Molecular Characterization Two Novel Crystal Protein …jb.asm.org/content/174/2/549.full.pdf · thompsoniandB. thuringiensis subsp. kurstaki HD-1-Dipel, separated in a 0.6% horizontal

JOURNAL OF BACTERIOLOGY, Jan. 1992, p. 549-557 Vol. 174, No. 20021-9193/92/020549-09$02.00/0Copyright © 1992, American Society for Microbiology

Molecular Characterization of Two Novel Crystal Protein Genesfrom Bacillus thuringiensis subsp. thompsonit

KIT L. BROWN* AND H. R. WHITELEYDepartment of Microbiology SC42, University of Washington, Seattle, Washington 98195

Received 5 September 1991/Accepted 12 November 1991

Two genes encoding the predominant polypeptides of Bacillus thuringiensis subsp. thompsoni cuboidalcrystals were cloned in Escherichia col and sequenced. The polypeptides have electrophoretic mobilities of 40and 34 kDa, with the deduced amino acid sequences predicting molecular masses of 35,384 and 37,505 Da,respectively. No statistically significant similarities were detected between the 40- or 34-kDa crystal protein andany other characterized B. thuringiensis crystal protein, nor were they detected between the 40- and 34-kDacrystal proteins. A 100-MDa plasmid carries both crystal protein genes, which appear to be part of an operon,with the 40-kDa gene 64 nucleotides upstream of the 34-kDa gene. Both crystal proteins are synthesized inapproximately the same amounts. Even though small compared with other crystal proteins, the 34-kDa crystalprotein has insecticidal activity against lepidopteran larvae (Manduca sexta). The 40-kDa polypeptide appearsto have no insecticidal activity, but it could have a role in crystal structure.

During sporulation, Bacillus thuringiensis produces pro-teinaceous crystals which are lethal to a variety of insectlarvae. These crystal proteins have been grouped into fourclasses on the basis of their host ranges and sequencehomologies: class I, lepidopteran specific; class II, lepi-dopteran-dipteran specific; class III, coleopteran specific;and class IV, dipteran specific (20). Significant amino acidsimilarities between the crystal proteins of the differentclasses exist, with the carboxy-terminal half of the crystalproteins containing most of the conserved sequences. Fivewell-defined regions are conserved among most of theknown crystal proteins; these are located in the N-terminalhalf of the protein, which is responsible for toxicity (21). Oneexception is CytA, a 28-kDa cytolytic toxin from B. thurin-giensis subsp. israelensis, which has no detectable sequenceidentity with the other crystal proteins (21, 41).The majority of the crystal proteins and all class I lepi-

dopteran-specific crystal proteins are synthesized as 130- to140-kDa protoxins, which are then proteolytically cleaved inthe insect midgut to 65- to 70-kDa active toxins. Somecrystal proteins, those of classes II and III, are produced as65- to 70-kDa toxins (reviewed in reference 21). The onlycrystal protein which falls outside these two size ranges isCytA, one of five crystal proteins from dipteran-specific B.thuringiensis subsp. israelensis (41). However, CytA has amode of action different from that of other crystal proteins(37, 38) and has recently been reported not to be essential formosquitocidal activity (12).Thus, to date, all the crystal proteins show some related-

ness and can be grouped into two size ranges. However,electrophoretic analysis of B. thuringiensis subsp. thomp-soni crystals (7), combined with crystal serotyping used todetermine crystal protein relatedness (23), suggested that thecrystal proteins of B. thuringiensis subsp. thompsoni wereunique.

This paper describes the cloning of two genes encodingnovel crystal proteins from B. thuringiensis subsp. thomp-soni. These crystal proteins are unique because of their size

* Corresponding author.t Dedicated to the memory of H. R. Whiteley.

(electrophoretic mobilities of 40 and 34 kDa) and, particu-larly, because their deduced amino acid sequences do notcontain any of the conserved regions observed in othercharacterized B. thuringiensis crystal proteins. For the in-sect assays used in this study, the 34-kDa crystal protein byitself demonstrates insecticidal activity, thereby making itthe smallest mature toxin reported to date.

MATERIALS AND METHODS

Bacteria and plasmids. B. thuringiensis subsp. thompsoniwas obtained from P. Baumann, University of California,Davis (originally from H. D. Burges; Dulmage designationHD-542). B. thuringiensis subsp. kurstaki HD-1-Dipel wasobtained from L. A. Bulla (22). Escherichia coli DH5a(Bethesda Research Laboratories) and JM103 (24) were thehosts for cloning purposes. E. coli DPWC and JGM (ob-tained from Melvin Simon via Kelly Hughes) were used for-y mobilization (36). Plasmids pTZ18R (Pharmacia) andpBluescript II KS+ (Stratagene) were used as cloning vec-tors. Plasmid pKK223-3 (Pharmacia) was used as an expres-sion vector for the crystal protein genes.

Molecular methods and enzymes. The standard molecularmethods used have been described previously (29). B. thu-ringiensis subsp. thompsoni plasmid DNA was isolated asdescribed previously (22) and purified by CsCl gradientcentrifugation. E. coli plasmid DNA was isolated by themethod of Bimboim and Doly (1). E. coli transformation wasby electroporation (15) or with competent cells (29). Restric-tion enzymes were purchased from New England BioLabs,Inc., Boehringer Mannheim, and Bethesda Research Labo-ratories. Calf intestinal alkaline phosphatase came fromBoehringer Mannheim. Klenow fragment, exonuclease III,and reverse transcriptase were purchased from BethesdaResearch Laboratories. All enzymes were used according tothe instructions of the manufacturers.

Electron microscopy. A synchronized culture of B. thu-ringiensis subsp. thompsoni was grown at 300C until thepresence of phase-dark prespores and the formation ofcrystalline inclusions were observed by phase-contrast mi-croscopy. The cells were harvested by centrifugation andresuspended in 0.1 M cacodylate buffer, pH 7.3. The cells

549

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

550 BROWN AND WHITELEY

were fixed with 3% glutaraldehyde and resuspended in 1.5%Noble agar. The agar-embedded cells were washed in ca-codylate buffer and dehydrated with ethanol. Finally, thecells were washed with propylene oxide and infiltrated andembedded in plastic resin. After thin sectioning, the sampleswere viewed by scanning electron microscopy. Purified B.thuringiensis subsp. thompsoni crystals were prepared asdescribed above but were viewed by transmission electronmicroscopy.B. thuringiensis subsp. thompsoni plasmid DNA hybridiza-

tion. Plasmids were isolated from B. thuringiensis subsp.thompsoni and B. thuringiensis subsp. kurstaki HD-1-Dipel,separated in a 0.6% horizontal agarose gel, and transferredto nitrocellulose as described previously (22). A DNA frag-ment specific to the 40- and 34-kDa protein genes wasradiolabelled by hexanucleotide random priming (16). Theradiolabelled DNA was subjected to alkaline denaturationand allowed to hybridize to the Southern blot DNA in 20%formamide-3x SSC (lx SSC is 0.15 M NaCl plus 0.015 Msodium citrate) for 20 h at 25°C. The blot was washed in 2xSSC-0.1% sodium dodecyl sulfate (SDS) for 15 min at 25, 37,and 42°C, air dried, and exposed to Kodak XAR film.

