JPURNAL OF BACTERIOLoGY, Feb. 1991, p. 1388-13980021-9193/91/041388-11$02.00/oCopyright 0 1991, American Society for Microbiology

Vol. 173, No. 4

The spoOK Locus of Bacillus subtilis Is Homologous to theOligopeptide Permease Locus and Is Required for

Sporulation and CompetenceDAVID Z. RUDNER, JOHN R. LEDEAUX, KEITH IRETON, AND ALAN D. GROSSMAN*

Department ofBiology 56-510, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

Received 26 October 1990/Accepted 5 December 1990

Spore formation in Bacilus subdfis a dramac response to environmental signals that is controle In partby a two-component regulatory system composed of a idbe protein kinase (SpoIEI) and a trriptinal

regulator (SpoOA). The spoOK locs plays an important but unde d role in the iitia of sporulation andin the development of genetic competence. spoIlJ spoOK double mutants had a more severe defect In sporulationthan either sngle mutant. Overproduction of the spoHIJ gene product resulted in the suppresson of thesporulation defect, but not the competence defect, caused by muttions in the spoOK locw. On the basis of thephenotype of the spolU spoOK double mutant and the effect of overproduction of the spoIIJ gene product, atrausposon insertion in the spoOK locus was Lisated. The spoOK locus was cloned and sequenced. spoOK provedto be an operon of five genes that is homologous to the olgopeptide permease (opp) operon of Salmonellatyphimuium and rdated to a large family of membrane transport systems. The requirement for the htansportsystem encoded by spoOK in the development ofcompetence was somewhat different than its requiremet in theiitiation f spornation. Disruption of the last open reading frme in the spoOK operon caused a defect incompetence but had little or no effect on sporulation. We hypothesize that the transport system encoded byspoOK may have a role in sensing extracedlular peptide factors that we have shown are required for efficientsporulation and perhaps in sensing similar factors that may be necessary for genetic competence.

Cells of the gram-positive bacterium Bacillus subtilis candifferentiate into a dormant cell type, the spore or endo-spore, upon nutrient deprivation (15, 52) at a high celldensity (18, 56). Seven genetic loci (spoOA, OB, OE, OF, OH,OJ, OK) that are required for the initiation of sporulation havebeen identified (41, 51). Mutations in these genes alter thenormal pattern of gene expression and prevent the earliestmorphological changes associated with sporulation. Cloningand DNA sequence analysis have been reported for most ofthe spoO genes (29, 41, 51). Most, if not all, of the spoO geneproducts are present during growth and are thought to beinvolved in sensing sporulation conditions and regulating thetransition from growth to sporulation.The spoOK locus was initially defined by a mutation that

causes a partial defect in sporulation (9, 42) and was laterfound to cause a defect in the development of geneticcompetence (44, 45). The sporulation defect caused by thespoOK mutation is at least partly suppressed by missensemutations in spoOA (49, 50). The same missense mutations inspoOA also bypass the need for several of the other spoOgenes (B, E, F, and J), indicating that the function of thesespoO gene products may be to activate SpoOA (23, 49, 50,53).While some missense mutations in spoOA can bypass the

need for some of the other spoO genes in sporulation, spoOAitself is required for sporulation. Null mutations in spoOAcompletely block sporulation and at least partially blockmany of the other processes associated with the end ofnormal growth, including the development of genetic com-petence and the production of extracellular proteases andantibiotics (29, 51). The spoOA gene product is a DNA-

* Corresponding author.

binding protein that controls the expression ofgenes that areinvolved in the transition from growth to the stationaryphase (14, 39, 40, 51, 52).Spo0A belongs to a family of prokaryotic regulatory

proteins (including OmpR and NtrC [GlnG, NRI]) that arehomologous in their amino termini and are involved in signaltransduction and the regulation of gene expression in re-sponse to changes in external conditions (14, 35). Theactivity of these regulatory proteins (called response regula-tors) is modulated by the phosphorylation of an aspartateresidue in the conserved amino terminus. The responseregulator is one component of the so-called two-componentregulatory systems. The other component is a histidineprotein kinase (also called the sensor component), whichautophosphorylates on a histidine residue and transfers thephosphate to the cognate regulator. The histidine proteinkinases share homology in their carboxy termini (reviewedin references 3 and 54).Two genes that encode histidine protein kinases that are

thought to lead to the activation of SpoOA in vivo have beenidentified in B. subtilis. spoIIJ (kinA), homologous to thekinase component of the two-component systems (6, 37), hasbeen shown to phosphorylate SpoOA in vitro (37). spoIIJmutants are partially defective in sporulation, sporulating ata frequency approximately 1 to 30%o of the wild-type fre-quency, depending on the strain background and the precisesporulation conditions (6, 37, 47, 58; unpublished data).comP, a gene required for the development of geneticcompetence (the ability to take up DNA), is also homologousto the kinase component ofthe two-component systems (58).Mutations in comP cause a modest defect in sporulation(approximately 10 to 50%o of the wild-type frequency) but,when combined with mutations in spoIIJ, cause a moresevere defect (58). This result indicates that both comP and

1388

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

B. SUBTILIS spoOK OPERON 1389

TABLE 1. B. subtilis strains used

Strain Genotype Comments, source, or reference

JH642 trpC2 pheAl J. HochPB2 trpC2 8PB69 PB2 trp+ spoOK141 8KS19 spoIIJ::Tn9J7QJHU19 (spoIIJ::Tn9J7) 47M0495 spoOK+ chr::Tn917flMO495 (chr::Tn917) P. Stragier; Tn917 -10% linked to spoOK by transformationM0496 spoOK+ chr::Tn917::pTV21A2 P. Stragier; M0495 transformed with pTV21A2 (Cmr MLSS)M01022 amyE::(Pspac-spoIIJ+ cat) P. StragierK195 JH642 amyE::(Pspac-spoIIJN cat) JH642 transformed with DNA from M01022K1102 PB2 amyE::(Pspac-spoIIJ+ cat) PB2 transformed with DNA from M01022AG522 JH642 spoIIJ::Tn917 JH642 transformed with DNA from KS19 (MLS')AG785 JH642 spoIIJ::Tn917::pTV21A&2 AG522 transformed with pTV21A&2 (Cmr MLS5)K1418 JH642 spoOK418::Tn9171ac (spoOK::Tn9171ac) 24aK1566 JH642 spoOK::Tn9l7lac::pTV21A2 K1418 transformed with pTV21A&2 (Cmr MLSS)AG880 PB2 trp+ spoOK141 chr::Tn917 PB69 transformed with DNA from M0495 (MLS' Spo-)K1246 JH642 spoOK141 chr::Tn917 JH642 transformed with DNA from AG880 (MLS' Spo-)KI250 JH642 spoOK141 chr::Tn917::pTV21A&2 KI246 transformed with DNA from M0496 (Cmr MLSS Spo-)K1261 JH642 chr::Tn9J7::pTV21A2 spoOK141 spoIIJ::Tn917 KI250 transformed with DNA from K1418 (MLSr)K1738 JH642 spoIIJ::Tn9J7::pTV21A2 spoOK::Tn917lac AG785 transformed with DNA from KI418 (MLS')KI741 JH642 amyE::(Pspac-spoIIJN cat) spoIIJ::Tn917 K195 transformed with DNA from K1418 (MLSr)K1742 JH642 spoOK141 chr::Tn917 amyE::(Pspac-spoIIJN cat) K1246 transformed with DNA from M01022 (Cm')K198 PB2 spoIIJ::Tn917 PB2 transformed with DNA from AG522 (MLS')DZR18 PB2 spoOK::Tn9J7lac PB2 transformed with DNA from K1418 (MLS')KI100 PB2 trp+ spoOK141 spoIIJ::Tn917 PB69 transformed with DNA from AG522 (MLS')K1776 PB2 spoIIJ::Tn917 spoOK::Tn9l71ac::pTV21A2 K198 transformed with DNA from KI566 (Cmr MLS')K122 PB2 trp+ spoOK141 amyE::(Pspac-spoIIJN cat) PB69 transformed with DNA from M01022 (Cmr Spo-)K1772 PB2 amyE::(Pspac-spoIIJ+ cat) spoOK::Tn9l7lac KI102 transformed with DNA from KI418 (MLSr)DZR84 PB2 spoOKE::pDR30 Disruption of spoOKE

spoIIJ may be involved in the activation of a responseregulator that is required for sporulation, perhaps SpoOA.As part of our work on signal transduction and the

initiation of sporulation, we set out to isolate insertionmutations that would cause spoIlIJ mutants to have a moresevere defect in sporulation (24a). In the same screen, wewere also able to isolate sporulation mutations that could besuppressed by the overexpression of spoIIJ from a heterol-ogous promoter. Such mutations could identify genes encod-ing additional protein kinases or other proteins that act in thesignal transduction pathways. During the course of theseexperiments, we isolated a Tn9J7lac insertion mutation thatseemed to be in spoOK. We describe the phenotypic charac-terization of strains containing this spoOK::Tn9171ac muta-tion, the effect of the spoIIJ mutation and spoIIJ overex-pression on spoOK mutants, and the cloning and DNAsequence of the spoOK locus. Our sequencing results indi-cate that the spoOK operon contains five genes and ishomologous to the oligopeptide permease (opp) operon ofSalmonella typhimurium. spoOK was independently clonedand sequenced by Perego et al., and they were the first tonote the similarity of spoOK to opp (38).

