phage-borne factors and host lexa regulate the lytic ...jb.asm.org/content/193/21/6008.full.pdf ·...

JOURNAL OF BACTERIOLOGY, Nov. 2011, p. 6008–6019 Vol. 193, No. 210021-9193/11/$12.00 doi:10.1128/JB.05618-11Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Phage-Borne Factors and Host LexA Regulate theLytic Switch in Phage GIL01�

Nadine Fornelos,1,2 Jaana K. H. Bamford,2* and Jacques Mahillon1

Laboratory of Food and Environmental Microbiology, Universite Catholique de Louvain, Croix du Sud, 2, Box L7.05.12,B-1348 Louvain-la-Neuve, Belgium,1 and Department of Biological and Environmental Science and Nanoscience Center,

University of Jyvaskyla, P. O. Box 35, 40014 Jyvaskyla, Finland2

Received 20 June 2011/Accepted 25 August 2011

The Bacillus thuringiensis temperate phage GIL01 does not integrate into the host chromosome but existsstably as an independent linear replicon within the cell. Similar to that of the lambdoid prophages, the lyticcycle of GIL01 is induced as part of the cellular SOS response to DNA damage. However, no CI-likemaintenance repressor has been detected in the phage genome, suggesting that GIL01 uses a novel mechanismto maintain lysogeny. To gain insights into the GIL01 regulatory circuit, we isolated and characterized a set of17 clear plaque (cp) mutants that are unable to lysogenize. Two phage-encoded proteins, gp1 and gp7, arerequired for stable lysogen formation. Analysis of cp mutants also identified a 14-bp palindromic dinBox1sequence within the P1-P2 promoter region that resembles the known LexA-binding site of Gram-positivebacteria. Mutations at conserved positions in dinBox1 result in a cp phenotype. Genomic analysis identified atotal of three dinBox sites within GIL01 promoter regions. To investigate the possibility that the host LexAregulates GIL01, phage induction was measured in a host carrying a noncleavable lexA (Ind�) mutation. GIL01formed stable lysogens in this host, but lytic growth could not be induced by treatment with mitomycin C. Also,mitomycin C induced �-galactosidase expression from GIL01-lacZ promoter fusions, and induction wassimilarly blocked in the lexA (Ind�) mutant host. These data support a model in which host LexA binds todinBox sequences in GIL01, repressing phage gene expression during lysogeny and providing the switchnecessary to enter lytic development.

Temperate phages are defined by their ability to infect asusceptible host cell and opt between either of two develop-mental pathways, the lytic or the lysogenic cycle. In the lyticcycle, the phage genome is replicated and expressed to pro-duce the structural components necessary to assemble capsids.Once engaged, this pathway inevitably ends with the death ofthe host cell and the release of a crop of newly formed virionsin the environment. In contrast, the lysogenic cycle is definedby the tight repression of lytic behavior, and the phage genomereplicates along with that of the host cell, either physicallyintegrated into the host genome or as an independent plasmidprophage. The quiescent prophage is thereby efficiently inher-ited as cells divide, until the lytic cycle is reactivated in re-sponse to alterations in host cell physiology. This transitionrequires the efficient interaction of regulatory components thatcontrol the passage from a highly stable standby mode tomultiple rounds of genome replication, vigorous gene expres-sion, virion assembly, and host lysis.

In the well-studied bacteriophage �, establishment of lysog-eny relies on the synthesis of CII, a transcriptional regulatorthat directs the expression of the lytic cycle repressor CI (15).Once produced, CI dimers bind cooperatively at operator sitesin the so-called switch region to prevent transcription from theearly promoters PL and PR and, subsequently, due to the lackof early gene products, from the late lytic promoters as well.

Expression of CI alone is sufficient to maintain lysogeny onceit has been established (26). Despite an intrinsically stablequiescent state, the � prophage can efficiently switch to analternate lytic cycle. The � CI repressor functions analogouslyto the bacterial SOS response master repressor LexA, withwhich it shares significant sequence similarity (21, 45). LexAbinds as a dimer to specific operator sites and negatively reg-ulates the expression of a host of genes that are essentiallyinvolved in DNA repair and cell survival (10, 18). Upon DNAdamage, such as that inflicted by UV irradiation or mitomycinC (MMC) treatment, the host recombinase RecA is convertedto an activated form when it binds single-stranded DNA. Ac-tivated RecA then acts as a coprotease to stimulate LexAautoproteolysis and, through a similar process, CI autoprote-olysis as well (43, 59). As a consequence, LexA dissociatesfrom its consensus sites to initiate the SOS response. Likewise,CI is no longer able to bind its operator sites in the � genome,leading to lytic gene expression and viral proliferation.

Alternative mechanisms of regulation have been describedfor several SOS-inducible phages. The repressors of coliphages186 and N15, for example, are not RecA sensitive but areinstead antagonized by phage-borne antirepressors that areunder LexA control (47, 61). In CTX�, the filamentous phagethat encodes cholera toxin in Vibrio cholerae, host LexA andthe phage repressor function as both activators and repressorsof gene expression to ensure the permanent production andsecretion of CTX� (36).

The Bacillus thuringiensis prophage GIL01 has been shownto be induced by DNA-damaging treatments such as UV irra-diation and MMC (69). The 15-kb linear prophage does notintegrate into the bacterial chromosome but instead replicates

* Corresponding author. Mailing address: Department of Biologicaland Environmental Science and Nanoscience Center, University ofJyvaskyla, P. O. Box 35, 40014 Jyvaskyla, Finland. Phone: 358 14 2604160. Fax: 358 14 260 2221. E-mail: [email protected].

� Published ahead of print on 2 September 2011.

6008

on May 17, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

independently within its host and occasionally experiences in-duction of its lytic cycle. GIL01 shares a similar genome orga-nization and size with PRD1, the model tectivirus infectingGram-negative hosts. However, unlike PRD1 and its closesiblings, which are all lytic phages, GIL01 is able to enter adormant state during which its lytic functions are turned off.Two additional tectiviruses infecting B. thuringiensis have alsobeen described: Bam35, which is almost identical to GIL01 (55,64) and has been the object of many studies on the internalmembrane that defines tectiviruses (20, 39–41), and GIL16,which shares considerable sequence identity (83.6%) withGIL01 (67). A more distant relative, pBClin15, has been se-quenced alongside the B. cereus reference strain ATCC 14579genome but is not yet known to be a prophage (29). Recently,the sequencing of AP50, a temperate phage highly specific toB. anthracis strains, added a new member to the subgroup ofTectiviridae that proliferates among members of the B. cereusgroup (62).

By analogy to other temperate phages, it has been assumedthat GIL01 controls its lysogenic cycle by expressing a lyticgene repressor. A potential candidate for this function waspredicted to be encoded by open reading frame 6 (ORF6) byvirtue of its N-terminal LexA-like DNA-binding domain (67).However, unlike the � phage CI repressor, ORF6 lacks anyrecognizable protease linker required for repressor cleavage orthe C-terminal dimerization domain. Furthermore, spontane-ous GIL01 clear plaque (cp) mutants—which are unable toestablish the lysogenic cycle—were previously mapped toORF7, but no such mutations were found to affect ORF6 (67).In this study, we demonstrated that ORF7 is indeed involved inthe regulation of the phage cycle and that it does so along withORF1. Importantly, we also showed that GIL01 is induced inresponse to SOS signals due to direct control by the host LexArepressor. LexA likely binds to a conserved operator, termeddinBox1 (for damage-inducible Box), that when mutated com-mits GIL01 to propagate exclusively as a virulent phage.dinBox1 lies within two promoters, P1 and P2, that direct theexpression of regulation and replication functions that arelikely important for lysogenic development and/or mainte-nance. Two additional LexA-binding sites, dinBox2 anddinBox3, were identified at promoter P3 regulating the down-stream structure and lysis module. Curiously, no cp mutantswere ever isolated that carried mutations within these operatorsites, possibly indicating that dinBox2 and dinBox3 are notrequired for lysogenic regulation but are important for lyticgrowth. Altogether, we identified one cis-acting element(dinBox1), two phage-borne genes (ORF1 and ORF7), and ahost regulator (LexA) that determine passage from the latentstate to lytic development in GIL01. The identification of thesefactors will ultimately help elucidate the gene circuit that reg-ulates GIL01 development.

MATERIALS AND METHODS

Bacterial strains, phages, and plasmids. The natural host of phage GIL01 isthe insect pathogen B. thuringiensis subspecies israelensis strain NB31 (2). Thepermissive plasmid-free strain GBJ002 (31), derived from strain 4Q2 and relatedto NB31 (32), was used to propagate GIL01. In order to work in identical geneticbackgrounds, we generated GBJ002 lysogens as follows. GIL01 filter-sterilizedsuspensions were spotted onto 0.7% top agar lawns seeded with GBJ002. Plateswere incubated overnight at 37°C. The center of a lysed spot was then recoveredwith a sterile inoculation loop and streaked on LB agar. After 16 h at 37°C,

individual colonies had grown and were screened by PCR with GIL01-specificprimers to confirm lysogeny. A positive colony (GBJ338) was selected and re-tained for the study.

GIL01 cp mutants were obtained spontaneously by infecting GBJ002 withwild-type GIL01 phage suspensions. Briefly, phage preparations from serial10-fold dilutions were added to 200 �l of mid-log-phase GBJ002 and incubatedat room temperature for 30 min. The infected cells were then plated on LB agarwith 4 ml of molten 0.7% top agar. The plates were incubated overnight at 37°Cand visually screened for the presence of clear plaques. Single clear plaques,isolated from independent experiments, were picked with micropipette tips andsubjected to three rounds of single-plaque purification on host GBJ002. Thewild-type pGIL01 genome sequence is available from the NCBI GenBank data-base under the accession number AJ536073.

