characterizationofthelocusresponsibleforthebacteriocin...

JOURNAL OF BACTERIOLOGY, Aug. 1996, p. 4472–4483 Vol. 178, No. 150021-9193/96/$04.0010Copyright q 1996, American Society for Microbiology

Characterization of the Locus Responsible for the BacteriocinProduction in Lactobacillus plantarum C11DZUNG BAO DIEP, LEIV SIGVE HÅVARSTEIN, AND INGOLF F. NES*

Laboratory of Microbial Gene Technology, Department of Biotechnological Sciences,Agricultural University of Norway, N-1432 Ås, Norway

Received 12 February 1996/Accepted 13 May 1996

Lactobacillus plantarum C11 secretes a small cationic peptide, plantaricin A, that serves as induction signalfor bacteriocin production as well as transcription of plnABCD. The plnABCD operon encodes the plantaricinA precursor (PlnA) itself and determinants (PlnBCD) for a signal transducing pathway. By Northern (RNA)and sequencing analyses, four new plantaricin A-induced operons were identified. All were highly activated inconcert with plnABCD upon bacteriocin induction. Two of these operons (termed plnEFI and plnJKLR) eachencompass a gene pair (plnEF and plnJK, respectively) encoding two small cationic bacteriocin-like peptideswith double-glycine-type leaders. The open reading frames (ORFs) encoding the bacteriocin-like peptides arefollowed by ORFs (plnI and -L, respectively) encoding cationic hydrophobic proteins resembling bacteriocinimmunity proteins. On the third operon (termed plnMNOP), a similar bacteriocin-like ORF (plnN) and aputative immunity ORF (either plnM or -P) were identified as well. These findings suggest that two bacteriocinsof two-peptide type (mature PlnEF and PlnJK) and a bacteriocin of one-peptide type (mature PlnN) could beresponsible for the observed bacteriocin activity. The last operon (termed plnGHSTUV) contains two ORFs(plnGH) apparently encoding an ABC transporter and its accessory protein, respectively, known to be involvedin processing and export of peptides with precursor double-glycine-type leaders. Promoter structure wasestablished. A conserved regulatory-like box encompassing two direct repeats was identified in the promoterregions of all five plantaricin A-induced operons. These repeats may serve as regulatory elements for geneexpression.

Bacteria respond to various environmental stimuli by regu-lating metabolic pathways through a network of so-called sig-nal transducing systems (4, 18, 30, 36, 44, 55). Many of thesesystems involve a common regulatory mechanism normally me-diated by two proteins of different functions: a histidine proteinkinase (HPK) serving as an environmental sensor, and a cyto-plasmic response regulator (RR) modulated by its cognateHPK to trigger an adapting response, in most cases by generegulation. Such a regulatory operon, consisting of a bacterio-cin-like peptide (encoded by plnA), an HPK (encoded by plnB),and two RRs (encoded by plnCD), has previously been clonedand sequenced in Lactobacillus plantarum C11 (8, 9). Tran-scriptional activation of this operon, which is thought to benecessary for the subsequent induction of bacteriocin produc-tion, is triggered by plantaricin A, a small cationic secreted pep-tide encoded by plnA. The pln regulatory system (PlnBCD) ismost similar to the Staphylococcus aureus accessory gene reg-ulatory (agr) system (26, 42, 45), whose HPK (AgrC) has beenproposed to be the signal transducer and whose RR (AgrA)has been proposed to the transcriptional activator responsiblefor both the activation of the agr circuit and the subsequentinduction of a set of exoproteins (including d-lysin) duringstationary growth phase (25, 27, 41). The agr locus has beenshown to be activated by an octapeptide processed from apolypeptide encoded by agrD (27, 42).The aim of this study was to identify other genes belonging

to the plantaricin A (pln) regulon, especially those responsiblefor the bacteriocin production. Bacteriocins are ribosomally