Oligopeptide permease from S. typhimurium transportspeptides from two to five amino acids in length and cell wallpeptides (17). Peptides and peptidelike factors are ofteninvolved in microbial differentiation and behavior. Previ-ously, we presented evidence that B. subtilis produces atleast one extracellular differentiation factor that is requiredfor efficient sporulation (18, 19). This factor is heat resistant,dialyzable, and pronase sensitive, indicating that it is at leastin part an oligopeptide. The finding that spoOK encodes anoligopeptide permease that is required for efficient sporula-tion raises the intriguing possibility that spoOK encodes areceptor or transport system for such a differentiation factor.

MATERIALS AND METHODSStrains and plasmids. The E. coli strains used for cloning

and maintaining plasmids were JM101 [supE thi A(proAB-lac)(F' traD proAB+ laclIq lacZMJ5)]; AG115, derived fromMC1061 [araD139 A(ara-leu)7697 AlacX74 galU galK rpsLhsdR] by mating in an F' laclIq lacZ::TnS; and XL1-Blue[recAl endAl gyrA96 thi hsdRI7 supE44 relAl lac(F'proAB+ laclIq lacZMIS TnlO)] (Stratagene).The B. subtilis strains used were derived from strain 168

and are listed in Table 1. The original spoOK141 mutantstrain apparently contained at least two mutations, onemutation in the spoOK locus and another mutation, unlinkedto spoOK by transformation, that enhances the sporulationdefect of the spoOK141 allele (27a). The spoOK141 mutationhas been separated from the second mutation (8), and thestrains used here seem only to contain the defective spoOKlocus.The spoOK418::Tn9171ac insertion mutation was isolated

from a transposon library kindly provided by R. Yasbin (30).This is the same library that was used to identify Tn9171acinsertions in din genes (30), com genes (20), and csh genes(25).The plasmids used are listed in Table 2 and described in

the text.Media. LB (10) medium was used for routine maintenance

and growth of E. coli and B. subtilis. DS medium (48) wasused as the nutrient sporulation medium. Minimal mediumcontained the S7 minimal salts described by Vasantha andFreese (57), except that MOPS buffer was used at a concen-tration of 50 mM rather than 100 mM. Minimal medium wassupplemented with glucose (1%) and glutamate (0.1%). Re-quired amino acids were added at 40 p,g/ml. Media weresolidified for plates with 15 g of agar (Difco Laboratories) perliter. Sporulation proficiency was visualized on DS or 2x SG

VOL. 173, 1991

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

1390 RUDNER ET AL.

TABLE 2. Plasmids used4Commentsb and source or reference

VectorspTV21A2 ............................ Ap Cm (61)pBluescriptlI KS' and SK+ ............................Ap; abbreviated pKS and pSK (Stratagene)pGEM3Zf+cat-i (pGEMcat) ............................ Ap Cm; vector to clone fragments for integration into B. subtilis chromosome (in-

tegrative vector) (60)pJH101 ............................ Ap Cm; integrative vector (13)

OtherspDRl............................ Initial Hindlll clone from K1566 (spoOK::Tn9171ac::pTV21A2)pDR2............................ Initial EcoRI clone from K1566 (spo0K::Tn9171ac::pTV21A2); used to sequence

DNA adjacent to the end of the transposonpDR3 ............................ Initial BgBIH clone from KL566 (spoOK::Tn917Iac::pTV21A2)pDR4............................ Initial PvuII clone from K1566 (spoOK::Tn917iac::pTV21A2); used to sequence

DNA adjacent to the end of the transposonpDR5............................ Initial BamHI clone from K1566 (spoOK::Tn917Iac::pTV21A2)pDR6............................ 2.5-kb EcoRI-BgIII fraginent from pDR3 cloned into pJH101 (EcoRI-BamHI); used

to "walk" to the end of the operon and to disrupt spoOKDpDR9............................ 6-kb EcoRI-SphI fragment in pJH101; cloned by "walking" from a pDR6 integrantpDR11 ............................ 8-kb EcoRI-Sall fiagment in pJH101; cloned by "walking" from a pDR6 integrantpDR13 ............................ PvuII-CiaI fragment from pDR4 cloned into pJH101 (EcoRV-ClaI); the ClaI site is

in the end of the transposon; used to clone wild-type DNA that was the site ofthe Tn9171ac insertion

pDR16 ............................ 480-bp Hindlll fiagment from pDR3, containing the EcoRl site, cloned intopGEMcat; used to sequence across the EcoRl site

pDR18 ............................ PvuII-EcoRl fragment in pJH101 between EcoRV and EcoRI; cloned by "walk-ing" from a pDR13 integrant

pDR20 ............................ 2-kb PvuII-EcoRI fraginent from pDR18 cloned into pKS; used for exonucleasedeletions

pDR21 ............................ 2-kb PvuH-EcoRl fragment from pDR18 cloned into pSK; used for exonucleasedeletions

pDR30 ............................ 540-bp EcoRV-ClaI fragment cloned into pJH101; used to disrupt spoOKEpJL1............................ 1.7-kb EcoRV fragment from pDR9 cloned into pGEMcat; used to map spoOK141

and to disrupt spoOKEpJL2............................ 3.5-kb EcoRI finr ent from pDR9 cloned into pKS in the same orientation as the

lacZ fragment; used to make exonuclease deletions for sequencingpJL3............................3.5-kb EcoRI fragment from pDR9 cloned into pKS in the opposite orientation as

the lacZ fragment; used to make exonuclease deletions for sequencingpJL4............................ 3.5-kb EcoRI fragment from pDR9 cloned into pGEMcat; used to map spoOK141

and to disrupt spoOKEpJL7............................ 2.6-kb EcoRV-XhoI fragment from pDR9 cloned into pSK; used to make exonucle-

ase deletions for sequencinga All plasmids can replicate in E. coli.b Indicated fragment sizes are approximate.

Plasmids

(28) plates. Ampicillin was used at 50 to 100 jg/ml, chlor-amphenicol was used at 5 iLg/ml, and erythromycin andlincomycin were used at 0.5 and 12.5 Fg/ml, respectively.The last two drugs were used in combination to select forresistance to macrolide-lincosamide-streptogramin B (MLS)antibiotics encoded by Tn917 and Tn9171ac.

Spore asays. Cells were grown in DS medium at 37°C(unless otherwise indicated), and spores were assayed atleast 12 h after the end of exponential growth. The number ofspores per milliliter culture was determined as the number ofheat-resistant (800C for 20 min) CFU on LB plates. Viablecells were measured as the total number of CFU (before heattreatment) under similar plating conditions. Colony mor-phology on DS plates or 2x SG plates was used as aqualitative indication of the sporulation phenotypes of vari-ous mutants. Spo- strains tend to be translucent while Spo+strains are opaque on these plates.

Transf_o Jra. E. coli cells were made competent andtransformed by standard procedures (46). Cells ofB. subtiliswere made competent by the two-step method essentially asdescribed previously (12). Alternatively, cells were made

competent by growth in defined minimal medium (S7 me-dium) supplemented with glucose and glutamate. At a den-sity of 150 to 170 Klett units, cells were removed from theculture and used for transformation. Cells grown in this way,in the absence of complex mixtures of amino acids(Casamino Acids and yeast extract), appear to be competentthroughout growth, peaking towards the end of exponentialgrowth and losing competence during the stationary phase(lla). Transformations were performed with excess DNAfor the quantitative determination of transformation frequen-cies.Clong of spoOK. DNA adjacent to the spoOK::Tn917lac

insertion was cloned into E. coli by methods described byYoungman et al. (59-61). In brief, the part of plasmidpTV21A2 containing an E. coli replicon and selectablemarkers for B. subtilis (Cm') and E. coli (Apr) anked bysequences from Tn917 was recombined into the centralregion of Tn917lac, resulting in Cmr MLS' transformants.DNA from these transformants was digested with one ofseveral restriction enzymes and ligated at a dilute concen-tration to promote intramolecular ligation. The ligation mix-

J. BACTERIOL.

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

B. SUBTILIS spoOK OPERON 1391

pDR5 pDR30pDR4/pDR13 pJLl

pDR1pDR2______pDR3

pDR6

pDR18/pDR20/pDR21- pDR16

pjL7

.yVL pDRl1pDR9

pJlI/pjL3/pjL4FIG. 1. Organization and restriction map of the spoOK operon. The locations of the likely promoter and terminator sites are indicated by

P and T, respectively, and by arrows. The location of the Tn917lac insertion is indicated by a triangle and Tn. The predicted open readingframes of spoOKD and spoOKE overlap slightly. The restriction sites are indicated as follows: B, BamHI; P, PvuII; H, HindIII; C, ClaI; E,EcoRI; R, EcoRV; G, BglIl; X, XhoI; S, SphI; and L, SalI. (E/H indicates an EcoRI site 3 bp upstream of a HindlIl site.) The map is onlypartial for the HindIII, ClaI, and PvuII sites; it is complete for all other sites shown.

ture was used to transform E. coli, with selection for Apr.Candidate clones were screened by restriction analysis, andthe cloned B. subtilis DNA was subcloned into appropriatevectors as needed.The 'initial clones and/or subclones were tested to see

whether the DNA extended past an end of the transcriptionunit needed for spoOK function or whether the DNA wasinternal to the transcription unit. This test was done byintegrating clones that contained the cat gene into thechromosome of wild-type B. subtilis cells. If the clonedDNA were internal to the locus, then integration (by singlecrossover) would disrupt the locus and cause a sporulationdefect. In contrast, if the cloned DNA extended past one end(either end) of the transcription unit, then integration wouldgenerate one truncated copy and one intact copy of spoOK,resulting in a wild-type phenotype.