Antibody production and N-terminal sequencing. Crystalswere isolated from B. thuringiensis subsp. thompsoni andpurified by Renografin gradient centrifugation (34). Thepurified crystals were solubilized in 4% SDS-10% P-mercap-toethanol and electrophoresed in a preparative SDS-10%polyacrylamide gel. The peptide bands were visualized bystaining in 0.15 M potassium chloride, and the individualprotein bands were excised from the gel. The proteins wereeluted from the polyacrylamide gel slices at 4°C with 4 W ofconstant power for 3 h by using the Isco electroelution-concentration device. The protein concentration of the sam-ples was determined by the method of Bradford (2). Eachsample, containing 20 ,ug of protein in 0.5 ml of 10 mM Tris-1mM EDTA, was emulsified 1:1 with Freund's completeadjuvant and injected into mice subcutaneously (20). Fiveweeks later, a booster of 10 ng of each sample was admin-istered intramuscularly, and ascites sarcoma 180/TG cellswere injected into the mice intraperitoneaily (31). The poly-clonal antiserum was harvested from the ascites fluid 10 to 14days later, and it was centrifuged to obtain a clarifiedsupernatant and used without further purification.

B. thuringiensis subsp. thompsoni crystals were purifiedand electrophoresed as described above and transferred toan Immobilon membrane (Millipore). The N-terminal aminoacid sequences of the 40- and 34-kDa crystal proteins weredetermined by automated Edman degradation.

Crystal protein gene cloning. On the basis of amino acidsequences, two degenerate DNA oligonucleotides were syn-thesized: a 38-mer for the 40-kDa crystal protein, 5'-ATGAA(C,T) TTC AAC AA(C,T) AT(C,T) ACI GGI AACTT(C,T) AAG GAC GT-3', and a 32-mer for the 34-kDacrystal protein, 5'-ATG AAT GA(C,T) AT(T,A) GCICA(G,A) GAT GCI GCI (C,A) GI GC-3'. Deoxyinosine wasused as a neutral base at the fourfold redundant positions toreduce base mismatching. B. thuringiensis subsp. thompsoniplasmid DNA was digested with EcoRI and subjected tosucrose gradient centrifugation. Fractions were separated byagarose gel electrophoresis, transferred to nitrocellulose,and hybridized to both oligonucleotides (29). Probe-reactivefractions were ligated into pTZ18R (which had been digestedwith EcoRI and treated with calf intestinal alkaline phos-phatase) and transformed into E. coli DH5a. Positive cloneswere selected by colony blot hybridization with the 38-merand 32-mer oligonucleotides and by the production of pro-

teins reacting with the 40- and 34-kDa antibodies. Twoclones were selected, both of which contained the same8.4-kb fragment but in opposite orientations, as indicated byrestriction enzyme mapping (data not shown). One of theclones, pKB100, was subcloned by removing an SstI-SstIfragment from the 5' end and ligating the 7.5-kb SstI-EcoRIfragment into pTZ18R, generating pKB102.

Subclone construction. Recombinant plasmid pKB102 wasused to generate the subclones pKB107 and pKB109.pKB102 was digested first with BstEII and then with exonu-clease III and treated with S1 nuclease to generate bluntends that were ligated to yield pKB107. To generatepKB109, the transposon -yb was mobilized by bacterialmating (36) into an SstI-NruI subclone of pKB102 (see Fig.3).To obtain specific clones of the genes encoding the 40- and

34-kDa crystal proteins, the polymerase chain reaction(PCR) was used to isolate their respective sequences. Thepriming oligonucleotides were used to amplify the regionencoding the 40-kDa protein (with the introduction of anApaI site upstream and a SalT site downstream) and also theregion encoding the 34-kDa protein (with the introduction ofan SmaI site upstream and an XbaI site downstream). Thesequences of the PCR primers were as follows: for the geneencoding the 40-kDa protein, sense strand 5'-GAG GGCCCA ATA AGG TGT CAG CT-3' and its nonsense strand,5'-GCG TCG ACT ATC ATT CCA TTA CAC-3'; for thegene encoding the 34-kDa protein, sense strand 5'-CTC CCGGGT GTA ATG GAA TGA TA-3' and its nonsense strand,5'-GCT CTA GAT CTT CAC AAT CCG GA-3'. Amplifica-tion was performed in a 100-,ul reaction volume containing 5ng of template, 1 ,ug of each primer, 200 ,.M (each) deoxy-nucleoside triphosphates, 2.5 U of Taq DNA polymerase(Promega), and Taq buffer. The 32 rounds of amplificationconsisted of 15 s at 940C, 15 s at 45°C, and 60 s at 72°C in aCoy thermal cycler.The PCR-generated fragments were digested with the

appropriate restriction enzymes and cloned into pBluescriptIT KS+ to generate pKB401 (40 kDa) and pKB341 (34 kDa).Expression vectors were then constructed by excising the40- and 34-kDa protein genes [as a 997-bp KpnI (blunt)-HindIII fragment and a 1,300-bp PstI-SstI (blunt) fragment,respectively] from these plasmids and ligating them intopKK223-3, generating pKB200 and pKB201. To generate asubclone containing both crystal protein genes, the geneencoding the 40-kDa protein was excised from pKB200 as anEcoRI-HindIII (blunt) fragment and ligated into pKB201digested with EcoRI-PstI (blunt), generating pKB202.DNA sequencing. Both strands of the DNA were se-

quenced by using the dideoxy-chain termination method ofSanger et al. (30). DNA fragments from pKB102 weresubcloned into pTZ18R, and successive unidirectional dele-tions were created by using exonuclease III (19). The secondstrand and any gaps in the first strand were sequenced byusing a series of complementary synthetic oligonucleotides.Sequencing templates were generated by subjecting CsCl-purified plasmid DNA to alkaline denaturation followed byethanol precipitation. Sequencing was accomplished with[o-35S]dATP (New England Nuclear) and the SequenaseVersion 2.0 kit (US Biochemical). Sequence similaritieswere analyzed with FASTDB (6) from the IntelliGeneticssoftware package.