Strain GBJ499 (a recA null mutant) was constructed by disrupting the recAgene of strain GBJ002 with the integrative vector pMutin4 (Bacillus GeneticStock Center). pMutin4 is unable to replicate in Bacillus and is therefore par-ticularly well suited for insertional mutagenesis in this host (66). Briefly, aninternal fragment of recA (coordinates 3,719,072 through 3,720,009 in B. cereusstrain G9842, GenBank accession number NC_011772) was amplified from strainGBJ002 with primers RecAmutF (5�-GCGGCCGCTCTTCACCAATTCCGTGAT) and RecAmutR (5�-GAATTCCAATTTGGTAAAGGTTCAATTA).The PCR product was digested with NotI and EcoRI (restriction sites areunderlined), purified, and cloned into NotI- and EcoRI-digested pMutin4 vector.The ligation was transformed into Escherichia coli (ER2925), and recombinantplasmids were extracted and introduced into competent GBJ002. Bacillus trans-formants, having integrated the vector by homologous recombination, were se-lected on 0.3 �g of erythromycin ml�1. The insertion of pMutin4 in the desiredlocus was confirmed by sequence analysis of recA. The downstream gene, whichencodes a phosphodiesterase (B1422), is separated from recA (B1421) by a500-bp noncoding region, including a Rho-independent terminator. It is there-fore unlikely that the insertion of pMutin4 in recA disrupts an operon.

Strain GBJ396 [lexA(A96D) mutant] was constructed by replacing an Ala(GCT) residue at position 96 in lexA with Asp (GAT). A 1.6-kb segment encom-passing lexA (coordinates 3,623,569 through 3,625,169 in B. cereus strain G9842,GenBank accession number NC_011772) was amplified from strain GBJ002 withprimers MutLex1 (5�-GGATCCTTGATATTCTTCAACAATAC) and MutLex2(5�-AAGCTTGTGTTAGAACACTTGCTGGA) and cloned into the pCR4-TOPO vector (Invitrogen). The ligation was transformed into E. coli (TOP10),and recombinants were extracted to undergo mutagenesis. A single-nucleotidemutation was generated by using a QuikChange II site-directed mutagenesis kit(Stratagene) with primer pairs LexindF (5�-GTCGGAAAAGTTACTGATGGTTTACCAATTACAGCG) and LexindR (5�-CGCTGTAATTGGTAAACCATCAGTAACTTTTCCGAC) (boldface type indicates the mutated nucleotide) ac-cording to the manufacturer’s instructions. DpnI-treated vector DNA wastransformed into XL1-Blue E. coli cells, and transformants were selected on 50�g of kanamycin ml�1. Mutants were confirmed by DNA sequence analysis withthe universal M13 sequencing primers. Mutated DNA from one selected clonewas restricted with BamHI and HindIII (restriction sites in MutLex1 and Mut-Lex2 are underlined), and the lexA-containing fragment was cloned into BamHI-and HindIII-digested pMutin4 plasmid DNA. The ligation was transformed intoE. coli (ER2925), and recombinant plasmids were extracted and introduced intocompetent GBJ002. Similarly, Bacillus transformants, having integrated the vec-tor by homologous recombination, were selected on 0.3 �g of erythromycin ml�1.One of these recombinants was grown for several consecutive cycles of 4 h inliquid LB, and samples were recovered at each cycle and plated on LB agar atdifferent dilutions. Single colonies were subjected to UV irradiation (wavelength,254 nm) for 10 s and left at room temperature for 2 days. UV-sensitive colonieswere amplified with primers MutLex1 and MutLex2, and the resulting PCRproducts were sequenced in order to screen for clones that had undergone asecond recombination event resulting in the lexA(A96D) genotype. LysogensGBJ500 and GBJ406 were generated by infecting the mutant strains GBJ499 andGBJ396 with GIL01 as explained above.

E. coli strain ER2925 (New England BioLabs), a dam and dcm mutant K12derivative, was used to propagate unmethylated constructs prior to transforma-tion into GBJ002. pCR4-TOPO-derived constructs were transformed into E. colistrain TOP10 (Invitrogen) or XL1-Blue (Stratagene). ORF1 and ORF7 werecloned into the multicopy plasmid pDG148, an E. coli-Bacillus spp. shuttle vectorwith an isopropyl-�-D-thiogalactopyranoside (IPTG)-inducible Pspac promoterupstream of the multiple cloning site (63). The pCR4-TOPO (Invitrogen) pos-itive selection vector was used to clone 5�-rapid amplification of cDNA ends(RACE)-generated fragments for sequencing as well as the amplified lexA geneand P1-P2 promoter segments for site-directed mutagenesis. pHT304-18Z is alow-copy-number shuttle vector used to generate lacZ fusions in B. cereus (1).

VOL. 193, 2011 PHAGE GIL01 IS UNDER LexA CONTROL 6009


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

Growth conditions and phage induction. B. thuringiensis and E. coli strainswere grown in modified lysogeny broth (LB) (1% tryptone, 0.5% yeast extract,0.5% NaCl) (6) at 37°C. B. thuringiensis liquid cultures were always grownovernight with shaking at 200 rpm before being diluted 1:100 in LB and allowedto grow exponentially. Transformation of B. thuringiensis and E. coli strains withthe different constructs was performed by electroporation (46, 57), and transfor-mants were selected on LB agar supplemented with 25 �g of kanamycin ml�1

(pDG148-derived constructs), 50 �g of kanamycin ml�1 (pCR4-TOPO-derivedconstructs), 25 �g of erythromycin ml�1, and 100 �g of ampicillin ml�1

(pHT304-18Z-derived constructs in B. thuringiensis and E. coli, respectively) or0.3 �g of erythromycin ml�1 and 100 �g of ampicillin ml�1 (pMutin4-derivedconstructs in B. thuringiensis and E. coli, respectively).

GIL01 was induced from GBJ338, GBJ500, and GBJ406 by UV irradiationand MMC treatment. In both cases, exponential lysogenic cultures were pelletedand resuspended in 5 ml of 10 mM MgSO4. For UV induction, the resuspendedcells were irradiated at 254 nm for 10 s at a distance of 10 cm. The irradiated cellswere then incubated at 37°C and 200 rpm for 15 min, pelleted, and resuspendedin liquid LB. Phage induction was allowed to proceed for 90 min at 37°C and 200rpm before culture supernatants were recovered and filtered (0.45 �m). ForMMC induction, MMC was added to 5 ml of resuspended cells at a concentra-tion of 50 ng ml�1. The cells were then incubated at 37°C and 200 rpm for 1 hbefore being pelleted and washed twice with 10 mM MgSO4. The pellets werefinally resuspended in liquid LB, and phage induction was allowed to proceed for90 min at 37°C and 200 rpm before culture supernatants were recovered andfiltered (0.45 �m). Control samples that were not irradiated or MMC inducedwere otherwise treated equally. GIL01 titers from induced and uninduced cul-tures were estimated by infecting GBJ002 in soft agar (0.7%).

DNA, enzymes, and reagents. Nucleotides and DNA modification enzymeswere purchased from Roche and Fermentas. Phusion Flash high-fidelity PCRmaster mix (Finnzymes) and GoTaq Flexi DNA polymerase (Promega) wereused in PCR amplifications. Oligonucleotides, antibiotics, mitomycin C, IPTG,and GenElute plasmid miniprep kits were purchased from Sigma-Aldrich.QIAquick gel extraction and PCR purification kits were purchased from Qiagen.

Determination of transcription start sites. Total RNA was extracted frommid-log-phase GBJ338 cultures with an Ambion RiboPure-bacteria kit andtreated with DNase DNA-free solution (Ambion). Transcription start sites weredetermined with the 5�-RACE system (Invitrogen) according to the manufac-turer’s instructions. Phage-specific primers hybridizing downstream of P1-P2(5�-TGTCCACTGTCATAATATCATTT, at coordinates 619 to 641) and P3(5�-TTGGAGATTATTTATTGACAAGA, at coordinates 5,233 to 5,255) wereused to synthesize cDNA strands from GIL01 transcripts. cDNAs were tailed andPCR amplified before being cloned into pCR4-TOPO and sequenced.

lacZ fusions and �-galactosidase assays. The region encompassing promotersP1 and P2 as well as dinBox1 (coordinates 39 through 413) was amplified usingprimers pHT1 (5�-AAGCTTGACATGTAATAACATTTATG) and pHT4 (5�-GGATCCTCTTTTTATGAACACGCA). The corresponding product was puri-fied and cloned into the selection vector pCR4-TOPO. The ligation was trans-formed in competent TOP10 cells, and recombinants were extracted andrestricted with enzymes HindIII and BamHI (restriction sites are underlined).The gel-purified fragments were then ligated into HindIII- and BamHI-digestedpHT304-18Z to obtain construct pDin1-2 in strain ER2925. A similar strategywas followed to generate a lacZ fusion with promoter P3. The region encom-passing the promoter sequence, dinBox2 and dinBox3 (coordinates 4,751 through4,977), was amplified with primers pHT5 (5�-AAGCTTAGACAAGCAAAACCGACA) and pHT6 (5�-GTCGACTATGATTAGTTCCGTTAAAG), clonedinto pCR4-TOPO, restricted with HindIII and SalI, and ligated to HindIII- andSalI-digested pHT304-18Z to obtain construct pDin3. Both constructs pDin1-2and pDin3 were confirmed by DNA sequence analysis and transformed to com-petent GBJ338 cells. To generate pDin1-2mut, in which dinBox1 contains thesame single-nucleotide mutation as Bam35c (CGAACaagcGTTTT to CGAACaagcTTTTT), a QuikChange II site-directed mutagenesis kit (Stratagene) wasused with primer pairs DinIndF (5�-GTTAAAAAACGAACAAGCTTTTTATAAGTGTTCGG) and DinIndR (5�-CCGAACACTTATAAAAAGCTTGTTCGTTTTTTAAC) (boldface letters represent the mutated nucleotide) on con-struct pCR4-TOPO, which contained the Din1-2 insert. After the desiredmutation was verified by sequence analysis, the same cloning steps as for thegeneration of wild-type pDin1-2 were followed.

