Vol. 167, No. 3JOURNAL OF BACTERIOLOGY, Sept. 1986, p. 952-9590021-9193/86/090952-08$02.00/0Copyright © 1986, American Society for Microbiology

Control of Lysogeny and Immunity of Bacillus subtilis TemperateBacteriophage SPI by Its d Gene

JANE R. McLAUGHLIN, HING C. WONG, YI E. TING, JANELLE N. VAN ARSDELL, AND SHING CHANG*Department of Microbial Genetics, Cetus Corporation, Emeryville, California 94608

Received 18 June 1985/Accepted 2 June 1986

The d gene from the Bacillus subtilis temperate bacteriophage SP,3 was isolated. When introduced into anSP,l-sensitive strain of B. subtilis, the cloned d gene directed the synthesis of a 22-kilodalton protein andconferred on the host immunity to SP,i phage. A frameshift mutation, designated d2, was introduced into thecloned d gene, and it was subsequently crossed back into the SPI8 phage genome. The resulting SPjI phage grewlyticaily and formed clear plaques on sensitive bacteria. Although the cloned d gene confers immunity to thehost, we could not detect complementation of the d gene by mixed infection with SPO d2 and various SP,I cmutants. The nucleotide sequence of the 1,033-base-pair PstI-to-EcoRI fragment containing the d gene wasdetermined; it includes an open reading frame that could potentially encode a protein of 227 amino acids. Thegene was mapped within the PstI H fragment on the phage genome, which positions the d gene about 25kilobases from the right end of the phage genome. It is transcribed from right to left.

Bacillus subtilis 168 and most of its derivatives arelysogenic for the temperate bacteriophage SP, (18). Thisomnipresence of the SP,B prophage implies the existence ofan efficient mechanism for the establishment and mainte-nance of lysogeny. The available data indicate that therepressor protein encoded by the c gene plays a central rolein the establishment of lysogeny by turning off lytic-cyclegenes (16, 18). Wild-type SP, (c+) phage can be induced bymitomycin C; it replicates at temperatures below 50TC andforms cloudy plaques on sensitive bacteria (16). The SP,phage carrying the thermoinducible repressor gene (SP, c2)also forms cloudy plaques under normal growth tempera-ture, but when in the prophage state it can be induced by abrief heat shock of the lysogens at 50°C (13). In contrast,clear-plaque mutants such as SP,B cl are unable tolysogenize sensitive B. subtilis strains and therefore formclear plaques. Double lysogens containing both the cl and c2prophages can be heat induced, while those containing c+and c2 cannot (13). Thus, the c gene encodes a trans-actingrepressor protein involved in the establishment and mainte-nance of lysogeny. All the clear-plaque mutants so far testedfall within the same complementation group (18). In ourattempt to clone the repressor gene from the 126-kilobase(kb) phage genome, we uncovered a new genetic elementinvolved in controlling phage lysogeny and immunity. Thebiochemical and biological characterization of the cloned dgene is reported here.

MATERIALS AND METHODS

Bacterial strains and media. The bacterial strains, phages,and plasmids used are listed in Table 1. The liquid mediaused were TB (1% tryptone [Difco Laboratories, Detroit,Mich.], 0.5% NaCI) and modified M (1% tryptone [Difco],0.5% yeast extract, 1% NaCl, 5 mM CaC12, 5 mM MgSO4,0.02 mM MnCl2) as described previously (12). Modified Mmedium was used for growing B. subtilis cells and preparingphage lysates. TBA soft agar (TB containing 0.7% agar) wasused when determining phage titers. Tryptose blood agarbase (TBAB; Difco), containing 5 ,ug of chloramphenicol per

* Corresponding author.

ml when needed, was used as a plating medium for phageand bacteria.Phage growth and DNA preparation. Phage SPI c2 was

prepared from strain CU1147 by the heat induction methodessentially as described by Rosenthal et al. (13). Wild-typeSP,B phage was prepared either by mitomycin C (0.8 ,ug/ml)induction or, as for SP3 clO (described below), by growingphages lytically by the method of Warner et al. (16) with thefollowing modifications: (i) the adsorption period at 37°C was15 to 20 min; (ii) the centrifugation step to remove unad-sorbed phage was omitted; and (iii) infected cells werediluted 30- to 100-fold into prewarmed (37°C) modified Mmedium.

Purified phage particles were prepared by the method ofFink et al. (3) with several modifications. Polyethyleneglycol 6000 was added to 10% (wt/vol) to the phage suspen-sion along with NaCl to 0.5 M to precipitate most of thephage particles. Polyethylene glycol-precipitated phage pel-lets were suspended in CsCl (1.497 g/ml, in a 10 mMtrischloride [pH 7.4]-2 mM MgSO4-10 mM CaC12 buffer)without attempting to remove the trace amount of polyeth-ylene glycol and centrifuged to equilibrium with a BeckmanTi 70.1 rotor at 24,500 rpm for 28 h or with a VTi8O rotor at60,000 rpm for 6.5 h at 18°C. After collection and dialysis ofthe phage band, phage suspensions were concentrated, andthe phage DNA was extracted with TE (10 mM Trischloride[pH 8.0], 1 mM EDTA) buffer-saturated phenol followed bytwo chloroform extractions and ethanol precipitation.

Isolation of clear-plaque SP13 c1O. During the preparationof SPI c+ phage from the SU+3 (SP1) strain, a spontaneousclear-plaque phage (designated SP, c1O) was fortuitouslyidentified. In all the tests performed, SP clO was indistin-guishable from SP, cl; they were both used in these studies.