Additional sequences from the spoOK locus were clonedby "walking" from existing clones. Plasmids were inte-grated into the chromosome (by single crossover), and DNAfrom such transformants was cut with appropriate restrictionenzymes to generate fragments containing the vector back-bone and adjacent chromosomal sequences. DNA was li-gated to promote circularization of fragments and trans-formed into E. coli. Again, candidate clones were examinedby restriction analysis and integration into the B. subtilischromosome to determine whether the'cloned DNA wasinternal to the spoOK locus.DNA sequencing. Nested deletions were made with exo-

nuclease III (New England BioLabs) and exonuclease VII(Bethesda Research Laboratories) essentially as describedpreviously (36). Plasmids used to make the deletions wereprepared by the Qiagen (Qiagen Inc.) procedure. Six sets ofdeletions were made to sequence the whole operon on bothstrands. Fragments of the operon were cloned into thepBluescriptIl KS and SK (pKS and pSK) plasmids (Strata-gene) to generate plasmids pDR20, pDR21, pJL2, pJL3, andpJL7 (Table 2 and Fig. 1). pJL7 was used to make two setsof deletions, one set from each end of the cloned fragment,starting from different ends of the polycloning site. Frag-ments were ordered by size and sequenced with the appro-priate primer. Primers included the reverse primer and the-40 and -20 forward primers (U.S. Biochemical Corp.).

Specific primers to fill in gaps in the sequence and tosequence DNA adjacent to the ends of the transposoninsertion were purchased from' Oligos Etc. Inc. (Guilford,Conn.). The sequence across the EcoRI site, located at theends of plasmids pJL2-pJL3 and pDR20-pDR21, was deter-mined from 'pDR16. Double-stranded plasmids used forsequencing templates were prepared by the alkaline lysismethod (46).DNA sequencing was done with the Sequenase kit (U.S.

Biochemical Corp.), with double-stranded plasmid DNA asthe template and [a-35S]dATP (Dupont, NEN ResearchProducts) as the label. The sequence was determined fromboth strands, and much of the sequencing was done multipletimes because of overlaps in the DNA fragments used. In afew places, there were differences in the sequence from onestrand to the other. These discrepancies were resolved byrepeating the sequencing with both dITP and dGTP.

Sequencing samples were electrophoresed on 6% poly-acrylamide gels with 8 M urea by standard procedures (46).Gels were fixed, dried, and exposed to Kodak X-OMAT ARfilm. DNA sequence analysis, manipulations, and compari-sons were done with the package of programs provided bythe Genetics Computer Group, University of Wisconsin (11).

Nucleotide sequence accession number. The spoOK se-quence has been assigned the GenBank accession numberM57689.

RESULTS

As part of our work on signal transduction and theinitiation of sporulation, we set out to isolate insertionmutations that would cause spoIIJ mutants to have a moresevere defect in sporulation (24a). In the same screen, wewere also able to isolate sporulation mutations that could besuppressed by the overexpression of spoIIJ from a heterol-ogous promoter. During the course of this work, we isolateda Tn9171ac insertion mutation that seemed to be in spoOK.We refer to the insertion mutation as spoOK418: :Tn9171ac orspoOK::Tn9171ac. While the transposon is Tn9171ac, theorientation of the insertion is such that the lacZ gene is notexpressed.spoOK was originally defined by the mutant allele

VOL. 173, 1991

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

1392 RUDNER ET AL.

spoOK141 (Z31 of Coote [9] and Piggot and Coote [42]). Thismutation causes a partial defect in the development ofgenetic competence (44, 45; unpublished results) as well as adefect in sporulation. We first established that the insertionmutation was likely to be in spoOK and then used theinsertion to clone the spoOK locus. The phenotypes causedby the insertion mutation were similar to those caused by thespoOK141 mutation. Definitive evidence that the transposonwas in the spoOK locus came from cloning and characterizingDNA surrounding the transposon. We demonstrated thatclones of DNA adjacent to the transposon could rescue thespoOK141 mutation by recombination and that other clonescould disrupt the spoOK locus when integrated into thechromosome. Clones were used to sequence the locus and tofurther characterize the function of spoOK.

spoOK418::Tn9l71ac and spoOK141 cause similar pheno-types. Strains containing the spoOK418::Tn9171ac mutationappeared Spo on sporulation plates, and the coloniesresembled those of an isogenic spoOK141 mutant. The blockin sporulation appeared to be the same as that in spoOK141,as judged by the appearance of the cells on plates and in thephase-contrast microscope.

Strains containing the insertion mutation were compareddirectly with strains containing the spoOK141 mutation fortheir effects on sporulation under several -conditions. Bothmutations caused an oligosporogenous phenotype, and thatphenotype was more severe in the PB2 genetic backgroundthan in the JH642 genetic background (Table 3). In addition,mutations in spoIlIJ caused both spoOK141 and spoOK::Tn9171ac mutants to be more severely Spo- (Table 3).Overproduction of the spolIJ gene product from the isopro-pyl-f3-D-thiogalactopyranoside-inducible promoter P, par-tially suppressed the sporulation defect ofboth the spoOK141mutant and the spoOK4l8::Tn9171ac mutant (Table 3). Thesuppression of spoOK by the overproduction ofspoIlJkinaseis consistent with the notion that spoO)K is involved in theactivation of SpoOA, perhaps by activating (either directly orindirectly) one of the histidine protein kinases.The strain differences seen with the spoOK141 mutants

were also observed with the insertion mutants (Table'3).Furthermore, disruptions of the spoOK locus generated byintegrative plasmids caused similar phenotypes and similarstrain differences. Therefore, the strain differences were notdue to multiple mutations in the spoOK141 mutant.The spoOK141 and spoOK::Tn9171ac mutants also had

similar defects in genetic competence. The ability of themutants to be transformed with exogenous DNA was re-duced to approximately 0.1 to 1.0%o of that of the isogenicwild type. Additional mutations in spoOK made by disruptingthe locus with cloned fragments in integrative plasmids alsocaused a defect in competence (see below). In all casestested, the defect in competence was seen in both strainbackgrounds. The fact that the transposon insertion anddisruptions with integrative plasmids caused defects in com-petence similar to that caused by the spoOK141 allele indi-cated that the competence defect was not due to additionalmutations outside spoOK and that spoOK is required for thenormal development of competence.

It is interesting to note that in contrast to the defect insporulation, the defect in competence in the spoOK mutantwas not relieved by the overproduction of spoIlJ (data notshown). This difference indicates that the role of spoOK incompetence may be different from its role in sporulation.Genetic mapping experiments indicated that the Tn9l7lac

insertion mutation mapped near the spoOK region of thechromosome. spoOK418::Tn9171ac was approximately 5 to

TABLE 3. Sporulation of strains containing spoOK141 andspoOK::Tn9l7Iac

Stra.ina spoQKb spoIIJc IPTGd Frequencye

JH642 + + - 1.0AG522 + Tn - 7.1 x 102KI250 141 + - 1.9 x 10-2KI418 Tn + - 1.8 x 10-2KI261 141 Tn - 4.0 x 10-4KI738 Tn Tn - 1.3 x 10-3KI742 141 Pspac-llJ - 1.5 x 10-2K1742 141 Pspa,-IJ + 1.5 x 10-1KI741 Tn Pspac-IJ - 1.9 X 10-2KI741 Tn Ps,-IIJ + 8.4 x 10-2

PB2 + + - 1.0KI98 + Tn - 6.5 x 10-2PB69 141 + - 6.0 x 10-5DZR18 Tn + - 3.7 x 10-4KI100 141 Tn - 2.5 x 10-6K1776 Tn Tn - 1.9 x 10-K122 141 Pspac4ilj - 1.8 x 10-4KI22 141 PspacdIIJ + 2.0 x 10-2KI772 Tn PspacIIJ - 2.9 x 10-4K1772 Tn Pspac4iIJ + 1.1 X 10-2

a The first 10 strains are all isogenic with the wild-type strain JH642. Thesecond 10 strains are all isogenic with the wild-type strain PB2.

b +, Wild-type allele; 141, spoOK141 allele; Tn, spoOK4l8::Tn9l7lac allele.+, Wild-type allele; Tn, spoIIJ::Tn917flHU19 allele; Pp,,-IIJ, amyE::

(Ps -sPoIIJ+ cat) construct in a spoIIJ+ strain.W=en present, isopropyl-,B-D-thiogalactopyranoside (IPTG) was added

during early- to mid-exponential growth at a concentration of 1 mM.Sporulation frequency (number of spores per viable cell), normalized to

that of the wild-type strain, JH642 or PB2.

