Vol. 176, No. 10JOURNAL OF BACrERIOLOGY, May 1994, p. 2835-28450021-9193/94/$04.00+0Copyright ©D 1994, American Society for Microbiology

Tn5401, a New Class II Transposable Elementfrom Bacillus thuringiensis

JAMES A. BAUM*Ecogen Inc., Langhome, Pennsylvania 19047-1810

Received 1 December 1993/Accepted 10 March 1994

A new class II (Tn3-like) transposable element, designated TnS401, was recovered from a sporulation-deficient variant of Bacillus thuringiensis subsp. morrisoni EG2158 following its insertion into a recombinantplasmid. Sequence analysis of the insert revealed a 4,837-bp transposon with two large open reading frames, inthe same orientation, encoding proteins of 36 kDa (306 residues) and 116 kDa (1,005 residues) and 53-bpterminal inverted repeats. The deduced amino acid sequence for the 36-kDa protein shows 24% sequenceidentity with the TnpI recombinase of the B. thunngiensis transposon Tn4430, a member of the phage integrasefamily of site-specific recombinases. The deduced amino acid sequence for the 116-kDa protein shows 42%sequence identity with the transposase of Tn3 but only 28% identity with the TnpA transposase of Tn4430. Twosmall open reading frames of unknown function, designated orfI (85 residues) and orJ2 (74 residues), were alsoidentified. Southern blot analysis indicated that TnS401, in contrast to Tn4430, is not commonly found amongdifferent subspecies of B. thuringiensis and is not typically associated with known insecticidal crystal proteingenes. Transposition was studied with B. thuringiensis by using plasmid pEG922, a temperature-sensitive shuttlevector containing TnS401. TnS401 transposed to both chromosomal and plasmid target sites but displayed anapparent preference for plasmid sites. Transposition was replicative and resulted in the generation of a 5-bpduplication at the target site. Transcriptional start sites within TnS401 were mapped by primer extensionanalysis. Two promoters, designated PL and PR direct the transcription of orfl-orJ2 and tnpI-tnpA, respectively,and are negatively regulated by Tnpl. Sequence comparison of the promoter regions of TnS401 and Tn4430suggests that the conserved sequence element ATGTCCRCTAAY mediates TnpI binding and cointegrateresolution. The same element is contained within the 53-bp terminal inverted repeats, thus accounting for theirunusual lengths and suggesting an additional role for TnpI in regulating TnS401 transposition.

The gram-positive bacterium Bacillus thunngiensis is wellknown for its insecticidal activity against the larvae of lepidopt-eran, dipteran, and coleopteran insects. This larvicidal activityis imparted primarily by a diverse family of parasporal crystalproteins, typically encoded on large plasmids, that displaydifferences in their spectrum of activity. The four major classesof insecticidal crystal proteins (ICPs), or 8-endotoxins, havebeen designated CryI, Cryll, CryIII, and CryIV, on the basis ofamino acid sequence homologies and insecticidal activity.These four classes of proteins, although highly divergent,contain conserved regions that suggest a common origin (20,23).

B. thuringiensis has also been a rich source of mobile geneticelements. Several insertion sequences, or IS elements, and aTn3-like transposon, Tn4430, have been identified and foundto be associated with ICP genes of either the cryI or crylV classbut not with genes of the cryII or cryIII class. The IS231 family,which displays homology with IS4 from Escherichia coli (27, 28,37), and IS232, which displays homology with IS21 fromPseudomonas aeruginosa (32), are closely associated with cryLAgenes on large plasmids in the B. thuringiensis subsp. thurin-giensis strains Berliner 1715 and HD2 and B. thuningiensissubsp. kurstaki strains HD1 and HD73 (21, 22, 25). TransposonTn4430 is also associated with cryL4 genes in these strains andis probably associated with cryI genes in other B. thuringiensisstrains as well (25). Another IS element, IS240, which displayshomology with the IS15 family (9), has been shown to flank the

* Mailing address: Ecogen Inc., 2005 Cabot Boulevard West, Lang-home, PA 19047-1810. Phone: (215) 757-1590. Fax: (215) 757-2956.Electronic mail address: jbaum@bionet.ig.com.

cryIVA gene of B. thuringiensis subsp. israelensis (7, 9). Asidefrom these associations, however, no evidence has been pro-vided to date that demonstrates a role for these elements in theevolution or mobility of ICP genes.Transposon Tn4430 is the only class II (Tn3-like [20a])

transposon to have been originally isolated from a Bacillusspecies (24, 25). This transposon contains genes encoding aresolvase or recombinase and a Tn3-like transposase that aretranscribed in the same direction, as is observed in the TnS01subclass of the Tn3 family (26, 36). The resolvase protein ofTn4430 is unique among the resolvases of Tn3-like transposonsin that it shows sequence similarity to the phage integrasefamily of site-specific recombinases and is unrelated to the Tn3resolvase-invertase family of site-specific recombinases (26). Inthis paper, I describe the isolation and characterization of anew Tn3-like transposable element indigenous to B. thuringien-sis subsp. morrisoni EG2158. This transposon, designatedTnS401, shows distant homology to Tn4430, shares a similarstructural organization with Tn4430, and encodes an integrase-like resolvase protein. Unlike Tn4430, however, Tn5401 is notcommonly found among B. thuningiensis subspecies and is notclosely associated with known ICP genes.

MATERIALS AND METHODS

Bacterial strains and plasmids. E. coli strains DH5a (Be-thesda Research Laboratories), GM2163 (New England Bio-labs, Inc.), and JM110 (Stratagene Corporation) were used ashost strains for molecular cloning experiments. Cloning vectorpTZ19u was purchased from U.S. Biochemical Corporation.Table 1 describes the B. thuringiensis strains and plasmids usedin this study. Strain EG7566 is a plasmid-free derivative of B.

2835

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

2836 BAUM

TABLE 1. B. thunngiensis strains and plasmids used in this study

Strain or plasmid Relevant features' Source orreference

StrainsEG7566 This studyEG10368 4.9 W. P. DonovanEG2424 130, 88, 60, 44, 43, 7.5, 5.4, 5.2, 5.0, 4.9, and 1.4 16EG2158 125, 105, 88, 72, and 35 11

PlasmidspEG491 pUC18/pE194ts replicon A.-M. MettuspEG853 B. thuningiensis-E. coli shuttle vector 4pEG854 B. thuringiensis-E. coli shuttle vector 4pEG911 cryIIIB2 subclone in pEG853 This studypEG922 Tn5401-based transposon vector This studypEG922 tnpAA Deletion of two internal HindIII fragments from tnpA This studypEG922 tnpI Frameshift mutation at the Alw44I site of tnpI This studyp76 pEG854 derivative containing NsiI fragment from TnS401 (nt 107-768) This studyp77 Seven-kilobase SstI fragment containing TnS401-tet, inserted into the SstI site of pTZ19u This studyp83 p76 derivative containing NspI fragment from Tn5401 (nt 109-926) with tet gene insertion This studypBC16 B. cereus plasmid with tet gene 6pEG640 pGEM-3Z containing cryIF 8

a Data given for strains EG7566, EG10368, EG2424, and EG2158 are sizes of the resident plasmid or plasmids in megadaltons.

thuringiensis subsp. kurstaki HD73. Strain EG10368, an HD73derivative containing a cryptic 4.9-MDa plasmid, was kindlyprovided by W. P. Donovan. B. thuringiensis subsp. morrisoniEG2158 is a naturally occurring strain that contains a cryIIL4insecticidal crystal protein gene on an 88-MDa plasmid (11).Plasmid pBR322::Tn4430 was kindly provided by D. Lereclus.Plasmid pEG491 was provided by A.-M. Mettus of Ecogen Inc.DNA manipulations and analyses. Standard recombinant

DNA procedures were performed essentially as described byManiatis et al. (30). The sequence of TnS401 was determinedon both strands from plasmid DNA by using the dideoxy chaintermination method (38). B. thuringiensis plasmids were re-

solved on vertical 0.52% agarose gels by using a modifiedEckhardt lysis procedure (18). B. thuringiensis strains were

transformed with the electroporation protocol described byMettus and Macaluso (33). Hybridization probes labeled with[ct-32P]dATP were prepared by the method of Feinburg andVogelstein (14).