Construction of SPO c2 phage DNA library and othermolecular cloning methods. Plasmid pLP1201 (11) replicatesin both Escherichia coli and B. subtilis, and it was used forthe construction of the phage gene library. BamHI and BglIIdoubly digested SP3 c2 DNA (2.8 ,ug) was ligated to BamHI-digested pLP1201 (1 ,ug) DNA at a final DNA concentrationof about 300 xug/ml. E. coli MM294 (9) was transformed withthe ligation reaction mixture and plated on Penassay agar(Antibiotic Medium 2; Difco) containing 50 ,ug of ampicillin

952

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

PHAGE SP,B d GENE 953

TABLE 1. Strain list

Strain Relevant Source/referencegenotype/properties

Bacillus subtilisCU1050 SPW' metBS sup-3 thr leu ade (su+3) S. A. Zahler; 13,16CU1147 trpC2 (SP,3 c2) S. A. Zahler; 13BGSC no. 1L4 Su+3 (SPP); source of wild-type SPP Bacillus Genetic Stock Center; 16CU1225 citKI- AilvBl kauAl attSP (SP, c2 dcitK+-I) S. A. Zahler; 13CU3082 ilvA2 ilvDl5 leuB6 leuB16 attSP (SPO c2 dilvD+-1) S. A. Zahler

Escherichia coliMM294 thi-J hsdRI7 endAl supE44 F- 9-DG99 thi-l hsdRJ7 endAl supE44 lacIq lacZAM15 proC::TnlO; host for D. Gelfand

pUC plasmid vetorsDG98 DG99/F' lacIq lacZAM15 proC+ (Pro' Tetr host for M13 phage) D. Gelfand

VectorspLP1201 E. colilB. subtilis shuttle vector; Cmr in B. subtilis, Apr and Tcr in 11

E. colipUC18 E. coli cloning vector with polylinker sequence; Apr 10M13mplO and M13mpll Coliphage vectors for cloning and sequencing 10

PhagesSP,B c+ Wild-type c gene 16SP3 c2 Temperature-sensitive mutation in c gene, inducible at 50°C, normal 13

in permissive temperature rangeSP3 cdO Inactive c gene This studySP, cl Inactive c gene, also contains dell 4,16SP,B: Tn9J7 cl Tn917 inserted in SP,B ci S. A. Zahler

per ml. Of 520 Ampr colonies tested for their sensitivity totetracycline (15 ,ug/ml), about 19% of the transformants weresensitive; presumably they contained inserts in the BamHIsite of the tetracycline resistance gene. The transformantsharboring recombinant plasmids were grown in pools of five,and the plasmid DNAs were isolated (6), digested withXhoII, and analyzed by agarose gel electrophoresis. On thebasis of the pattern of the restriction fragments, we chose sixpools of these DNA preparations for transformation into B.subtilis. Competent cells were prepared (1) from the phage-sensitive strain CU100S and transformed with the pooledplasmid DNAs. The transformants were then tested forimmunity to SP, clO phage by the cross-streak test de-scribed below.

Restriction endonucleases were purchased from eitherNew England BioLabs, Inc. (Beverly, Mass.) or BethesdaResearch Laboratories, Inc. (Gaithersburg, Md.) and usedaccording to the specifications of the suppliers. Preparationof plasmid (5) and chromosomal (2) DNAs was essentially aspreviously reported. DNA restriction fragments were sepa-rated on either polyacrylamide or agarose slab gels, depend-ing on their size, and electroeluted as described previously(7), except that the buffers used were 0.05 x TBE foracrylamide and 0.1 x TEA (4 mM Trischloride [pH 8.0], 0.1mM EDTA, 0.5 mM sodium acetate) for agarose gels.

Site-directed mutagenesis was carried out as describedpreviously (20). A synthetic oligonucleotide with the se-quence 5'-GATATATTTTCATAAGCTTGGAATCAGAT-3' was used to modify the d gene sequence around nucleotideposition 405. Other methods for molecular cloning in E. coliwere essentially as described previously (7).

Determination of DNA sequence. Appropriate DNA frag-ments were cloned into the M13 phage vectors mplO andmpll (10). Single-stranded DNA templates were prepared,and the DNA sequence was determined on both strands bythe dideoxyribonucleotide chain termination method (14, 15)with [a-35S]dATP obtained from New England Nuclear

Corp. (Boston, Mass.). A synthetic 17-mer with the se-quence 5'-GGTCAAATCTGTCACGC-3', designed accord-ing to the preliminary sequencing data, was subsequentlyused as an additional sequencing primer to allow sequencingacross the ClaI site at position 666 within the d gene.

Tests for SPO lysogeny and immunity. To look for thepresence of SPi prophage, we tested the B. subtilis strainsby the method described by Rosenthal et al. (13) for theirimmunity to clear-plaque mutants of SP,, their ability torelease phage particles, or their ability to produce betacin.Transformants of CU1050 carrying various plasmids weretested for their immunity to SP13 by streaking transformantcolonies across a vertical line along which 1 x 108 to 2 x 108SP clO phage had been applied on a TBAB plate containing5 ,ug of chloramphenicol per ml (19). CU1050 cells carryingonly the vector pLP1201 were lysed on plates after infectionby SP, clO phage and thus served as the negative control;they grew only on the first half of the cross streak.CU1050(SPP c2) cells carrying pLP1201 served as the posi-tive control which grew across the entire streak.