Bacillus transformants were grown at 30°C in liquid cultures supplementedwith 10 �g of erythromycin ml�1 to exponential phase, at which point 50 ng ofMMC ml�1 was added to one-half of the cultures. Cultures were further incu-bated for one additional hour, and samples (20 �l) from each different conditiontested were taken. �-Galactosidase activities were assayed as previously de-scribed (16), with some modifications. Bacterial cells were permeabilized with

980 �l of buffer Z (containing 50 mM �-mercaptoethanol) and 10 �l of toluene.After 5 min at 37°C, 200 �l of buffer Z supplemented with o-nitrophenyl-�-D-galactopyranoside (ONPG; 4 mg ml�1) were added to each sample. The treatedsamples were incubated at 37°C for a time period ranging from 20 to 30 min, andreactions were stopped by adding 500 �l of Na2CO3 1 M. The samples werecentrifuged for 7 min at 13,000 rpm, and the supernatants were collected. Opticaldensities (ODs) were measured at a wavelength of 420 nm. The �-galactosidaseactivity values are expressed as Miller units and were calculated with the formula(1,000 � OD420)/(OD595 � time [min] � volume [ml]).

pDG1, pDG7, and pDG1/7 plasmid constructions and complementation tests.ORF1 (coordinates 356 to 532) was amplified from GIL01 DNA with primersDG1F (5�-GCTCTAGATGAGTAAACTACTGACTGCG) and DG1R (5�-TGCGCATGCTTAATTGTCCAGTTTATTTT) and digested with XbaI and SphI(restriction sites are underlined). The restriction fragment was purified andligated to XbaI- and SphI-digested pDG148 plasmid DNA to generate pDG1.pDG7 was constructed similarly by using primer pairs DG7F (5�-GCTCTAGATGCGTGACAAATTGCTCG) and DG7R (5�-TGCGCATGCTCAGTCATCCTTCTTCCCC) (coordinates 4,564 through 4,716). An ORF1 to ORF7 transcrip-tional unit was generated by amplifying ORF1 with primers DG1_rbsF (5�-TGCAAGCTTGaaggtggtgCATATGTAAT) and DG1_rbsR (5�-GCTCTAGATTAATTGTCCAGTTTATTTTTA) (coordinates 336 through 533, including theORF1 ribosome-binding site shown in Fig. 1C in lowercase), restricting the PCRproduct with HindIII and XbaI, and cloning the purified product into HindIII-and XbaI-digested pDG148 plasmid, resulting in construct pDG1_rbs. ORF7was amplified with primers DG7_rbsF (5�-GCTCTAGAaaggagggAAAAGGTATG) and DG7_rbsR (5�-GCTCTAGATCAGTCATCCTTCTTCC) (coordi-nates 4,549 through 4,717, including the ORF7 ribosome-binding site shown inlowercase). After digestion with XbaI and gel purification, the fragment wasligated into XbaI-digested pDG1_rbs plasmid, resulting in construct pDG1/7.pDG1, pDG7, and pDG1/7 ligations were transformed into competent E. coli(ER2925), and recombinant plasmids were extracted and confirmed by DNAsequence analysis. Competent GBJ002 was transformed with unmethylatedpDG1, pDG7, and pDG1/7 (containing ORF7 in the correct orientation) to yieldthe clones GBJ250, GBJ223, and GBJ467, respectively. Protein expression wasinduced by adding IPTG to a final concentration of 1 mM to one-half of theliquid cultures in the early exponential growth phase. Control cultures weregrown without the addition of IPTG. After an additional 3-h incubation at 37°Cwith shaking, 200 �l of induced cultures were infected with three 10-fold serialdilutions of cp mutant phage stocks and plated with molten top agar (0.7%) onLB plates supplemented with IPTG to a final concentration of 1 mM. The sametreatment was followed with uninduced cultures, but they were plated on LBplates devoid of IPTG. The plates were incubated overnight at 37°C, and plaquecounts were determined. Plate pictures were taken with the Bio-Rad MolecularImager Gel Doc XR system using Quantity One version 4.6.3. imaging software.

Sequencing and sequence analyses. Constructs were sequenced at the Flan-ders Interuniversity Institute for Biotechnology (VIB) Genetic Service Facility(University of Antwerp, Belgium) by double-strand primer walking, with indi-vidual quality control of sequence reads included. cp mutants were amplified withGIL01-specific primer pairs designed to cover the 15-kb genome in segments of1 kb with 100-bp overlaps. PCR products were purified and sequenced with aBigDye Terminator v3.1 cycle sequencing kit on an ABI Prism 3130xl geneticanalyzer instrument (both from Applied Biosystems). cp genomes were se-quenced on both strands to provide 3- to 4-fold sequence coverage. Sequenceswere analyzed with the EMBOSS package at the Belgian EMBL node andEMBL-EBI tools. GBJ002 sequences were compared to the genome of strainG9842 (GenBank accession number NC_011772), with which GBJ002 sharesmost sequence identity.

Phylogenetic footprinting. To discover evolutionary conserved elements, weused the footprint discovery program from RSAT (Regulatory Sequence AnalysisTools) (65) and followed the protocol as described for study case 2 (17). Thisstrategy has been systematically evaluated on E. coli and gives good results atdifferent taxonomical levels for well-conserved genes such as lexA (30). Thealgorithm was run on the lexA and recA promoter sequences from the B. cereusreference strain ATCC 14579 (GenBank accession number AE016877) and theGIL01 promoters.

RESULTS

The GIL01 genome has two major functional domains. Likemost phage genomes sequenced to date, the GIL01 genomecontains closely packed genes with few intergenic spaces. AllGIL01 genes are transcribed in the same rightward direction,

6010 FORNELOS ET AL. J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

and the ORFs are frequently overlapping, suggesting thatGIL01 genes are organized into several operons (Fig. 1A).Interestingly, genes with similar functions cluster together,forming two functional modules. The left arm of approximately4.8 kbp contains replication and regulatory genes, such as thepredicted terminal protein (ORF4), the DNA polymerase(ORF5), and two proteins with predicted DNA-binding do-mains (ORF1 and ORF6). The larger downstream module (10kbp) contains all the structural and assembly genes as well astwo endolysins (ORF26 and ORF30) involved in cell wall deg-radation (68). Two noncoding segments stand out by theirlength and position in the genome as containing potentialpromoter/regulatory elements. The first one lies in the leftextremity, upstream of ORF1 (355 bp), while the second oneseparates ORF8 from ORF9 (118 bp) (Fig. 1A). Based on thisgenetic organization, both noncoding segments were examinedfor the presence of promoters, using RACE PCR, a rapidmethod for the amplification of 5�-cDNA ends. The technique

relies upon amplifying RNA transcripts between a definedinternal site and the unknown 5� extremity of interest, thusallowing for the identification of the transcription start site.Using RNA extracted from a lysogenic culture in the exponen-tial growth phase, two transcription start sites were identifiedupstream of ORF1, at positions 175 and 278, and one initiationsite was identified in the intergenic region between ORF8 andORF9, at coordinate 4,915 (Fig. 1B). Centered approximately�10 and �35 nucleotides upstream of each transcription startsite are conserved sigma A-dependent promoter sequences(Fig. 1A and C). The two potential promoters on the left werenamed P1 and P2, and the internal promoter, P3. It is inter-esting to note that the �10 hexamer in P1 is preceded by a TGdinucleotide, positioned one base upstream of the �10 se-quence (Fig. 1C), a feature found in so-called extended �10promoters. In E. coli and B. subtilis, the TGn extension (wheren is any nucleotide) allows for closer contacts between theRNA polymerase sigma subunit and its promoters, especially

FIG. 1. Partial physical map of phage GIL01 with SOS-regulated promoters. (A) Predicted ORFs are depicted as open boxes, and left-to-rightarrows indicate the direction of transcription of the left and right functional domains. Probable gene functions are indicated above certain ORFs.Putative Rho-independent transcription terminators are shown as open stem loops downstream of ORF8 and ORF30. Promoters are shown asfilled arrowheads, and each promoter region is enlarged below the map. Boxes for �35 and �10 are depicted as black rectangles, and angled arrowsrepresent transcription start sites. Gray boxes show the locations of conserved LexA-binding sites in relation to the three promoters. (B) Tran-scription start site mapping of GIL01 ORF1 and ORF9 by 5� RACE. cDNAs were prepared from GIL01 mRNA, tailed with dCTP, and PCRamplified. The purified products were cloned into pCR4-TOPO and sequenced using the universal M13 primers. The electropherograms show thetranscription start sites (indicated with arrows; sequences correspond to the GIL01 negative strand) fused to their oligo(dC) tails. The promoterfrom which transcription is initiated is indicated for each electropherogram. Note that for the transcription start at P3, an additional thymineresidue that is not found in the GIL01 sequence has been inserted upstream of the dC tail. (C) Detailed nucleotide sequences of promotersdepicted in panel A. Transcription start sites detected in panel B are indicated with angled arrows (P1, P2, and P3). Putative ribosome-binding sitesare shown in lowercase, start codons are shown in bold, and SOS boxes are shaded gray. A TG dinucleotide motif characteristic of �10 extendedpromoters is double underlined within P1. Genome coordinates are indicated above each sequence.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

when a well-conserved �35 box is absent (5, 11, 34, 37, 58, 70).This could be the case for P1, where the �35 box does notbegin with T, the predominant base found at the beginning ofthe �35 hexamer in Gram-positive bacteria (70).