10o linked by transformation to a marker that is also 5 to10%O linked to spoOK141 (data not shown). The marker usedwas a plasmid integrated into the chromosome near spoOK.The plasmid contained cloned DNA adjacent to a silentTn917 insertion (from strain M0495) linked to spoOK (datanot shown).The phenotypic and preliminary mapping results indicated

that the Tn9171ac mutation was most likely in spoOK.Definitive evidence that the transposon was in the spoOKlocus came from experiments that demonstrated that clonesof DNA adjacent to the transposon could rescue thespoOK141 mutation by recombination and that other clonescould disrupt the spoOK locus when integrated into thechromosome on an integrational plasmid (see below).

Cloning and DNA sequencing of spoOK. DNA adjacent tothe transposon was cloned into E. coli by the methodsdescribed by Youngman et al. (59-61) and summarized inMaterials and Methods. Clones to the right of the transposon(as drawn in Fig. 1) were obtained with HindIII (pDR1),EcoRI (pDR2), and BglII (pDR3) sites at the end; clones tothe left were obtained with PvuII (pDR4) and BamHI (pDR5)sites (Fig. 1 and Table 2).

All three clones to the right were internal to the transcrip-tion unit required for normal sporulation, while both clonesto the left seemed to extend past one end of the transcriptionunit. These results were obtained by integrating the fourclones (HindlIl, BglII, PvuH, and BamHI) that still con-tained the cat gene into the chromosome of wild-type B.subtilis cells. If the cloned DNA were internal to the locus,then integration (by single crossover) would disrupt thelocus and cause a sporulation defect. This in fact was theresult with the HindlIl (pDR1) and BglII (pDR3) clones. Weassumed that the EcoRI (pDR2) clone was also internal, as

J. BACTERIOL.

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

B. SUBTILIS spoOK OPERON 1393

the EcoRI site was contained within the larger BglII clone.In contrast, integration of the PvuII (pDR4) and BamHI(pDR5) clones resulted in a Spo+ phenotype, indicating thatthese clones extended past the end of the locus.

Preliminary DNA sequence information was obtained forDNA adjacent to both ends of the transposon in pDR2 andpDR4. We used oligonucleotide primers complimentary tothe ends of the transposon to determine the sequence.Analysis of this DNA sequence indicated an open readingframe going from left to right in Fig. 1. We concluded fromthe integrational mapping and the preliminary DNA se-quence data that the 5' end of the spoOK locus is between thePvuII site and the transposon (Fig. 1).

Additional clones that extended past the 3' end of thespoOK locus were made by "walking" downstream withintegrational plasmids. The EcoRI-BglII fragment from theBglII (pDR3) clone was subcloned into the integrationalvector pJH101. The subclone (pDR6) was integrated intowild-type cells, and DNA was made, restricted with SphI orSail, ligated, and transformed into E. coli. When integratedinto B. subtilis, the clones (pDR9 and pDR11) gave Spo+transformants, indicating that they extended past the 3' endof the locus required for spoOK function.

Subclones were made in the pBluescriptIl KS and SKvectors and used to generate variously sized fragments byexonuclease deletions for sequencing the spoOK locus.pDR20, pDR21, pJL2, pJL3, and pJL7 were used to makeexonuclease-generated nested deletions, and the resultingclones were used for sequencing.The DNA sequence of spoOK was determined by the

dideoxy chain termination method on double-stranded DNAtemplates. The sequence was determined on both strands,and much of the sequence was determined multiple times.Analysis of the sequence revealed five open reading frames,called spoOKA through spoOKE (Fig. 1 and 2). The transpo-son insertion was in spoOKA, the first open reading frame.Immediately downstream of spoOKE, the last open readingframe, was a sequence that looked like a typical factor-independent terminator (nucleotides 6085 to 6119).spoOK is homologous to the S. typhimurium opp operon.

Comparison of the five spoOK open reading frames with theGenBank data base (using FastA) revealed significant simi-larity to the proteins encoded by the opp operon of S.typhimurium (22) (Fig. 3). opp encodes a membrane trans-port system required for the uptake of oligopeptides and cellwall material (17). Other similarities between the two oper-ons include the order of the genes, a potential stem-loopstructure between the first and second reading frames, and apossible overlap in the coding regions of the last two readingframes (Fig. 2 and 3).

Preliminary experiments indicated that there was a pro-moter located between the PvuII site and the transposon. AHindIII-ClaI restriction fragment of approximately 900 bp(the HindlIl site was approximately 100 bp downstream ofthe PvuII site, and the ClaI site was in the end of thetransposon) was subcloned into a promoter probe plasmid(pDG268 [6]), and the resulting lacZ transcriptional fusionwas crossed into the amyE locus ofB. subtilis. This fragmentcaused a considerable level of expression of lacZ (44a),indicating the presence of a promoter. Analysis of the DNAsequence between the HindIII site and the start of spoOKArevealed several sequences similar to promoters recognizedby the major form of the RNA polymerase holoenzyme. Thesequence that most resembles the consensus promoter rec-ognized by RNA polymerase containing sigma-A is TTGCAA-17 bp-TATAAT. This sequence, found at nucleotides

295 to 323 (Fig. 2), matches the consensus -35 sequence infour of six positions and matches the consensus -10 se-quence and the spacing between the two sequences per-fectly. In addition, the sequence immediately upstream ofthe putative -35 region is extremely A+T-rich, a character-istic of many B. subtilis promoters.Mapping of spoOK141. We used integrational mapping to

determine the approximate location of the spoOK141 muta-tion relative to various subclones. When pDR9 was inte-grated into the spoOK141 mutant, approximately 35% of thetransformants became Spo+, indicating that at least onemutation required for the spoOK141 mutant phenotype liesbetween the upstream EcoRI site and the SphI site.

Similar integrational mapping experiments were done withpJL4 (containing the -3.5-kb EcoRI fragment) and pJL1(containing the 1.7-kb EcoRV fragment). These experimentswere a bit more complicated to interpret because integrationof these plasmids caused disruption of spoOKE. However,disruption of spoOKE, in otherwise wild-type cells, by inte-gration of pJL1 or pJL4 (also pDR30) resulted in little or nodefect in sporulation (see below). This phenotype, whilesubtly different from that of the wild type, was clearlydistinguishable from the phenotypes caused by the spoOK::Tn9171ac insertion, the spoOK141 mutation, and disruptionsin the other spoOK open reading frames. Integration of pJL4or pJL1 into the spoOK14l mutant resulted in approximately60 to 70% and approximately 30% "Spo+" transformants,respectively, as judged by colony morphology on sporula-tion plates, indicating rescue of the spoOK141 mutation. Weconclude that we have indeed cloned the spoOK locus andthat at least one mutation that is necessary for the spoOK141mutant phenotype lies betweenthe two EcoRV sites.

Disruption of the spoOKE open reading frame. Disruption ofspoOKE, the last gene in the spoOK operon, did not cause asignificant defect in sporulation. We disrupted the spoOKEgene by integrating pDR30, which contains an internalfragment of the gene. This mutation caused a small (-30%,relative to the wild type) but reproducible defect in sporula-tion (Table 4). In contrast, the same mutation had a dramaticeffect on competence. The defect in competence caused bythe spoOKE mutation was similar to that caused by thetransposon insertion (Table 4). Thus, spoOKE is required forthe development of competence but has only a minor role inthe initiation of sporulation.

DISCUSSION

The five proteins encoded by the spoOK locus are homol-ogous to the five proteins encoded by the opp (oligopeptidepermease) operon of S. typhimurium (Fig. 3). There areimport and export systems in both prokaryotes and eukary-otes that are similar to those encoded by opp and sjioOK (5,7, 24). In gram-negative bacteria, the uptake of maltose,maltodextrins,'phosphate, histidine, branched-chain aminoacids, and many other substrates is mediated by transportsystems analogous to that encoded by opp (5). In thegram-positive bacterium Streptococcus pneumoniae, the up-take of maltose, maltodextrins, and oligopeptides is medi-ated by similar transport systems (4, 16).The export of a variety of compounds is also dependent on

transport systems that are similar in structure to thoseencoded by spoOK and opp. Hemolysin secretion frompathogenic strains of E. coli depends on the hlyB and hlyDgene products, which constitute a membrane-bound translo-cation complex that is structurally similar to the importsystems (7). In eukaryotic cells, the similar transport sys-

VOL. 173, 1991

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

Co oC 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Co kD (N 0) w 0 oCD 0 lq 0) kO 0) lw 0 CN OD IT 0 'IO (N4 O) lqO0) 0) 0 1- 4 (N C(4 en) lw %V ) Lr r- C- 0) 0C) 0 0 1 -4 CN en)

1- (N (N4 CN (N (N (N N (N N (N(N 44(N N(N e) en)