synthesized peptides with antimicrobial activity. Most bacte-riocins are active only against bacteria closely related to theproducer organism (29). A subgroup, so-called nonlantibioticbacteriocins (nonlantibiotics), consists of small heat-stable cat-ionic peptides made up of unmodified amino acids (29). Theircounterparts, the lantibiotics, contain posttranslationally mod-ified amino acids, such as didehydroalanine, didehydrobu-tyrine, lanthionine, and b-methyllanthionine (52). Some bac-teriocins require the complementary action of two differentpeptides to achieve biological activity; this two-peptide groupincludes lactacin F (2), lactococcin G (39), and most probablylactococcin M (59) as well. The primary translation product ofmost nonlantibiotics, some lantibiotics, and colicin V containsa leader peptide of double-glycine type which might serve as arecognition signal for protein export (13, 17). Such precursorpeptides are processed and secreted by a dedicated exportsystem made up by an ABC transporter and its accessory pro-tein (16). For the nonlantibiotics, the genetic determinantswhich have been proposed or confirmed to confer immunityare frequently found in the bacteriocin operons, often locatedjust downstream of the bacteriocin structural gene(s) (19, 29,37). This conserved organization is believed to be necessary toprotect the producer from being killed by its own bacteriocin.These immunity proteins normally have a high pI. Further-more, those associated with two-peptide bacteriocins consist of110 to 154 amino acids (aa), containing several (usually four)potential transmembrane helices (PTHs) (14, 37, 59), whilethose of one-peptide bacteriocins are generally smaller in size(51 to 113 aa) and contain few (one to two) or no PTHs (3, 35,48, 57–60). Still, immunity proteins with no PTHs are thoughtto be associated with the membrane (38, 62).The L. plantarum C11 bacteriocin activity has been reported

to arise from a small heat-stable proteinaceous compound (7).

* Corresponding author. Mailing address: Laboratory of MicrobialGene Technology, Agricultural University, P.O. Box 5051, 1432 Ås,Norway. Phone: 47-64-949471. Fax: 47-64-941465. Electronic mail ad-dress: [email protected].

4472

on May 6, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

However, the results achieved from the initial biochemical andgenetic studies on bacteriocin activity are not in full agreementwith recent work concerning the regulatory function of plan-taricin A (8, 9, 40). In sum, the observed bacteriocin activity inthis bacterium was at first attributed to plantaricin A on thebasis of the following facts: (i) plantaricin A was the majorcomponent enriched during the purification (40) and (ii) itsprecursor peptide (PlnA) contains a typical double-glycine-type leader (9). However, when plantaricin A was obtained byeither chemical synthesis or heterologous gene expression, nosignificant bacteriocin activity was found (8). On the otherhand, synthetic or expressed plantaricin A was able to inducebacteriocin production at very low concentrations. Our work-ing hypothesis for explaining these contradictory results is thatfactors other than plantaricin A are required for bacteriocinactivity. In this study, we have identified four new plantaricinA-induced operons: two operons encoding possible two-pep-tide bacteriocins, an operon encoding a possible one-peptidebacteriocin, and an operon encoding an ABC transporter sys-tem. These findings support our theory that plantaricin Aserves primarily as a regulatory factor and that the observedbacteriocin activity is due to one or more of these three puta-tive bacteriocins. Transcription and promoter sequence analy-ses on these plantaricin A-induced operons strongly suggestthe presence of a common regulatory mechanism implicated intheir transcription.

MATERIALS AND METHODS

Bacterial strains and growth conditions. L. plantarum C11 possesses an in-ducible bacteriocin production triggered by plantaricin A (8). The various phe-notypes of C11 cultures with regard to bacteriocin production have been de-scribed as follows (8): Bac1 C11 and Bac2 C11 refer to cultures with and withoutproduction, respectively, whereas induced Bac2 C11 denotes cultures with in-duced production. L. sake Lb706B(pVH52), a transformant containing thecloned plantaricin A operon (plnABCD) (8), was used as source for plantaricinA in induction experiments. L. plantarum 965 was used as indicator strain inbacteriocin assay (7). All strains were grown on MRS plates or in MRS broth(Oxoid, Basingstoke, Hampshire, United Kingdom) at 308C without agitation.Erythromycin (final concentration, 5 mg/ml) was added into the growth mediumof Lb706B(pVH52) to stabilize the cloned plasmid (pVH52). Subcultivation ofBac1 C11 and Bac2 C11, induction of Bac2 C11, and preparation of cell-freesupernatant from Lb706B(pVH52) were performed as previously described pre-viously (8).DNA manipulations and sequencing. Standard DNA cloning techniques were