In addition to the reported promoter sequences, potentialribosome-binding sites were identified 10 and 6 nucleotidesupstream of the start codons of ORF1 and ORF9, respectively.Also, Rho-independent transcription terminators were identi-fied downstream of ORF8, where transcription of the replica-tion/regulation operon would be expected to terminate, anddownstream of ORF30, where transcription of the lysis/assem-bly operon is expected to terminate (Fig. 1A). We confirmedthat transcription does not span the regulatory region betweenORF8 and ORF9 by performing reverse transcription (RT)-PCR with primer sets that span this region (data not shown).Finally, lacZ reporter fusions were generated to the regionscontaining P1-P2 and P3 in order to verify that these arebiologically active promoters. As shown in Table 1, both pro-moter regions directed the expression of �-galactosidase in theGIL01 B. thuringiensis host. Such transcriptional organizationand the corresponding gene distribution further strengthen theidea that the GIL01 genome is transcribed as two major oper-ons, one responsible for genome replication and regulation,the other for virion structure, assembly, and host cell lysis.

GIL01 cp mutants map to three distinct loci. A chief objec-tive of this study was to identify the components of the geneticcircuit governing GIL01 development. The only obvious can-didates for transcription regulation were ORF1 and ORF6.PSI-BLAST searches identified a potential helix-turn-helix(HTH) DNA-binding motif in ORF1 similar to the HTHfound in the MerR superfamily of transcription regulators. InORF6, the same database searches revealed a potential wingedHTH DNA-binding domain similar to that found in the Nterminus of LexA, the master repressor of the SOS regulon.We reasoned that since GIL01 is induced in DNA-damagingconditions, most likely by way of the cellular SOS response, itis potentially under the transcriptional control of a cleavable

repressor analogous to CI in the phage �. However, neitherORF1/ORF6 nor any other protein-coding region in GIL01displays the typical structure of the CI/LexA class of repres-sors: an N-terminal DNA-binding domain, followed by an Ala-Gly or Cys-Gly cleavage site and a C-terminal protease do-main. Yet we did not rule out the existence of a distinct type ofSOS-dependent induction mechanism, such as the LexA-reg-ulated antirepressors found in coliphages 186 (8, 38) and N15(47), or that GIL01 induction relies on a yet-unknown SOSpathway.

To investigate these possibilities, we took advantage ofknowledge acquired from a previous study of spontaneousGIL01 cp mutants defective for the lysogenic cycle (67). SinceORF6 was a strong candidate for encoding a repressor of thelytic cycle, the first cp mutants had been sequenced within andin the vicinity of ORF6. Unexpectedly, no mutations weredetected in ORF6. Instead, mutations were found in the down-stream ORF7, the function of which was unknown. ORF7 cpmutants displayed either a deletion or insertion of 11 nucleo-tides in an almost perfect direct repeat (DR) (5�-CAAGTCGTAaG CAAGTCGTAtG; the lowercase letters represent theonly mismatch) at coordinates 4,615 through 4,636.

In this study, a more extensive collection of cp mutants wasassembled from multiple independent experiments and ana-lyzed. GIL01 cp mutants arose at a frequency of approximately10�4 and displayed otherwise normal growth properties. Inaddition to the expected DR mutations in ORF7 (deletions incp04 and cp18, insertions in cp25 and cp29), a mutant (cp08)carrying a single nucleotide substitution in the second repeat,resulting in a premature terminator codon, was also isolated.Another ORF7 mutant, cp27, has a single-nucleotide insertiondownstream of the DR (Fig. 2A). In every instance, the mu-tation altered the predicted protein sequence: cp04, cp08, cp18,cp25, and cp29 all resulted in truncation of ORF7, while cp27resulted in a modified C terminus (Fig. 2B).

Additionally, a class of GIL01 cp mutants that lacked mu-tations in ORF7 was isolated, suggesting that mutations inother regions of the GIL01 genome might account for their cpphenotype. This class of mutants all contained single-nucle-otide substitutions, deletions, or insertions within ORF1 (Fig.2A) or they contained mutations immediately upstream ofORF1 between promoters P1 and P2 (Fig. 1C and 3A). Inter-estingly, mutations in ORF1 cluster in two regions: the pre-dicted HTH DNA-binding motif (cp32, cp33, and cp36) or themost C-terminal residues (cp06 and cp10) (Fig. 2B). Mutationsin the noncoding region upstream of ORF1, hereafter termeddinBox1, resulted in base substitutions (cp16, cp19, cp23, andcp26) or single-nucleotide deletions (cp17 and cp37) at eitherof two nucleotide positions within a 14-bp sequence that bearsstrong similarity to the LexA-binding site in Gram-positivebacteria (Fig. 3B). As outlined below, a single mutation indinBox1 is sufficient to abolish the lysogenic pathway of GIL01.

To rule out the possibility that additional mutations, outsidethe sequenced areas, contributed to, or were responsible for,the observed cp phenotype, we sequenced the entire GIL01genome of cp26 (mutated in dinBox1), cp27/cp29 (mutated inORF7), and cp32/cp33 (mutated in ORF1). cp26 (dinBox1),cp27 (ORF7), and cp33 (ORF1) displayed only the initiallyidentified nucleotide substitutions (cp26 and cp33) and inser-tion (cp27). cp29 and cp32, however, each contained one ad-

TABLE 1. P1-P2 and P3 promoter activities in lexA� andlexA(A96D) backgrounds

Promoter fusion lexA allele

Mean �-galactosidaseactivity (Miller

units SD) in indicatedgrowth conditionsa

�MMC �MMC

P1-P2 lexA� 75 13 303 106lexA(A96D) 91 5 78 5

P1-P2 (dinBox1mut) lexA� 345 28 351 10lexA(A96D) 339 19 326 12

P3 lexA� 3 1 26 10lexA(A96D) 3 1 6 2

a lacZ fusions to P1-P2, P1-P2 mutated in dinBox1 (CGAACaagcTTTTT;boldface letter represents the mutated nucleotide), and P3 promoters weregenerated, and �-galactosidase activity was measured in both lexA� andlexA(A96D) lysogens in normal growth (�MMC) and SOS-inducing (�MMC)conditions. Cultures were grown for 3 h at 30°C when MMC50 was added toone-half of the cells, and growth was allowed to proceed for one additional hour.Cells were then permeabilized and assayed for �-galactosidase. Activity valuesare the averages of three independent experiments.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

FIG. 2. cp mutants in ORF1 and ORF7. Gene (A) and protein (B) sequences of cp mutants were aligned and compared to their respective wild-typesequences. Nucleotide coordinates (A) and the corresponding protein sizes (B) are indicated for each sequence. Start and stop codons are shown in bold,and mutations appear on a light gray background. (A) The direct repeat in ORF7 is highlighted against a black background in the wild-type sequence.(B) ORF1 harbors a putative helix-turn-helix DNA-binding motif in its N-terminal domain, and the two helices, 1 and 2, are shown boxed in blackin the wild-type sequence. 1 and 2 were predicted using the software programs Phyre (35) and ProteDNA (14).

6013


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

ditional mutation. In cp29, a single-nucleotide substitutionchanged a Gln to Lys (CAA3AAA) at the ORF29 residue 19,the function of which is currently unknown. cp32 contained anadditional deletion of three nucleotides at codon 64 withinORF19 (function unknown), resulting in the in-frame deletionof a single Ile residue. Yet cp29 and cp32 display the sameinfectious behavior as their cognate cp27 and cp33 mutants,which are mutated solely in ORF7 and ORF1, respectively (seecomplementation experiments below and Table 2). Thisstrongly suggests that ORF18 and ORF29 are not involved inthe regulatory network of GIL01. Taken together, our resultsindicate that dinBox1, ORF1, and ORF7 are critical for theestablishment or maintenance of the GIL01 lysogenic cycle.

Ultimately, we wished to classify cp mutants according totheir degree of virulence on lysogen and nonlysogen hosts.However, none of the cp mutants tested produced plaques onlawns of the lysogen strain GBJ338, indicating that GIL01lysogens are immune to infection by cp mutants. This could bedue to the activity of phage- and/or host-encoded immunityfactors that shift GIL01 toward lysogenic development. Alter-natively, we cannot rule out the possibility that infection wasbarred by a yet-unidentified phage-borne superinfection exclu-sion system.

FIG. 3. dinBox1 is a conserved LexA operator. (A) A class of GIL01 cp mutants harbor single-nucleotide mutations (shaded gray) indinBox1, a 14-bp predicted SOS box located between promoters P1 and P2 (Fig. 1A and C). The Bam35c operator differs from that of GIL01by one single nucleotide, and this difference is believed to account for the opposed propagation styles of the two phages (see the text fordetails). The GIL01 nucleotide coordinates are indicated above the wild-type sequence. (B) Putative LexA operators were detected upstreamof SOS genes in B. cereus strain G9842 (GenBank accession number NC_011772) and compared to the consensus sequence for LexA bindingin B. subtilis (24). In addition to dinBox1, two more potential SOS boxes, dinBox2 and dinBox3, were identified within promoter P3 (Fig. 1Aand C). The outer nucleotides, shown in capital letters, are highly conserved in SOS boxes of Gram-positive bacteria and are particularlyimportant for LexA binding (24). (C) Comparison of the N-terminal DNA-binding region of the B. cereus LexA protein to the consensussequence determined for Gram-positive bacteria (48). The three helices directly involved in DNA binding are underlined, and amino aciddifferences are shaded gray (24).