Insect toxicity assays. Toxicity of purified B. thuringiensissubsp. thompsoni crystals and E. coli inclusions (32) toneonate larvae of the tobacco hornworm, Manduca sexta,was tested as described previously, except that the crystals

J. BACTERIUOL.

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

B. THURINGIENSIS SUBSP. THOMPSONI CRYSTAL PROTEIN GENES 551

A B C

0i 0-MC - 0 CCl)ClCCD

oc Mcx C O M0COm CO CO _ m

in m m° - a m ma m m° m m

97 -

68 -

40 -

34 -

10,

1t;S

Vj; E,

v0 ~I0

FIG. 1. Electron micrographs of crystals from B. thuringiensissubsp. thompsoni. (A) Thin section of a sporulating B. thuringiensissubsp. thompsoni cell containing a crystalline inclusion; (B) three-dimensional view of a B. thuringiensis subsp. thompsoni crystal.Bar in each panel, 100 nm.

and/or inclusions were not solubilized prior to the assay (25,33). Protein concentrations were estimated by densitometry(202 Ultrascan laser densitometer; LKB Instruments, Inc.)of Coomassie blue-stained SDS-polyacrylamide gels, withdilutions of bovine gamma globulin used as standards.

Nucleotide sequence accession number. The sequence datapublished in this paper have been assigned GenBank acces-sion number M76442.

RESULTS

Crystal structure. To determine the shape of B. thurin-giensis subsp. thompsoni crystals, cells were grown untilphase-dark prespores and crystals were observed. The elec-tron micrograph of a thin-sectioned cell (Fig. 1A) showed theproduction by B. thuringiensis subsp. thompsoni of a squarecrystal alongside the spore. To determine whether the crys-tals were flat or cuboidal, purified B. thuringiensis subsp.thompsoni crystals were viewed by transmission electronmicroscopy. The electron micrograph in Fig. 1B shows thethree-dimensional structure of a B. thuringiensis subsp.thompsoni crystal as cuboidal rather than flat.

Cloning of the B. thuringiensis subsp. thompsoni genes in E.coli. Electrophoretic analysis of purified B. thuringiensissubsp. thompsoni crystals showed only two peptide bands,which migrated at estimated electrophoretic mobilities of40,000 and 34,000 Da (Fig. 2A, lane 1). These sizes varysomewhat from those reported by Calabrese and Nickerson(7), 48,000 and 40,000 Da; however, this discrepancy might

1 2 1 2 3 4 5 6 7 1 2 3 4 5 6 7

FIG. 2. Electrophoretic analysis of crystals and inclusions fromrecombinant E. coli clones. (A) Coomassie blue-stained SDS-poly-acrylamide gel of purified B. thuringiensis subsp. thompsoni crystals(lane 1) and purified inclusions from E. coli clone pKB102 (lane 2).(B) Immunoblot with antibody specific for the 40-kDa crystalprotein. Lanes: 1, B. thuringiensis subsp. thompsoni crystals; 2 to 7,purified E. coli inclusions pKB102 (lane 2), pKB107 (lane 3),pKB109 (lane 4), pKB200 (lane 5), pKB201 (lane 6), and pKB202(lane 7). (C) Immunoblot with antibody specific for the 34-kDacrystal protein (lane designations are the same as in panel B). E. coliDH5a was the host for all the plasmids except pKB107, which wasin E. coli JGM. Molecular masses (in kilodaltons) are indicated tothe left of panel A.

be explained by the different procedures used for sizeestimation. Immunoblots showed the specificity of the twocrystal protein antibody preparations (Fig. 2B and C, lane 1).Additionally, these two antibodies did not cross-react withrepresentative members from each of the four crystal proteinclasses or with CytA (data not shown).To facilitate cloning of the B. thuringiensis subsp. thomp-

soni crystal proteins, the N-terminal sequences of the 40-kDa polypeptide [MNFNNITGNFKDVTELFTDYANQX(S)XQNG] and the 34-kDa polypeptide [AIMNDIAQDAARAXDIIAGPFIRPGT(T)PXN(N)QLF(N)(Y)X(I)(G)(N)] were determined. Residues in parentheses are uncer-tain, and unknown residues are represented with an X. Fromthese amino acid sequences, DNA oligonucleotides weresynthesized. Both oligonucleotides hybridized to a 7.5-kbSstI-EcoRI fragment cloned from B. thuringiensis subsp.thompsoni plasmid DNA as described in Materials andMethods. Recombinant E. coli carrying pKB102 (Fig. 3)synthesized two polypeptides which reacted with the 40- and34-kDa antibodies, respectively (Fig. 2B and C, lane 2). Themobilities of the polypeptides produced from pKB102 wereidentical or very similar to those of polypeptides present inpreparations of purified B. thuringiensis subsp. thompsonicrystals (Fig. 2A, lanes 1 and 2). A ca. 100-kDa peptide,visible in immunoblots of B. thuringiensis subsp. thompsoniand pKB102, cross-reacted with both antibody preparations

VOL. 174, 1992

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

552 BROWN AND WHITELEY

Reported Sequence

40 kDa 34 kDa

S Ns CI I I

B CNs N NsC1_1 III

SI_

SV

N-soft- I

A Sa

M X

EI

(S)I

500 bpFIG. 3. Restriction maps showing the locations of the crystal protein genes on the recombinant plasmids used in this study. The positions

and orientations of the two crystal protein genes are indicated by arrows. The open area in pKB107 is a deleted region. The triangle in pKB109is the site of the -yf insertion. The diagrammed size of pKB109 does not account for the additional 5.7 kb of -y8 DNA. The followingabbreviations were used for restriction sites: A, ApaI; B, BstEII; C, ClaI; E, EcoRI; H, HindlIl; M, SmaI; N, NruI; Ns, NsiI; S, SstI; Sa,Sall; and X, XbaI. The restriction sites shown on pKB200 and pKB201 were obtained from PCR amplification. The SstI site in parentheseson pKB202 was lost during cloning.

(Fig. 2B and C, lanes 1 and 2). This peptide species ispossibly an aggregate of the 40- and 34-kDa crystal proteins,or it might represent another polypeptide, present on theDNA cloned in pKB102, which cross-reacts with both anti-body preparations.