performed as described by Sambrook et al. (49). HindIII fragments (H3 and H5)and EcoRI fragments (E5, E7, and E9) (Fig. 1A) were obtained by digestion ofsome selected recombinant l clones (9) and subsequently ligated into pGEM-7Zf(1) (Promega, Madison, Wis.) to give rise to plasmids pGH3, pGH5, pGE5,pGE7, and pGE9, respectively. Subfragments EH2 (EcoRI plus HindIII), EC1(EcoRI plus ClaI), and E7S (KpnI) were digested from these plasmid clones andligated into pGEM, giving rise to plasmids pGEH2, pGEC1, and pGE7S, re-spectively. Escherichia coli DH5a was used as host for plasmid amplification. Forsequencing, nested clones of pGE5 and pGE7 were made by using an Erase-a-Base kit (Promega), and the Sequenase system (U.S. Biochemicals, Cleveland,Ohio) was used to sequence both strands completely. Synthetic oligonucleotideswere used as primers to sequence into part of pGEH2 (both strands) to completethe 16,139-bp sequence shown in Fig. 3.Probes, RNA isolation, Northern (RNA) analysis, and primer extension. Re-

striction fragments used as probes are depicted in Fig. 1A. Oligonucleotides usedfor the same purpose and in primer extension experiments were S14 (59-GCAGTTGCCCCCATCTGCAAAGAATACGCACTACTC-39), complementary toplnA (accession number X75323 [9]); oligoJ (59-GGTGCAAATGCATCTAC-39), complementary to plnJ; oligoM (59-CAAGCATCTTCCCACTGC-39), com-plementary to plnM; oligoN (59-CCACCTTCAACGGTAGTC-39), complemen-tary to plnN; oligoE (59-ggaattccGTTGATCTCCCCCAAGAAAATTAACG-39;capital letters indicate complementary bases), complementary to the regionbetween plnE and -F; and oligoG (59-CTCGTCAACTTGCGCAAC-39), com-plementary to plnG (see Fig. 3).A time course induction experiment for transcription analysis was carried out

as previously described (8). However, the plantaricin A-containing cell-free su-pernatant from Lb706B(pVH52) was used as a source of inducer (final concen-tration, 2% [vol/vol]) instead of synthetic plantaricin A. In short, a freshlyinoculated Bac2 C11 culture was grown at 308C until the optical density at 600nm reached 0.1; then plantaricin A was added to induce bacteriocin production.Samples were collected during both exponential (time points of 15 min, 2 h, 3 h,4 h, and 5 h) and stationary (7, 10, and 24 h) growth phases, and RNA wasisolated by the method of Igo and Losick (24). RNA (about 5 mg loaded ontoeach well) was separated on 1.4% (wt/vol) agarose gels containing 2.2 M form-aldehyde, then blotted onto a GeneScreen Plus membrane filter (DuPont, Bos-ton, Mass.), and probed with either end-labeled oligonucleotides or random-labeled restriction fragments. 32P-labeling reactions were performed by standardmethods (49), and hybridization was carried out as described previously (6, 33).Primer extension was performed by the method of Alam et al. (1) with the

following modifications: 3 to 5 mg of total RNA was annealed with about 1 pmolof prelabeled 32P-primer in a total 11-ml volume (by cooling from 65 to 308Cwithin 15 to 20 min), then 53 avian myeloblastosis virus reverse transcriptasebuffer (Promega), deoxynucleoside triphosphates (each at a final concentrationof 5 mM) and the enzyme (5 U; Promega) were added to the mixture (to a finalvolume of 20 ml), and the mixture was incubated at 428C for 60 min for poly-merization. The reaction was stopped with 10 ml of Sequenase stop buffer andanalyzed on a sequencing gel.Computer analysis of DNA and protein sequences. The PC/Gene program

package (version 6.8; IntelliGenetics, Inc., Mountain View, Calif.) and GeneticsComputer Group package (version 8; University of Wisconsin) were used forDNA and protein analyses. With the PC/Gene package, SOAP was used for

FIG. 1. Genetic map of the pln locus. (A) Locations of the various restriction fragments that were used as probes. (B) The 16,139-bp pln locus encompasses 21 ORFsand an incompletely sequenced ORF (plnV) (indicated by open arrows); 20 of the ORFs are organized into five operons, plnJKLR, plnMNOP, plnABCD, plnEFI, andplnGHSTUV. Their transcription patterns are indicated by filled arrows in panel C; see Results for a detailed description. Above the ORFs is the restriction map ofthe pln locus (H, HindIII; C, ClaI; E, EcoRI; K, KpnI); on the map, promoters and potential transcription terminators are indicated by P’s and T’s, respectively.