TABLE 2. Functions of defective gp1 and gp7can be complemented

Host(expressedprotein�s�)

Ratio of plating efficienciesa for indicatedcp mutant (relevant mutation)

cp23/cp26(dinBox1)

cp32/cp33(ORF1)

cp27/cp29(ORF7)

GBJ250 (gp1) 0.5 0.1b 0.2GBJ223 (gp7) 1 1b 0.5c

GBJ467 (gp1/gp7) 0.2 10�8 0.2c

a The GIL01-cured host strain (GBJ002) was transformed with plasmids ex-pressing either gp1 (GBJ250), gp7 (GBJ223), or gp1 and gp7 (GBJ467). Theresulting strains were grown exponentially before 1 mM IPTG was added toone-half of the culture, and growth was allowed to proceed for three additionalhours. Uninduced and induced cultures were then infected with GIL01 cp mu-tants in dinBox1 (cp23/cp26), ORF1 (cp32/cp33), or ORF7 (cp27/cp29). Twodifferent cp mutants at each locus were tested independently, and similar infec-tion patterns were obtained for each pair of mutants. The results shown arerepresentative of four independent experiments and give the ratio between titersproduced on induced cells and titers counted on the same uninduced strains(PFU on host � IPTG/PFU on host � IPTG).

b cp32 and cp33 formed small clear plaques on the cured host strain. There wasno systematic change in plaque morphology between induced and uninducedcells infected with ORF1 mutants.

c The plaques observed on induced lawns were turbid and significantly smallerthan those produced on the uninduced lawns (Fig. 4).



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

gp1 and gp7 regulate lysogenic development. Since muta-tions in either ORF1 or ORF7 commit GIL01 to multiplyexclusively as a lytic phage, we investigated more closely theroles of their respective gene products, gp1 and gp7, in GIL01development. The corresponding genes were cloned into the E.coli-B. subtilis shuttle vector pDG148, placing them under thetranscriptional control of the IPTG-inducible Pspac promoter.Three different plasmids, whereby gp1 and gp7 were expressedindividually or together as a single transcriptional unit, weregenerated. The permissive B. thuringiensis host GBJ002 wastransformed with the resulting plasmids, generating strainsGBJ250 (gp1), GBJ223 (gp7), and GBJ467 (gp1/gp7). Each ofthese strains was induced for protein expression and subjectedto infection with GIL01 cp mutants in dinBox1, ORF1, orORF7. The results of these experiments are summarized inTable 2. Expression of gp1 alone reduced the infective titers ofORF1 mutants cp32 and cp33 approximately 10-fold. Surpris-ingly, gp1 expression also reduced the plating efficiency ofdinBox1 and ORF7 mutants, although to a lesser extent (2-foldand 5-fold PFU reductions, respectively). This was not the casewhen gp7 was expressed in the host strain. In this case, areduction in phage titers was observed only when inducedGBJ223 cells were infected with the ORF7 mutants cp27 andcp29. Although titers showed only a 2-fold decrease, theplaques were now turbid and considerably smaller than thoseseen on the uninduced strain GBJ223 (Fig. 4). This observationwas less remarkable when ORF1 mutants were plated on in-duced cultures of GBJ250, as ORF1 mutants tend to producesmaller plaques in general.

Infection of these same induced cell lines with wild-typeGIL01 yielded titer reductions that varied between 2-fold and10-fold of those of infections of uninduced cultures, whileplaque morphology remained unaltered (data not shown). Thisnoticeable reduction in titers, as well as the considerable ex-periment-to-experiment variability in titers, could be explainedby variations in the cellular concentrations of gp1 and gp7generated with our plasmid-borne, IPTG-inducible system.Wild-type GIL01 probably expresses balanced levels of gp1and gp7 inside the host cell, and disrupting that balance islikely to affect phage behavior in many ways. These data areconsistent with the idea that gp1 and gp7, and possibly theproper expression rates of these two factors, play importantroles in the choice between lytic and lysogenic development.

Given the above results, we investigated whether there wasany effect due to the simultaneous expression of gp1 and gp7 in

the cell. There was indeed a substantial reduction in phagetiters when induced GBJ467 cells were infected with dinBox1mutants, and no plaques at all were observed with ORF1mutants (Table 2). In contrast, no difference in plating effi-ciency was observed between infections of induced GBJ250and GBJ467 with ORF7 mutants. Hence, coexpressing gp1 andgp7 in the same strain enhances complementation of dinBox1and ORF1 mutants but does not result in enhanced rescue ofORF7 mutants.

Since complementation was associated with sharp reduc-tions in infective titers, we wished to verify that expression ofgp1 and/or gp7 conferred immunity by preventing the infectingmutant from propagating, rather than by blocking infectionaltogether. In order to investigate the fate of the cp mutantspostinfection, bacterial samples were recovered from infectedspots and streaked on fresh plates containing IPTG. PCRs withGIL01-specific primers were performed on isolated colonies inwhich gp1 and/or gp7 was expressed and which had been in-fected with mutants cp26 (dinBox1), cp27 (ORF7), and cp33(ORF1). Induced GBJ223 colonies were found to be positivefor GIL01 only when infected with cp27, showing that expres-sion of gp7 complements the missing immunity function of thecorresponding cp mutant. GIL01-positive colonies were alsoobtained from GBJ250 spot infections with cp33. Similar re-sults were obtained when induced GBJ467 was infected witheach of the three cp mutants. These results are in agreementwith the infection patterns described above (Table 2), andaltogether, they support the idea that gp1 and gp7 act in im-munity against infection and, together with the dinBox1 ele-ment, play important roles in the regulatory network of GIL01.

dinBox1 is a conserved LexA box. dinBox1 showed strongsimilarity to the consensus LexA-binding site (CGAACn4GTTCG, where n is any nucleotide) in B. subtilis (13, 30,72). Its sequence, CGAACaagcGTTTT, is centered at posi-tions �17 and �86 with respect to the P1 and P2 transcriptionstart sites, respectively (Fig. 1C). cp mutants at dinBox1 wereunable to enter lysogeny, and mutations always involvedchanges to strongly conserved positions within the SOS box.Interestingly, Bam35c, a cp mutant of phage Bam35 (55), dif-fers from GIL01 at only a few nucleotide positions, one ofwhich is located within the critical bases for LexA binding todinBox1 (Fig. 3A). In agreement with this finding, experimen-tal data for the LexA box sequence requirements in B. subtilisshow that changes to any one of the five outer nucleotides canconsiderably reduce or prevent LexA binding (24). In addition

FIG. 4. gp7 is involved in immunity against infection. GBJ223 was grown exponentially, and gp7 expression was induced by adding 1 mM IPTGto one-half of the culture. Uninduced and induced cultures were then allowed to grow for three more hours. Cells were then infected with cp29,a mutant in ORF7 (Fig. 2), and plated.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

to conserved dinBox1, two motifs with similar characteristicswere found near promoter P3. dinBox2 and dinBox3 are cen-tered at �22 and �11 relative to the start of transcription (Fig.1C). It is worth noting that, while numerous mutations indinBox1 were isolated, none of the cp mutants isolated in thisstudy carried mutations in dinBox2 or dinBox3.

To further consolidate our observations, we searched forSOS box sequences in the upstream regions of SOS genes in B.cereus strain G9842, which is closest to strain GBJ002 in DNAsequence. We compared the LexA operators found upstreamof genes involved in transcriptional regulation (lexA), recom-bination (recA and ruvAB), and DNA repair (uvrAB) to theconsensus sequence for LexA binding in B. subtilis and to thethree dinBox sequences identified in GIL01. As shown in Fig.3B, the B. cereus SOS boxes and the three GIL01 dinBoxsequences closely match the B. subtilis consensus SOS box.Additionally, an in silico approach using the program RSAT(17) identified the same three dinBox sites found by visualanalysis of the GIL01 promoter sequences (Fig. 3B). Lastly,alignment of the B. cereus LexA protein sequence with theGram-positive consensus LexA sequence revealed strong con-servation of the N-terminal regions that make close contactswith DNA (Fig. 3C). Taken together, these observationsstrongly suggest that the dinBox sequences identified in GIL01function as LexA-binding sites in the B. cereus group of bac-teria.

GIL01 is induced in DNA-damaging conditions in a LexA-dependent manner. In a previous study, we reported thatGIL01 was induced by DNA-damaging agents such as MMC,nalidixic acid, or exposure to UV irradiation (69). Such treat-ments are known to elicit the cellular SOS response in a RecA-dependent manner (71). To investigate if RecA is involved inthe induction cycle of GIL01, we disrupted the B. thuringiensisrecA gene through insertional mutagenesis (66). The resultingstrain, GBJ499, was particularly sensitive to UV and MMCtreatment and displayed a slower growth rate than the wild-type strain (data not shown). GBJ499 was infected with GIL01to generate the lysogen GBJ500. Phage titers produced byspontaneous induction of GBJ500 were reduced approximately10-fold compared to those of the wild-type lysogen GBJ338(Table 3). In addition, whereas GIL01 was strongly inducedafter UV or MMC treatment of the wild-type lysogen strain,there was no increase in phage titers when GBJ500 was simi-larly induced (Table 3, row 2). These results indicate thatGIL01 induction is RecA dependent and potentially linked tothe B. thuringiensis SOS response pathway. To investigate thishypothesis, the role of cellular LexA in the lytic switch wasassessed. We were unable to isolate a lexA null mutant of B.

thuringiensis. This result was not entirely unexpected, consid-ering that LexA represses the expression of cell division inhib-itors in other bacterial groups (sulA in E. coli and yneA in B.subtilis). In E. coli, lexA null mutants are not viable unless sulAis first inactivated (52). In B. subtilis, lexA-deficient cells growinto long filaments (25) as a consequence of YneA upregula-tion (33). The B. cereus LexA is 70% identical to its counter-part in B. subtilis, and most of the shared residues lie in theN-terminal DNA-binding domain (Fig. 3C). Most importantly,B. cereus LexA contains an Ala96-Gly97 peptide sequence thatin B. subtilis and E. coli (at Ala91-Gly92 in both organisms) isthe site of RecA-stimulated autoproteolysis (28, 51). Certainamino acid changes at the Ala-Gly cleavage site are knownto abolish LexA proteolysis (19, 42). We used a similarapproach to construct a noncleavable lexA mutant by alleleexchange in B. thuringiensis, replacing the Ala codon with anAsp codon [lexA(A96D)]. As expected, the resulting strain,GBJ396, was hypersensitive to DNA damage conditions,displaying a 300-fold reduction in survival after treatmentwith MMC, compared to a 9-fold reduction in survival ofMMC-treated wild-type cells. Furthermore, a lexA-lacZ reporterfusion, when introduced into GBJ002 (lexA�), showed an ex-pected 4-fold increase in lacZ expression after treatment withMMC. However, there was no increase in lacZ expression afterSOS treatment of GBJ396 [lexA(A96D)] carrying the same lexA-lacZ reporter plasmid, indicating that LexA was not proteo-lyzed and therefore not released from its operator sites by SOSinduction signals (data not shown). The lexA(A96D) mutantwas infected with GIL01 to generate the lysogen strainGBJ406. Similar to what we observed for the recA mutantlysogen, GIL01 induction from GBJ406 was strongly reducedin comparison to that of the wild-type host strain (Table 3, row3). The data obtained with both recA and lexA mutants showthat GIL01 is induced as part of the host SOS response path-way.