Crossing the d2 frameshift mutation back into the SPI8phage genome. Cells of CU1147, a strain lysogenic for SP13 c2phage, were made competent and then transformed withpJM172, the plasmid containing the ClaI-to-NruI +2 frame-shift mutation in d. One of the transformants was used toinoculate 5 ml of TB containing 5 ,ug of chloramphenicol perml and grown for 5 to 6 h. The titer of the supernatantcontaining phage particles was determined on a lawn ofCU1050(pLP1201) cells. Spontaneous release of SP13 phagewas between 106 and 107 PFU/ml. Of approximately 104plaques screened visually, 2 clear plaques were picked. Oneof these was plaque purified through three plating cycles.The clear-plaque phage isolated was then grown up (500 ml)and gradient purified as described above. Phage DNA wasextracted, and the presence of the d2 allele was confirmed byanalysis of restriction endonuclease digestions for the pres-ence of the diagnostic NruI site.

VOL. 167, 1986

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

954 McLAUGHLIN ET AL.

To construct the SP, cl d2 double mutant, CU1050(pJM172) cells were spotted with SP, cl lysate on a plate.After overnight incubation, phage from the cleared zonewere collected and plated on CU1050 indicator cells.Cloudy, but distinctively smaller, plaques appeared at afrequency of about lo-3, These phage were plaque purified,and the presence of the d2 allele on the genomes wasconfirmed as described above.

RESULTSMolecular cloning of the d gene from SPj( c2. A gene library

of SP, c2 genome cloned into the bifunctional plasmidpLP1201 was constructed (see Materials and Methods).Competent B. subtilis CU1050 cells were transformed withplasmids prepared from the pooled cultures. Individualtransformants were then tested for SP, clO immunity by thecross-streaking test. Of 234 transformants tested, 15 showedimmunity. Plasmid DNA from six of these was prepared andanalyzed. Three of the six appeared to have identical restric-tion patterns for EcoRI, PstI, and Sall, while the other threediffered but appeared to be related clones that had suffereddeletions presumably outside of the cloned immunity genesequence. The restriction map of one of the undeletedcandidates, designated pJM135, is shown in Fig. 1. Since thephage genome was digested with BamHI and BglII and thevector was cleaved with BamHI, we checked pJM135 plas-mid DNA for the presence of regenerated BamHI sites. NoBamHI recognition sites were present at either junction withpLP1201 DNA. Therefore, most likely the cloned fragmentin pJM135 was derived from a 5.9-kb BglII fragment in theSP, c2 genome. The cloned gene that confers immunity wasdesignated the d gene.Mapping of the d gene fragment. The approximate location

of the 5.9-kb fragment on the SPP genome was determinedby Southern hybridizations. The restriction maps of the SPPgenome for the AvaI, BamHI, SacI, and Sall endonucleaseshave been reported previously (4). SPPc2 DNA digestedindividually with each of these enzymes and fractionated ona gel was hybridized with 32P-labeled Sall fragments ofpJM135. The cloned SP,B DNA hybridized with the AvaI Dfragment, the BamHI A fragment, the Sacl F fragment, theSacI B fragment (very weakly), and the SalI C and Dfragments of SP, (data not shown). This hybridizationpattern is consistent with the assignment of the location ofthe d gene in the SP, genome as shown in Fig. 2.To localize the d gene within the cloned phage DNA, we

constructed various subclones. A deletion of the 1.2-kbEcoRV fragment (Fig. 1) had no effect on the immunityphenotype of the transformants. In contrast, transformantscarrying plasmids lacking either the 3.9-kb ClaI fragment[Clal(l) to ClaI(2) in Fig. 1] or the 4.5-kb ClaI fragment[ClaI(2) to ClaI(3) in Fig. 1] from pJM135 lost the immunity tophage infection. It therefore seemed likely that the d genespans the ClaI(2) site.

In vitro construction of frameshift mutations in the d gene.Mapping data revealed the presence of multiple ScaI andPstI sites in the cloned SPP fragment in pJM135 (Fig. 1). The2.9-kb Pst(l) to ScaI(2) fragment (Fig. 1) generated'by partialPstI digestion and complete Scal digestion was cloned intoPstI- and SmaI-digested pUC18 plasmid (10). The resultingplasmid was designated pJM154. This plasmid has a uniqueClaI restriction site within the SP, DNA insert (Fig. 1).After ClaI digestion and repair with E. coli DNA polymeraseKlenow fragment, these molecules were ligated and used fortransforming E. coli MM294. The transformants werescreened for the generation of a NruI site that would result

Sal 1(2)FIG. 1. Restriction map of the plasmid containing the d gene.

The shaded region is the SP, c2 DNA insert in the pLP1201 vector.Marks at 1-kb intervals are shown inside the circle. Some restrictionenzyme sites are numbered with subscripts to facilitate identifica-tion. The location for the chloramphenicol resistance (Cm) and theampicillin resistance (Ap) genes as well as the replication origins forE. coli (E) and B. subtilis (B) are marked.

from the ligation of completely repaired ClaI ends. Themajority of the plasmids had lost the ClaI site and gained anNruI site; one of these clones was designated pJM169. Theminority of the resulting plasmids had neither a ClaI nor anNruI site; one of these was designated pJM158. The 1,033-base-pair (bp) PstI-EcoRI fragment from each of theseplasmids (pJM154, pJM158, and pJM169) was purified byelectroelution from 4% acrylamide gels and ligated to PstI-and EcoRI-digested pLP1201 to replace the small PstI-EcoRI fragment carrying most of the pBR322 ,B-lactamasegene. Aps Tcr transformants were screened by restrictiondigestion to ensure the presence of the SPP DNA insert, andthe resulting plasmids were designated pJM160 (d+; withClaI site), pJM170 (d-; without CMa), and pJM172 (d-; withNruI replacing CMl).