Plasmid constructions. To create plasmid p76, a small NsiIfragment from TnS401 (nucleotides [nt] 107 to 768) was

inserted into the unique PstI site of cloning vector pEG854 (4).This fragment contains the tnpI promoter region, orfl, andorJ2. To create plasmid p83, a 2.5-kb NspI fragment frompEG922 was inserted into the unique SphI site of p76. The2.5-kb insert includes the 3' end of orf2, the tet gene of pBC16inserted into the ClaI site of orf2, orfi, the tnpI promoterregion, and the 5' end of tnpI (extending to the NspI site at nt926). To create plasmid pEG491, an -6.5-kb BamHI fragmentfrom pTV32ts (44), containing a chloramphenicol acetyltrans-ferase gene and the temperature-sensitive replication origin ofpE194ts (42), was inserted into the unique BamHI site ofpUC18. To create plasmid pEG911, a 4.3-kb XbaI-SspI frag-ment containing cryIIIB2 (12) was blunt ended with Klenowpolymerase and ligated into the unique SmaI site of shuttlevector pEG853 (4). Subsequently, a 293-bp Dral fragmentcontaining the transcription terminator of cryIF was isolatedfrom pEG640 (8) and inserted into a blunt-ended Asp718 site3' to cryIIIB2. For details on the creation of plasmids p77 andpEG922, see Results and Table 1. A deletion of two smallHindlIl fragments within the transposase gene (see Fig. 3)resulted in plasmid pEG922 tnpAA. This was accomplished by

digesting the transposon plasmid p77 with SstI and HindIII,purifying the two SstI-HindIII fragments containing tet, orfl,tnpI, the 5' end of tnpA, and the 3' end of tnpA, and ligatingthese into the SstI site of pEG491. To create plasmid pEG922tnpI, a frameshift mutation was introduced at the uniqueAlw44I site within the tnpI gene to generate the pEG922derivative pEG922 tnpI. This was accomplished by digestingthe transposon plasmid p77 with Alw44I, blunt ending it withKlenow polymerase, digesting it with SstI, purifying the twoAlw44I-SstI fragments, and ligating them together into the SstIsite of pEG491. Plasmid pEG922 tnpI tnpAA, a double mutant,was constructed by ligating the SstI-HindIII fragment frompEG922 tnpI (containing the tnpI frameshift mutation) withthe HindIII-SstI fragment from pEG922 tnpAA (containing the3' end of tnpA) into the SstI site of pEG491.RNA analyses. Total RNA was isolated from B. thuringiensis

DSM cultures (12) grown to mid-logarithmic phase (150 Klettunits, red filter) at 30°C. RNA was prepared from 5-ml aliquotsaccording to the method of Zuber et al. (45) and suspended in100 RI of sterile water. Relative RNA concentrations wereestimated from the intensity of the rRNA bands detected byethidium bromide staining after agarose gel electrophoresis ofglyoxyl-treated RNA (31). Primer extension reactions wereperformed with the Primer Extension System kit and proce-dures supplied by the Promega Corporation. Primer extensionproducts were resolved on 6% sequencing gels. The followingprimers were used for the RNA analyses: pr3, 5'-ATTCGAAAGGTTCCAATC-3' (nt 195 to 212); pr5, 5'-TCTTCTTGACCGTTGAAC-3' (nt 501 to 518); prlO, 5'-CTTCTFGAGATAAGCTAG-3' (nt 839 to 822); prl4, 5'-TGCATAAATACCTCCTCT-3' (nt 768 to 751); prlS,5'-GCCATGTGTTACACCTAC-3' (nt 604 to 621); prl6, 5'-CTGCTCTGATGTTGGAGTGTCAC-3' (nt 1830 to 1808); and pr9llBamR3, 5'-GTACATlTGTAGGATCAG-3' (nt 1972 to 1955). Primers wereprepared by using an Applied Biosystems 380B DNA synthe-sizer or were obtained from Lofstrand Laboratories Ltd.(Gaithersburg, Md.) or Integrated DNA Technologies, Inc.(Coralville, Iowa). Sequencing reactions using the same prim-ers were performed to provide standards. Primer prlO, used forthe mapping of the PR transcriptional start site, is complemen-tary only to RNA transcribed from the NspI insert in p83 and

J. BACTERIOL.

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

B. THURINGIENSIS CLASS II TRANSPOSABLE ELEMENT Tn5401 2837

is not complementary to RNA transcribed from the NsiI insertin p76 and p83.Computer analyses. Data base searches via on-line servers

were performed with the programs FASTDB and BLASTP (1)provided by Intelligenetics Inc. (Mountain View, Calif.) andthe National Center for Biotechnology Information (Bethesda,Md.), respectively. FASTDB searches were run with thePAM150 matrix, while the BLASTP searches were run with theBLOSUM62, PAM40, PAM120, and PAM250 matrices. Themultiple sequence alignment program -MACAW (39) waskindly provided by the National Center for BiotechnologyInformation. Other computer-assisted analyses of nucleotideand amino acid sequences were performed with programs(CLUSTAL, PALIGN, NALIGN, REPEATS, COD_FICK)found in the Intelligenetics PC/GENE sequence analysis pack-age.

RESULTS

Isolation of TnS401. Transposon TnS401 was initially recov-ered as a spontaneous insertion within the recombinant plas-mid pEG911 (Fig. 1), a derivative of shuttle vector pEG853 (4)containing the cryIIIB2 gene (12), after its introduction byelectroporation into a sporulation-deficient (Spo -) variant ofB. thuringiensis subsp. morrisoni EG2158. Restriction enzymeanalysis of plasmid DNAs extracted from the Spo- EG2158transformants indicated that a subpopulation of the pEG911molecules in each transformant contained an -5-kb insertion.Subsequently, two transformation experiments with strainEG2158 yielded similar pEG911 derivatives. Plasmid DNAsfrom three independent transformants were used in transfor-mation experiments to recover the pEG911 derivatives in E.coli DH5a. Restriction enzyme analysis of plasmid DNAsisolated from individual E. coli transformants was used toidentify pEG911 clones containing the cryptic insertion. Threeindependent pEG911 clones were recovered and designatedpEG911-1, pEG911-2, and pEG911-3. Subsequent restrictionenzyme analysis indicated that all three plasmids contained acommon insertion element and that the sites of integrationmapped close together within a region located -1.8 to 2.0 kbupstream of the ciyIIIB2 coding region near the multiplecloning site (Fig. 1A). The orientation of the pEG911-3 insertis opposite that of the pEG911-1 and pEG911-2 inserts.DNA sequence analysis of plasmids pEG911-1, pEG911-2,

and pEG911-3 revealed that the insertion element had inte-grated at three different sites within a 175-bp region of plasmidpEG911 (Fig. 1B). As is typical for a Tn3-like transposon, a5-bp A+T-rich sequence was duplicated at the site of integra-tion: AGAAT for pEG911-1, AAT1T for pEG911-2, andAATTA for pEG911-3. The pEG911-3 target site is only 64 bpfrom the SmaI-XbaI junction of pEG911 (Fig. 1A), while themost distant pEG911-1 target site is 234 bp from the junction.Interestingly, a pEG911 derivative lacking the pTZ19u region(removed by NotI digestion and religation of the vectorfragment) readily yielded transposition events at distant siteson the plasmid after introduction into the EG2158 back-ground. The approximate locations of these integration sitesare indicated by the asterisks in Fig. 1A. These results suggestthat the transposition hot spot defined by pEG911-1,pEG911-2, and pEG911-3 is not dependent on a particularnucleotide sequence but is dependent on some other charac-teristic of the plasmid (e.g., DNA topology).

Structural organization of Tn5401. Sequence analysis of theinsertion element in plasmid pEG911-1 revealed a new trans-poson of 4,837 nt (Fig. 2). A structural map of the transposon,designated Tn5401, is shown in Fig. 3. Tn5401 contains four

A

N

pTZl9u cat ori6O cr

pEG911 -1

pEG911-2

*

Bs C

Bs C

C BspEG911-3

BGGATCCTAGG CCCCTAGTA CCTGGAAAGC AGATGGGGGA TATTGCTTTG AAAAAGAGGA 60

SmaI/XbaI

TGTAAAAAAG CTAGAGGA&AITACAAACA TTGTTTAACA TTAAAAGAAG CTGCTGAATT 120pEG911-3

TCTAAACAAA TCTAAGACAT ACGTCTATAA TGCAGCTAAA GATGGGATAT TGCCTTrrAA 180

AGAGATTGCT AAAGGGAAAT CTACTGAACG ATTGTACTTA AAAAGTGATT TAG=rr 240pEG911-2

CAAAGAAAgAATMAAAATA GGTCAAAGGA GGAGTCGAAA GCAAAAAAAG CAGCATTTAA 300pEG911-1

FIG. 1. (A) Structural map of plasmid pEG911 and derivatives(pEG911-1, -2, -3) containing insertions of Tn5401 upstream of thecryIIIB2 ICP gene. The orientation of TnS401 in pEG911-3 is oppositethat of TnS401 in pEG911-1 and pEG911-2. Restriction endonucleasesites: A,Asp718; B, BamHI; Bl, BlnI; Bs, BssHII; C, ClaI; D, DraI; N,NotI; Sf, Sfil; Sm, SmaI; Ss, SspI; St, SstI; X, XhoI, and Xb, XbaI. Otherdesignations: open arrow, cryIIIB2; solid box, cryI transcription termi-nator region; shaded box, oM6O replication origin region; open box, E.coli phagemid vector pTZ19u; darkly stippled boxes, Tn5401 inserts.(B) Tn5401 insertion sites. The sites of Tn5401 insertion in plasmidspEG911-1, pEG911-2, and pEG911-3 were determined by DNAsequence analysis. The 5-bp target sites duplicated upon Tn5401insertion are underlined. The SmaI-XbaI (Sm/Xb) junction shown inpanel A is doubly underlined.

potential coding regions and 53-bp inverted repeats (IRs) atthe termini. The A+T content of the transposon is 63.7%. Thetwo largest open reading frames, encoding putative proteins of306 and 1,005 amino acids, are oriented in the same direction.These genes have been designated tnpI and tnpA, respectively(described below). A third open reading frame, designatedorfi, encodes a putative protein of only 85 amino acids and isoriented opposite to tnpI and tnpA. A 74-amino-acid openreading frame, designated orf2, is located immediately down-stream of orfl. The tandem organization of tnpI and tnpA onTnS401 resembles that of the TnS01 subclass of Tn3-liketransposons, which includes the Enterococcus faecalis transpo-son Tn917 (36).Terminal IRs. The 53-bp perfect IRs at the ends of TnS401

are large for a Tn3-like transposon, which typically containsterminal IRs of only 38 to 40 bp (19). The outermost 38 bp ofthe TnS401 IRs share similarity with the terminal IRs of theTn3 transposon family, while the remaining 15 bp are unique.The TnS401 IRs show the greatest similarity to the IRs of Tn3