Sequence determination of the cloned crystal protein genes.The region of pKB102 containing the genes encoding the 40-and 34-kDa crystal proteins was sequenced (Fig. 3). TheDNA sequence (Fig. 4) revealed the presence of two openreading frames (ORFs) in the same orientation and separatedby 64 nucleotides, with each ORF preceded by a potentialribosome-binding site (24). The sequence of the first ORFmatched the N-terminal amino acid sequence of the 40-kDaprotein determined previously and could encode a polypep-tide with a predicted molecular mass of 35,384 Da. TheN-terminal amino acid sequence predicted from the se-quence of the second ORF corresponded to the N-terminalamino acid sequence of the 34-kDa protein. This ORF couldcode for a polypeptide with a predicted molecular mass of37,505 Da. Since both strands of the DNA were sequenced,the discrepancies between the predicted molecular masses ofthe two crystal proteins and those estimated by electropho-retic mobility most likely resulted from anomalous migrationduring electrophoresis. Molecular mass discrepancies havealso been observed for CrylIA and CrylIB: each sequencepredicts a polypeptide of ca. 71 kDa, but the proteins migrateas 65- and 50-kDa polypeptides, respectively (42).The sequences were searched for any regions that resem-

bled B. thuringiensis, Bacillus subtilis, or E. coli promoterstructures (4, 5, 13, 18, 27, 43). No likely promoter se-quences were found in the sequenced region, either in theregion between the two genes or for approximately 700nucleotides upstream of the 40-kDa crystal protein gene. It ispossible that there is a B. thuringiensis promoter in thisregion but that it does not match consensus sequencesreported for B. thuringiensis or B. subtilis RNA polymerases(4, 5, 18, 27, 43). Alternatively, both genes may be tran-scribed from a promoter located further upstream. Addi-tional experiments will be required to locate the transcrip-tional initiation site of the B. thuringiensis subsp. thompsonicrystal protein genes.The sequenced region was searched for the presence of

inverted repeat structures which could result in transcriptiontermination (28). A sequence which could serve as a rho-independent transcriptional terminator was found down-stream of the gene encoding the 34-kDa protein. Transcrip-tion of this region could lead to the formation of a stem-and-loop structure with a AG' of -19.1 kcal (ca. -79.9 kJ),calculated according to the rules of Tinoco et al. (39).However, this inverted repeat structure does not have theclassical GC-rich region of other terminators or a run of T'safter the stem-and-loop; thus, this structure may not serve toterminate transcription. The organization of the two ORFs,the presence of individual ribosome-binding sites, the lack ofa promoter between the two genes, and the putative tran-scription terminator suggest that these genes are likely to be

E

EI

pKB102

pKB1 07

pKB109

pKB200

pKB201

pKB202

m s m 0 0 Mj

0 a miammj

J. BACTERIOL.

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

VOL. 174, 1992 B. THURINGIENSIS SUBSP. THOMPSONI CRYSTAL PROTEIN GENES 553

1 CCAATAAGGTGTCAGCTGAATCTAAATTCGAAAGGAGAATAAAAATGAATTTTAACAATATCACAGGAAATTTTAAAGATGTCACAGAACTATTTACAGAM N F N N I T G N F K D V T E L F T D

101 TTACGCTAATCAATGGAGTCGGCAAAATGGGGGTGGTAAGCCTGAAATTTCTCTTATAGTACCAGGGTATGAGGCTTATGCTGTGACTAGTTCTGATGATY A N Q W S R Q N G G G K P E I S L I V P G Y E A Y A V T S S D D

201 AGAACTATCTATCATCATCCAAAGAAAAAGAAACGAGAAAAATCGAGATCACATCATGGTTGCTCAAGCGAGAACCAGGAACGTATATATGAAGATACGTR T I Y H H P K K K K R E K S R S H H G C S S E N Q E R I Y E D T Y

301 ATGAGACTAATCTATCTTTCAATCATGATCCTAATTTAATGGAAGAATGCGAAAAGGAAATTGAACTTGCAACTGAGACGTATGAAAATGCCAGTTGTCAE T N L S F N H D P N L M E E C E K E I E L A T E T Y E N A S C H

401 TGAAAAGAAAATAAAGATACAAATCGGTGGAAATGTAGAAAATTATGGTGAGTGGTTCGTTTACGAAGGGGCAACTTTATCAGGAAAAGACCTTCTATCCE K K I K I Q I G G N V E N Y G E W F V Y E G A T L S G K D L L S

501 ATTGATGTATTTGGACATGAACCAGTAGACATTGATCAAGTCCCTGTTTCATTGCATCCAGGAGAAATAGAAGTATTAAAGCGACCATTGGAAGTTGACAI D V F G H E P V D I D Q V P V S L H P G E I E V L K R P L E V D T

601 CATATCGCTCTTATAAGATTCGCCCTCGTTCTAAGGTCACTGCGACTTTGAAAGTAAAACAAAAACATTTTAAACAGTGCTTTGATGTGGAAACAGATGTY R S Y K I R P R S K V T A T L K V K Q K H F K Q C F D V E T D V

701 ATCTGGTTATGTTGCAATTATACAAAAACAAAAAGATTGTGATGTACAGACATCATTTCACCATGTTGCCGCTATTTTACAACGGTATTACAGTCCCTTTS G Y V A I I Q K Q K D C D V Q T S F H H V A A I L Q R Y Y S P F

801 ATTCGTATCAATGGAGATGAAGTAACTTTACTGTGTAAAGGAGTATTTAAAGGGGTTAAGATTACGGATATATATATTCATATCCAGATAGAAAGTTTAGI R I N G D E V T L L C K G V F K G V K I T D I Y I H I Q I E S L D

901 ATATTCCTGGATTGATTGAAGAGTATAACATTTATGATGTGAATCAACGAAATATAGGTGTAATGGAATGATAGTACAAAACTCATAAATTAGATTGATGI P G L I E E Y N I Y D V N Q R N I G V M E

1001 AGAATCTGATTTATATTTTAAAGGAGGAATTTATAATGGCAATTATGAATGATATTGCACAAGATGCAGCAAGAGCTTGGGATATAATAGCAGGGCCATTM A I M N D I A Q D A A R A W D I I A G P F

1101 TATACGACCGGGAACAACTCCTACCAATCGACAATTATTTAATTATCAAATTGGAAATATAGAGGTTGAACCTGGAAATCTTAATTTTTCAGTCGTCCCTI R P G T T P T N R Q L F N Y Q I G N I E V E P G N L N F S V V P

1201 GAACTAGACTTTAGTGTCTCTCAAGACCTTTTCAACAATACAAGTGTGCAGCAAAGTCAAACAGCATCATTTAACGAATCAAGAACGGAAACGACTTCAAE L D F S V S Q D L F N N T S V Q Q S Q T A S F N E S R T E T T S T

1301 CGGCCGTTACTCATGGCGTAAAATCTGGGGTTACCGTTTCTGCTTCAGCAAAATTTAATGCCAAAATATTAGTAAAATCCATTGAGCAAACTATTACAACA V T H G V K S G V T V S A S A K F N A K I L V K S I E Q T I T T