B. subtilis CUlOSO transformants carrying pJM160 wereimmune to SP, c+ as well as SP clO and other SPBc-phage. In contrast, CU1050(pJM170) and CU1050(pJM172)did not confer immunity to these phage. From these data weconclude that the d gene resides within the 1-kb PstI-EcoRIfragment and that the ClaI site is located within the d gene.DNA sequence of the PstI-EcoRI fragment of SP( phages.

The sequencing strategy used to determine the DNA se-quence of the SP, PstI-EcoRI fragment is shown in Fig. 2.The 1,033-base sequence of the antisense strand of thisfragment isolated from SPpi c2 phage is shown from the5'-end PstI site to the 3'-end EcoRI site in Fig. 3. A longopen reading frame exists from position 135 through 815within which is located the ClaI site (position 666). FourAUG (Met) and one GUG (Val) codons cou,ld be the initia-tion site for the d gene protein (Fig. 3).

J. BACTERIOL.

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

PHAGE SPI d GENE 955

PI G I L I D I (E,l,J,M) I B I C IN IRISIIQI F 1t KI HII A Pst I

10 kb A E B D C I F Sal I

|_K j H T A

PstI NruI Pst I BgIII SacI Eco stI PstI Sal I BgIl 1CIa II1 kb I la

_________ II 1-_Eco R I Taq I Xmn I Cla I Taq I Ban 11 Pst I(1028) (930) (858) (666) (328) (258) (1)

FIG. 2. Map location of the cloned d gene on the SP, phage genome. PstI and Sall restriction maps of SP, DNA are taken from publishedwork (4). The order of the PstI K, H, and T fragments was determined experimentally. The shaded region is the 5.9-kb BglII fragment clonedin pJM135. The arrows represent the direction of transcription of the d gene. The EcoRI-PstI fragment containing the d gene is expanded;the positions of the restriction sites within this fragment are indicated according to the DNA sequence shown in Fig. 3. The sequencingstrategy is depicted at the bottom.

In addition to the sequence of the d gene shown in Fig. 3,the sequence of the insertion mutants at the ClaI site wasalso determined. A synthetic 17-mer oligonucleotide wasused as the sequencing primer to determine the nucleotidesequence surrounding the ClaI site regions in pJM154,pJM158, and pJM169. The sequence across the ClaI site inpJM154 was 5'. . .TAT,ATC,GAT,TTT,CTC,. .3' (Fig. 3).The corresponding sequences for pJM158 and pJM169 were5' . . .TAT,ATC,CGA,TTT,TCT,C. . .3' and5'. . .TAT,ATC,GCG,ATT,TTC,TC . . .3', respectively.

The ClaI site was apparently only partially repaired, result-ing in a + 1 frameshift mutation in the d gene on pJM158. Theproperly repaired ClaI ends were ligated and generated a +2frameshift mutation and an NruI site (TCGCGA) in the dgene on pJM169. We designate the alleles with the +1 and+2 frameshift mutations dl and d2, respectively.The same PstI-EcoRI fragment was also isolated from the

SP,B cl, c1O, and c+ genomes by cloning into the coliphageM13mplO vector. The sequences of these fragments wereanalyzed. They were all identical to that isolated from SPP

CTGCAGATTAT(,TCAAAGAACTGCTCATTCAGATAAAATCAGAAAAGAACTAATGATTTCACTTTCCATrTATACAGrTCTCCATGACTGGAAACAATGGTATAATGTAGGAAAATGATAp

121 CCGGCGGTGGTTGAATGGGATATAAGTTTATGGCTTATGGTGGCTATTTT rrATTcrGTcTTTrTcTrTTGCTAATGGACGGCTGGAGAGGCATGGGAATTTGCTTAATAATTGrAGGTTM G Y K F M A Y G G Y F L F C L F F L L M D G W R G M G C L V G 35

241 TAGCCCTATTAGCACTTGAGCCCTATAAAATTAAAGCCCAAAAAAATATAGATAAACTAAAAGAAAATGCGGAAACACTTAAGCACTTCGACGGTGGTTTFAATCCAGACAATTTCrTTAl A L L A L E P Y K K A Q K N D K L K E N A E T L K H F D G G F N P D N F F 75

361 ATACTTACAAAACTAAAATTGCATTTAAGGAATCTGATTCCCTTGTGAAAATATATCAGCTTAATAAAGATGAACATATTGAAGAATACACAATCCCTTTTTCCAATGTTATAGAATCCGN T Y K T K A F K E S D S L V K Y Q L N K D E H E E Y T P F S N V t- S 115

481 AAATTGC r T rAGACAATCAAATAATTTCFAAAGTATCAAAATCAGGAATCGTAGCTGGTGGCCTGTTAGCTGGAGGAATTGGAGCTGCAATTGGAGGGTTGTCTGCCrCTTCAATAOAAAE A L D N Q S K V S K S G V A G G L L A G G G A A G G L S A S S Q 155