VOL. 176, 1994

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

2838 BAUM

GUGGTATGTACATlGGAACAGATCCCAACAAGCATTAGGAATTTCGCAC

ACAAAAzAAGGAAGGTTCTGTTCAAGACTTTCTTAATCATGTTGCTT

ATTTATAGATGTCACCACGATTTCCAATTGCTTGTATGTATATGACTCTATCATGATK Y I D G R N G I A Q I Y I V K E D H N

TTATTTCAAATAAATTCGAAAGGTTCCAATCCGTAATCGATATAGTTCTGTGTAACCTTI E F L I R F T G I R L R Y L E T Y G K

TCATACTTTTAATATCTCCTTCAGGAGGAATCTTAAGAAGTCCCTTCAATCCTTCTGCAAM S K I D G E P P I K L L G K L G E A I

TTCTTTTTTGAATCCCTTTTTCTTGCTTTGCAATAAATTTCACCGCGGACTTATGGTAAAR K Q I G K E Q K A I F K N < orf2 P L

TCAATTTGTAGTCCGAATTCACGTTTTGCGTCCTCCCCTGATACATATCCTTCTTCACT;D I Q L G F E R K A D E G S V Y G E E S

TTTAACTGTTCTAACTCTTGTGTAGACAGCGGTTCATGATCAGGATCTGCCATATCAAT;N L Q E L E Q T S L P E H D P D A M D I

TTTTCCCATTCTTTAGGTTTTCTTCTTGACCGTTGAACAAGAAATTCTAAAAAGTCAAATK E W E K P K R R S R Q V L F E L F D F

GCTGCTTTTTCATCTTGTTGATCCAGGTGATCAATTAACCGATACAATTCATCTTTACGAA A K E D Q Q D L H D I L R Y L E D K R

ATAGCCATGTGTTACACCTACTTTCGAGATAGTTTTAAATGTCCACT ITATAGTI A M < orf 1 ------>-----

-35 -10

GGACATGAAGTGTGGGAAAATAAATGT=&CCGCTAACATAATTGArTMgAGATVa______ -----------> <_*

PLPR

ATATCATGTCCGCTAATGTAAGTCIAATAAaAAAGGAGG.TATTTATGCATTCCACTAAAAC---------->> tnpI> M H S T K T

-10 -35

AATTTCTATACAAGCAACATCTTTGATTCCGATTTTATTTCTAGCTTATCTCAAGAAGGI S I Q A T S L I S D F I S S L S Q E G

AGATTGCA;ACAAAAAcACTAAAGAAAGGGTTAAGTTGATD L H T K T L K E Y T S D L K D F V F W

GTTGAAAATGTGTGGGGAA.AACATGCTGAGGATACTCTTTTTCATCCAATAGAAGTTACF E N V W G K H A E D T L F H P I E V T

CGCTCGCACTATTGCTCGATAT_AGGTAAACCTTCA R T I A R Y R G H N Q V T R L L K P S

TACGATTAA;CGGCGCATTAATCAAT CG ATTAT TTCAAAT I N R R I N S I K R Y F D W A K Q E G

ACTGGTACAAACAAATTATTCAAAATCAATTAAGTTTGTACCAACAGAAAAAACGAGTCCL V Q T N Y S K S I K P V P T E K T S P

CAAACGCATGTCAGATAAAGAAGAAGCCGCTTTAATGCATGCCGTTGAAAAATACGGCACK R M S D K E E A A L N H A V E K Y G T

ACTACGTGACAGGGCAATGATTATTTTTATGCTTCATACTGGCCTTCGTTCAATGGAAGTL R D R A M I I F N L H T G L R S M E V

GTGTGATGTTCAAATAGAGGATGTTATCATGAGAAAAAGAGGCGGCTATGTTGTTGTTCGC D V Q I E D V I M R K R G G Y V V V R

ATCTGGAAAACGAAATAAACAGAGGGAAGTGCCTTTGAATAGTACAGCTCGTTGTGCACTS G K R N K Q R E V P L N S T A R C A L

AGAAGAACATATCAGATTAAGTGAGATTTCACAGAGTTATTTGTTTCCTTCTTCTAAAACE E H I R L S E I S Q S Y L F P S S K T

AGGAAAACGCCTACAAGAAAGAGCGATCCGCCATATTCTTCAGAAGTATTAACTTGCG K R L Q E R A I R H I L Q K Y I R L A

AAAGTTAGAAGGATTTAGTGCCCATGATTTAAGGCATCGCTTTGGTTATGTGATGGCTGAK L E G F S A H D L R H R F G Y V N A E

ACGCACACCATTACATCGTCTTGCACAAATTATGGGACACGATAACTTGAATACCACGATR T P L H R L A Q I N G H D N L N T T M

GATTTATGTAAGAGCTA ACA CAG A TAGAAAAGATTGCCTGGAAI Y V R A T Q E D L Q G E V E K I A W N

CTAAAGAATGCACATTATCCTACTCATTTGGTCATGTGATACAAAATAAGAATTGTAAC=

GGAGGAACAAGGGTTATGCCTGTAGATTTTTAACACCTGAACAAGAAGAAAAATATGGTtnpA > M P V D P L T P E Q E E K Y G

TGTTTTTGTGACACTCCAACATCAG AGCAT TTAGATGATACAC F C D T P T S E Q L A K Y F W L D D T

GACAAAGAACTGATATGGAATCGTCGTGGAGAGCATAATCAACTTGGTTTCGCTGTTCAAD K E L I W N R R G E H N Q L G F A V Q

TTAGGAACCGTTAGGTTCTTAGGAACATTTTTATCTGATCCTACAAATGTACCACAATCGL G T V R F L G T F L S D P T N V P Q S

GTTATTACATATATGGCAAATCAACTTCATCTAGATGCCAAAGCTTITCTCGTTATGGAV I T Y M A N Q L H L D A Q S F S R Y R

60

120

180

240

300

360

420

480

540

600

660

720

780

840

900

960

1020

1080

1140

1200

1260

1320

1380

1440

1500

1560

1620

1680

1740

1800

1860

1920

1980

2040

AATAAACGAAGTCAGTGGGATCAAATGCAAGAGATACGTTCTGTTTATGGATATAAAAACN K R S Q W D Q M Q E I R S V Y G Y K N

TTTACAGATAAATCAACACATTGGCGATTCATCAGATGGCTATATGCA5TGCTTGGCTAF T D E S T H W R F I R W L Y A R A W L

TATAATGAACGGCCAAGTGT AT T G A A CAAAAAY N E R P S V L F D L A T A R C I E Q K

ATTTTACTACCTGGTGTATCTGTATTAACAAGGCCAGTATCAACGGTTcGTGATCGTTCAI L L P G V S V L T R L V S T V R D R S

GCAGAAAATATATGGAAAAAGCTCTCTAGTCTTCCGGATAATGTTCAGAAAAAACAATTAA E N I W K K L S S L P D N V Q EEX Q L

GAAAACCTTCTTCAGATAGATCAAAAAACAAAGAAAACGTATTTAGAGCGTCTAAGTAATE N L L Q I D Q K T K K T Y L E R L S N

CCCCCTGTTCCGATTAGTGTTACGGGCATTAAGAATACGCTGATTCGTTTACAAGAGCTTP P V P I S V T G I K N T L I R L Q E L

CGTCAATTGAACACTGAAAATTGGGATATGTCTAGAATTCCTTCGAAAA ACAACAAR Q L N T E N W D N S R I P S K R L Q Q

TTCGCGCGTCACACAGTCGCTGTTAGATCACAAGCAATTGCTAGAATGCCCGATCAACGAF A R H T V A V R S Q A I A R N P D Q R

CGTATGGCTATGTTAGTTGCATTTGCTAAAATGTATACACAAAGTGCTCAGGATGATGTCR M A M L V A F A K M Y T Q S A Q D D V

ATTGATATTTTTGATAGATATTTAACAGATTTATTTGCTAACAAI D I F D R Y L T D L F A K T Y R E E Q

AAAGAACGTCTTCGTACAATTAAGGATTTAGATAAGGCAGCGCGCCAATTACGGGAAGCTK E R L R T I K D L D K A A R Q L R E A

TGTGTAATATTATTAGAACATACGGATCCTTCTGTCCATCCAAAACGGCAGTGTTTGAAC V I L L E H T D P S V H P K T A V F E

AAAATTTCAGAAAAGGATTTAATACAAGCTGTCCAAATTGTTGATTCACTCACCTATTCAK I S E K D L I Q A V Q I V D S L T Y S

CCAAATCAAACACTAGCCTATTCAGGATTGTTACAACATTATGGCATAATCCGAAAATTTP N Q T L A Y S G L L Q H Y G I I R K F

CTTCCTTTACTCATGGAAGAAATTGAATTACAAGCAACGCCTGCTGGATTACCCATCTTGL P L L N E E I E L Q A T P A G L P I L

CAAGCATGGAATTTTGTAAAAGAGCATGGGAAATCCAATAAGAAAAGATGGAAAAATGCTQ A W N F V K E H G K S N K K R W K N A

CCTCTTGCCGGTTTGAATGCAAATTGGTCTAAGGTTGTAATTGATAAAGATTCCGGAACTP L A G L N A N W S K V V I D K D S G T

GTAAATCATCGAGCATATACGTTTTGGATGCTCGAACAAGTATTAGAAGCTTTGCACCGAV N H R A Y T F W M L E Q V L E A L H R

CATGATCTATATATAGTAGGAAGTGAAAAATATGGGGACCTTCGCGCACAATTATTACAAH D L Y I V G S E K Y G D L R A Q L L Q

GACGAAGAATGGAAAAGTATTCGTCCTAGTATTCTTCGCTCATTAGACTGGTCAATAGATD E E W K S I R P S I L R S L D W S I D

TCTTATGAATCATTGACACCGTTAAAAGAAGAGTTAGACAAAACTTATCATCAAGTCATTS Y E S L T P L K E E L D E T Y H Q V I

GAGAATGGGAGAATAATCCTGCGGTGCAAATAGACAC T AAE N W E N N P A V Q I D T F A G E E R I

GTTTTGACACCTTTAGACAAACAACCAGAACCTGAATCACTACAAAAACTAAAACAACAAV L T P L D K Q P E P E S L Q K L K Q Q

ATACATACGATGTTGCCAAATATAGATATTCCTCAATTATTACTCGAAGTAAATCGTTGGI H T M L P N I D I P Q L L L E V N R W

ACGGGATTTATGGATGGTTTTCGACATATAGTGAGGCAT G F M D G F R H I S E A K S R I N E L