1401 AACGGTTTCTACAGAATATAATTTTAGTAGTACTACAACTAGAACAAATACTGTAACAAGGGGATGGTCAATTGCTCAGCCTGTATTAGTTCCTCCTCATT V S T E Y N F S S T T T R T N T V T R G W S I A Q P V L V P P H

1501 AGTAGAGTAACAGCAACATTGCAAATTTATAAAGGGGATTTTACAGTGCCCGTTCTATTATCACTTAGAGTTTATGGTCAAACAGGAACACTTGCAGGGAS R V T A T L Q I Y K G D F T V P V L L S L R V Y G Q T G T L A G N

1601 ATCCTAGTTTTCCTTCTTTATATGCAGCCACATATGAAAACACACTTTTGGGAAGAATTAGAGAGCATATTGCTCCACCTGCTCTTTTCAGAGCCTCCAAP S F P S L Y A A T Y E N T L L G R I R E H I A P P A L F R A S N

1701 CGCATACATTTCGAATGGCGTTCAGGCAATTTGGAGAGGAACAGCAACGACGAGAGTTTCGCAAGGTCTGTATTCCGTTGTAAGAATCGATGAAAGACCTA Y I S N G V Q A I W R G T A T T R V S Q G L Y S V V R I D E R P

1801 TTAGCAGGTTATTCAGGAGAAACAAGAACGTATTATTTACCAGTGACACTTTCAAATTCAAGTCAAATCCTTACACCTGGTTCTTTAGGAAGTGAGATTCL A G Y S G E T R T Y Y L P V T L S N S S Q I L T P G S L G S E I P

1901 CAATTATCAATCCAGTTCCGAATGCATCTTGTAAAAAGGAAAACTCGCCTATTATCATTCATCATGATCGAGAGAAGCATCGTGAACGCGATTATGATAAI I N P V P N A S C K K E N S P I I I H H D R E K H R E R D Y D K

2001 AGAGCATATTTGTCATGATCAAGCTGAGAAGTATGAACGCGATTATGATAAAGAATAACTAATTATGTAAGAGATTTGTAAACAAGAGAAATAGCATTTTE H I C H D Q A E K Y E R D Y D K E

201

2101 ACTATTTCTCTTGTTTTTAATCTATAATTAGAATGGTAGACGCTCTTTAAATTAAATGTAAAAAAAGGGGGCTAAGATTATAGAATAAA TCAATCCAAAA

2201 CAATATAAGCTAATTATTTTACCTCTTTTCATGTATTGGTCCGGATTGTGAAGATCAFIG. 4. Nucleotide and deduced amino acid sequences of the 40- and 34-kDa crystal protein genes. Ribosome-binding sites are underlined

twice, and the inverted repeats which form a potential transcription terminator are marked by long arrows over the sequences.

part of an operon. This operon may have additional ORFs were analyzed to identify any similarities to other knownupstream of the 40-kDa crystal protein gene and possibly protein sequences. No statistically significant similaritiesdownstream of the 34-kDa crystal protein gene as well. between the 40- or 34-kDa crystal protein sequence and anyThe predicted amino acid sequences of the two ORFs other protein sequence, including any other crystal protein

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

554 BROWN AND WHITELEY

-

C:I

1I

MUQ2

O

tn 0co 0

150-

47-

30-

If-

5.5-

1 2 3

FIG. 5. Electrophoretic analysis of plasmids and location ofcrystal protein genes in B. thuringiensis subsp. thompsoni. Lanes 1and 2, ethidium bromide-stained agarose gel of plasmids isolatedfrom HD-1-Dipel (lane 1) and B. thuringiensis subsp. thompsoni(lane 2). The numbers to the left indicate sizes in megadaltons. Lane3, autoradiograph of lane 2 DNA transferred to nitrocellulose andhybridized with a 32P-labelled fragment from the region encoding theB. thuringiensis subsp. thompsoni crystal protein genes. The blackdot to the right of lane 2 indicates the plasmid carrying the crystalprotein genes. Some hybridization was observed with the linearizedfragments; this most likely resulted from breakage of the crystalprotein gene-containing plasmid.

sequence, were detected. Additionally, comparison of the40- and 34-kDa amino acid sequences showed no substantialregions of identity between the two proteins.

Plasmid profile. To examine the plasmid profile of B.thuringiensis subsp. thompsoni, purified plasmids were sep-

arated in an agarose gel. The sizes of the B. thuringiensissubsp. thompsoni plasmids were determined by comparisonwith sized plasmids isolated from B. thuringiensis HD-1-Dipel (Fig. 5, lanes 1 and 2) (22). Five plasmids in B.thuringiensis subsp. thompsoni, ranging from 40 MDa togreater than 150 MDa, were observed, with four of theplasmids clustered around 100 to 150 MDa (Fig. SA, lane 2).This profile differs from that reported, but not shown, byCarlton and Gonzalez (9), who reported four plasmids in B.

thuringiensis subsp. thompsoni ranging from 4 to 100 MDa.These authors may have used a different strain, or the largerplasmids may have been broken during isolation; in ourstudy, no plasmids smaller than 30 MDa were ever observed.To determine which plasmid encoded the crystal protein

genes, B. thuringiensis subsp. thompsoni plasmid DNA wastransferred to nitrocellulose and hybridized with a radioac-tive probe specific to the genes encoding the 40- and 34-kDaproteins. The crystal protein genes were located on thepredominant plasmid species of -100 MDa (Fig. 5, lane 3).This observation agrees with the results of Carlton andGonzalez (8), who showed that an acrystalliferous strain ofB. thuringiensis subsp. thompsoni resulted from the loss of a100-MDa plasmid.

Description of subclone constructions. As described inMaterials and Methods, subclones of pKB102 were gener-ated in which the gene expression of each crystal protein wasindividually eliminated. In order to leave any regulatoryelements on pKB102 intact, insertional inactivation or dele-tion was used. Because of the lack of unique restrictionenzyme sites within the gene encoding the 40-kDa protein,-yb transposition was used to inactivate the expression of itsgene product (36). Plasmid pKB109 (Fig. 3) had a -y5 inser-tion in the 40-kDa crystal protein gene after nucleotide 161(Fig. 4), resulting in the synthesis of only the product of the34-kDa crystal protein gene (Fig. 2B and C, lane 4). Thoughpart of an operon, the 34-kDa crystal protein gene wasexpressed in pKB109, either because -y5 does not containany transcriptional terminators or because it contains apromoter which was used for transcription of the 34-kDagene. A different approach was used to inactivate the 34-kDacrystal protein gene; a unique restriction enzyme site,BstEII, within the 34-kDa crystal protein gene was used toinitiate exonuclease III digestion. A resulting subclone,pKB107 (Fig. 3), had a deletion in the 34-kDa crystal proteingene ranging from nucleotides 1190 to 1920 (Fig. 4). Thisclone expressed only the 40-kDa crystal protein, as demon-strated by immunoblot analysis (Fig. 2B and C, lane 3).To determine the insecticidal activity of the crystal pro-

teins, subclones were constructed that contained only thegene encoding the 40-kDa protein (pKB200), only the geneencoding the 34-kDa protein (pKB201), or both genes(pKB202) (Fig. 3). In these subclones, the genes wereexpressed under the control of the strong tac promoter.Each gene had its own Shine-Dalgarno sequence present (24)and was cloned upstream of the E. coli rrnB T1 and T2terminators to ensure transcriptional termination. Recombi-nant E. coli with pKB200 expressed only the 40-kDa crystalprotein (Fig. 2B and C, lane 5), whereas the subclonepKB201 synthesized only the 34-kDa crystal protein (Fig. 2Band C, lane 6). Both crystal proteins were expressed in E.coli carrying pKB202 (Fig. 2B and C, lane 7).

