cyt1Aa protein ofBacillus thurin-giensisis toxic to the cottonwood leaf beetle,Chrysomela scripta,and suppresses high levels of resistance to Cry3Aa. Appl EnvironMicrobiol 64:4368–4371

10. Feitelson JS, Payne J, Kim L (1992)Bacillus thuringiensis: insectsand beyond. Bio/Technology 10:271–275

11. Gould F, Martinez-Ramirez A, Anderson A, Ferre J, Silva FJ, andMoar WJ (1992) Broad-spectrum resistance toBacillus thuringien-sistoxins in heliothis virescens. Proc Natl Acad Sci USA 89:7986–7990

Table 3. Toxicity of ICPs produced by the wild-type strain S184 and the engineered strain TnY toSpodoptera lituralarvae

Bt strainsMean LC50 (mg/ml)a

(95% fiducial limits) Slope (6SE) Activity ratio

Btab 29.65 (23.55;37.33) 2.3006 0.118 1S184 9.74 (7.63;12.43) 2.1666 0.125 3.04TnX .2000c n.d. n.d.TnY 4.68 (3.67;5.98) 2.1846 0.124 6.33

a Against third instars.b The strain Bta were used as positive controls forSpodotera lituralarvae.c No toxicity detected at 2 mg/ml, the highest concentration tested.

Table 4. Toxicity of ICPs produced by the two engineered strains of TnX and TnY toCulex pipiens fatiganslarvae

StrainsMean LC50 (ng/ml)a

(95% fiducial limits) Slope (6SE) Activity ratio

Btic 9.30 (6.22;13.90) 1.326 0.20 1S184

.6000bn.d. n.d.TnY 103.20 (63.78;166.99) 1.106 0.246 0.09TnX 5.12 (4.24;6.18) 2.806 0.096 1.82

a Against third instars.b No toxicity detected at 6mg/ml, the highest concentration tested.c The strain Bti was used as positive control forCulex pipiens fatiganslarvae.

118 CURRENT MICROBIOLOGY Vol. 43 (2001)

12. Hofte H, Whiteley HR (1989) Insecticidal crystal proteins ofBacillus thuringiensis. Microbiol Rev 53:242–55

13. Knowles BH and Ellar DJ (1987) Colloid-osmotic lysis is a generalfeature of the mechanism of action ofBacillus thuringiensisdelta-endotoxin with different insect specificity. Biochim Biophys Acta924:509–518

14. Knowles BH, Blatt MR, Tester M, Horsnell JM, Carroll J, Me-nestrina G, Ellar DJ (1989) A cytolytic delta-endotoxin fromBacillus thuringiensisvar. israelensisforms cation-selective chan-nels in planar lipid bilayers. FEBS Lett 244(2):259–262

15. Knowles BH, Dow JAT (1993) The crystald-endotoxins ofBacil-lus thuringiensis: Models for their mechanism of action on theinsect gut. Bioessays 15:469–476

16. Laemmli UK (1970) Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 227:680–685

17. Lereclus D, Arantes O (1992) spbA locus ensures the segregationalstability of pHT1031, a novel type of gram-positive replicon. MolMicrobiol 6:35–46

18. Li J, Koni PA, Ellar DJ (1996) Structure of the mosquitocidaldelta-endotoxin CytB fromBacillus thuringiensissubsp.kyushuen-sis and implications for membrane pore formation. J Mol Biol257(1):129–152

19. Mclean KM, Whiteley HR (1987) Expression inEscherichia coliof a cloned crystal protein gene ofBacillus thuringiensissubsp.israelensis. J Bacteriol 169:1017–1023

20. Pang Y, Frutos R, Federici BA (1992) Synthesis and toxicity offull-length and truncated bacterial CryIVD mosquitocidal proteinsexpressed in lepidopteran cells using a baculovirus vector. J Gen-eral Virology 73:89–101

21. Park H, Ge B, Bauer LS, Federici BA (1998) Optimization ofCry3A yields in Bacillus thuringiensisby use of sporulation-dependent promoters in combination with the STAB, SD mRNAsequence. Appl Environ Microbiol 64:3932–3938

22. Sandman K, Losick R, Youngman P (1987) Genetic analysis ofBacillus subtilisspo mutants generated by Tn917-mediated insec-tional mutagensis. Genetics 117:603–617

23. Sangadala S, Walters FS, English LH, Adang MJ (1994) A mixtureof Manduca sexta aminopeptidase and phosphatase enhancesBa-

cillus thuringiensis insecticidal CryIA(c) toxin binding and86Rb(1) 2 K1 efflux in vitro. J Biol Chem 269(13):10088–10092

24. Thomas WE, Ellar DJ (1983) Mechanism of action ofBacillusthuringiensisvar israelensisinsecticidal delta-endotoxin. FEBSLett 154:362–368

25. Wirth MC, Deleclude A, Federici BA, Walton WE (1998) Variablecross-resistance to Cry11B fromBacillus thuringiensissubsp.je-gathesanin Culex quinguesfasciatus(Diptera: Culicidae) resistantto single or multiple toxins ofBacillus thuringiensissubsp.is-raelensis. Appl Environ Microbiol 64:4174–4179

26. Wirth MC, Georghiou GP, Federici BA (1997) CytA enablesCryIV endotoxins ofBacillus thuringiensisto overcome high lev-els CryIV resistance in the mosquito,Culex quinguesfasciatus.Proc Natl Acad Sci USA 94:10536–10540

27. Wu D, Chang FN (1985) Synergism in mosquitocidal activity of 26and 65 kDa proteins fromBacillus thuringiensissubsp.israelensiscrystal. FEBS Lett 190:232–236

28. Wu D, Federici BA (1993) A 20-kDa protein preserves cell via-bility and promotes cytA crystal formation during sprulation inBacillus thuringiensis. J Bacteriol 175:5276–5280

29. Wu D, Federici BA (1995) Improved production of the insecticidalCryIVD protein inBacillus thuringiensisusing cry1A (c) promot-ers to express the gene for an associated 20-kDa protein. ApplMicrobiol Biotechnol 42:697–702

30. Wu D, Johnson J, Federici BA (1994) Synergism of mosquitocidaltoxicity between CytA and CryIVD proteins using inclusions pro-duced from cloned genes ofBacillus thuringiensis. Mol Microbiol13:965–972

31. Yu JX, Pang Y, Li JH, Yu RJ, Tang MJ (1999) Construction ofinsecticidal engineeredBacillus thuringiensisusing the transposon.Acta Scientiarum Naturalium Universitatis Sunyatseni 38:52–57

32. Yu JX, Zhang YJ, Pang Y, Xu MQ (2000) A replication origin ofBacillus thuringiensis. Curr Microbiol 44:123–127

33. Zhuang M, Pang Y, Deng R, Long Q, Tang C (1995) Expansion ofthe toxic spectrumBacillus thuringiensisby eletroporation. Supplto J Sun Yatsen Univ 2:83–88

J. Yu et al: Engineered Strain ofB. thuringiensis 119