CCTATAAGTATCTGTGCATTGCTTATATCTCAAGCATGCAATATTGGGTTAAGACCTTTAP I S I C A L L I S Q A C N I G L R P L

GTTCAAGATGGGGGTCCTTCATTAGAACGTGATCGTCTTACATGGATTGAACAAAATTATV Q D G V P S L E R D R L T W I E Q N Y

TTTCGTGCAGAAACACTTTCAGAATCAAACGCGAAACTTGTAGATTTTCATAGCCAATTAF R A E T L S E S N A K L V D F H S Q L

CAGCTGGCTAAAATGTGGGGTGGTGGAGAAATTGCTTCAGCTGATGGATTACGTTTCATCQ L A K M W G G G E I A S A D G L R F I

ACACCAGTAAAATCCGTACACACTGGTCCAAATCCTAAATATTTCGGTTCTGGTCGTGGTT P V K S V H T G P N P K Y F G S G R G

GTTACGTATTACAACTATACGAGCGATCAATTTACCGGAATCCACGGTTGGTGATTCCAV T Y Y N Y T S D Q F T G L H G L V I P

GGCACAATTCGTGATTCATTATACTTACTTCAATGTGTGTTAGAACAAAATACGAACTTAG T I R D S L Y L L Q C V L E Q N T N L

CAGCCAAAAGAAATTATGACAGATAQ P K E I N T D T A G Y S D I I F G L F

GGATTATTAGGATATCAATTTAGTCCTCGTACGTT_TATAG_T_G L L G Y Q F S P R L A D I S E S R L W

CGTTTTGATGCGAACTCAGATTATAGCATTGTR F D A N S D Y S M L N N L S K S R I R

FIG. 2.

(23 of 38 identical matches, no gaps) and the least similarity to Transposase gene. The tnpA gene of TnS401 extends from ntthe rightward IR of Tn917 (15 of 38 matches). The IRs of 1756 to nt 4770 and encodes a protein of 1,005 amino acids,Tn4430 and Tn2l show an intermediate level of similarity (17 with ATG as the likely start codon. The ribosomal binding siteof 38 to 19 of 38 matches). AGGAGGa is positioned 9 nt 5' to the start codon. The

J. BACTERIOL.

2100

2160

2220

2280

2340

2400

2460

2520

2580

2640

2700

2760

2820

2880

2940

3000

3060

3120

3180

3240

3300

3360

3420

3480

3540

3600

3660

3720

3780

3840

3900

3960

4020

4080

4140

4200

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

B. THURINGIENSIS CLASS II TRANSPOSABLE ELEMENT Tn5401 2839

GAAGAACTCATACATCGTCATTGGGAAGACATGCTTCGTGTTGCGGGATCTTTGAAACTA 4260E E L I H R H W E D M L R V A G S L K L

AATAAAATAAATGCAACACATCTTATCCAAGCACTTCAGTATAATGGGAAACCAACTATG 4320N K I N A T H L I Q A L Q Y N G K P T N

TTAGGGCGAGCAATTGGAGAATTGGGGAGACTCT8AAAACAG_ 4380L G R A I G E L G R L F K T R Y L L L Y

TTACATGATGAAAATTATCGTCGTAAATTTTAAATCAA 4440L H D E N Y R R K I L N Q L N R G E A R

CATAGTTTAGCGAGGGCTGTATTTTACGGCAAACGA CGA 4500H S L A R A V F Y G K R G E L H Q S Y R

GAAGGACAAGAAGAGCAATTAGGTGCATTGGGTTTAGTAGTAAATGC AATGG 4560E G Q E E Q L G A L G L V V N A I I V W

AATACACGATATATAGAATCTGCGTTACAAGTACTCCGAAATCGCGGTCATACAATTGAT 4620N T R Y I E S A L Q V L R N R G H T I D

AATGATGATATATCTAGACTTTCACCATTAGGCCATAAACACATTAACATAGTAGGTCGG 4680N D D I S R L S P L G H K H I N I V G R

TATTCATTTGTTCTCCCAGAAGAAGTAAAAGATGGGCAATTACGTACACTAACATATGAA 4740Y S F V L P E E V E D G Q L R T L T Y E

GAAACAAACAAAAAGGAACCTGATTCTTTATAAGAATAGGTTCCTAATGTCCGCTAATGC 4800E T N K K E P D S L

TTGTTGCGTGATTTTGTTCCATTGCTACACATACCCC 4837

TABLE 2. Multiple sequence alignment of transposasesa

% Amino acid sequence identity of':Transposase

TnS401 Tn3 Tn4556 Tn5O1 Tn2J Tn4430 Tn917

Tn5401 42.0 28.6 27.5 26.4 27.7 25.1Tn3 42.0 31.9 29.9 29.1 27.6 26.0Tn4556 28.6 31.9 22.9 22.1 22.0 20.3Tn501 27.5 29.9 22.9 68.1 36.2 32.9Tn2l 26.4 29.1 22.1 68.1 36.8 32.0Tn4430 27.7 27.6 22.0 36.2 36.8 40.4Tn9J7 25.1 26.0 20.3 32.9 32.0 40.4

a Sequences were aligned by the CLUSTAL multiple sequence alignmentprogram (Parameters: K-tuple value, 1; gap penalty, 1; window size, 10; filteringlevel, 2.5; open gap cost, 10; unit gap cost, 10; matrix, MDM-78).

b Percentages of amino acid sequence identity were estimated from pairwisesimilarity scores and a consensus sequence length of 1,049 residues.

FIG. 2. Nucleotide sequence of Tn5401 and deduced coding re-gions. Tn5401 contains 53-bp terminal IRs (solid arrow) and four openreading frames. The deduced ribosomal binding sites for orfl, tnpI, andtnpA are doubly underlined. The conserved sequence element ATGTCCRCTAAY is indicated by the dashed arrows. Transcriptionalstart sites corresponding to the promoters PL and PR are indicated bythe asterisks. The -10 and -35 positions for PL and PR are indicatedby the colons, and the proposed - 10 and - 35 regions are underlined.The GenBank accession number for this sequence is U03554.

deduced amino acid sequence displays significant sequencesimilarity with the transposases of the Tn3 family. The resultsof the pairwise sequence alignments with representative trans-posase proteins from the Tn3 family are shown in Table 2. Thehierarchy of sequence similarity resembles that observed forthe terminal IRs, with the TnpA protein of TnS401 sharing thegreatest similarity with the transposase of Tn3 and the leastsimilarity with the transposase of Tn917. Surprisingly, thetransposases of Tn4430 and Tn5401 show only -28% sequenceidentity.

Resolvase gene. The tnpI gene of TnS401 extends from nt764 to nt 1681 and encodes a protein of 306 amino acids, withATG as the likely start codon. A ribosomal binding site,AGGAGGT, is positioned 4 nt 5' to the start codon. Thededuced amino acid sequence displays 24% sequence identitywith the TnpI protein of Tn4430 (PALIGN parameters: uni-tary matrix, open gap cost of 1, unit gap cost of 3) and nosignificant homology to the resolvase proteins of other Tn3-like transposons. Unlike these resolvase proteins, the TnpIprotein of Tn4430 shows sequence similarity to the phageintegrase family of site-specific recombinases (26). A multiplesequence alignment of the TnpI proteins of Tn4430 andTn5401 with other integrase family proteins is shown in Fig. 4.The amino acid blocks of regions A and B in the carboxyl half

orf2 tnpl tnpA

FIG. 3. Structural map of TnS401. The four open reading framesare designated by arrows, and the terminal IRs are designated by solidarrowheads. Restriction sites used in the subcloning and manipulationof TnS401 are shown.

of the proteins correspond to those identified by Argos et al.(2) as being conserved in the integrase protein family. Specif-ically, residues His-254, Arg-257, and Tyr-288 are highlyconserved among the integrase family proteins. Presumably,Tyr-288 forms a 3' phosphodiester bond to the DNA duringstrand exchange (see reference 41 and references therein).