< z C)OCa < D H E-C C4 C) CD Hl L) H H-

El C)DE- H H40 EH C)-( H HDU H CD- D D E44 4 CD (E-iiU(D 0 0 E1 (D ( H- 4 CD Hl D C-)

iE44 .4 0 < E4 < C C) Hq H, C) 4 D E) 4

CD F CD> CDW C)a H0() E- 4< H ( DC 4L ( )0 D> 4~ CD>

e4 C) CD H Hl CD CD 4 CD < CD 0CEH1> HE- 4 ~ 4~ -C F,W D CD CD )C.) -1 EiE4H- H4E

4 4 ~ Hi CD < 4 4 CD 4 CD CD CD Hl

CD C C ~4 ~

C D0 H>s CH CE4 e4 I) H- CD H H H CD C) CD H

-CD (CD CEC4 CD H

Q

)-E-

0) (D00 (D LI)> (D D

4 HE4U) 4 H

C) U)CICDCD C> (D CD~z ~

nCDU U LC)H)E- CDC)U4E4 E4 2 4F 4 F4E4C)C 4 HD C)F4 C C)a0:4 H

)E-4 E- U) E- C) A D < (HE- 4 (D E- ~H ) C)4 E 4 E CE H CD4 E- -4UUF4(D) a CD(D(D E4H Hl l4 4 U D4H4H H:E44<2 4 N

4 4 D CD H-4 ) 4 D ElC D

Z) Z U 4 0 EH 444 D (DH- CD (D CD H4 H- H CD C)UD ( 4E4 C D C D > F 4E4U 4E(DCH (D > ( CD z

HL)LI)CCC C-) H4 CD 4 H H4 H4 H C.) CD 4C).- D- D E-4~H 4 Hi H C.) CD 4 C-) 4 4 4 CD

uC)HCD E- (DHECD UC C) CD < H CD 4~4u a4 D W u 0 H (D H 4 CD Hl C)

(-D u 4 F CD (CD Hq C)

CD0 ( D0HMCEC4NC,U 4 Ei E- , LC&H -4 E-H<HC H CD CLDC)E- CID CD > CD Z~ > -4(D4 CE ( DU 4 lE- F>HH C.D CD LI) CD HE rCCr 4 4 CD P4 CD < CD CD C)Cx. D

4 FI O 0 CD) D HE- 4 CD CD C-) CD )EC4)

C)u4 CD ECDC)

(D E-H C) C) ).C

H4 Cq (u E-4 H0 CDDU E- E-4C) E-4C) UC CD4

C) C) H4 D C)4 CD)CuC)CY

E-

E-

C) H CDu CD C) C)D CD CD 4~ HCD HH UC) HDUCE-)E-CD CD CD Cu H4 H CDH

0(D0C 0 rl a D (D ID CD El U

C) CD CD4 H4 0 C)4 E4 C-) CD4(.DFi (.D E-4 UC) CLD C) C) Hl C.) E

H-4 CU)~ l CD C4 CD H- CD C.D EH CD CD CD H4 C) C)

Du 4 C) HE4D 4 (D WCD > D Z HEC444 D P4CZD U CD< DH-I C) C,~urLD0E-4 < H EH C-) CD C-) CD C) C.D D

C.) C.D C) 4 4 iu a44HCLDHILDC DD CD>C)CL>U 0 HE4 2 EH- 4 C.D E H4

IDC (D<~4

D CLD CD H C)u u E1 Lz E-40)HEC4 DC4 D CE--4 EiH u C.) < C) H4 L) E HE HE H-4 E4LDD E-C E) U 4 a .) ( 4 E- ~ C)04 U 4F l(D D> E4 CD

D HE H F H4 Hi D

DQ CU ~ 9 C D0E H4>4 HE- O, ) CD H- CD E ( D

CC) H CD C-)vH Z ) ~EH (D > HE4 H DCDC) D >HLE HCI)<EC/)W4(n < HE- CD > HF 44 U F )

C.D CD l QC CD CD Hl CD C) HE-

lw 0 %0) (N OD) 0 (N 0) 0 (N4 0) 0 (N4 O) 0 'IO (N, 0)

r- 0) ON 0) 0 -4 - N en(1 'w(CL)u) OD 0) 0 (N0

-4 14 - -4 -4 (li N (N (N (N4 (N (N (N (N (N (N (N (N N(N en)

00000 00C)C 0 0 0 0 0 0 0C 0 0 0 0 0 0)CC)) 0 0 0 0) 0 0

0)(N4 0)V '0D0C N 0)lq 00)C4o k (N OD w 0) O0) T 0) CN O0) v

"E-4 0 4 00) 0) w > 0> E- ~4 ( 4( (Ns 4 E- )E4 renE 4(0 Ur-4

E CD4<0CD4 CD u E-Cf u U C-)HCDE( 4 Hi EC C CD4

IC)CELIDH HH< X (nE4L <ED C)4 C D0 4 E4E4WU(DC)H4H4 (. W E- H4E- C) H4E C-D ( QU E CD CLD (D E- 'D CD H-

CD E HE 4 EH C-) H E- E-H EH E-H C)

E-CDHHHCD HD ( <QE4~ D 0 U 0 D > (

C)4 E- -4(D U L E-

C) CCD CDUU(DF~~E-4E-4C 4 < C) CID I

E C)E HHC)iE-L 4 c 4 E-0CDEH >4( )a W ~ 4(DID 9 C)0 D<D 4 C.) C.D CD E F

C)H CD4 C)C)HUE-

E-

C) C) CD C) C)

E-

EHC.) CD HH4E- C) (DC(D0 4: z~UDCD<4 (DCD) <HCD<H.-4r24EH EHC ).-

D < CD (D N4CD)EH4 EH C.D D C(D H4 HD CD C) C_) E-4-4H C-) CD

El CD H CrD C4 D CrD ~ E.DCD>EHE--4 ) DCFCHCC),4H 4 H ( D ~ D

<< CrD E44CD u D<.)C)q u(D E-

C DHC)4 D CD CD CD s< (D

CECD4<D Hu 0

CDL 1~ z SD 0 ci

0

HgHH HUH C)DH CD C) C) < < <H0uuHC)CDHC) NoHCDHV44)QHHU CL5E-< L< a C)LLDCD&H< CE)H U (4E4UC)4CC0E)

HHoC 4 H HU E4E D)g E- 4 CD0 E< D'

E-q 4 U E- b4< 4 L D D CD H C)D>~ W U 4 E-

E- )HE4 ( H H4LC) ( EH<~~~~U ~4 CD IL HE- CrL W -L4ZE) C.D> >C4DW )0CD>F4E- 4 ~C1 D E- >4 ~

CED-CHD CD EC EH EC4D CD 4 H CD

H

E HUiF E4 FCD C)

E-4<C) Z E

< C) H4 CD CD XCDW E

E-CEDF< CD P< -4'HE-1 EC) I)< < C ) H

U E-C) C) C) C HCD<< C ) C C)4

H

E4C) H rL CD rDC) D ID F,F

H H H CDECD<E54C)EUH CCD H <

HHC)-C)' CD )

4 Z L U

C) C H) CD H C) C) H C) HODU M F UCD

E 4 U U

HCDI<<C)>C CD4

CD CD C) <DC)4E4> ) 4 ID W u c

CD

F4 <4 4CE)4 D H C) LD )EDC4C

ElC)CE C4 C)DCD CDu 4 DuHH4 C)D ( 4 4 H 4 z 0 fD E1~cCD4 0 C L L <~

C)H

UCD F u U E CH CD HCDCDD)C CD FHH UC). ED< D0)C <CDEHHEC)( H-- - - )4U D( z( -

CD C CD D Z> CDC) H H

>C)

> C)4 C)EHH C) < H C) H C) H CD- -

D H H- EDC)

HHH<CD<C) D C.D CD H CD < C) H 4C CH)ED QC E-fCE C ) < < H

4C.)C)E-HH- E< E4 )ECD CDCD4 E< D 0)ECZ 4 E4 4 0ECDCD< E-

CCHj H-C1C N4 co C)1 C)L CZ

E F4 E0 0 (N (' .

E-44Enen0 F D C.) E (0 0-)

1394 RUDNER ET AL. J. BACrERIOL.

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

B. SUBTILIS spoOK OPERON 1395

o0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 000

0 lw 0 %D CNJ 0 %D CNj v 0 %D C14 0) 0 ko(0 clJ0) 0 (0 1- -4 CN C14 CY) lw W LO U') kD r- r- 0)% 0 4 1(YC=C'

lw LO Un U') LO Un LO U') U) U') U) U) kD %D W o Dw C3

004~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-r.0O)) U D W ( DU~I) iL~)CD L 0 (D D00 (D( H HH4E

ElH 4 LO > 4 LI CD 0 0D C) Hl ( 0 EC) 0H)E-4 ID C E 4 H C.) E L 1

)~~~~HCD C) H~4E 4 i-4 E-U) >C0W ( ~E4~ 4( 0in 0D E- HE UODOH>H E 4 D C 0 -1EH-HH1 0DC) H

H~~~~~~~~Q)H Q El C 0 4CD 0-H00 4C I

E- C.) C.) () ) E-O1E-4~H CDU0 #C) 0 H( C)4E4

C)4 0 ( W C.0U>0E0 EC-4&Ci I

)U 4-0

C)D4 4 H4~Ua40 -4r

HDr0) 4 H E-

E-4 CDC.4) EC4 C)4 C)

4

C) C)L'H0UuE- E- E- 4 (DCDCH( U 01C u a4u M:4 E 00CD0

C)4 0 - C-)E-4H()C.)HH CD EHC) C

HD C) CD) UE-'ECEOC 0- H- H 4 )

C)~~~O'c*O>C~~~~~~~~~~ 4

0l 0 C)4 u 2: b n I) M:L iLHCD H HCD 4 0

C.0 D4 C~~> Ca)F04 ( C))U HD LI) C E- E

E-O E-4 U 0 0i>L C) H4(D 0.)C) C)0D [-)E4W ( D(Dc

E-4 E-84 HFHlC U F H HD HcnL)E1H U bD

Z 0 W DW E4uH -i i- H-oC)~ H u) 4I) )E00C

0EH 4 HE 0E- ) 4 C00 r0D( 0) 2:~ 04 Hi 0 HCD HDWC C) D> E40 4U a )F0 - C

C )D4C)OCDi, .) 0CDH H C)4 HI C. C) E- 4 CHDE4 E-4 H- 0D E- H, C.