In order to assess if promoters P1 and P2 are regulated byLexA, we determined whether a 375-bp region encompassingboth promoters and dinBox1 could direct transcription of apromoterless lacZ gene upon SOS induction. The 375-bp seg-ment was cloned into the low-copy-number pHT304-18Z plas-mid (1), transcriptionally fusing the 5� end of ORF1 to the lacZreporter gene. The construct was transformed to GBJ338(lexA�) and GBJ338 [lexA(A96D)], and �-galactosidase expres-sion from the P1-P2-lacZ fusion was measured with and with-out SOS induction by MMC. Table 1 shows that expressionfrom promoters P1 and P2 was induced approximately 4-fold inthe wild-type host in SOS conditions, whereas no increase in�-galactosidase levels was observed in the lexA(A96D) strain.

TABLE 3. GIL01 induction is recA and lexA dependenta

Strain [relevant genotype] Phage titerwithout MMC

Phage titer withMMCb Ratio Phage titer

without UVPhage titerwith UVc Ratio

GBJ338 �GBJ002 (GIL01)� 3.1 � 107 1.9 � 109 61.3 4.0 � 107 2.7 � 109 67.5GBJ500 (GBJ338 recA::pMutin4) 4.0 � 106 3.0 � 106 0.8 1.1 � 106 2.0 � 106 1.8GBJ406 �GBJ338 lexA(A96D)� 1.5 � 105 1.8 � 105 1.2 6.7 � 104 5.3 � 104 0.8

a Phage titers were determined by plating serial dilutions of GIL01 on the cured host GBJ002. Induction ratios were determined by dividing phage titers of inducedcultures by the titers of uninduced cultures. Results are averages from three independent experiments.

b The MMC concentration used was 50 ng ml�1, and cultures were induced for 1 h.c Cells were irradiated at 254 nm for 10 s.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

Moreover, �-galactosidase expression from a P1-P2-lacZ fu-sion in which dinBox1 was mutated at a critical residue was nolonger repressed in either GBJ338 (lexA�) or GBJ338[lexA(A96D)], and MMC treatment did not further increaseLacZ expression. Taken together, these data strongly indicatethat LexA regulates P1 and P2 and that dinBox1 is necessaryfor this LexA-mediated regulation. Parallel experiments withpromoter P3, in which a 226-bp fragment that includes dinBox2and dinBox3 was fused to lacZ, yielded a 9-fold increase in�-galactosidase activity in the wild-type host after MMC treat-ment. Once again, the increase in lacZ expression levels afterMMC treatment was eliminated when the same experimentwas performed in a host carrying the lexA(A96D) mutation.These experiments demonstrate that host LexA regulates pro-moters P1, P2, and P3, most likely by binding to the dinBoxsequences and preventing transcription initiation.

DISCUSSION

In this study, we used a genetic approach to identify centralcomponents of the system that regulates the temperate state inphage GIL01. We provided evidence that the host SOS repres-sor LexA is directly involved in the switch but that phage-bornefactors gp1 and gp7 are also necessary for lysogeny. LexA islikely to act by binding to dinBox1, a conserved SOS box lyingbetween two newly identified promoters upstream of lysogenyand replication genes. Two additional conserved LexA-bindingsites were also identified upstream of structure and lysis genes,further suggesting that LexA plays a central role in GIL01development. Our results led to the conclusion that GIL01uses a combination of host- and phage-derived regulators tocontrol its cycle.

At first glance, GIL01 does not appear to display the char-acteristic modular organization usually encountered in temper-ate phages. Its 15-kbp genome is smaller than most sequencedprophage genomes, and all 30 ORFs are transcribed in thesame rightward direction, a rarely observed feature. In addi-tion, only a few genes have attributed functions, making itdifficult to define the boundaries of gene clusters involved inrelated functions. Most notably, a lysogeny module is not read-ily apparent, as there are only two gene candidates, ORF1 andORF6, predicted to code for DNA-binding proteins that couldfunction as transcription regulators. Nevertheless, the identi-fication of promoters in large noncoding segments and Rho-independent terminators at the end of transcriptional unitsrevealed a functional organization in two major modules (Fig.1A). Two adjacent promoters, P1 and P2, at the left extremityof the phage genome, control transcription of lysogeny andreplication functions, whereas a probable single internal pro-moter, P3, directs the expression of structural and lytic genes.Such transcriptional organization typically allows for complex,multilayered regulation of gene transcription. It would, forinstance, permit the expression of regulatory factors and phageDNA polymerase to levels that are required for lysogeny whilekeeping lytic genes fully repressed. Conversely, upon percep-tion of an inducing signal, synthesis of terminal protein andDNA polymerase would be significantly enhanced along withinduction of structural and lytic genes.

A similar overall genome organization can be found in the B.subtilis lytic phage �29, with the notable difference that early

and late genes are divergently transcribed. Tandem promotersA2b and A2c jointly direct the expression of the leftward reg-ulation and replication genes, which are similarly distributed inGIL01 (49). Both promoters are strong, and their repression bytwo early factors coincides with late gene transcription activa-tion from the flanking divergent promoter A3, a determiningstep in the switch from early to late transcription (11, 50). Thecoupling of tandem promoters to the same operator is fre-quently observed in bacteria (examples are the rRNA operons[23, 73], crp gene [22], and fis operon [53]) and offers a certainflexibility for differential gene regulation. In the autonomouslyreplicating coliphage N15, for example, two adjacent promotersdirect gene expression from the anti-immunity locus immA, andtheir differential control is important for the choice between lysisand lysogeny (56). Similarly arranged promoters regulate anti-immunity operons in distinct phages, such as P1, P4, and P7 (56),suggesting that this regulatory strategy is widespread among tem-perate phages. It is therefore conceivable that in GIL01, promot-ers P1 and P2 display distinct strengths, are subject to variablelevels of regulation, or give rise to different transcripts.

GIL01 lytic development is induced by DNA-damagingtreatments that elicit the cellular SOS response, and yet GIL01does not code for a cleavable CI-like repressor. Instead, wefound that regulation is mediated by host LexA, since thephage lytic pathway was no longer induced in a noncleavablelexA (Ind�) background. Accordingly, three putative LexA op-erators that bore a strong resemblance to the consensus LexA-binding site in Gram-positive bacteria were identified betweenpromoters P1 and P2 (dinBox1) and at P3 (dinBox2 and din-Box3). We showed that the SOS induction of both promoterregions was blocked by a noncleavable lexA (Ind�) host muta-tion. Moreover, a single-nucleotide mutation at a strongly con-served position within dinBox1 completely eliminated tran-scription repression at P1-P2. Although additional experimentswill be needed in order to confirm LexA binding to dinBox1,these data lend credence to the idea that host LexA is directlyinvolved in the GIL01 control circuit.

Whereas numerous cp mutants were isolated in dinBox1, nomutations were ever detected in the downstream dinBox2 ordinBox3. This suggests that dinBox2 and dinBox3 are eitheressential for phage development or, if such mutants are indeedviable, they cannot be detected as cp formers. Also, dinBox2and dinBox3 could perform an additional function in the reg-ulatory system, such as in LexA-mediated DNA looping be-tween dinBox2/3 and the dinBox1 promoter region. In this case,a cp mutant might not be obtained unless both boxes weresimultaneously mutated, which is likely to occur at too low afrequency to be detected. In any event, LexA binding to bothsites provides the basis for a plausible model in which a cleav-able repressor would be immediately responsible for lytic geneexpression. However, it is unlikely that LexA is the sole regu-lator of the switch, as this would lead to inefficient GIL01induction. Due to the negative autoregulation of lexA tran-scription, normal LexA levels are restored once DNA damageinside the cell has been repaired (7, 44). LexA would then beable to bind its operators in GIL01 again and interrupt lyticdevelopment. In the � phage, for instance, this is prevented bystrong repression of CI synthesis once lytic development com-mences (60). It is therefore conceivable that LexA plays anancillary role only in the regulatory network by controlling the



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

expression of phage-borne repressor(s) and activator(s), thesein turn being responsible for the lytic switch per se. This hy-pothesis is strongly supported by the isolation of GIL01 cpmutants deficient for gp1 or gp7. Both proteins are requiredfor lysogeny, and their absence can be complemented by ex-pressing the missing functions in trans. Interestingly, cross-complementation is observed with gp1, whereas it is not seenwith gp7. The observed differences could be attributed to dis-tinct and/or independent roles of gp1 and gp7, gp1 potentiallyhaving a broader activity range. In addition, since the expres-sion system used in this study does not provide any insight intothe protein concentrations generated, it is quite probable thatthey differ from the intracellular levels produced by GIL01during infection. This could also explain why wild-type GIL01is not fully repressed when infecting cells that express both gp1and gp7. One would expect the wild-type phage cycle to behindered by the two factors known to be involved in lysogeny,and yet titers are only slightly reduced. This observation sug-gests that there is a need for precise levels of both factors and,probably, a determined timing of expression as well.

gp1 displays a DNA-binding domain similar to the onefound in the MerR (repressor of mercury Resistance operons)superfamily of transcription regulators. MerR itself is knownto mediate both repression and activation of transcription byvirtue of its ability to distort the DNA helix (3, 4). A role forgp1 as a modulator of DNA structure is an attractive idea, asit is a common theme among transcription regulators. gp7, onthe other hand, has no predicted function or recognizablesequence motif. It might cooperate with gp1 in the same reg-ulatory process or it could act independently, at a completelydifferent level of immunity regulation.