Small open reading frames. Two small open reading frames,encoding putative proteins of only 85 and 74 amino acids, areoriented in tandem opposite to tnpI and tnpA. Both openreading frames are predicted to be coding regions on the basisof the Fickett method (15). No other predicted coding regionswere found in the six reading frames. Searches of the Swiss-Prot, PIR, and GenPept data bases with the FASTDB andBLASTP programs failed to identify protein sequences withsignificant sequence similarity to orfi or orJ2. The 85-amino-

lambda int

Tn4430

Tn5401

Tn916

Tn554 tnpA

Arm. _r-N_M~

Bm

UZZZZZIIZ11 IZZLJ LIIIIZ. E-UJ 35611 EEB=====ff=j[ 284

11 EE=== :l= 306

II 11II II [III []1>1 40511 T-T7F7} {§ EI = [=[I= 361

A RLAMELA LC WS IVd--- *YIATLLAY I ALS K Fnl--- RAMIIFML S C I Imtysa DEILILL Is FG L Ldf--- ^LILMLMY I L IVt

D_

j 5 lsarlyekqisdkfaqh-L SDT SQ- rdd___ d /||FFCTNAIEKGFSIHEVANQ SNIH LL- tnp--- n _ RFgyvmaert-PLHRLAQI NLN I- ralp 5 S _TFCTNYANAGMNPKALQYI I LN- ah___ a ||lHATQLIREGWDVAFVQKR HVQLNQt vh

FIG. 4. MACAW multiple sequence alignment of the TnS401recombinase protein (TnpI) with several recombinases from the phageintegrase family, including the Int protein of bacteriophage lambda,the TnpI protein of Tn4430, the Int protein of Tn9O6, and the TnpAprotein of TnS54. In the schematic alignment, the large boxes corre-spond to blocks of sequence similarity and the vertical lines withinboxes indicate conserved amino acid residues. Below the schematicalignment, amino acid residues within the sequence similarity blocks ofregions A and B are indicated in capital letters and conserved residues,defined by the PAM120 matrix, are highlighted in black.