Insect toxicity assays. Purified B. thuringiensis subsp.thompsoni crystals and inclusions isolated from severalrecombinant E. coli clones were tested for toxicity againstthe larvae of the lepidopteran M. sexta. The samples werenot solubilized prior to testing for toxicity against insectlarvae, because the various methods (pH >11 or 0.2%sarkosyl [16] in 0.1 M carbonate buffer) which were success-ful in solubilizing the crystals or inclusions destroyed theirbiological activity (data not shown). Solutions that normallysolubilize other crystal proteins and inclusions (e.g., 0.1 Mcarbonate buffer [pH 10]-0.2% 3-mercaptoethanol) resultedin concentrations of soluble protein that were too low to usein the insect assays. Perhaps expressing the various recom-

J. BACTERIOL.

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

B. THURINGIENSIS SUBSP. THOMPSONI CRYSTAL PROTEIN GENES 555

TABLE 1. Assay for toxicity of B. thuringiensis subsp.thompsoni crystals and of inclusions purified from recombinant

E. coli clones

Expressed gene(s) LC50b for 34-kDa proteinOrganism or clone (kDa)a M. sexta concnc

B. thuringiensis subsp. 40, 34 0.40 0.25thompsoni

pKB102 40, 34 0.40 0.25pKB107 40 >3.0pKB109 34 0.25 0.25pKB200 40 >3.0pKB201 34 0.25 0.25pKB202 40, 34 0.98 0.25

aThe molecular mass in kilodaltons of the protein which the gene encodes.b LC50 values (in micrograms per square centimeter + 50%) are based on

the total concentration of expressed polypeptide(s). Protein concentrationswere estimated as described in Materials and Methods.

c Estimated protein concentration (in micrograms per square centimeter ±50%) of the 34-kDa polypeptide at the LC50 values of the various clones.

binant clones in either Bacillus cereus or B. subtilis wouldproduce inclusions that could be solubilized more easily.The recombinant E. coli carrying pKB102, which pro-

duced both crystal proteins, exhibited toxicity levels equiv-alent to those of the B. thuringiensis subsp. thompsonicrystals, with a 50% lethal concentration (LC50) of 0.40p.g/cm2 (Table 1). The clone, pKB202, had an LC50 of 0.98pLg/cm2; however, the 34-kDa polypeptide concentration wasthe same as that of B. thuringiensis subsp. thompsoni andpKB102, 0.25 p.g/cm2. When only the 34-kDa crystal proteinwas expressed, in E. coli carrying pKB109 and pKB201, theLC50 of the 34-kDa polypeptide, 0.25 p.g/cm2, was similar tothat observed for clones expressing both crystal proteins,pKB102 and pKB202 (Table 1). However, E. coli carryingpKB107 and pKB200 that expressed only the 40-kDa crystalprotein was not toxic to the larvae at a concentration of 3p.g/cm2 (Table 1). B. thuringiensis subsp. thompsoni crystalsalso demonstrated toxicity against the lepidopteran Artogeiarapae (Cabbage white) but did not kill larvae of Heliothisvirescens or Trichoplusia ni at concentrations of 3 p.g/cm2(concentration of both crystal proteins), although growth ofthese larvae was severely stunted. Purified B. thuringiensissubsp. thompsoni crystals were also tested for toxicityagainst the dipteran Aedes aegypti and the coleopteranLeptinotarsa decemlineata (Colorado potato beetle), but notoxicity was observed.

DISCUSSION

In this paper, we report the DNA sequences of two genesencoding unique crystal proteins from B. thuringiensissubsp. thompsoni. These genes appear to be part of anoperon, the other members of which have yet to be eluci-dated. To date, almost all of the known crystal protein genesare monocistronic, except crylIA, which is contained in anoperon that has two additional ORFs whose products areunidentified (42). This is the first report of genes encodingthe predominant polypeptides of a B. thuringiensis crystalwhich are located in the same operon.

Since the cloned crystal proteins expressed in E. coli (i.e.,any of the clones carrying the 34-kDa crystal protein gene)had the same levels of toxicity against lepidopteran larvae asdid the purified crystals from B. thuringiensis subsp. thomp-soni, it is unlikely that another toxic crystal protein ispresent on the unsequenced, cloned DNA of pKB102 or

elsewhere in the B. thuringiensis subsp. thompsoni genome.If other ORFs exist within the crystal protein gene operon,they might code for structural components of the crystal orfor regulatory proteins which control or stabilize expressionof the crystal proteins. In this regard, it is interesting to notethat the construct pKB202 resulted in the production of the40-kDa protein at levels three times those of the 34-kDaprotein. This is unlike the pattern of expression seen with B.thuringiensis subsp. thompsoni and pKB102, in which thenet amount of the 34-kDa protein is 1.5 times that of the40-kDa protein. These observations suggest the presence inB. thuringiensis subsp. thompsoni of a regulatory elementwhich is present in pKB102 and absent in pKB202. Analternative explanation would be that the RNA is unstable atthe 3' end in pKB202, so that the 34-kDa protein is expressedat reduced levels. In pKB202, each gene has its nativeShine-Dalgarno sequence; therefore, the different levels ofexpression of the 40-kDa polypeptide and the 34-kDa poly-peptide cannot be explained by more efficient translationalinitiation in pKB202 than in pKB102.The toxicity assays in this study focused on only a few

lepidopteran species-primarily M. sexta, but also A. rapae,H. virescens, and T. ni. For these lepidopterans, the 34-kDacrystal protein appeared to account for the insecticidalactivity because the 40-kDa protein demonstrated no toxic-ity when tested alone. The 34-kDa polypeptide is the small-est mature crystal protein with insecticidal activity reportedto date. Some small toxic molecules have been reported forB. thuringiensis subsp. israelensis; however, these have allarisen from digesting the large mature toxins with trypsin(11). It is not known whether proteolytic fragments derivedfrom the B. thuringiensis subsp. thompsoni 34-kDa crystalprotein will retain full toxicity.Although there was no observable difference in toxicity