Temperate phages that are under the direct control of hostLexA utilize a variety of regulatory mechanisms. Coliphages186 and N15 code for LexA-regulated antirepressors (8, 47),and gene homologues that are preceded by LexA-binding siteshave been identified in other phages (9, 12, 27). A uniqueexample where LexA exerts a dual function is provided by theVibrio cholerae filamentous phage CTX. The phage-encodedrepressor RstR acts together with host LexA to activate andrepress gene transcription (36, 54). Such a mechanism allowsthe phage to respond to small variations in RstR/LexA levelsand switch from a low-particle-formation mode to higher yieldsand vice versa. This regulatory system appears to be well suitedfor a filamentous phage, the production of which is maintainedthroughout the cell cycle.

GIL01 cp mutants were isolated in three distinct loci—dinBox1, ORF1, and ORF7—each of which is necessary for thelysogenic cycle. Surprisingly, none of the isolated mutants pro-duced plaques on lysogens, a result that we expected to seewith at least the operator mutants in dinBox1. There are manypossible explanations for the observed immunity effect. One isthat GIL01 codes for an as-yet-unidentified superinfection ex-clusion system that blocks entry of newly infecting phages. Amore attractive explanation, however, would be that dinBox1 isnot the only operator needed for lysogeny. If other regulatorysites are present on the phage genome, potentially respondingto gp1 and/or gp7, these sites could constitute themselves aLexA-independent pathway for regulation and compensate forthe loss of the dinBox1 pathway. In this case, a mutant, super-infecting phage would still be sensitive to repression upon

infection of a lysogen, provided that the appropriate immunityfactors are readily available in the cell. It is noteworthy thateven though the dinBox1, ORF1, and ORF7 alleles were inde-pendently isolated multiple times, they probably do not repre-sent a saturated screen of all possible mutations leading to thecp phenotype. We therefore do not exclude the possibility thatyet-unidentified factors also participate in GIL01 regulation.

GIL01 is regulated by a multicomponent system composedof at least host LexA, gp1, and gp7. The combination of host-and phage-encoded functions in the lytic switch has alreadybeen seen in other phages and enables a fine regulation of lyticfunctions. LexA is used as an internal sensor of stress andallows for the rapid escape from potentially deleterious dam-age to the host. The addition of phage-borne functions couldmodulate the switch by conferring more precision to the systemor by adding a further layer of control. GIL01 has developedan elaborate network to control its development, and futurestudies will address how the different components identified inthis study intervene in the switch.

ACKNOWLEDGMENTS

We are grateful to B. Clement and S. Degand for their contributionsto the early stages of this work and to A. Toussaint and A. Aertsen forstimulating discussions. We thank R. Janky for continuous encourage-ment and skillful advice with RSAT and R. Leplae for assistance withprotein analysis. We are particularly indebted to H. Kimsey for insight-ful observations and comments on the manuscript. We thank C. Con-don for kindly providing the pDG148 vector.

This work was supported by grants from the National Fund forScientific Research and the Universite Catholique de Louvain (grantsto N.F. and J.M.). Part of this work was also supported by the FinnishCentre of Excellence Program of the Academy of Finland (grant 1129648to J.K.H.B.), an EMBO short-term fellowship, and the National ResearchFund in Luxembourg, cofunded under the Marie Curie Actions of theEuropean Commission (FP7-COFUND) (grants to N.F.).

REFERENCES

1. Agaisse, H., and D. Lereclus. 1994. Structural and functional analysis of thepromoter region involved in full expression of the cryIIIA toxin gene ofBacillus thuringiensis. Mol. Microbiol. 13:97–107.

2. Andrup, L., J. Damgaard, and K. Wassermann. 1993. Mobilization of smallplasmids in Bacillus thuringiensis subsp. israelensis is accompanied by specificaggregation. J. Bacteriol. 175:6530–6536.

3. Ansari, A. Z., J. E. Bradner, and T. V. O’Halloran. 1995. DNA-bend modula-tion in a repressor-to-activator switching mechanism. Nature 374:371–375.

4. Ansari, A. Z., M. L. Chael, and T. V. O’Halloran. 1992. Allosteric under-winding of DNA is a critical step in positive control of transcription byHg-MerR. Nature 355:87–89.

5. Barne, K. A., J. A. Bown, S. J. Busby, and S. D. Minchin. 1997. Region 2.5of the Escherichia coli RNA polymerase �70 subunit is responsible for therecognition of the �extended-10� motif at promoters. EMBO J. 16:4034–4040.

6. Bertani, G. 1951. Studies on lysogenesis. I. The mode of phage liberation bylysogenic Escherichia coli. J. Bacteriol. 62:293–300.

7. Brent, R., and M. Ptashne. 1980. The lexA gene product represses its ownpromoter. Proc. Natl. Acad. Sci. U. S. A. 77:1932–1936.

8. Brumby, A. M., I. Lamont, I. B. Dodd, and J. B. Egan. 1996. Defining theSOS operon of coliphage 186. Virology 219:105–114.

9. Bunny, K., J. Liu, and J. Roth. 2002. Phenotypes of lexA mutations inSalmonella enterica: evidence for a lethal lexA null phenotype due to theFels-2 prophage. J. Bacteriol. 184:6235–6249.

10. Butala, M., D. Zgur-Bertok, and S. J. Busby. 2009. The bacterial LexAtranscriptional repressor. Cell. Mol. Life Sci. 66:82–93.

11. Camacho, A., and M. Salas. 2001. Mechanism for the switch of �29 DNAearly to late transcription by regulatory protein p4 and histone-like proteinp6. EMBO J. 20:6060–6070.

12. Casjens, S. R., et al. 2004. The pKO2 linear plasmid prophage of Klebsiellaoxytoca. J. Bacteriol. 186:1818–1832.

13. Cheo, D. L., K. W. Bayles, and R. E. Yasbin. 1991. Cloning and character-ization of DNA damage-inducible promoter regions from Bacillus subtilis. J.Bacteriol. 173:1696–1703.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

14. Chu, W. Y., et al. 2009. ProteDNA: a sequence-based predictor of sequence-specific DNA-binding residues in transcription factors. Nucleic Acids Res.37:W396–401.

15. Court, D., L. Green, and H. Echols. 1975. Positive and negative regulation bythe cII and cIII gene products of bacteriophage lambda. Virology 63:484–491.

16. Cutting, S. M., and P. B. Vander Horn. 1990. Genetic analysis, p. 27–74. InC. R. Harwood and S. M. Cutting (ed.), Molecular biological methods forBacillus. John Wiley & Sons, Chichester, England.

17. Defrance, M., R. Janky, O. Sand, and J. van Helden. 2008. Using RSAToligo-analysis and dyad-analysis tools to discover regulatory signals in nucleicsequences. Nat. Protoc. 3:1589–1603.

18. Erill, I., S. Campoy, and J. Barbe. 2007. Aeons of distress: an evolutionaryperspective on the bacterial SOS response. FEMS Microbiol. Rev. 31:637–656.

19. Fabret, C., S. D. Ehrlich, and P. Noirot. 2002. A new mutation deliverysystem for genome-scale approaches in Bacillus subtilis. Mol. Microbiol.46:25–36.

20. Gaidelyte, A., V. Cvirkaite-Krupovic, R. Daugelavicius, J. K. Bamford, andD. H. Bamford. 2006. The entry mechanism of membrane-containing phageBam35 infecting Bacillus thuringiensis. J. Bacteriol. 188:5925–5934.

21. Gicquel-Sanzey, B., and P. Cossart. 1982. Homologies between differentprocaryotic DNA-binding regulatory proteins and between their sites ofaction. EMBO J. 1:591–595.

22. Gonzalez-Gil, G., R. Kahmann, and G. Muskhelishvili. 1998. Regulation ofcrp transcription by oscillation between distinct nucleoprotein complexes.EMBO J. 17:2877–2885.

23. Gourse, R. L., H. A. de Boer, and M. Nomura. 1986. DNA determinants ofrRNA synthesis in E. coli: growth rate dependent regulation, feedback in-hibition, upstream activation, antitermination. Cell 44:197–205.

24. Groban, E. S., et al. 2005. Binding of the Bacillus subtilis LexA protein to theSOS operator. Nucleic Acids Res. 33:6287–6295.

25. Haijema, B. J., et al. 1996. Regulated expression of the dinR and recA genesduring competence development and SOS induction in Bacillus subtilis. Mol.Microbiol. 22:75–85.

26. Hendrix, R. W., J. W. Roberts, F. W. Stahl, and R. A. Weisberg (ed.). 1983.Lambda II. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

27. Hertwig, S., et al. 2003. Sequence analysis of the genome of the temperateYersinia enterocolitica phage PY54. J. Mol. Biol. 331:605–622.