C.) E- 04 -0 W 0) 0 0 )> U >E 4)- E C.) )) E HC

-4 4 40COCH H4( 4 H1 i)4 0'sUH) H. -)CD -)0E HD U E- C 0HEHOEOH U 4C C) E4 40CDC.)) CD E

D 0 0 C ) % ( E-HH ( HCD L E4 4 CH C)4 0

4O0 L) E-4 C) 0U % C. Q ) C E4 l > 0> HHE--4(L C C) C))E HD C.) ) 0E- H

-4 E4 U C)>0(D W 44 4O Zi-H4CE)0H-O C.) W E

CD ( UHD 4 H- CD E- HC

L) C1 E CD0E- 0D U0eJ 4' 4O Ea4 CD0 E-D ECD 0 --4 C)4 ~ (0 H' Ea) E-) CD 0- 4 C~)

Eq U U U)D U) U)4 U) U U)D CD EO4HO4O0

4Z5~~~~CO

0~~~~~~~E- 0> D C 0a 0- D I 0C. 00D 0C W 0- 0E~~C E-4 LI) Ea)CD0EAeHH E-4 E0UCLfin,) UE-U) (D (- E-) 00DC-

C')E')'E')-C'CDC')-C')

)FHD>0 )CDD Z -4> 4) CiFt

o H1 C)4 OD q 1 00 V 0 % 0)ccV 0 Dco ( n 0 14 r4 C1 1 1) V V f 0, O 0D CA 4 4C

0r V L n L Hl) L0OH) L H' o W k

~ ~ ~ ~ ~~ C

~~~~~0 0 > 0 CD 0 > ) > C H) 0C C C) C> 0 4)$.

Hr CV) C1) C)m V V VO0CV"C) V )HW-4cW

HF00 H CD CD >4 u 04 u0>u 4U .H 00 F

HI E-40H44Hw~4uuz g 0> 00 4HC)H0 0 rCAC)

0 0 C)0 H CIOCD CD OCD CD 0 H E H C z 4 H 0 CD~C H C)0 AO%H 0 a H 0uC D H H Ha H H C)a

CDH0 H E-1 I p H 0 H~ 0 H 4 0 0H0O

H w H~c0H u 0 HH C CD

CI~HO W co aH HO A H HE 0H

H~O

H co

u 4H c D> C ~0 H oH H W u0~c C) 40 H 0 C)i~~~~~~~0F,

E U-0H4

C)4 0OCaH0H

C) H~~0 C

~ ~ ~ ~ ~ ~ ~~~~~~H H U H t

EOH CD .CE (1 C D > 4 2 4 0:0D>E- C) 0

EC HC0)CH- E H-1 OEl HuC) HCZElE)-OCl4HC 4 U CD> D > X C) 0 E4(HC0O4C)El HD 4

H40 0 ( D 3 D L i>4C)H(HU0 0 4 HD E0 E-H_) OsH~ D U(E-HC)Cl4 C) )

E- 'H.) E OOEHi4CD)(4u' (L.)) ~4 C.) 0 0 > (0 I..CD E-wC) 0 0 0 HH0i C) C)

2 E-4 L~O C.) CC.) )C E- HI HI 0C) C)

E EC 0E- E4E-HO4U C)4 H) &q C.) 4 E 4 >

H~ (1 4,04 C)4 H- HC) CD H UC)E- C) CHCOH4 E- C*E4E4 E4 4 -

ECD~~~~~~~~~~~~~~C

E-C 0UC E-0 (N9 4 H 0I CD (C)WE- 4 H D >H' A CD a ~CD E-40Z 0)40 0 0 F4U)U)4O CDl E-4 U) r.

E') C'L C))C.DC') CD) CD CD U 0 ) .)U O0)

VOL. 173, 1991

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

1396 RUDNER ET AL.

I spoOKA |IspOKBIspWKA spoOKD IspooKEl545aa 311aa 305aa 358aa 308aa

% Identity; 34; 54

Similarity48;73 43;67 51;69 53;69

|oppA | |oppB |oppC |oppD | oppF542aa 306aa 301 aa 335aa 334aa

FIG. 3. Homology between the spoOK operon of B. subtilis andthe opp operon of S. typhimurium. Percent identity and similarityare given as determined by the BestFit program (11). aa, Aminoacids.

tems seem to be composed of a single large polypeptide thathas domains that are related to the individual polypeptides inthe prokaryotic systems. These systems include, amongothers, STE6 of Saccharomyces cerevisiae, involved in theexport of the a factor mating pheromone (31); MDR orP-glycoprotein, responsible for multiple drug resistance inmammalian tumors; CQR, responsible for chloroquine resis-tance in Plasmodium falciparum; and CFTR, the cysticfibrosis transmembrane regulator (reviewed in reference 7).These systems play important physiological roles, and manyare involved in controlling differentiation and gene expres-sion.One of the components of the import systems is a high-

affinity substrate-binding protein (SpoOKA, OppA) whichpresumably interacts with integral membrane proteins thattransport the substrate into the cell. In gram-negative bac-teria, the substrate-binding proteins are located in the peri-plasmic space (5). In contrast, gram-positive bacteria do nothave a periplasmic space because they do not have an outermembrane. The substrate-binding proteins that are localizedto the outside of the cell are thought to be attached to thecytoplasmic membrane through a lipoamino acid anchor (4,16). The NH2 terminus of SpoOKA has the characteristics ofa signal sequence and has a sequence (L-S-A-C-G, starting atamino acid 18) that is similar to the consensus sequence oflipoprotein modification sites (4, 16, and references therein),indicating that SpoOKA may be a lipoprotein. The work ofPerego et al. (38) has shown that some of the spoOKA geneproduct is found free in the culture medium while some isassociated with the cell.The other components of the transport systems include

two integral membrane proteins and one or two ATPases.The membrane proteins (SpoOKB and SpoOKC, OppB andOppC) are hydrophobic, have characteristic transmembranedomains and, in some cases, are similar to each other. ThespoOKB and spoOKC gene products are approximately 20%identical and -51% similar to each other. The energy forimport presumably comes from the hydrolysis of ATP byone or two proteins that are on the cytoplasmic side of themembrane and associated with the integral membrane pro-

TABLE 4. Effects of a spoOKE mutation on sporulationand competence

Strain Mutation Sporulationa Competencea

PB2 None (wild type) 1.0 1.0DZR18 spoOK418::Tn917lac 7.6 X 10-3 2.9 x 1O-3DZR84 spoOKE::pDR30 0.34 <1 X 1o-3

a Frequency relative to th^at in wild-type cells.

teins (21, 24, 32). SpoOKD and SpoOKE are the ATP-bindingproteins, as determined on the basis of the sequence homol-ogy to OppD and OppF and a region of homology tonucleotide-binding sites. SpoOKD and SpoOKE are homolo-gous to each other, having -42% identical amino acids.

Mutations in spoOKE did not cause a SpoOK- phenotype.The spoOKE mutation caused little or no defect in sporula-tion but did cause a defect in competence similar to thatcaused by mutations in other parts of the spoOK operon(Table 4). Thus, spoOKE is required for the development ofcompetence but is dispensable (or almost dispensable) forsporulation. It is interesting to note that oppF of S. typhi-murium (the homolog of spoOKE) is required for Opp func-tion. That is, mutations in oppF cause an Opp- phenotype(22). Perego et al. have also shown that spoOKE is notrequired for sporulation (38). In addition, they have shownthat spoOKE is not required for the transport of a toxicoligopeptide into B. subtilis (38). Again, this result is incontrast to the requirement for oppF in oligopeptide trans-port in S. typhimurium (22).The role of spoOK in sporulation and competence presum-

ably is related to its role in transport or in recognizingmolecules for transport. The opp system in S. typhimuriumtransports oligopeptides from two to five amino acids inlength and cell wail peptides (17). The role of spoOK insporulation and/or competence could be in the transport orsensing of extracellular peptides. There are many examplesof extracellular molecules that act as signals for differentia-tion in microbes, including the multiple factors involved inmyxobacterium development (27). We have described ex-periments indicating that B. subtilis produces at least oneextracellular factor that is required for efficient sporulation(18, 19). This factor accumulates in the culture medium asthe cells reach a high density. It is sensitive to pronase, heatresistant, and dialyzable, indicating that it is at least in partan oligopeptide (18). It is tempting to speculate that thespoOK gene products may be involved in sensing and/ortransporting such a factor.

It is possible that an extracellular factor also exists forcompetence development in B. subtilis. There is a relativelywell defined competence factor in S. pneumoniae (33, 55),and there have been reports of possible competence factorsin B. subtilis (1, 2, 26), although definitive evidence islacking. Again, it is tempting to speculate that if there is acompetence factor in B. subtilis, the spoOK-encoded systemmay be involved in sensing and/or transporting such a factor.The factors controlling competence could be the same ordifferent from those controlling sporulation.The effects of spoOK on sporulation seem to be mediated