601 ATGAAATGGTCAAATCTGTCACGCTAAAGATTACTGTTGAAGACCTCGGTAAGCCTATCCATTATATCGATTT[1 rcCCcCACACAAGAAGTTGAAGGGTATAATATTCAGGGGTATAAAAN E M V K S V T L K T V E n L G K P H Y D F L P T Q E V E G Y N Q G Y K 195

C721 AAGATAGCAATGTCATTCAACAAGCACTTACGAATGCAGAATATTGGCATGGTGT[AFGGATGTAATTA[TAAGAAAGCAAACAAAGTCGCTCAATAACTGAGTGGCTTT r T TCT TTGTC

K D S N 'V Q Q A L T N A E Y W H G V M D V K K A N K V A Q * 227

841 CTCTCCCTACT(IAAAGGAAGTGATTCrTACTTGAGTCAAAACCTCAAAATTATACTAACCCCGCAAGCTGATACCTCATCCAAAACTGTCGAACAGTTAAATCAGCAAATTAAATCCCTx

961 GGAAAAGAAACTCAACTCCCTCAAGCTCAATACAAACATTGATTCTACAACCTTAAAAGCTCTGCAAGAA rTCE

FIG. 3. DNA sequence of the 1,033-bp PstI-EcoRI fragment of SPP c2. Restriction sites are designated by upper-case letters as follows:P, PstI; C, ClaI; X, XmnI; and E, EcoRI. The potential translational initiation sites (in-phase AUG and GUG codons) are underlined. Thededuced amino acid sequence for the entire open reading frame is shown in single-letter code. The amino acids are numbered on the right,and nucleotide positions are numbered on the left.

VOL. 167, 1986

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

956 McLAUGHLIN ET AL.

c2. Although the d gene was identified by its ability to conferon its host immunity to superinfecting SP, phage, this pieceof biochemical data clearly established the fact that thecloned immunity gene is distinct from the repressor c gene.

Analyses of the d gene protein. By comparing the sizes ofthe proteins produced in the E. coli in vitro transcription-translation system directed by plasmids pJM154, pJM158,and pJM169, a protein of 15 kilodaltons (kDa) was identifiedinitially as the d protein. Sequence analysis of the amino-terminal portion of this protein (data not shown) indicatedthat it corresponded to the peptide initiated at the GUG (Val)codon at nucleotide position 405 shown in Fig. 3.The sequence downstream from the GUG codon at nucle-

otide position 405 consists of an open reading frame of 137codons. This 137-amino acid polypeptide was used as theantigen to raise antiserum. To achieve this, we carried outoligonucleotide-directed site-specific mutagenesis to convertthe GUG codon at position 405 (Fig. 3) to an AUG codon andintroduced a HindlIl site immediately 5' to the ATG se-quence (see Materials and Methods). The sequence betweenthe newly created HindIII site and the XmnI site (nucleotideposition 858) beyond the termination codon was then clonedbehind the coliphage lambda gene N Shine-Dalgarno se-quence and the PL promoter (H. C. Wong and Y. E. Ting,unpublished data). The plasmid containing this sequencewas introduced into a proper E. coli strain and, under theconditions that induced the PL promoter, a large amount ofthe 15-kDa protein was produced. It was isolated and usedfor raising rabbit antiserum against the d gene protein.The d gene proteins encoded by the SPi prophage and by

various plasmids were analyzed by the immunoblottingmethod, using the rabbit antiserum raised against the 15-kDaprotein. Cell extracts were prepared from CU1050, CU1147,and CU1050 derivatives harboring pLP1201, pJM160, orpJM172. The results (Fig. 4) indicated that the in vivo-synthesized d gene protein is about 22 kDa. Furthermore,the d gene protein was only detected in CU1O50(pJM160).Although the method used was sensitive enough to detectone molecule of the 15-kDa peptide per cell in a testexperiment, no signal was obtained in the extracts fromCU1147 that carries the SP, c2 prophage. We also did notdetect immunoreactive proteins from the extracts ofCU1O50(pJM172). This suggests that the d2 protein probablyis unstable in vivo.We also fused the DNA fragment encoding the amino-

terminal portion of the d gene (positions 1 to 540; Fig. 3) tothe E. coli lacZ coding sequence and obtained results (notshown) indicating that the fusion gene directed the synthesisof enzymatically active ,B-galactosidase in vivo in both E.coli and B. subtilis CU1050. The lacZ fusion data show thata promoter and a translation initiation signal that function inboth E. coli and B. subtilis transformants are containedbetween the PstI and HaeIII sites located in this region, butthe precise location of these regulatory sequences has notbeen further defined.SPI phage does not lysogenize cells carrying the cloned d

gene. The introduction of the two frameshift mutations intothe d gene (as described for pJM170 and pJM172) wascorrelated with the loss of the immunity phenotype. To testwhether the apparent immunity conferred to the host by thed gene is due to efficient lysogeny of the infecting SP,B phage,we performed the following tests. SPP cl lysate was spottedon a lawn of CU1O50(pJM160) and incubated overnight.Although no lysis was evident, cells within the zone wherephage lysate was spotted were isolated and tested for thepresence of SPp prophage. Of 50 isolates tested, none gave

abcdefa b c d ef~~~~muie$.~~~~~~~~~~~~~T!N!slNn..

- .