28. Horii, T., et al. 1981. Regulation of SOS functions: purification of E. coliLexA protein and determination of its specific site cleaved by the RecAprotein. Cell 27:515–522.

29. Ivanova, N., et al. 2003. Genome sequence of Bacillus cereus and compara-tive analysis with Bacillus anthracis. Nature 423:87–91.

30. Janky, R., and J. van Helden. 2008. Evaluation of phylogenetic footprintdiscovery for predicting bacterial cis-regulatory elements and revealing theirevolution. BMC Bioinformatics 9:37.

31. Jensen, G. B., L. Andrup, A. Wilcks, L. Smidt, and O. M. Poulsen. 1996. Theaggregation-mediated conjugation system of Bacillus thuringiensis subsp.israelensis: host range and kinetics of transfer. Curr. Microbiol. 33:228–236.

32. Jensen, G. B., et al. 1995. The genetic basis of the aggregation system inBacillus thuringiensis subsp. israelensis is located on the large conjugativeplasmid pXO16. J. Bacteriol. 177:2914–2917.

33. Kawai, Y., S. Moriya, and N. Ogasawara. 2003. Identification of a protein,YneA, responsible for cell division suppression during the SOS response inBacillus subtilis. Mol. Microbiol. 47:1113–1122.

34. Keilty, S., and M. Rosenberg. 1987. Constitutive function of a positivelyregulated promoter reveals new sequences essential for activity. J. Biol.Chem. 262:6389–6395.

35. Kelley, L. A., and M. J. Sternberg. 2009. Protein structure prediction on theWeb: a case study using the Phyre server. Nat. Protoc. 4:363–371.

36. Kimsey, H. H., and M. K. Waldor. 2009. Vibrio cholerae LexA coordinatesCTX prophage gene expression. J. Bacteriol. 191:6788–6795.

37. Kumar, A., et al. 1993. The minus 35-recognition region of Escherichia colisigma 70 is inessential for initiation of transcription at an “extended minus10” promoter. J. Mol. Biol. 232:406–418.

38. Lamont, I., A. M. Brumby, and J. B. Egan. 1989. UV induction of coliphage186: prophage induction as an SOS function. Proc. Natl. Acad. Sci. U. S. A.86:5492–5496.

39. Laurinavicius, S., D. H. Bamford, and P. Somerharju. 2007. Transbilayerdistribution of phospholipids in bacteriophage membranes. Biochim. Bio-phys. Acta 1768:2568–2577.

40. Laurinavicius, S., R. Kakela, P. Somerharju, and D. H. Bamford. 2004.Phospholipid molecular species profiles of tectiviruses infecting Gram-neg-ative and Gram-positive hosts. Virology 322:328–336.

41. Laurinmaki, P. A., J. T. Huiskonen, D. H. Bamford, and S. J. Butcher. 2005.Membrane proteins modulate the bilayer curvature in the bacterial virusBam35. Structure 13:1819–1828.

42. Lin, L. L., and J. W. Little. 1988. Isolation and characterization of noncleav-able (Ind�) mutants of the LexA repressor of Escherichia coli K-12. J.Bacteriol. 170:2163–2173.

43. Little, J. W. 1984. Autodigestion of lexA and phage � repressors. Proc. Natl.Acad. Sci. U. S. A. 81:1375–1379.

44. Little, J. W., and J. E. Harper. 1979. Identification of the lexA gene productof Escherichia coli K-12. Proc. Natl. Acad. Sci. U. S. A. 76:6147–6151.

45. Luo, Y., et al. 2001. Crystal structure of LexA: a conformational switch forregulation of self-cleavage. Cell 106:585–594.

46. Mahillon, J., and D. Lereclus. 2000. Electroporation of Bacillus thuringiensisand Bacillus cereus, p. 242–252. In J. Teissie and N. Eynard (ed.), Electro-transformation of bacteria. Springer Lab Manual. Springer-Verlag, Berlin,Germany.

47. Mardanov, A. V., and N. V. Ravin. 2007. The antirepressor needed forinduction of linear plasmid-prophage N15 belongs to the SOS regulon. J.Bacteriol. 189:6333–6338.

48. Mazon, G., et al. 2004. Reconstruction of the evolutionary history of theLexA-binding sequence. Microbiology 150:3783–3795.

49. Meijer, W. J., J. A. Horcajadas, and M. Salas. 2001. Phi29 family of phages.Microbiol. Mol. Biol. Rev. 65:261–287.

50. Mencía, M., M. Monsalve, M. Salas, and F. Rojo. 1996. Transcriptionalactivator of phage phi 29 late promoter: mapping of residues involved ininteraction with RNA polymerase and in DNA bending. Mol. Microbiol.20:273–282.

51. Miller, M. C., J. B. Resnick, B. T. Smith, and C. M. Lovett, Jr. 1996. TheBacillus subtilis dinR gene codes for the analogue of Escherichia coli LexA.Purification and characterization of the DinR protein. J. Biol. Chem. 271:33502–33508.

52. Mount, D. W. 1977. A mutant of Escherichia coli showing constitutive ex-pression of the lysogenic induction and error-prone DNA repair pathways.Proc. Natl. Acad. Sci. U. S. A. 74:300–304.

53. Nasser, W., M. Rochman, and G. Muskhelishvili. 2002. Transcriptionalregulation of fis operon involves a module of multiple coupled promoters.EMBO J. 21:715–724.

54. Quinones, M., H. H. Kimsey, and M. K. Waldor. 2005. LexA cleavage isrequired for CTX prophage induction. Mol. Cell 17:291–300.

55. Ravantti, J. J., A. Gaidelyte, D. H. Bamford, and J. K. Bamford. 2003.Comparative analysis of bacterial viruses Bam35, infecting a gram-positivehost, and PRD1, infecting gram-negative hosts, demonstrates a viral lineage.Virology 313:401–414.

56. Ravin, N. V., A. N. Svarchevsky, and G. Deho. 1999. The anti-immunitysystem of phage-plasmid N15: identification of the antirepressor gene and itscontrol by a small processed RNA. Mol. Microbiol. 34:980–994.

57. Sambrook, J., and D. Russell. 2001. Molecular cloning: a laboratory manual,3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

58. Sanderson, A., J. E. Mitchell, S. D. Minchin, and S. J. Busby. 2003. Substitu-tions in the Escherichia coli RNA polymerase sigma70 factor that affect recog-nition of extended �10 elements at promoters. FEBS Lett. 544:199–205.

59. Sauer, R. T., M. J. Ross, and M. Ptashne. 1982. Cleavage of the lambda andP22 repressors by recA protein. J. Biol. Chem. 257:4458–4462.

60. Schubert, R. A., I. B. Dodd, J. B. Egan, and K. E. Shearwin. 2007. Cro’s rolein the CI Cro bistable switch is critical for �’s transition from lysogeny to lyticdevelopment. Genes Dev. 21:2461–2472.

61. Shearwin, K. E., A. M. Brumby, and J. B. Egan. 1998. The Tum protein ofcoliphage 186 is an antirepressor. J. Biol. Chem. 273:5708–5715.

62. Sozhamannan, S., et al. 2008. Molecular characterization of a variant ofBacillus anthracis-specific phage AP50 with improved bacteriolytic activity.Appl. Environ. Microbiol. 74:6792–6796.

63. Stragier, P., C. Bonamy, and C. Karmazyn-Campelli. 1988. Processing of asporulation sigma factor in Bacillus subtilis: how morphological structurecould control gene expression. Cell 52:697–704.

64. Stromsten, N. J., S. D. Benson, R. M. Burnett, D. H. Bamford, and J. K.Bamford. 2003. The Bacillus thuringiensis linear double-stranded DNAphage Bam35, which is highly similar to the Bacillus cereus linear plasmidpBClin15, has a prophage state. J. Bacteriol. 185:6985–6989.

65. Thomas-Chollier, M., et al. 2008. RSAT: regulatory sequence analysis tools.Nucleic Acids Res. 36:W119–W127.

66. Vagner, V., E. Dervyn, and S. D. Ehrlich. 1998. A vector for systematic geneinactivation in Bacillus subtilis. Microbiology 144:3097–3104.

67. Verheust, C., N. Fornelos, and J. Mahillon. 2005. GIL16, a new gram-positive tectiviral phage related to the Bacillus thuringiensis GIL01 and theBacillus cereus pBClin15 elements. J. Bacteriol. 187:1966–1973.

68. Verheust, C., N. Fornelos, and J. Mahillon. 2004. The Bacillus thuringiensisphage GIL01 encodes two enzymes with peptidoglycan hydrolase activity.FEMS Microbiol. Lett. 237:289–295.

69. Verheust, C., G. Jensen, and J. Mahillon. 2003. pGIL01, a linear tectiviralplasmid prophage originating from Bacillus thuringiensis serovar israelensis.Microbiology 149:2083–2092.

70. Voskuil, M. I., and G. H. Chambliss. 1998. The �16 region of Bacillus subtilisand other gram-positive bacterial promoters. Nucleic Acids Res. 26:3584–3590.

71. Walker, G. C. 1984. Mutagenesis and inducible responses to deoxyribonu-cleic acid damage in Escherichia coli. Microbiol. Rev. 48:60–93.

72. Winterling, K. W., et al. 1998. The Bacillus subtilis DinR binding site: redef-inition of the consensus sequence. J. Bacteriol. 180:2201–2211.

73. Young, R. A., and J. A. Steitz. 1979. Tandem promoters direct E. coliribosomal RNA synthesis. Cell 17:225–234.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

phage-borne factors and host lexa regulate the lytic ...jb.asm.org/content/193/21/6008.full.pdf ·...

Documents