through a two-component regulatory system composed of ahistidine protein kinase and a response regulator. The re-quirement for spoOK in sporulation can be partly bypassedby mutations in spoOA (49, 50; unpublished results) and bythe overproduction of the kinase encoded by spoIIJ (Table3). It is possible that the normal role of spoOK is to activateone or more protein kinases that are involved in the initiationof sporulation. This activation could be either indirect ordirect, conceivably through a specific interaction with akinase, perhaps SpoIl and/or other kinases. Competencedevelopment is also regulated in part by histidine proteinkinases, those encoded by comP (58) and degS (34). Again,the activity of one or both of these kinases could also becontrolled by spoOK.The precise mechanisms by which spoOK affects compe-

tence and sporulation must be different. Overproduction ofSpoIIJ partly suppressed the sporulation defect, but not the

J. BACTERIOL.

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

B. SUBTILIS spoOK OPERON 1397

competence defect, caused by spoOK mutations. Further-more, mutations in spoOKE caused a significant defect incompetence but not in sporulation. It is possible that thespoOKE gene product is not part of the postulated transportcomplex composed of the other spoOK gene products. Per-haps SpoOKE is recruited by other proteins that are requiredfor competence but not sporulation. In any case, if there iscoupling between the spoOK gene products and histidineprotein kinases, perhaps SpoOKE affects the activity of akinase required for competence and SpoOKD affects theactivity of a kinase required for sporulation.Membrane transport systems and two-component regula-

tory systems are both highly conserved and play importantroles in cellular physiology. Our results indicate that theactivity of a histidine protein kinase (perhaps SpoIIJ) may becoupled to the transport system encoded by spoOK. Theactivity of at least one other two-component regulatorysystem seems to be coupled to one of the conserved trans-port systems. The expression of genes involved in phosphatemetabolism (the pho regulon) in E. coli is regulated by thetwo-component PhoR-PhoB system. PhoR is a membrane-bound protein kinase that phosphorylates PhoB, the re-sponse regulator (reviewed in references 3 and 54). Theexpression of genes in the pho regulon is also controlled bythe products of the pstSCAB genes, which encode compo-nents of a high-affinity transport system for the uptake ofphosphate. The phosphate transport system is similar tothose encoded by spoOK and opp and seems to be coupled(directly or indirectly) to the activity of the two-componentregulatory system, perhaps through the phoU gene product(reviewed in reference 43). It will be interesting to seewhether this potential interaction between two differenttypes of highly conserved systems is a general scheme incontrolling gene expression in response to external signals.

ACKNOWLEDGMENTS

We thank P. Stragier for clones of spoIIJ, including the Pspac-spoIIJ fusion, and for the silent Tn917 insertion linked to spoOK; C.Price for providing strains; and R. Yasbin for providing the Tn917lacinsertion library. We are grateful to J. Hu, P. Stragier, D. Dubnau,R. Losick, L. Sonenshein, and the reviewers for comments andsuggestions on the manuscript.

J.R.L. was supported by an NSF Predoctoral Fellowship, andK.I. was supported in part by an NIH Predoctoral Training Grant.A.D.G. is a Lucille P. Markey Scholar in Biomedical Sciences. Thiswork was supported in part by a grant from the Lucille P. MarkeyCharitable Trust and by NIH Public Health Service grant GM41934to A.D.G.

REFERENCES

1. Akrigg A., and S. R. Ayad. 1970. Studies on the competence-inducing factor of Bacillus subtilis. Biochem. J. 117:397-403.

2. Akrigg, A., S. R. Ayad, and G. R. Barker. 1967. The nature ofa competence-inducing factor in Bacillus subtilis. Biochem.Biophys. Res. Commun. 28:1062-1067.

3. Albright, L. M., E. Huala, and F. M. Ausubel. 1989. Prokaryoticsignal transduction mediated by sensor and regulator proteinpairs. Annu. Rev. Genet. 23:311-336.

4. Alloing, G., M. C. Trombe, and J. P. Claverys. 1990. The amilocus of the gram positive bacterium Streptococcus pneumoniaeis similar to binding protein-dependent transport operons ofgram-negative bacteria. Mol. Microbiol. 4:633 644.

5. Ames, G. F. L. 1986. Bacterial periplasmic transport systems:structure, mechanism, and evolution. Annu. Rev. Biochem.55:397-425.

6. Antoniewski, C., B. Savelli, and P. Stragier. 1990. The spoIIJ

gene, which regulates early developmental stages in Bacillussubtilis, belongs to a class of environmentally responsive genes.J. Bacteriol. 172:86-93.

7. Blight, M. A., and I. B. Holland. 1990. Structure and function ofhaemolysin B, P-glycoprotein and other members of a novelfamily of membrane translocators. Mol. Microbiol. 4:873-880.

8. Boylan, S. A., K. T. Chun, B. A. Edson, and C. W. Price. 1988.Early-blocked sporulation mutations alter expression of en-zymes under carbon control in Bacillus subtilis. Mol. Gen.Genet. 212:271-280.

9. Coote, J. G. 1972. Sporulation in Bacillus subtilis. Characteri-zation of oligosporogenous mutants and comparison of theirphenotypes with those of asporogenous mutants. J. Gen. Mi-crobiol. 71:1-15.

10. Davis, R. W., D. Bostein, and J. R. Roth. 1980. Advancedbacterial genetics. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

11. Devereux, J., P. Haeberli, and 0. Smithies. 1984. A comprehen-sive set of sequence analysis programs for the VAX. NucleicAcids Res. 12:387-395.

11a.Dubnau, D. Personal communication.12. Dubnau, D., and R. Davidoff-Abelson. 1971. Fate of transform-

ing DNA following uptake by competent B. subtilis. J. Mol.Biol. 56:209-221.

13. Ferrari, F. A., A. Nguyen, D. Lang, and J. A. Hoch. 1983.Construction and properties of an integrable plasmid for Bacil-lus subtilis. J. Bacteriol. 154:1513-1515.

14. Ferrari, F. A., K. Trach, D. LeCoq, J. Spence, E. Ferrari, andJ. A. Hoch. 1985. Characterization of the spoOA locus and itsdeduced product. Proc. Natl. Acad. Sci. USA 82:2647-2651.

15. Freese, E. 1981. Initiation of bacterial sporulation, p. 1-12. InH. S. Levenson, A. L. Sonenshein, and D. J. Tipper (ed.),Sporulation and germination. American Society for Microbiol-ogy, Washington, D.C.

16. Gilson, E., G. Alloing, T. Schmidt, J.-P. Claverys, R. Dudler,and M. Hofnung. 1988. Evidence for high affinity binding-protein dependent transport systems in gram-positive bacteriaand in Mycoplasma. EMBO J. 7:3971-3974.

17. Goodell, E. W., and C. F. Higgins. 1987. Uptake of cell wallpeptides by Salmonella typhimurium and Escherichia coli. J.Bacteriol. 169:3861-3865.

18. Grossman, A. D., and R. Losick. 1988. Extracellular control ofspore formation in Bacillus subtilis. Proc. Natl. Acad. Sci. USA85:4369-4373.

19. Grossman, A. D., and R. Losick. 1988. Role of extracellularfactors in the control of sporulation in Bacillus subtilis, p.175-179. In A. T. Ganesan and J. A. Hoch (ed.), Genetics andbiotechnology of bacilli, vol. 2. Academic Press, San Diego,Calif.

20. Hahn, J., M. Albano, and D. Dubnau. 1987. Isolation andcharacterization of Tn917lac-generated competence mutants ofBacillus subtilis. J. Bacteriol. 169:3104-3109.

21. Higgins, C. F., I. D. Hiles, G. P. C. Salmond, D. R. Gill, J. A.Downie, I. J. Evans, I. B. Holland, L. Gray, S. D. Buckel, A. W.Bell, and M. A. Hermodson. 1986. A family of related ATP-binding subunits coupled to many distinct biological processesin bacteria. Nature (London) 323:448-450.

22. Hiles, I. D., M. P. Gallagher, D. J. Jamieson, and C. F. Higgins.1987. Molecular characterization of the oligopeptide permeaseof Salmonella typhimurium. J. Mol. Biol. 195:125-142.

23. Hoch, J. A., K. Trach, F. Kawamura, and H. Saito. 1985.Identification of the transcriptional suppressor sof-l as analteration in the SpoOA protein. J. Bacteriol. 161:552-555.