VOL. 178, 1996 CHARACTERIZATION OF THE pln BACTERIOCIN LOCUS 4473


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

prediction of pI, hydrophobicity, and potential transmembrane segments of pro-teins; with the Genetics Computer Group package, FASTA, TFASTA, BLAST,MOTIFS, and PILEUP were used for DNA and protein similarity searches,identification of known protein sequence patterns, and analysis of similarityamong multiple sequences.Nucleotide sequence accession number. The nucleotide sequence presented in

this report has been assigned EMBL accession number X94434.

RESULTS

Identification of plantaricin A-induced transcripts. In ourprevious work (8), we showed that the L. plantarum C11 bac-teriocin activity in Bac1 C11 cultures was produced at a fairlyconstant rate throughout exponential growth phase. It reacheda maximum in early stationary growth phase and then gradu-ally decreased (i.e., there was no net production) later in thestationary growth phase. On the other hand, the bacteriocinactivity in induced Bac2 C11 cultures was first detected about2.5 h (approximately in mid-exponential growth phase) afterthe addition of plantaricin A and then peaked in early station-ary growth phase as observed with Bac1 C11. Thus, it seemslikely that transcription of the genetic determinants responsi-ble for bacteriocin production was activated upon addition ofplantaricin A to the culture. Gene clustering is often observedfor genes involved in bacteriocin production and regulation (3,51, 54, 59). Therefore, the flanking regions of the bacteriocinregulatory unit (plnABCD) were examined to see if they con-tain genes involved in bacteriocin production. Northern anal-yses were performed with two restriction fragments, E5 (5 kb)and E7 (7 kb), located on either side of plnABCD, as probes(Fig. 1A). As shown in Fig. 2A and B and described in moredetail below, nine different transcripts were detected, six byprobe E5 and three by probe E7. Most of these transcriptswere present at high levels over almost the entire exponentialgrowth phase of the bacteriocin-producing culture (Bac1 C11;Fig. 2A and B, lane set a) compared with that of Bac2 C11 (seebelow). Then levels rapidly declined in stationary growthphase. In the nonproducing culture (Bac2 C11; Fig. 2A and B,lane set b), the levels of all of these transcripts were very lowthroughout its growth. However, when plantaricin A wasadded to the Bac2 C11 culture to induce bacteriocin produc-tion (induced Bac2 C11; Fig. 2A and B, lane set c), an increasein the synthesis of some of these transcripts was readily ob-served only 15 min after the addition of plantaricin A, and alltranscripts were fully induced 2 h after addition of this induc-tion factor. These results clearly show that both regions E5 andE7 harbor genes induced by plantaricin A and that expressionof these genes seems to be coordinated with the bacteriocinproduction.Sequence analysis of plantaricin A-induced operons. The

DNA sequence of the plnABCD-flanking regions was subse-quently determined as described in Materials and Methods (Fig.3). The contiguous 16,139-bp sequence obtained revealed, inaddition to the operon plnABCD, 17 new open reading frames(ORFs) and an incompletely sequenced ORF (Fig. 1B), allpreceded by a plausible ribosome binding site (Fig 3). As de-monstrated below, 17 of the new ORFs are organized into fourdifferent operon structures designated plnJKLR and plnMNOP,both located upstream of plnABCD, and plnEFI and plnGH-STUV, both located in the downstream region; the latter (plnGHSTUV) is partial, containing the incompletely sequencedORF (plnV) at its end. The last ORF (designated orf1) is lo-cated in the region between plnMNOP and plnABCD (Fig. 1B).(i) plnGHSTUV. The first two ORFs (plnGH) in plnGH-