VOL. 176, 1994

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

2840 BAUM

tnpI wtS I I------

tnpI w tS I . 1

Pstl aSstEcoRI

~~~~~~~~IRrep

pEG922 tnpA

BssHII

cat ^, IR

Smal

Kpn teNspSstl /\ Np

Nsp /AclC\(Clal/Accl/(AcCa)

FIG. 5. Structural map of the transposon vector pEG922. Therelevant coding regions are indicated by the wide arrows or boxes. Theterminal IRs of Tn5401 are indicated by the wide arrowheads. TnS401-tet contains a tetracycline resistance (tet) gene inserted into the ClaIsite of orf2 (Fig. 3) and is flanked by SstI sites. Restriction sites forBamHI, EcoRI, HindIII, SphlI, and XbaI within Tn5401-tet are notshown. Open reading frames orfl and orf2 are also not shown.

acid orf] extends from nt 608 to nt 354, with ATG as theputative start codon and the sequence GAAAGtAGGT as thelikely ribosomal binding site positioned 7 nt 5' to the startcodon. The 74-amino-acid o-f2 extends from nt 343 to nt 122,immediately following orf1, with GTG as the putative startcodon. No obvious ribosomal binding site could be detectedimmediately 5' to the start codon of orJ2.

Transposition in B. thuringiensis. In order to study thetransposition of Tn5401 in B. thuringiensis, an antibiotic resis-tance gene was inserted into Tn5401 and the tagged transpo-son was subsequently inserted into a temperature-sensitiveshuttle vector to generate a transposon vector akin to theTn917-based transposon vector pTV32's (44). To accomplishthis, the 3-kb BssHII-SstI fragment from plasmid pEG911-1and the 2.3-kb BssHII-SstI fragment from pEG911-3 (Fig. 1)were cloned together into the SstI site of the E. coli cloningvector pTZ19u, thereby reconstructing the entire transposonwith flanking SstI sites. The tetracycline (tet) resistance genefrom the Bacillus cereus plasmid pBC16, contained on a 1.7-kbAccI fragment, was then inserted into the unique ClaI site ofTnS401 (Fig. 3). The resulting plasmid was designated p77, andthe tagged transposon was designated TnS401-tet. The tet geneinsertion in p77 disrupts orJ2, but this potential coding regionis not essential for transposition (see below). The transposoncontaining tet was subsequently recovered from p77 on a 7-kbSstI fragment and inserted into the unique SstI site of thetemperature-sensitive plasmid pEG491 to yield the transposonvector pEG922, shown in Fig. 5. Plasmid pEG491 is a pUC18derivative that contains a chloramphenicol acetyltransferase(cat) gene and the replicon of pE194ts (42).To demonstrate transposition in B. thuringiensis, plasmid

pEG922 was introduced into the plasmid-free B. thuningiensisEG7566 by electroporation. Transposition frequencies were

measured by the procedure described by Youngman et al. for

7 kb-

A BSstI SstI

tet : : tetkb.....

.....

n5401 T 50

cat

probe A 1probe B * ElFIG. 6. Southern blot analysis of SstI-digested genomic DNA from

strain EG10368 isolates containing chromosomal insertions of Tn5401-tet (wt) or a Tn5401-tet tnpI mutant (tnpI). DNAs from three isolateseach were probed successively with 32P-labeled pEG491 (probe A andpanel A) and a 658-bp NsiI fragment derived from TnS401 (probe Band panel B). S, pEG922 DNA digested with SstI. The schematicdrawing shows the cointegrate structure expected from transposition ofTn5401-tet tnpI. The dashed line denotes chromosomal DNA, and theshaded boxes indicate the tetracycline resistance gene (tet) insertedinto Tn5401 (open boxes).

Tn917 (43, 44). Luria broth cultures of EG7566/pEG922 were

grown to mid-logarithmic phase (150 Klett units), and cellswere plated for single colonies on Luria broth plates plus 3 ,ugof chloramphenicol per ml at 30°C and on Luria broth platesplus 20 pg of tetracycline per ml at 41°C. Tetracycline-resistantcolonies that grow at 41°C should contain only a transposedcopy of Tn5401-tet. This was confirmed in three ways. (i) Thetetracycline-resistant colonies recovered at 41°C were typically(>95%) chloramphenicol sensitive, as would be expected forcolonies containing only a transposed copy of TnS401-tet. (ii)Southern blot analysis of chromosomal DNA isolated fromputative Tn5401-tet integrants showed transposition ofTnS401-tet but not of pEG922 vector sequences (Fig. 6). (iii)Direct chromosomal sequencing of individual TnS401-tet inte-grants showed that the transposon had integrated into thechromosome and was no longer flanked by pEG922 vectorsequences (29). By this method, transposition frequencies of 2X 10-3 to 2 x 10-4 were obtained in strain EG7566 (Table3), similar to the 10 transposition frequencies reported forthe Tn917 transposon vectors (43).

Transposition was also studied with B. thuringiensis subsp.kurstaki EG2424, a strain harboring 11 resident plasmids,several of which encode insecticidal crystal proteins. Transpo-sition frequencies were much higher in strain EG2424 than inthe plasmid-free strain EG7566, ranging from 8 x 10 - 3 to 2 x10' (Table 3). Strain EG2424 derivatives containing an

integrated copy of Tn5401-tet were analyzed by agarose gelelectrophoresis to determine if Tn5401 exhibits a preferencefor plasmid target sites. Of the 34 isolates examined, 33 showedtransposition of Tn5401-tet to large resident plasmids, primar-ily the 43-, 44-, and 60-MDa plasmids of strain EG2424 (3). Inseveral isolates, apparent multiple transposition events involv-ing large plasmids were detected. These results suggest thatTnS401 exhibits a strong preference for plasmid target sitesand that this preference accounts for the higher transposition

9.2 kb-

J. BACTERIOL.

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

B. THURINGIENSIS CLASS II TRANSPOSABLE ELEMENT Tn5401 2841

TABLE 3. Transposition of Tn5401 in B. thuringiensis

Strain Transposon Expt Transpositionplasmid frequency'

EG7566 pEG922 1 5 x 10-42 x 10-3

2 2x10-44 x 10-4

EG10368 pEG922 1 2 x 10-4EG10368 pEG922 tnpAA 1 3 x 10-6

2 x 10-52 1 x 10-5

1 x 10-5EG10368 pEG922 tnplb 1 7 x 10-6

2 4x10-64 x 10-5

EG2424 pEG922 1 8 x 10-35 x 10-2

2 1 x 10-2 x 10-1

Transposition frequency was calculated as the number of tetracycline-resistant colonies at 41'C/the number of chloramphenicol-resistant colonies at30°C. Most experiments involved measurements of transposition frequency induplicate cultures.

b For pEG922 tnpI, transposition frequency was calculated as the number oftetracycline-resistant colonies at 41'C/the number of tetracycline-resistant colo-nies at 30°C.

frequencies observed in strain EG2424. Tn3 also appears totranspose at a higher frequency into plasmids, but Tn3-likeelements from gram-positive bacteria typically do not (seereferences 19, 36, and 40 and references therein).

Effect of mutations on transposition. Mutations were intro-duced into the tnpI and tnpA genes of pEG922 in order toassess their effect on TnS401 transposition (see Materials andMethods). Transposition frequencies obtained with pEG922tnpAA were 10- to 1,000-fold lower than those obtained withpEG922, indicating that disruption of the tnpA gene doesimpair transposition (Table 3). In this case, the tetracycline-resistant colonies recovered at 41°C were also chloramphenicolresistant. Since the tnpI gene encodes the recombinase thatresolves the cointegrates formed during transposition (seebelow) and since this gene is presumably functional in pEG922tnpAA, replicative transposition of TnS401 tnpAA should resultin loss of the chloramphenicol resistance gene cat. The reten-tion of the cat gene therefore suggests that the low frequencyof integration obtained with pEG922 tnpAA either involves analternative pathway for transposition or is merely the result ofectopic recombination events.The effect of the tnpI mutation on the frequency of Tn5401

transposition could not be determined precisely because thepEG922 derivative harboring this mutation (pEG922 tnpI)shows an apparent 40-fold-lower copy number than pEG922(described below) and appears to be unstable under selectionfor chloramphenicol resistance. Nevertheless, transpositionfrequencies of 4 x 10-6 to 4 x 10' were obtained byselecting for tetracycline-resistant colonies at both 30 and 41°C(Table 3), a range of frequencies 50-fold lower than thatobtained with pEG922. These values correspond to the fre-quency of cointegrate formation, since the tnpI mutationimpairs cointegrate resolution (described below).

Evidence for cointegrate resolution. On the basis of the Tn3model of replicative transposition, mutations in tnpI leading toa defective resolvase should allow for the detection upontransposition of cointegrate molecules containing two copies ofthe transposon (19, 40). To test this, transposition events instrains EG10368/pEG922 and EG10368/pEG922 tnpI were

obtained by selecting for tetracycline-resistant colonies at41°C. EG10368/pEG922 colonies isolated in this way werechloramphenicol sensitive, whereas 40 of 44 EG10368/pEG922tnpI colonies were chloramphenicol resistant, as would beexpected for isolates deficient in cointegrate resolution. SinceB. thuningiensis can resolve cointegrate structures by homolo-gous recombination (10), these chloramphenicol-resistant iso-lates were maintained and evaluated under selection at 37 to41°C. Total DNA from these EG10368 derivatives was iso-lated, digested with SstI, and resolved on a 1% agarose gel forSouthern blot analysis. Since TnS401-tet in plasmids pEG922and pEG922 tnpI is flanked by SstI sites (Fig. 5), transpositionevents leading to an unresolved cointegrate should retain the9.2-kb SstI fragment that contains the pEG491 fragmentconferring chloramphenicol resistance and should yield twocopies of Tn5401-tet (schematic drawing in Fig. 6). As ex-pected, total DNA from the EG10368/pEG922 isolates did nothybridize to the pEG491 probe (Fig. 6A, wt lanes) but didhybridize as a single band to the Tn5401 probe (Fig. 6B, wtlanes). Southern blot analysis of total DNA from severalEG10368/pEG922 tnpI isolates revealed a 9.2-kb fragment thathybridizes to the pEG491 probe (Fig. 6A, tnpI lanes) and,typically, two larger fragments that hybridize to the TnS401probe (Fig. 6B, tnpI lanes). Thus, TnS401 transposition ap-pears to be replicative, proceeding via a cointegrate interme-diate that is resolved by the TnpI recombinase. The weakhybridization of the TnS401 probe to the 7-kb SstI fragment inFig. 6B, corresponding to the 7-kb Tn5401-tet fragment ofpEG922 tnpI, suggests that some resolution of the cointegrateis occurring in the absence of recombinase, presumably viahomologous recombination (10), and is resulting in release ofthe transposonr plasmid from the chromosomal site.

Transcriptional mapping. Transcriptional start sites withinTn5401 were mapped by primer extension analysis of totalRNA isolated from EG10368 recombinant strains containingthe transposon vector pEG922 and derivatives thereof withmutations in either tnpI or tnpA (Table 1). In addition, totalRNA was isolated from EG10368 recombinant strains contain-ing plasmid subclones of TnS401 in which the tnpI and tnpAgenes are absent. These subclones, designated p76 and p83,contain portions of the tnpI upstream region (Table 1).Two transcriptional start sites were mapped within the