24. Hyde, S. C., P. Emsley, M. J. Hartshorn, M. M. Mimmack, U.Gileadi, S. R. Pearce, M. P. Gallagher, D. R. Gill, R. E.Hubbard, and C. F. Higgins. 1990. Structural model of ATP-binding proteins associated with cystic fibrosis, multidrug resis-tance and bacterial transport. Nature (London) 346:362-365.

24a.Ireton, K., and A. D. Grossman. Unpublished data.25. Jaacks, K. J., J. Healy, R. Losick, and A. D. Grossman. 1989.

Identification and characterization of genes controlled by thesporulation regulatory gene spoOH in Bacillus subtilis. J. Bac-teriol. 171:4121-4129.

VOL. 173, 1991

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from

1398 RUDNER ET AL.

26. Joenje, H., M. Gruber, and G. Venema. 1972. Stimulation of thedevelopment of competence by culture fluids in Bacillus subtilistransformation. Biochim. Biophys. Acta 262:189-199.

27. Kim, S. K., and D. Kaiser. 1990. C-factor: a cell-cell signalingprotein required for fruiting body morphogenesis of M. xanthus.Cell 61:19-26.

27a.L1ghton, T. and C. Price. Personal communication.28. Leighton, T. J., and R. H. Doi. 1971. The stability of messenger

ribonucleic acid during sporulation in Bacillus subtilis. J. Biol.Chem. 252:268-272.

29. Lodck, R., P. Youngnan, and P. J. Piggot. 1986. Genetics ofendospore formation in Bacillus subtilis. Annu. Rev. Genet.20:625-669.

30. Love, P., M. J. Lyle, and R. E. Yasbin. 1985. DNA-damage-inducible (din) loci are transcriptionally active in competentBacillus subtilis. Proc. Natl. Acad. Sci. USA 82:6201-6205.

31. McGrath, J. P., and A. Varshavsky. 1989. The yeast STE6 geneencodes a homologue of the mammalian multidrug resistanceP-glycoprotein. Nature (London) 340:400404.

32. Mhmmack, M. L., M. P. Gallher, S. R. Pearce, S. C. Hyde,I. R. Booth, and C. F. Hlgns. 1989. Energy coupling toperiplasmic binding protein-dependent transport systems: stoi-chiometry of ATP hydrolysis during transport in vivo. Proc.Natl. Acad. Sci. USA 86:8257-8261.

33. Morrison, D. A., and M. F. Baker. 1979. Competence for genetictransformation in pneumococcus depends on synthesis of asmall set of proteins. Nature (London) 282:215-217.

34. Msadek, T., F. Kunst, D. Henner, A. Kler, G. Rapoport, and R.Dedonder. 1990. Signal transduction pathway controlling syn-thesis of a class of degradative enzymes in Bacillus subtilis:expression of the regulatory genes and analysis of mutations indegS and degU. J. Bacteriol. 172:824-834.

35. Nixon, B. T., C. W. Ronson, and F. M. Ausubel. 1986. Two-component regulatory systems responsive to environmentalstimuli share strongly conserved domains with the nitrogenassimilation regulatory genes ntrB and ntrC. Proc. Natl. Acad.Sci. USA 83:7850-7854.

36. Ozkaynak, E., and S. D. Putney. 1987. A unidirectional deletiontechnique for the generation of clones for sequencing. Biotech-niques 5:770-773.

37. Perego, M., S. P. Cole, D. Burbulys, K. Trach, and J. A. Hoch.1989. Characterization of the gene for a protein kinase whichphosphorylates the sporulation-regulatory proteins SpoOA andSpoOF of Bacillus subtilis. J. Bacteriol. 171:6187-196.

38. Perego, M., C. F. Higgins, S. R. Pearce, M. P. Galagber, andJ. A. Hoch. Mol. Microbiol., in press.

39. Perego, M., and J. A. Hoch. 1988. Sequence analysis andregulation of the hpr locus, a regulatory gene for proteaseproduction and sporulation in Bacillus subtilis. J. Bacteriol.170:2560-2567.

40. Perego, M., G. B. Spe , and J. A. Hoch. 1988. Structureof the gene for the transition state regulator abrB: regulatorsynthesis is controlled by the spoOA sporulation gene in Bacillussubtilis. Mol. Microbiol. 2:689-699.

41. Piggot, P. J. 1989. Revised genetic map of Bacillus subtilis 168,p. 1-41. In I. Smith, R. Slepecky, and P. Setlow (ed.), Regula-tion of procaryotic development. American Society for Micro-biology, Washington, D.C.

42. Plggot, P. J., and J. G. Coote. 1976. Genetic aspects of bacterialendospore formation. Bacteriol. Rev. 40:908-962.

43. Rao, N. N., and A. Torrlani. 1990. Molecular aspects of phos-phate transport in Escherichia coli. Mol. Microbiol. 4:1083-1090.

44. Roal, M., J. Hahn, and D. Dubnau. 1990. Suppression ofearly competence mutations in Bacillus subtilis by mec muta-

tions. J. Bacteriol. 172:40564063.44a.Rudair, D. Z., and A. D. Grossman. Unpublished data.45. Sadale, Y., and T. Kada. 1983. Formation of competent Bacillus

subtilis cells. J. Bacteriol. 153:813-821.46. Sambrook, J., E. F. Fritsch, and T. ManItis. 1989. Molecular

cloning: a laboratory manual, 2nd ed. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

47. Sandn, K., R. Losick, and P. Youngman. 1987. Geneticanalysis of Bacillus subtilis spo mutations generated by Tn917-mediated insertional mutagenesis. Genetics 117:603-617.

48. Schaeffer, P., J. Met, and J. Aubert. 1965. Catabolite repres-sion of bacterial sporulation. Proc. Natl. Acad. Sci. USA54:704-711.

49. Sharock, R. A., S. Rubenstein, M. Chun, and T. LeIgbton. 1984.Intergenic suppression of spo0 phenotypes by the Bacillussubtilis mutation rvtA. Mol. Gen. Genet. 194:260-264.

50. ShoJi, K., S. H9ratsuka, F. Kawamur, andY. Kobayashi. 1988.New suppressor mutation surOB of spoOB and spoOF mutationsin Bacillus subtilis. J. Gen. Microbiol. 134:3249-3257.

51. Smith, I. 1989. Initiation of sporulation, p. 185-210. In I. Smith,R. Slepecky, and P. Setlow (ed.), Regulation of procaryoticdevelopment. American Society forMicrobiology, Washington,D.C.

52. Sonenshein, L. 1989. Metabolic regulation of sporulation andother stationary-phase phenomena, p. 109-130. In I. Smith, R.Slepecky, and P. Setlow (ed.), Regulation of procaryotic devel-opment. American Society for Microbiology, Washington, D.C.

53. S p ,G., B. Van Hoy,M.Preg J. Day, K. Trach, andJ. A. Hoch. 1990. Structural alterations in the Bacillus subtilisSpoOA regulatory protein which suppress mutations at severalspoO loci. J. Bacteriol. 172:5011-5019.

54. Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989. Proteinphosphorylation and regulation of adaptive responses in bacte-ria. Microbiol. Rev. 53:450-490.

55. Tomasz, A., and R. D. Hotchkiss. 1964. Regulation of thetransformability of pneumococcal cultures by macromolecularcell products. Proc. Natl. Acad. Sci. USA 51:480-487.

56. Vasantha, N., and E. Freese. 1979. The role of manganese ingrowth and sporulation of Bacillus subtilis. J. Gen. Microbiol.112:329-336.

57. Vasantha, N., and E. Freese. 1980. Enzyme changes duringBacillus subtilis sporulation caused by deprivation of guaninenucleotides. J. Bacteriol. 144:1119-1125.

58. Wednrauch, Y., R. Penchev, E. Dubas, L. Smith, and D.Dubnsu. 1990. A Bacillus subtilis regulatory gene product forgenetic competence and sporulation resembles sensor proteinmembers of the bacterial two-component signal-transductionsystems. Genes Dev. 4:860-872.

59. Youngman, P., J. B. Perkins, and K. Sanan. 1984. Newgenetic methods, molecular cloning strategies, and gene fusiontechniques for.Bacillus subtilis which take advantage of Tn917insertional mutagenesis, p. 103-111. In J. A. Hoch and A. T.Ganesan (ed.), Genetics and biotechnology of bacili. AcademicPress, Inc., New York.

60. You an, P., H. Poth, B. Green, K. York, G. Olmedo, and K.Smith. 1989. Methods for genetic manipulation, cloning, andfunctional analysis of sporulation genes in Bacillus subtilis, p.65-87. In I. Smith, R. Slepecky, and P. Setlow (ed.), Regulationof procaryotic development. American Society for Microbiol-ogy, Washington, D.C.

61. Youngman, P. J., J. B. Perkins, and R. Louick. 1984. A novelmethod for rapid cloning in Escherichia coli of Bacillus subtilischromosomal DNA adjacent to Tn917 insertions. Mol. Gen.Genet. 195:424-433.

J. BACTERIOL.

at Harvard Libraries on M

ay 16, 2008 jb.asm

.orgD

ownloaded from