between the 34-kDa crystal protein alone and the twoproteins together for the insects tested, the possibility thatthe 40-kDa protein may affect toxicity in some cases has notbeen excluded. The cooperative effect of two small proteins(51 and 42 kDa) in producing toxicity against mosquitoes hasbeen seen for Bacillus sphaericus (3). Additionally, therehave been no reports that any nontoxic proteins are major B.thuringiensis crystal components, so it would not be surpris-ing if the 40-kDa crystal protein was required for toxicityagainst some insects or was involved in toxicity at a level notdetectable in these experiments.Although the 40-kDa crystal protein appears to be non-

toxic to the insects examined, perhaps it has insecticidalactivity against some lepidopterans or other orders of in-sects. Clearly, crystal proteins, even those within the sameclass, have different host ranges. For example, CryIC isspecific for Spodoptera littoralis and Mamestra brassicae,but it has little toxicity toward M. sexta or H. virescens (21),and a newly described crystal protein, CryIE, is specific forS. littoralis but is not toxic toward M. brassicae (40). Ratherthan containing a toxic domain, the 40-kDa crystal protein inB. thuringiensis subsp. thompsoni may be required forcrystal formation or structure. The 40-kDa polypeptidemight provide the necessary scaffolding, thus allowing it andthe 34-kDa polypeptide to form toxic cuboidal crystals.The deduced amino acid sequences of the 40- and 34-kDa

crystal proteins have no significant similarity to those of anyother crystal proteins or to each other. This finding isunusual, because to date all the characterized crystal pro-teins share some degree of homology (reviewed in reference21). Even recently reported crystal proteins which fall intonew subclasses-CryIE (40), CryIF (10), and CryllIB (35)-

VOL. 174, 1992

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

556 BROWN AND WHITELEY

have extensive homologies to crystal proteins in other sub-classes. To date, the most divergent insecticidal crystalproteins are CryllA and CryIVD from B. thuringiensissubsp. kurstaki and B. thuringiensis subsp. israelensis, re-spectively (14, 42). Though divergent, these crystal proteinsstill have sequences that are conserved with respect to othercrystal proteins (21). Another crystal protein, CytA, from B.thuringiensis subsp. israelensis does not have any of thecrystal protein conserved sequences; however, CytA is not aspecific insecticidal toxin but instead demonstrates a generalcytolytic activity (37, 38, 41). Because the sequences for theB. thuringiensis subsp. thompsoni 40- and 34-kDa crystalproteins are so divergent from that of any known crystalprotein, it is unclear how they might be evolutionarilyrelated to other crystal proteins. These two crystal proteingenes cannot be easily classified in any of the currentlycharacterized crystal protein gene classes; therefore, nogene designations will be given at this time. Perhaps, afteradditional crystal proteins have been identified, a pattern ofrelatedness between the 40- and 34-kDa crystal proteins andthe other crystal proteins will emerge.

ACKNOWLEDGMENTSWe are grateful to J. C. Lara of the Department of Microbiology,

University of Washington, for the electron micrographs; Earl Davieof the Department of Biochemistry, University of Washington, forthe N-terminal sequencing; and Lynn Riddiford of the Departmentof Zoology, University of Washington, for providing M. sexta. Wethank Stephen Lory, Jon Visick, and Howard Clarke for criticallyreading the manuscript. K.L.B. acknowledges the technical assis-tance of Elizabeth K. Baker with the N-terminal sequencing andAugust A. Zitzka with the insect toxicity assays.

This research was supported by Public Health Service grantGM-20784 from the National Institute of General Medical Sciences.H.R.W. was the recipient of Research Career Award K6-GM-442from the National Institute of General Medical Sciences. K.L.B.was supported in part by the Dr. Helen Riaboff Whiteley GraduateFellowship in Microbiology.

REFERENCES1. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction

procedure for screening recombinant plasmid DNA. NucleicAcids Res. 7:1513-1523.

2. Bradford, M. M. 1976. A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 72:248-254.

3. Broadwell, A. H., L. Baumann, and P. Baumann. 1990. The 42-and 51-kilodalton mosquitocidal proteins of Bacillus sphaericus2362: construction of recombinants with enhanced expressionand in vivo studies of processing and toxicity. J. Bacteriol.172:2217-2223.

4. Brown, K. L., and H. R. Whiteley. 1988. Isolation of a Bacillusthuringiensis RNA polymerase capable of transcribing crystalprotein genes. Proc. Natl. Acad. Sci. USA 85:4166-4170.

5. Brown, K. L., and H. R. Whiteley. 1990. Isolation of the secondBacillus thuringiensis RNA polymerase that transcribes from acrystal protein gene promoter. J. Bacteriol. 172:6682-6688.

6. Brutlag, D. L., J.-P. Dautricourt, S. Maulik, and J. Relph. 1990.Improved sensitivity of biological database searches. Comput.Appl. Biosci. 6:237-245.

7. Calabrese, D. M., and K. W. Nickerson. 1980. A comparison ofprotein crystal subunit sizes in Bacillus thuringiensis. Can. J.Microbiol. 26:1006-1010.

8. Carlton, B. C., and J. M. Gonzalez, Jr. 1984. Plasmid-associateddelta-endotoxin production in Bacillus thuringiensis, p. 387-400. In A. T. Ganesan and J. A. Hoch (ed.), Genetics andbiotechnology of bacilli. Academic Press, New York.

9. Carlton, B. C., and J. M. Gonzalez, Jr. 1985. The genetics andmolecular biology of Bacillus thuringiensis, p. 211-249. In D.Dubnau (ed.), The molecular biology of the bacilli. Academic

Press, New York.10. Chambers, J. A., A. Jelen, M. Pearce Gilbert, C. S. Jany, T. B.

Johnson, and C. Gawron-Burke. 1991. Isolation and character-ization of a novel insecticidal crystal protein gene from Bacillusthuringiensis subsp. aizawai. J. Bacteriol. 173:3966-3976.

11. Chilcott, C. N., and D. J. Ellar. 1988. Comparative toxicity ofBacillus thuringiensis var. israelensis crystal proteins in vivoand in vitro. J. Gen. Microbiol. 134:2551-2558.

12. Delecluse, A., J. Charles, A. Klier, and G. Rapoport. 1991.Deletion by in vivo recombination shows that the 28-kilodaltoncytolytic polypeptide from Bacillus thuringiensis subsp. israe-lensis is not essential for mosquitocidal activity. J. Bacteriol.173:3374-3381.