STUV potentially encode two relatively large proteins (716 and458 aa, respectively). Some other relevant predicted physico-chemical properties of PlnG and -H and of other ORFs de-

scribed below are summarized in Tables 1 and 2. Homologysearches identified PlnG and PlnH as members of two proteinfamilies, the so-called ABC transporters and their accessoryproteins, respectively, known to be involved in transport ofvarious substrates across the bacterial membrane (12). Highestsimilarity was found with proteins mediating maturation andexport of peptides with precursor double-glycine-type leaders(16). Such peptides include not only bacteriocins such as lac-tococcin A of Lactococcus lactis (19, 56) but also the compe-tence factor of Streptococcus pneumoniae (15, 21, 22). Theputative ABC transporter PlnG and its accessory protein PlnHshow 58.9 and 41.1% identity to their homologs ComA andComB, respectively. Similarly, they show 52.6 and 29.7% iden-tity to their homologs LcnC and LcnD, which together makeup the dedicated lactococcin A secretion machinery.The remaining four ORFs (plnS, -T, -U, and -V) of this par-

tial operon encode polypeptides of 99, 140, 222, and 44 aa,respectively, of which PlnV apparently represents the N ter-minus of a larger polypeptide. The 39 end of plnS overlaps thefirst 10 codons of plnT (Fig. 3). The deduced products fromthese four ORFs are predicted to be hydrophobic and containPTHs, suggesting a membrane location. However, homologysearches with the deduced gene products of these ORFsshowed no obvious sequence similarity to any known proteins.(ii) plnEFI, plnJKLR, plnMNOP, and orf1. Two of the three

remaining operons (plnEFI and plnJKLR) start with a genepair (plnEF and plnJK, respectively) encoding small (52- to57-aa) cationic polypeptides that each appear to contain an N-terminal double-glycine-type leader sequence, which is proba-bly removed by the putative PlnGH export system during thematuration process. The amino acid sequence and some otherrelevant physicochemical data of the deduced mature productsare listed in Table 2. All of these peptides appear to be rela-tively small (25 to 34 aa) and to have relatively high pIs (10.7to 12.06). Furthermore, when displayed on an Edmundsona-helical wheel (50), all of the peptides were shown to containa segment that could potentially form an amphiphilic helix(data not shown). Such amphiphilic structures are believed tobe associated with pore-forming toxins creating cell membranechannels through a “barrel-stave” mechanism (43), as sug-gested for the toxin d-lysin (43) and lactococcin G (both a andb peptides) (39) and for the induction factor plantaricin A(40). Thus, being small, cationic and amphiphilic peptides ma-tured from precursors with double-glycine-type leaders, theseputative peptides possess the characteristics of known nonlan-tibiotics (29). Both of these two bacteriocin-like gene pairs(plnEF and plnJK) are followed by ORFs encoding polypep-tides (PlnI of 257 aa and PlnL of 138 aa) with high pI, hydro-phobicity, and three to seven PTHs. The last ORF (plnR) ofplnJKLR encodes a small (50-aa) polypeptide, also predicted tobe highly cationic and hydrophobic and to have a PTH.As found for plnEFI and plnJKLR, the last operon (pln

MNOP) also starts with two ORFs (plnMN) encoding small(55- to 66-aa), cationic polypeptides. However, only PlnN ap-pears to have a double-glycine-type leader. The deduced ma-ture product of plnN was shown to resemble the other bacte-riocin-like peptides mentioned above in being small andcationic (Table 2) and containing a fragment which potentiallycan form an amphiphilic structure (data not shown). PlnMappeared to be highly hydrophobic and to have two PTHs.plnO and -P, the two following ORFs on this operon, encoderelatively large polypeptides: PlnO (399 aa) is predicted to berather hydrophilic and to have an acidic pI, while PlnP (247 aa)shows physicochemical properties resembling those of PlnI,being cationic and hydrophobic and containing seven predictedPTHs (Table 2).

4474 DIEP ET AL. J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

The single orf1 encodes a small (62-aa), cationic hydropho-bic polypeptide with two PTHs.Homology searches for all ORFs within these three operons

and for orf1 showed no obvious similarity to known proteins indata banks. An exception was plnO, whose deduced polypep-tide contains one of two consensus motifs typical for aspartylproteases (Fig. 3). Such proteins belong to a widely distributedfamily of proteolytic enzymes known to exist in eukaryotes andviruses but not in bacteria. Sequence comparisons between all

ORFs were also performed; only PlnI and -P, which are ofabout the same size (257 and 247 aa), show significant similar-ity (55.9% similarity and 29.3% identity over their entirelengths).Transcript mapping and promoter sequence analyses. The

genetic locations of the various plantaricin A-induced tran-scripts detected by the probes E5 and E7, shown in Fig. 2A andB, were subsequently mapped by Northern analyses using anarray of oligonucleotides and restriction fragments as probes