.1

-43

-25

-18

-14

-6.2

FIG. 4. Immunoblotting analysis of the d proteins synthesized invivo. Cellular extracts prepared from B. subtilis CU1050 (lane b),CU1147 (lane c), CU1050(pLP1210) (lane d), CU1050(pJM160) (lanee), and CU1050(pJM172) (lane f) were fractionated by sodiumdodecyl sulfate-poylacrylamide gel electrophoresis, transferred tonitrocellulose paper, and then probed with the antiserum raisedagainst the purified 137-amino acid peptide (lane a) derived from thed protein. The immunoreactions were visualized by the horseradishperoxidase-TMB-dextran sulfate method (kindly prepared and pro-vided by W. Bloch et al. of Cetus Corporation). The molecularweight of the protein markers (x 103) is shown on the right.

a positive result. We also used an alternative test whichpermits direct selection of lysogens. Phage SP: :Tn917 cl isa cl-derived clear-plaque phage containing the Tn917 trans-poson (17). Cells lysogenic to this phage are resistant toerythromycin and lincomycin because of the presence of theresistance gene on the transposon. Cells from strainCU1050(pJM160) were infected with SP,::Tn9J7 cl at amultiplicity of infection of 1. After absorption and incuba-tion, 2 x 108 infected cells were plated on selective medium.No antibiotic-resistant colony was obtained. Assuming thatthe Tn917 transposon in this phage does not inactivate anySPP gene that is involved in immunity and lysogeny, theseresults indicate that the presence of the cloned d gene in thehost prevents lysogeny as well as lytic growth of the infect-ing SPI cl phages.

Genetic analyses of SPjI phage carrying the d2 mutation.We crossed the d2 allele into the SP, c2 genome and studiedthe phenotype of the SPP c2 d2 phage (see Materials andMethods). Two clear-plaque phages were identified amongthe spontaneously released phages from CU1147(pJM172)transformants. Phage DNA from one of these clones wasisolated and analyzed. The presence of an additional NruIsite that is diagnostic of the d2 allele confirmed that thephage indeed carried the d2 mutation. The genome of theclear-plaque phage contained two NruI sites approximately6.6 kb apart (Fig. SA, lane b) instead of a single site found inthe c2 phage genome (Fig. SA, lane b). Thus, the d2 mutationcauses the clear-plaque phenotype in the SP3 c2 d2 phage.We also crossed the d2 mutation into the lytic phage SPP cl

J. BACTERIOL.

f

..

.. .......

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

PHAGE SP, d GENE 957

Aa b c d e f g Bh i j k I

. _

FIG. 5. Electrophoretic analysis of SPt3 c2 and SPf3 c2 d2 phage DNAs. Molecular weight markers are EcoRI-HindIII-digested coliphageA DNA (lanes c and h) and Sal- and PstI-digested SP,I phage DNA (lanes d and i, respectively) as previously reported (4). (A) A 0.4% agarosegel fractionation of SP, c2 and SPP c2 d2 DNAs digested with NruI (lanes a and b, respectively), with Sall (lanes d and g, respectively), andwith Sall and NruI (lanes e and f, respectively). (B) A 0.8% agarose gel fractionation of SP,B c2 and SPO c2 d2 DNAs digested with PstI (lanesi and 1, respectively) and with PstI and NruI (lanes j and k, respectively).

(see Materials and Methods). The SP,B cl d2 phage formedsmaller and cloudy plaques on a CU1050 lawn. The cellswithin the plaques were isolated and further characterized.They appeared to be lysogenic to SPP phage since theyspontaneously released phage particles and producedbetacin. However, these lysogens were very unstable. Uponseveral transfers by streaking on plates, cured cells appearedat high frequency. This cl d2 phage is phenotypically similarto the int- SP,B phage that forms abortive lysogens asdescribed by Zahler (18). The main difference is that the cld2 phage forms cloudy plaques in the absence of a functionalrepressor gene owing to the cl mutation.The clear-plaque SPP c2 d2 phage was further tested for

complementation with SPP cl by mixed infection. When themixed lysate was spotted on the CU1050 lawn and incubatedovernight, the clear zone appeared to be the same as thoseseem with either phage lysate alone; only scattered phage-resistant colonies grew within the lysis zones. Upon furtherincubation for a few hours, a thin film of bacteria could bedetected in the lysis zone spotted with the mixed lysate, butnot with the individual lysates. These cells were picked andpurified by repeated streaking. All of them spontaneouslyreleased cloudy-plaque phage morphologically similar to theSPP c2 phage. In addition, of 10 independently isolatedphage tested, none carried the diagnostic NruI site unique tothe d2 mutation. Thus, these cells were lysogenic for SP, c2that was generated by recombination between the cl and thec2 d2 phage.

Physical mapping of the d gene. The sequencing primer

complementary to the internal sequence of the d gene(nucleotide positions 608 to 624 in Fig. 3) was used to probethe genomes of strains carrying defective SP3 prophages.Southern analysis data are shown in Fig. 6. The probereacted to the total DNA isolated from CU1050(SPP c2)(strain CU1147). No hybridization was detected with strainscarrying either SP13 c2 dcitK+-J or SP3 c2 dilvD+-1. It isinteresting to note that the citK and ilvD markers are locatedon the opposite side of the attachment site for SP, on thechromosome; thus, each of these two defective phage pre-sumably has a different phage arm deleted. Out data suggestthat one of the defective phage might carry an additionaldeletion previously undetected by genetic analyses.The precise physical location of the d gene on the phage