13. Doi, R. H., and L.-F. Wang. 1986. Multiple procaryotic ribonu-cleic acid polymerase sigma factors. Microbiol. Rev. 50:227-243.

14. Donovan, W. P., C. C. Dankocsik, and M. P. Gilbert. 1988.Molecular characterization of a gene encoding a 72-kilodaltonmosquito-toxic crystal protein from Bacillus thuringiensissubsp. israelensis. J. Bacteriol. 170:4732-4738.

15. Dower, W. J., J. F. Miller, and C. W. Ragsdale. 1988. Highefficiency transformation of E. coli by high voltage electropora-tion. Nucleic Acids Res. 16:6127-6145.

16. Feinberg, A. P., and B. Vogelstein. 1983. A technique forradiolabeling DNA restriction endonuclease fragments to highspecific activity. Anal. Biochem. 132:6-13.

17. Frankel, S., R. Sohn, and L. Leinwand. 1991. The use ofsarkosyl in generating soluble protein after bacterial expression.Proc. Natl. Acad. Sci. USA 88:1192-1196.

18. Helmann, J. D., and M. J. Chamberlin. 1988. Structure andfunction of bacterial sigma factors. Annu. Rev. Biochem. 57:839-872.

19. Henikoff, S. 1984. Unidirectional digestion with exonuclease IIIcreates targeted breakpoints for DNA sequencing. Gene 28:351-359.

20. Herbert, W. J. 1968. The mode of action of mineral-oil emulsionadjuvants on antibody production in mice. Immunology 14:301-318.

21. Hofte, H., and H. R. Whiteley. 1989. Insecticidal crystal proteinsof Bacillus thuringiensis. Microbiol. Rev. 53:242-255.

22. Kronstad, J. W., H. E. Schnepf, and H. R. Whiteley. 1983.Diversity of locations for Bacillus thuringiensis crystal proteingenes. J. Bacteriol. 154:419-428.

23. Lynch, M. J., and P. Baumann. 1985. Immunological compari-sons of the crystal protein from strains of Bacillus thuringiensis.J. Invertebr. Pathol. 46:47-57.

24. McLaughlin, J. R., C. L. Murray, and J. C. Rabinowitz. 1981.Unique features in the ribosome binding site sequence of thegram-positive Staphylococcus aureus P-lactamase gene. J. Biol.Chem. 256:11283-11291.

25. McLean, K. M., and H. R. Whiteley. 1987. Expression inEscherichia coli of a cloned crystal protein gene of Bacillusthuringiensis subsp. israelensis. J. Bacteriol. 169:1017-1023.

26. Messing, J., R. Crea, and P. H. Seeburg. 1981. A system forshotgun DNA sequencing. Nucleic Acids Res. 9:309-321.

27. Moran, C. P. 1989. Sigma factors and the regulation of tran-scription, p. 167-184. In I. Smith, R. A. Slepecky, and P. Setlow(ed.), Regulation of procaryotic development. American Soci-ety for Microbiology, Washington, D.C.

28. Rosenberg, M., and D. Court. 1979. Regulatory sequencesinvolved in the promotion and termination of RNA transcrip-tion. Annu. Rev. Genet. 12:319-353.

29. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecularcloning: a laboratory manual, 2nd ed. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

30. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequenc-ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci.USA 74:5463-5467.

31. Sartorelli, A. C., D. S. Fischer, and W. G. Downs. 1966. Use ofsarcoma 180/TG to prepare hyperimmune ascitic fluid in themouse. J. Immunol. 96:676-682.

32. Schnepf, H. E., K. T. Tomczak, J. P. Ortega, and H. R.Whiteley. 1990. Specificity-determining regions of a lepi-

J. BACTERIOL.

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

B. THURINGIENSIS SUBSP. THOMPSONI CRYSTAL PROTEIN GENES 557

dopteran-specific insecticidal protein produced by Bacillus thu-ringiensis. J. Biol. Chem. 265:20923-20930.

33. Schnepf, H. E., and H. R. Whiteley. 1985. Delineation of atoxin-encoding segment of a Bacillus thuringiensis crystal pro-

tein gene. J. Biol. Chem. 260:6273-6280.34. Sharpe, E., K. W. Nickerson, L. A. Bulla, Jr., and J. N.

Aronson. 1975. Separation of spores and parasporal crystals ofBacillus thuringiensis in gradients of certain X-ray contrastingagents. Appl. Microbiol. 30:1052-1053.

35. Sick, A., F. Gaertner, and A. Wong. 1990. Nucleotide sequenceof a coleopteran-active toxin gene from a new isolate of Bacillusthuringiensis subsp. tolworthi. Nucleic Acids Res. 18:1305.

36. Strathmann, M., B. A. Hamilton, C. A. Mayeda, M. I. Simon,E. M. Meyerowitz, and M. J. Palazzolo. 1991. Transposon-facilitated DNA sequencing. Proc. Natl. Acad. Sci. USA 88:1247-1250.

37. Thomas, W. E., and D. J. Ellar. 1983. Mechanism of action ofBacillus thuringiensis var israelensis insecticidal 8-endotoxin.FEBS Lett. 154:362-367.

38. Thomas, W. E., and D. J. Ellar. 1983. Bacillus thuringiensis var

israelensis crystal &-endotoxin: effects on insect and mammaliancells in vitro and in vivo. J. Cell Sci. 60:181-197.

39. Tinoco, I., P. N. Borer, B. Dengler, M. D. Levine, 0. C.Uhlenbeck, D. M. Crothers, and J. Gralla. 1973. Improvedestimation of secondary structure in ribonucleic acids. Nature(London) New Biol. 246:40-41.

40. Visser, B., E. Munsterman, A. Stoker, and W. G. Dirkse. 1990.A novel Bacillus thuringiensis gene encoding a Spodopteraexigua-specific crystal protein. J. Bacteriol. 172:6783-6788.

41. Waalwik, C., A. Dullemans, M. van Workum, and B. Visser.1985. Molecular cloning and the nucleotide sequence of the M,28,000 crystal protein gene of Bacillus thuringiensis subsp.israelensis. Nucleic Acids Res. 13:$R07-8217.

42. Widner, W. R., and H. R. Whiteley. 1989. Two highly relatedinsecticidal crystal proteins of Bacillus thuringiensis subsp.kurstaki possess different host range specificities. J. Bacteriol.171:965-974.

43. Wong, H. C., H. E. Schnepf, and H. R. Whiteley. 1983. Trfsn-scriptional and translational start sites for the Bacillus thurin-giensis crystal protein gene. J. Biol. Chem. 258:1960-1967.

VOL. 174, 1992

on June 14, 2018 by guesthttp://jb.asm

.org/D

ownloaded from