transposon, corresponding to overlapping divergent promoterslocated immediately upstream of the tnpI gene. The leftwardpromoter, designated PL, apparently directs the transcriptionof orfi from a single initiation site at nt 711 (Fig. 7A). Therightward promoter, designated PR, apparently directs thetranscription of tnpI from a single initiation site at nt 723 (Fig.7B). These were identified by using two primers (pr5 and prlSfor PL and prlO and prl4 for PR) to confirm the position ofeach start site. The start site for PL lies within the - 10 regionof PR and vice versa (Fig. 2). The -10 regions of PL(TATITlT) and PR (TAAGAT) and the -35 regions of PL(TTGACT) and PR (TTGATG) show similarity to the consen-sus -10 (TATAAT) and -35 (TTGACA) regions of vegeta-tively expressed genes in Bacillus subtilis (34, 35). Additionalprimer extension analyses failed to identify transcriptional startsites between orfi and orJ2 (primer pr3) or within the inter-genic region between tnpI and tnpA (primers prl6 andpr911BamR3), suggesting that these genes are cotranscribed.

Transcription from both promoters is derepressed on plas-mids containing deletions of both tnpI and tnpA (Fig. 7A, p76versus pEG922 wt, and 7B, p83 versus pEG922 wt) but not ona plasmid containing an internal deletion of tnpA (pEG922 Aversus pEG922 wt). The copy numbers of plasmids p76, p83,pEG922, and pEG922 tnpAA are all comparable (Fig. 7C and

VOL. 176, 1994

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

2842 BAUM

p76 pEG922C G T A w I wt A I

p83C G T Al I

pEG922wt A I

relative

amounts * 8 4 2 1 8 4 2 1 40 8

FIG. 7. RNA analysis of Tn5401. (A and B) Primer extensionanalyses were performed with total RNA isolated from strain EG10368recombinants containing plasmids p76, p83, pEG922 (wt), pEG922tnpAA (A), and pEG922 tnpI (I). Primers pr5 and prlO were used forthe analyses shown in panels A and B, respectively. The followingprimer extension products were loaded for panel A: p76 (2 and 20 ,ul),pEG922 wt (20 ,ul), pEG922 A (20 RI), and pEG922 I (15 [LI). Thefollowing primer extension products were loaded for panel B: p83 (2and 20 pul), pEG922 wt (20 ,ul), pEG922 A (20 ,ul), and pEG922 I (15[LI). Transcriptional start sites are indicated by the asterisks. (C)Southern blot analysis of EG10368 recombinant strains showingrelative copy numbers for plasmids p76, pEG922 (wt), pEG922 tnpAA,and pEG922 tnpI. Plasmids were isolated from the cultures used forRNA extraction by the alkaline lysis method (30), digested with SstI,resolved by agarose gel electrophoresis, blotted to nitrocellulose, andhybridized to a 32P-labeled 658-bp NsiI fragment from Tn5401 (Fig. 3)common to all four plasmids.

reference 3), indicating that the elevated transcript levelsdetected for plasmids p76 and p83 are not due to differences inplasmid (template) copy number. In contrast, plasmid pEG922tnpI shows an apparent 40-fold reduction in copy numbercompared with pEG922 and pEG922 tnpAl\ in strain EG10368(Fig. 7C), indicating that the PL and PR transcripts detected byprimer extension analysis of EG10368/pEG922 tnpI RNA (Fig.7A and B, pEG922 I) represent a significant elevation intranscript levels per plasmid copy. The extent of this derepres-sion appears greater for PR than for PL. Accordingly, theseresults indicate that disruption of tnpI, but not tnpA, results inderepression of PL and PR, providing evidence that the recom-binase encoded by tnp! functions as a transcriptional repressor.

TABLE 4. B. thuringiensis plasmids that hybridize to thetransposons Tn4430 and Tn5401

B. thuringiensis Hybridization probe"sus.

Strain Source"subsp. Straln SOUrCea Tn4430 Tn540]aizawai EG2175 1 46, 43, 33/31,' 8.0aizawai EG6345 1 48aizawai HDI1 2 8.0darmstadiensis HD146 2 9.5darmstadiensis HD498 2 9.5finitimus HD3 2 90, 77galleriae HD8 2 10.3, 8.7kenyae HD5 2 41kenyae HD63 2 48, 38, 9.5, 6.1kumamotoensis EG4961 1 70kurstaki HD1 2 53/51,' 44kurstaki HD73-6 3 50kurstaki HD119 2 50, 44, 5.2kurstaki HD263 2 44kurstaki HD269 2 69, 43kurstaki HD279 2 60, 44kurstaki EG3260 1 57, 50, 42kurstaki EG3899 1 49, 45kurstaki EG6429 1 30, 9.9morrisoni HD597 2 48, 6.8, 5.7morisoni EG2158 1 72, 35morisoni EG2832" 1 50, 44sotto Type strain 4 43, 38, 9.5thuringiensis HD2 2 57/54,' 37, 32, 6.2thulringiensis HD931 2 60, 52, 41, 10thuringiensis EG6430 1 60, 35, 9.6tolworthi HD13 2 44, 28tolworthi HD537 2 28yunnanensis HD977 2 100, 80, 52, 46

" Sources: 1, Ecogen strain collection; 2, Dulmage collection (13); 3, variant ofstrain HD73 (17); 4, H. de Barjac, Institut Pasteur.

"' Values are apparent molecular masses of hybridizing plasmids in megadal-tons.

' Unresolved plasmid doublet.dEG2832 is also known as B. thuringiensis subsp. tenebrionis.

The effect of the tnpl mutation on the copy number ofpEG922 tnpI is apparently not the consequence of unregulatedtransposition by pEG922 tnpl (Table 3). In support of thisconclusion, plasmid pEG922 tnpI tnpAA, containing mutationsin both genes, also shows the 40-fold-lower copy number (3).Perhaps the higher-level transcription proceeding from PL andPR as a result of the tnpI mutation interferes with the replica-tion of the pE194ts replicon.

Distribution of TnS401 among B. thuringiensis subspecies. A2.56-kb ClaI-BamHI fragment from Tn5401, containing tnpI,ofl], and portions of orJ2 and tnpA, was used as a hybridizationprobe to survey (by Southern blot analysis) the residentplasmids of different B. thuringiensis strains for TnS401. Forcomparison, a KpnI fragment from plasmid pBR322::Tn4430containing the B. thuringiensis transposon Tn4430 (24) was alsoused as a hybridization probe. The 54 strains included in thesurvey, representing 27 different subspecies of B. thuringiensisand containing over 400 resident plasmids, are listed in Table2 of reference 5. Table 4 lists those strains containing hybrid-izing plasmids (indicated by approximate molecular masses inmegadaltons). The two transposons show markedly differentdistributions among the strains included in the survey, withTn4430 being far more widespread. Only the B. thuringiensissubsp. morrisoni strains EG2158 and EG2832 contain plasmidsthat hybridized strongly to the TnS401 probe. The sole ICPgene in strains EG2158 and EG2832, cryIILA, is located on

A TT

Tp TL *A

TCA

p T

T

B A

AAp T

R * AT

A

GL

f

J. BACTERIOL.

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

B. THURINGIENSIS CLASS II TRANSPOSABLE ELEMENT Tn5401 2843

plasmids of 88 and 90 MDa, respectively, and therefore is notassociated with Tn5401. Relatively weak hybridization signalsto plasmids in strains HD3, HD537, and HD977 may suggestthe presence of a related transposon or transposons.

DISCUSSION

The transposable element TnS401, derived from B. thurin-giensis subsp. mornsoni EG2158, is the second indigenousTn3-like transposon to be isolated from a Bacillus species.TnS401 and the B. thuringiensis transposon Tn4430 are crypticand share a similar structural organization, but TnS401 is 688bp longer and contains two additional potential coding regions,orfl and orJ2, located upstream of tnpI. The terminal IRs ofTnS401 are unusually long (53 bp) for a Tn3-like transposon,the outermost 38 bp of which can be aligned with the IRs ofTn3-like transposons, and show the greatest sequence similar-ity with the IRs of Tn3. The TnS401 transposase exhibits thegreatest similarity to the Tn3 transposase and the least simi-larity to the Tn917 transposase, while for the Tn4430 trans-posase, the similarities are nearly the reverse. The integrase-like recombinase proteins of Tn4430 and TnS401, althoughpresumably derived from a common ancestor, show only 24%amino acid sequence identity. In summary, these transposonsare highly divergent at the amino acid and nucleotide sequencelevels, despite their apparent common origin and structuralorganization.TnS401 and Tn4430 also differ with respect to their distri-

bution among subspecies of B. thuringiensis and their associa-tion with ICP genes. The Southern blot analysis data seem torule out the possibility that TnS401 is typically associated withknown ICP genes or is commonly found among differentsubspecies.

Computer-assisted analyses of the tnpI promoter regionrevealed an interesting arrangement of sequence elements thathas probable bearing on the function of the tnpI-encodedrecombinase protein. As shown in Fig. 2, the 12-bp sequenceATGTCCRCTAAY occurs four times in the intergenic regionbetween orfi and tnpI, two copies of which form a 28-bpimperfect IR (nt 639 to 666). The remaining two elements arepositioned between the - 35 and - 10 sites of PL and PR. Thissequence arrangement is strikingly similar to that observedupstream of the tnpI gene of Tn4430 (26), but the amount ofnucleotide sequence identity between the two regions is low(Fig. 8), as is the homology between their respective conservedsequence elements. Both promoter regions display a sequencewith imperfect dyad symmetry followed by direct repeatslocated near the tnpI coding region. The presence of aconserved sequence arrangement but little sequence homologysuggests that the 12-bp repeats of TnS401 may serve somefunction.These observations, together with the results presented in

this paper showing that the TnpI recombinase serves as atranscriptional repressor of PL and PR, suggest a model forunderstanding the regulation of TnS401 transposition. Assuggested previously for the TnpI recombinase of Tn4430 (26),the Tn5401 recombinase probably binds the direct repeatswithin PL and PR and represses transcription by preventing thebinding of RNA polymerase. The fact that the 12-bp directrepeats are opposite in orientation with respect to theirpromoters may have bearing on the differential repressionobserved for PL and PR by the recombinase. It also seems likelythat the IRs serve as the site at which recombination actuallyoccurs during cointegrate resolution, since the crossover sitesfor integrase family proteins occur within imperfect dyads -30bp in length (41). Deletion analysis of the tnpI promoter region

------------> <-----------

Tn5401 - ATAGTTTTAAATGTCCACTA-ATTAATATTAGTGGACATGAAGTGTGGGAII II llIll11I1111111111I111111

Tn4430 - AAAAAAT-AATACAACACAATATTAAT-TGTGTTGT-ATTAGGTGTTATA

<-PLTn5401 - A--AATA--AATGTTTGA-TGTC-CGCTAACATAATTGATAAGATTAAAA

Tn4430 - ATAAATATAAATCTAGGGGTTTAACGC-AACACAATTTATC-GATAAATA

PR*-Tn5401 - TAT-CATGTC--CGCTAATGTAAGTCAATAAAAGAGGAGGTATTT--ATG

Tn4430 - AATACTTTTAGACGC-AACACAATTTA-TAGACGCGGAGGAAATCACATG______________--- >

FIG. 8. Nucleotide sequence alignment (top strand) of the tnpIpromoter regions of Tn4430 and Tn5401. The conserved sequenceelements referred to in the text are indicated by the arrows above(TnS401) and below (Tn4430) the alignment. The transcriptional startsites for PL and PR of Tn5401 are indicated by the asterisks. The -10and - 35 regions of PR are underlined. The - 10 and - 35 regions ofPL (inverse complement shown) are doubly underlined. The ATGinitiation codons are indicated in boldface. NALIGN parameters(open gap cost of 5, unit gap cost of 10) were chosen to minimize largegaps in the alignment.

indicates, however, that removal of nt 714 to 747, containingthe proximal 12-bp element (nt 726 to 737), blocks cointegrateresolution (3), providing evidence that the internal resolutionsite region extends beyond the IR sequence.