FIG. 2. (A and B) Transcription analysis of the plantaricin A-induced operons in cultures of Bac1 C11 (lane set a), Bac2 C11 (lane set b), and induced Bac2 C11(lane set c), using restriction fragments E5 (A) and E7 (B) as probes. The conditions for growth and induction are described in Materials and Methods. After theaddition of plantaricin A to induce bacteriocin production, RNA was isolated at 15 min, 2 h, 4 h, 7 h, 10 h, and 24 h (lanes 1 to 6, respectively). The first three timepoints (15 min, 2 h, and 4 h) are from exponential growth phase, whereas the later time points (7, 10, and 24 h) are from stationary growth phase. (A) RNA in lanea/4 is the same as in lane 4 of lane set a but with better electrophoretic resolution. Arrows in panel A, lane set a, lane 4 indicate the estimated sizes of the inducedtranscripts: MNOP-RNA (2.8 kb), NOP-RNA (2.5 kb), JKLR-RNA (1.5 kb), MN-RNA (0.7 kb), and N-RNA (0.4 kb). In panel B, sizes are as follows: GHSTUV-RNA,7 to 8 kb; EFI-RNA, 1.3 kb; and EF-RNA, 0.4 kb. Note that GHSTUV-RNA appears as a smeared band such that it overshadows the signal of EFI-RNA, which isdemonstrated in lane set g. Standard (SD) rRNA molecule weight markers (2.9, 1.5, and 0.12 kb) are indicated on the left. (C) Transcript mapping. The RNA usedin this experiment was isolated from the same Bac1 C11 culture as used in panels A and B (see above). Lanes 1 to 4 therefore correspond to the time points specifiedabove, while lanes 39 and 49 (in lane sets i and j) correspond to time points 3 and 5 h, respectively. See Results for a detailed description of the mapping strategy andinterpretation of the results. The arrows indicate the following transcripts: JKLR-RNA (d); MNOP- and MN-RNAs (e); MNOP-, NOP-, MN-, and N-RNAs (f); EFI-and EF-RNAs (g); GHSTUV-RNA (h); N-D-RNA (7 kb) and the others corresponding to those indicated in panel A, lane set a, lane 4 (i); and N-D- (7 kb) andABCD-RNA (3.3 kb) (j). The standard (SD) rRNA marker indicated in lane set d corresponds to that is lane sets e to h but not i and j.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

FIG. 3. Nucleotide sequence and deduced proteins in the pln locus. Promoter 235 and 210 sites, ribosome binding sites (RBS), transcriptional starts (TS; 1 1),and the 59 ends of NOP- and N-RNAs are underlined; direct repeats (L, R, and R9) within the CRLBs and sequences of dyad symmetry with potential to serve astranscription terminators are overlined. Positions of the oligonucleotide probes and primers are indicated by boldface letters. The amino acid sequence containing theconsensus motif for aspartyl proteases in PlnO is also underlined. The 4,000-bp sequence from coordinates 5246 to 9245 was characterized earlier (9) and assignedaccession number X75323.

4476


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

FIG. 3—Continued.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

(see Materials and Methods and Fig. 1A). The results areshown in Fig. 2C. The probes oligoJ, EC1, and EH2, specificfor plnJ, plnJKL, and plnR, respectively, all hybridized to atranscript corresponding to the full length of a plnJKLR tran-script (1.5 kb). The result for EC1 is shown in Fig. 2C, lane setd. The plnM-specific probe (oligoM) hybridized to two tran-scripts (Fig. 2C, lane set e), one which gave rise to a fairly weak

band corresponding to the full length of a plnMNOP transcript(2.8 kb) and a stronger band corresponding to a transcript ofpossibly plnMN only (0.7 kb). However, the plnN-specificprobe (oligoN) hybridized to several transcripts (Fig. 2C, laneset f), of which two correspond to those detected by oligoM(plnMNOP and plnMN) and two correspond to transcripts ofpossibly plnN (0.4 kb) and plnNOP (2.5 kb). The other oligoN-