genome was determined by mapping the NruI site unique tothe d2 allele. The NruI site present in the SPP c2 genome islocated in the Sall D fragment. The SalI-NruI double diges-tion resulted in the approximately 18-kb SalI D fragment(Fig. 5A, lane d) being cleaved into two fragments of about9.7 and 7.5 kb (Fig. 5A, lane e). The additional NruI site inthe clear-plaque mutant c2 d2 also maps in the SalI Dfragment. Similar double digestion produced two fragmentsof about 6.6 and 3 kb (Fig. 5A, lane f) from the original 9.7-kbSaII-NruI fragment of SP,B c2 DNA (Fig. 5A, lane e).Further mapping of the NruI sites with respect to the PstIrecognition sequences indicated that the unique NruI site inthe SP, c2 genome was located in the PstI K fragment (Fig.SB, lanes i and j). The approximately 5-kb K fragment (lanei) disappeared when doubly digested with NruI (lane j). The

VOL. 167, 1986

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

958 McLAUGHLIN ET AL.

a b c d e

FIG. 6. Detection of d gene sequence in the SP3 prophages byhybridization. Plasmid pJM160 DNA (lane a) and chromosomalDNAs from CU1050 (lane b), CU1147 (lane c), CU1225 (lane d), orCU3082 (lane e) were digested with PstI and EcoRI, fractionated byagarose gel electrophoresis, transferred to nitrocellulose, and thenprobed with the 32P-labeled oligonucleotide complementary to theantisense strand of the d gene (see Materials and Methods for thesequence of the oligonucleotide). The sizes (in kilobases) of theDNA markers are shown on the right.

additional NruI site introduced in the c2 d2 mutant waslocated in the PstI H fragment, which was shortened toabout 5 kb when double digested with PstI and NruI (Fig.SB, lanes k and 1). Since the H fragment was reported to be6 kb in size (4), NruI cleavage resulted in the 5-kb productreplacing the cleaved K fragment and a 1-kb fragment thatwas not resolved in the gel analysis. These data coupled withthe NruI and SalI-NruI restriction data made it possible tomap the two NruI sites in the SPP c2 d2 genome (Fig. 2),where the ClaI site shown in the d gene was modified togenerate the second NruI site in SPi c2 d2.

DISCUSSIONA 5.9-kb BglII fragment of SPP c2 phage genome was

cloned into a plasmid, and it was shown to confer immunityto the SP, cdO clear-plaque phage and to other SPP phage.This gene, designated d, was further localized to the1,033-bp PstI-EcoRI fragment within the BglII fragment bysubcloning. The location of this 5.9-kb BglII fragment on thephage genome was determined by Southern hybridizationanalysis, and the mapping data are summarized in Fig. 2.The PstI T fragment and a portion of the PstI H fragmentthat map in this region of the phage genome (4) werecontained within the cloned BglII fragment (data not shown).By in vitro conversion of the ClaI restriction site within thed gene to an NruI site, we were able to position the d genewithin the PstI H fragment and to localize precisely the dgene on the SP, genome. These mapping data also clarifiedthe relative order of the clustered PstI K, H, T, and Afragments (4) on the genome (Fig. 2).The 1,033-bp d gene fragment contains an open reading

frame (227 codons) which begins with the methionine codonAUG. Immunoblotting data indicated that the d gene proteinis 22 kDa, an estimate based on its mobility in a gel. Thisdoes not help to identify unambiguously which one of thefour clustered AUG codons is the initiation codon for the d

protein. The fourth Met codon (nucleotide position 213 inFig. 3) is preceded by a sequence closely resembling theShine-Dalgarno sequence (8), and it would encode a proteinof 201 amino acids (-22 kDa). Whether this is indeed theinitiation codon for the d gene remains to be determined.The transcriptional initiation site for the d gene is likely

also contained within the PstI-EcoRI fragment. InCU1050(pJM160) cells, both biological expression of the dgene (by immunity test) and synthesis of the 22-kDa dprotein were detected. In addition, the DNA fragment en-coding the amino-terminal portion of the d gene (positions 1to 540, Fig. 3) was fused to the E. coli lacZ coding sequence,and the fusion gene directed the synthesis of enzymaticallyactive P-galactosidase in vivo in both E. coli and B. subtilis(unpublished data). These data also suggest that the d geneprotein does not positively regulate its own expression.The expression of the d gene is probably regulated in vivo.

We detected a fair amount of d protein in CU1050(pJM160)cells by immunoblotting, but could not detect any d proteinin CU1147, the strain carrying the SP, c2 prophage. Eitherthe d gene is not normally expressed in the lysogen or the dprotein is synthesized but sequestered or degraded so it isnot detectable under the conditions used for the preparationof the cell extracts.Cloned d gene confers on its host immunity to superinfect-

ing phages. The lack of observable lytic growth is not due toefficient lysogeny of the infecting SP, phage as was shownby using the SPP::Tn917 cl phage to infect CU1050(pJM160). In this experiment, no lysogenic clone was ob-tained. Thus, functional expression of the cloned d gene ona plasmid totally prevents superinfecting SPP from lyticgrowth and, at least for the clear-plaque phage, it alsoprevents SPt from lysogenic growth. These experimentsindicate that the d gene protein acts in trans to conferimmunity.