Surprisingly, the 12-bp sequence element is also foundwithin the terminal IRs (Fig. 2), thus accounting for theirunusual length and suggesting a third function for the TnpIrecombinase. The recombinase may serve not only as a tran-scriptional repressor and as a site-specific recombinase but alsoas a negative regulator of the first step in transposition. Forinstance, recombinase binding to the terminal IRs couldmodulate transposase activity or prevent the binding of thetransposase to the terminal IRs, thereby providing a secondlevel of control over transposition. Since tnpI and tnpA areapparently cotranscribed, this mechanism could ensure thatthe transposase inevitably produced as a result of transcriptionfrom PR can still be regulated by the recombinase. In thisinstance, disruption of tnpI should result in elevated frequen-cies of cointegrate formation, an effect that was not observedeven when the lower copy number of pEG922 tnpI was takeninto account (Table 3). Alternatively, the recombinase couldserve a positive role and facilitate the first step in transpositioncatalyzed by transposase at the terminal IRs. In this instance,disruption of tnpI would have both positive and negative effectson the frequency of cointegrate formation. Evidence for eitherrole awaits more accurate measurements of the effect of thetnpI mutation on TnS401 transposition and the demonstrationthat the recombinase does bind to the 12-bp repeats and thatrecombinase binding to the terminal IRs affects transposition.The TnS401-based transposon vector pEG922 can be used

to construct transposon insertion libraries of the B. thuringien-sis chromosome in the same way that the temperature-sensitiveTn91 7-based transposon vectors are used in B. subtilis (43, 44).One such library has been constructed by using B. thuringiensisEG10368 (29), a strain which is nearly devoid of nativeplasmids (Table 1). The analysis of TnS401-tet transposition instrain EG2424 suggests that Tn5401 transposes at a higherfrequency into plasmids than into the chromosome. Thisapparent preference for plasmid target sites may make TnS401useful for structure-function studies of large B. thuringiensisplasmids.

VOL. 176, 1994

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

2844 BAUM

ADDENDUM

The TnpI recombinase protein of TnS401 has been ex-pressed in E. coli and has been shown to bind to the 12-bprepeat sequence by gel mobility shift and DNase I footprintinganalyses.

ACKNOWLEDGMENTS

I thank Cynthia Gawron-Burke, Tom Malvar, and Chris Egolf fortheir critical reading of the manuscript and C. Gawron-Burke and T.Malvar for their helpful advice during the course of this study. I alsothank Diane Rutkowski for assisting with the Eckhardt gel analysis andMark Rupar for providing me with synthetic oligonucleotides.

REFERENCES

1. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman.1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.

2. Argos, P., A. Landy, K. Abremski, J. B. Egan, E. Haggard-Ljungquist, R. H. Hoess, M. L. Kahn, B. Kalionis, S. B. L.Narayana, L. S. Pierson III, N. Sternberg, and J. M. Leong. 1986.The integrase family of site-specific recombinases: regional simi-larities and global diversity. EMBO J. 5:433-440.

3. Baum, J. A. Unpublished data.4. Baum, J. A., D. M. Coyle, M. P. Gilbert, C. S. Jany, and C.

Gawron-Burke. 1990. Novel cloning vectors for Bacillus thuringien-sis. Appl. Environ. Microbiol. 56:3420-3428.

5. Baum, J. A., and J. M. Gonzalez, Jr. 1992. Mode of replication,size and distribution of naturally occurring plasmids in Bacillusthuringiensis. FEMS Microbiol. Lett. 96:143-148.

6. Bernhard, K., H. Schrempf, and W. Goebel. 1978. Bacteriocin andantibiotic resistance plasmids in Bacillus cereus and Bacillus subti-lis. J. Bacteriol. 133:897-903.

7. Bourgouin, C., A. Delecluse, J. Ribier, A. Klier, and G. Rapoport.1988. A Bacillus thuringiensis subsp. israelensis gene encoding a125-kilodalton larvicidal polypeptide is associated with invertedrepeat sequences. J. Bacteriol. 170:3575-3583.

8. Chambers, J. A., A. Jelen, M. P. Gilbert, C. S. Jany, T. B. Johnson,and C. Gawron-Burke. 1991. Isolation and characterization of anovel insecticidal crystal protein gene from Bacillus thuringiensissubsp. aizawai. J. Bacteriol. 173:3966-3976.

9. Delecluse, A., C. Bourgouin, A. Klier, and G. Rapoport. 1989.Nucleotide sequence and characterization of a new insertionelement, IS240, from Bacillus thuringiensis israelensis. Plasmid21:71-78.

10. Delecluse, A., J.-F. Charles, A. Klier, and G. Rapoport. 1991.Deletion by in vivo recombination shows that the 28-kilodaltoncytolytic polypeptide from Bacillus thuringiensis subsp. israelensis isnot essential for mosquitocidal activity. J. Bacteriol. 173:3374-3381.

11. Donovan, W. P., J. M. Gonzalez, Jr., M. P. Gilbert, and C.Dankocsik. 1988. Isolation and characterization of EG2158, a newstrain of Bacillus thuringiensis toxic to coleopteran larvae, andnucleotide sequence of the toxin gene. Mol. Gen. Genet. 214:365-372.

12. Donovan, W. P., M. J. Rupar, A. C. Slaney, T. Malvar, M. C.Gawron-Burke, and T. B. Johnson. 1992. Characterization of twogenes encoding Bacillus thuringiensis insecticidal crystal proteinstoxic to Coleoptera species. Appl. Environ. Microbiol. 58:3921-3927.

13. Dulmage, H. T., C. C. Beegle, H. de Barjac, D. Reich, G.Donaldson, and J. Krywienczyk. 1982. Bacillus thuringiensis cul-tures available from the U.S. Department of Agriculture. USDA-ARS agricultural reviews and manuals, ARM-S-30/October 1982.U.S. Department of Agriculture, Agricultural Research Service,New Orleans.

14. Feinburg, A. P., and B. Vogelstein. 1984. A technique for radiola-beling DNA restriction endonuclease fragments to high specificactivity. Anal. Biochem. 137:266-267.

15. Fickett, J. W. 1982. Recognition of protein coding regions in DNAsequences. Nucleic Acids Res. 10:5303-5318.

16. Gawron-Burke, C., and J. A. Baum. 1991. Genetic manipulation of

Bacillus thuringiensis insecticidal crystal protein genes in bacteria,p. 237-263. In J. K. Setlow (ed.), Genetic engineering. PlenumPress, New York.

17. Gonzalez, J. M., Jr. 1984. Ph.D. thesis. University of Georgia,Athens.

18. Gonzalez, J. M., Jr., H. T. Dulmage, and B. C. Carlton. 1981.Correlation between specific plasmids and delta endotoxin pro-duction in Bacillus thuringiensis. Plasmid 5:351-365.

19. Heifron, F. 1983. Tn3 and its relatives, p. 223-260. In J. A. Shapiro(ed.), Mobile genetic elements. Academic Press, Orlando, Fla.

20. Hofte, H., and H. R. Whiteley. 1989. Insecticidal crystal proteins ofBacillus thuringiensis. Microbiol. Rev. 55:242-255.

20a.Kleckner, N. 1981. Transposable elements in prokaryotes. Annu.Rev. Genet. 15:341-404.

21. Kronstad, J. W., and H. R. Whiteley. 1984. Inverted repeatsequences flank a Bacillus thuringiensis crystal protein gene. J.Bacteriol. 160:95-102.

22. Kronstad, J. W., and H. R. Whiteley. 1986. Three classes ofhomologous Bacillus thuringiensis crystal-protein genes. Gene 43:29-40.

23. Lereclus, D., C. Bourgouin, M.-M. Lecadet, A. Klier, and G.Rapoport. 1989. Role, structure, and molecular organization ofthe genes coding for the parasporal delta-endotoxins of Bacillusthuringiensis, p. 255-276. In I. Smith, R. Slepecky, and P. Setlow(ed.), Regulation of procaryotic development. American Societyfor Microbiology, Washington, D.C.

24. Lereclus, D., J. Mahillon, G. Menou, and M.-M. Lecadet. 1986.Identification of Tn4430, a transposon of Bacillus thuringiensisfunctional in Escherichia coli. Mol. Gen. Genet. 204:52-57.

25. Lereclus, D., J. Ribier, A. Klier, G. Menou, and M.-M. Lecadet.1984. A transposon-like structure related to the delta-endotoxingene of Bacillus thuringiensis. EMBO J. 3:2561-2567.

26. Mahillon, J., and D. Lereclus. 1988. Structural and functionalanalysis of Tn4430: identification of an integrase-like proteininvolved in the co-integrate-resolution process. EMBO J. 7:1515-1526.

27. Mahillon, J., J. Seurinck, J. Delcour, and M. Zabeau. 1987.Cloning and nucleotide sequence of different iso-IS231 elementsand their structural association with the Tn4430 transposon inBacillus thuringiensis. Gene 51:187-196.

28. Mahillon, J., J. Seurinck, L. Van Rompuy, J. Delcour, and M.Zabeau. 1985. Nucleotide sequence and structural organization ofan insertion sequence element (IS231) from Bacillus thuringiensisstrain berliner 1715. EMBO J. 4:3895-3899.

29. Malvar, T., and J. A. Baum. Unpublished data.30. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular

cloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

31. McMaster, G. K., and G. C. Carmichael. 1977. Analysis of single-and double-stranded nucleic acids on polyacrylamide and agarosegels by using glyoxal and acridine orange. Proc. Natl. Acad. Sci.USA 74:4835-4838.

32. Menou, G., J. Mahillon, M.-M. Lecadet, and D. Lereclus. 1990.Structural and genetic organization of IS232, a new insertionsequence of Bacillus thuringiensis. J. Bacteriol. 172:6689-6696.

33. Mettus, A.-M., and A. Macaluso. 1990. Expression of Bacillusthuringiensis 8-endotoxin genes during vegetative growth. Appl.Environ. Microbiol. 56:1128-1134.

34. Moran, C. P., Jr. 1989. Sigma factors and the regulation oftranscription, p. 167-174. In I. Smith, R. A. Slepecky, and P.Setlow (ed.), Regulation of procaryotic development. AmericanSociety for Microbiology, Washington, D.C.

35. Moran, C. P., Jr., N. Lang, S. F. J. LeGrice, G. Lee, M. Stephens,A. L. Sonenshein, J. Pero, and R. Losick. 1982. Nucleotidesequences that signal the initiation of transcription and translationin Bacillus subtilis. Mol. Gen. Genet. 186:339-346.

36. Murphy, E. 1989. Transposable elements in gram-positive bacte-ria, p. 269-288. In D. E. Berg and M. M. Howe (ed.), MobileDNA. American Society for Microbiology, Washington, D.C.

37. Rezsohazy, R., B. Hallet, and J. Delcour. 1992. IS231D, E, and F,three new insertion sequences in Bacillus thuringiensis: extensionof the IS231 family. Mol. Microbiol. 6:1959-1967.

38. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing

J. BACTERIOL.

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

B. THURINGIENSIS CLASS II TRANSPOSABLE ELEMENT Tn5401 2845

with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA85:5463-5467.

39. Schuler, G. D., S. F. Altschul, and D. J. Lipman. 1991. Aworkbench for multiple alignment construction and analysis. Pro-teins Struct. Funct. Genet. 9:180-190.

40. Sherratt, D. 1989. Tn3 and related transposable elements: site-specific recombination and transposition, p. 163-184. In D. E.Berg and M. M. Howe (ed.), Mobile DNA. American Society forMicrobiology, Washington, D.C.

41. Stark, W. M., M. R. Boocock, and D. J. Sherratt. 1992. Catalysis bysite-specific recombinases. Trends Genet. 8:432-439.

42. Villafane, R., D. H. Bechhofer, C. S. Narayanan, and D. Dubnau.1987. Replication control genes of plasmid pE194. J. Bacteriol.169:4822-4829.

43. Youngman, P. 1987. Plasmid vectors for recovering and exploitingTn917 transpositions in Bacillus and other gram-positive bacteria,p. 79-103. In K. G. Hardy (ed.), Plasmids: a practical approach.IRL Press, Oxford.

44. Youngman, P., H. Poth, B. Green, K. York, G. Olmedo, and K.Smith. 1989. Methods for genetic manipulation, cloning, andfunctional analysis of sporulation genes in Bacillus subtilis, p.65-87. In I. Smith, R. Slepecky, and P. Setlow (ed.), Regulation ofprocaryotic development. American Society for Microbiology,Washington, D.C.

45. Zuber, P., J. M. Healy, and R. LosicLk 1987. Effects of plasmidpropagation of a sporulation promoter on promoter utilizationand sporulation in Bacillus subtilis. J. Bacteriol. 169:461-469.

VOL. 176, 1994

on August 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from