FIG. 3—Continued.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

detected transcripts of sizes between 1.5 and 2.9 kb were alsodetected when the same probe was used in a hybridizationexperiment with control RNA (isolated from a nonproducingBac2 C11 culture) (Fig. 2C, lane k). These signals are henceinterpreted as being due to unspecific hybridization, which isfrequently observed when small oligonucleotide probes areused. Probe oligoE, specific for the region between plnE and-F, hybridized to two transcripts (Fig. 2C, lane set g): a weakband corresponding to the full length of a plnEFI transcript(1.3 kb) and a strong band corresponding to a possible plnEFtranscript (0.4 kb). The plnG- and plnHSTUV-specific probes(oligoG and E7S, respectively) hybridized to a highly unstabletranscript (appearing as a smeared band). The result for E7S isshown in Fig. 2C lane set h. This transcript (approximately 7 to8 kb) is 2 to 3 kb larger than the sequenced part of plnGH-STUV and is therefore believed to start with plnG and termi-nate 2 to 3 kb downstream of the incompletely sequenced ORF(plnV). Note that this large transcript seemed to disappearsomewhat earlier than other transcripts (Fig. 2A and B).plnABCD has previously been shown to be transcribed as a3.3-kb unit (9). In addition, a large transcript (about 7 kb) wasobserved when probe E5 was used to hybridize RNA isolatedfrom the late-exponential growth of Bac1 C11 or inducedBac2 C11 (Fig. 2C, lane set i). This transcript was also detectedwhen probe H5 and the same growth conditions were used(Fig. 2C, lane set j); H5 spans the adjoining 5-kb region har-

boring the entire operon plnABCD (Fig. 1A and B). A likelyexplanation for this large transcript is that it is due to a read-through transcript which starts at plnM and proceeds to thepotential transcriptional terminator situated at the end of plnD(9), thus traversing a stretch of about 7 kb. Two band signals inFig. 2B, lane set c, lane 1, one of 2.5 to 2.8 kb and the other of1.3 to 1.4 kb, are believed to be due to artifacts; they apparentlyresulted from a combination of the instability of the hybridizedmRNA (GHSTUV-RNA) that gave rise to the smear and thepresence of the relatively large amounts of rRNAs (2.9 and 1.5kb) that could have caused biased distribution of the degradedGHSTUV-RNA. Similarly, the band signals in Fig. 2C, lanesets h (2.5 to 2.8 kb) and j (2.5 to 2.8 and 1.3 to 1.4 kb), couldhave resulted from such artifacts.To examine the region downstream of plnJKLR, the restric-

tion fragment E9 was used as a probe. E9 spans the wholelength of E5 and the adjoining 4-kb region downstream ofplnJKLR (Fig. 1A). This fragment gave rise to the same hy-bridization pattern as obtained with the E5 probe (data notshown), thus indicating that no other plantaricin A-inducedtranscript was present in this 4-kb flanking region.The results given above together with the transcription anal-

ysis of plnABCD (9) indicate that plnJKLR, plnMNOP,plnABCD, and plnEFI represent four different operon struc-tures, whereas plnGHSTUV represents a partially sequencedoperon lacking the last 2 to 3 kb. The operons plnJKLR,plnABCD, and plnGHSTUV each seemed to give rise only to afull-length transcript (termed JKLR-, ABCD- and GHSTUV-RNAs, respectively). On the other hand, the transcription ofplnMNOP and plnEFI seemed to be more complex. These twooperons gave rise to four (termed MNOP-, MN-, NOP- andN-RNA) and two (termed EFI- and EF-RNA) transcripts,respectively. All transcripts are depicted in Fig. 1C. To furtherascertain this transcription pattern, the 59 end of each tran-script was subsequently determined by primer extension anal-ysis using appropriate primers (Fig. 4). The results suggest thatthe mRNAs of all five operons have transcriptional startingpoints preceded by plausible 235 and 210 promoter se-quences (Fig. 5). The results also showed that the 59 ends ofthe two truncated transcripts possibly representing plnNOPand plnN (see above) most probably start just upstream of plnN(Fig. 3 and 4C). However, no obvious promoter elements werefound at these locations. Consequently, we assume that thesetwo transcripts were formed by processing of the transcriptinitiated from the promoter upstream of plnM.Regions of dyad symmetry (inverted repeats) with potential

to serve as transcriptional terminators were found at the end of

TABLE 1. Some relevant physicochemical properties of thededuced gene products from the pln bacteriocin locusa

Deducedprotein Size (aa) No. of

PTHs pI GRAVY

PlnI 257 7 9.73 9.97PlnL 138 3 9.31 7.26PlnR 50 1 10.35 7.3PlnM 66 2 9.95 11.21PlnP 247 7 9.86 10.67PlnO 399 1 5.27 24.48PlnG 716 6 6.3 1.93PlnH 457 1 10.04 23.37PlnS 140 4 7.29 7.49PlnT 99 2 7.16 2.87PlnU 222 6 7.87 9.68PlnVb 44 1 10.28 11.72ORF1 62 2 10.58 5.4

a Potential PTHs, pIs, and grand average of hydropathy (GRAVY) values (32)are predicted by the SOAP program.b Presumed to represent the N terminus of a larger protein; see Results.