In addition to controlling immunity, the d gene is alsoinvolved in lysogeny. The SPP c2 phage normally formscloudy plaques, as does the wild-type phage. But SPI c2 d2phage forms clear-plaque phage on indicator cells. Thissuggests that either the d gene mutation blocks the repressorfunction or it is involved in the regulation of an essential stepin the establishment of lysogeny.Mixed infection experiments with c2 d2 and other clear-

plaque c- phage suggest that there is no d gene complemen-tation between these two types of clear-plaque phage, al-though the d gene confers immunity in trans. When CU1050was infected with SP, c2 d2 and SP, cl, only two types ofcells were recovered from the lysis zone in this experiment:SPP-resistant cells, and cells lysogenic for the recombinantphage SP1 c2. The other recombination product, phage SPPcl d2, was not obtained. Since we have shown that B.subtilis is unable to stably maintain SP,B cl d2 at theprophage state, this result was not unexpected. On the otherhand, it has been shown that mixed infection of SPP cl andc2 produces double lysogens (13). If the d gene product actsin trans and d2 is a recessive mutation, then one would alsoexpect to obtain double lysogens carrying both SP, c2 d2and SPI cl. Our failure to obtain such double lysogenssuggests that the regulation of the lysogeny and immunityprocesses in SPP is complex. Possibly the d gene producthas more than one regulatory role in controlling phageimmunity and lysogenization.

ACKNOWLEDGMENTSWe thank Stanley A. Zahler (Cornell University) for providing

many strains and for valuable consultation on SP3 genetics, Ken

J. BACTERIOL.

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

PHAGE SP, d GENE 959

Watt for the amino-terminal sequencing of the purified d geneprotein, and David Gelfand for reviewing the manuscript and forproviding host strains DG98 and DG99 and T4 DNA ligase. We alsoacknowledge the secretarial assistance of Edna McCallan andSheldon Brown.

LITERATURE CITED1. Anagnostopoulos, C., and J. Spizizen. 1961. Requirements for

transformation in Bacillus subtilis. J. Bacteriol. 81:741-746.2. Chater, K. F., D. A. Hopwood, T. Kieser, and C. J. Thompson.

1982. Gene cloning in Streptomyces. Curr. Top. Microbiol.Immunol. 96:69-95.

3. Fink, P. S., R. Z. Korman, J. M. Odebralski, and S. A. Zahler.1981. Bacillus subtilis bacteriophage SPIc1 is a deletion mutantof SPJ. Mol. Gen. Genet. 182:514-515.

4. Fink, P. S., and S. A. Zahler. 1982. Restriction fragment maps ofthe genome of Bacillus subtilis bacteriophage SPI. Gene19:235-238.

5. Katz, L., D. T. Kingsbury, and D. R. Helinski. 1973. Stimulationby cyclic adenosine monophosphate of plasmid deoxyribonu-cleic acid replication and catabolite repression of the plasmiddeoxyribonucleic acid-protein relaxation complex. J. Bacteriol.114:577-591.

6. Ish-Horowicz, D., and J. F. Burke. 1981. Rapid and efficientcosmid cloning. Nucleic Acids Res. 9:2989-2998.

7. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning, a laboratory manual, p. 156-164, 243, and 382-386.Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

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

9. Meselson, M., and R. Yuan. 1968. DNA restriction enzyme fromE. coli. Nature (London) 217:1110-1114.

10. Norrander, J., T. Kempe, and J. Messing. 1983. Construction ofimproved M13 vectors using oligodeoxynucleotide-directed mu-tagenesis. Gene 26:101-106.

11. Ostroff, G. R., and J. J. Pene. 1984. Molecular cloning withbifunctional plasmid vectors in Bacillus subtilis. II. Transfer ofsequences propagated in Escherichia coli to B. subtilis. Mol.Gen. Genet. 193:306-311.

12. Palefski, S., H. E. Hemphill, P. E. Kolenbrander, and H. R.Whiteley. 1972. Dominance relationships in mixedly infectedBacillus subtilis. J. Virol. 9:594-601.

13. Rosenthal, R., P. A. Toye, R. Z. Korman, and S. A. Zahler.1979. The prophage of SPt3c2dcitK,; a defective specializedtransducing phage of Bacillus subtilis. Genetics 92:721-739.

14. Sanger, F., A. R. Coulson, B. G. Barrell, A. J. H. Smith, andB. A. Roe. 1980. Cloning in single-stranded bacteriophage as anaid to rapid DNA sequencing. J. Mol. Biol. 143:161-178.

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

16. Warner, F. D., G. A. Kitos, M. P. Romano, and H. E. Hemphill.1977. Characterization of SP,B: a temperate bacteriophage fromBacillus subtilis 168M. Can. J. Microbiol. 23:45-51.

17. Youngman, P., P. Zuber, J. B. Perkins, K. Sandman, M. Igo,and R. Losick. 1985. New ways to study developmental genes inspore-forming bacteria. Science 228:285-291.

18. Zahler, S. A. 1982. Specialized transduction in Bacillus subtilis,p. 269-305. In D. A. Dubnau (ed.), The molecular biology of thebacilli. Academic Press, Inc., New York.

19. Zahler, S. A., R. Z. Korman, R. Rosenthal, and H. E. Hemphill.1977. Bacillus subtilis bacteriophage SP,B: localization of theprophage attachment site, and specialized transduction. J. Bac-teriol. 129:556-558.

20. Zolier, M. J., and M. Smith. 1983. Oligonucleotide-directedmutagenesis of DNA fragments cloned into M13 vector. Meth-ods Enzymol. 100:468-500.

VOL. 167, 1986

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from