TABLE 2. Prepeptides containing double-glycine-type leader sequences

Protein Amino acid sequence of prepeptidesa Size (aa), pI ofprepeptide

Size (aa), pI of putativemature peptide

PlnA mkiqikgmkqlsnkemqkivgg kssayslqmgataikqvkklfkkwgw 48, 11.23 26, 10.02PlnE mlqfeklqysrlpqkklakisagg fnrggynfgksvrhvvdaigsvagirgilksir 56, 11.09 33, 12.06PlnF mkkflvlrdrelnaisgg vfhaysargvrnnyksavgpadwvisavrgfihg 52, 11.7 34, 10.7PlnJ mtvnkmikdldvvdafapisnnklngvvgg gawknfwsslrkgfydgeagrairr 55, 10.68 25, 11.5PlnK mkikltvlnefeeltadaeknisgg rrsrkngigyaigyafgaveravlggsrdynk 57, 10.04 32, 10.98PlnN mksldkiaglgiemaekdlttvegg knysktwwyksltllgkvaegtssawhglg 55, 9.28 30, 10.26

Consensus 1b LS--EL- -I-GGConsensus 2 * * ±±± * *

a Capital letter, conserved residue; *, hydrophobic residue; ±, hydrophilic residue. Arrow indicates the conserved cleavage site; the putative mature peptides areindicated by italic letters.b Suggested consensus sequences according to reference 17.

➤



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

the operons plnJKLR, plnMNOP, and plnEFI and of the ORFsplnF and plnN, indicated in Fig. 3. Such a potential transcrip-tional terminator has also been identified at the end ofplnABCD (9). All of these features seem to be consistent withthe transcription pattern obtained above.Further sequence analysis revealed that in all promoters, the

210 consensus sequences were located 6 to 7 bp upstream ofthe transcriptional starting points. The corresponding 235consensus sequences, however, were more difficult to identify.When these promoter sequences were aligned (Fig. 5), all wereseen to contain two direct repeats located just upstream of theputative 235 sites. Each repeat consists of 9 bp with the con-sensus sequence 59-TACGTTAAT-39. The two repeats arespaced by an AT-rich stretch of 12 bp. These features resembleDNA sequence elements identified as binding sites for thetranscription regulator AraC (5, 23, 34) (see Discussion). Therepeat closest (6 to 9 bp) to the 235 site was termed right (R),whereas the second repeat further upstream (12 bp) wastermed left (L). It is noteworthy that the plnGHSTUV pro-moter contain three direct repeats, all separated by 12 bp. Thethird repeat (termed R9) is located downstream of the R re-peat between the 235 and 210 promoter elements. The align-ment of the repeats (11 in total) from the five promotersrevealed that the base pairs in positions (from 59) 2 (adenine),3 (cytosine), 6 (thymine), and 8 (adenine) are invariant in allcases.

DISCUSSION

Nature of the bacteriocin activity in L. plantarum C11. Thepresent results provide new evidence supporting our hypothe-sis (8) that plantaricin A is an induction factor and that oth-er peptides are required for the bacteriocin activity observed.We have identified three operons, plnEFI, plnJKLR, andplnMNOP, all encoding different bacteriocin-like peptideswith double-glycine-type leaders. In the case of plnEFI andplnJKLR, the genes (plnEF and plnJK) encoding the bacterio-cin-like prepeptides are followed by genes (plnI and -L) en-coding cationic hydrophobic polypeptides with PTHs. This

gene order seems to follow the conserved genetic organizationof all known two-peptide bacteriocins reported to date (14,37, 59). In these cases, the corresponding cationic hydropho-bic determinants are thought to serve an immunity role. Thetwo operons (plnEFI and plnJKLR) are thus potential to en-code two bacteriocins of two-peptide type. The third operon(plnMNOP) contains a bacteriocin-like ORF, plnN (see Re-sults), that could encode a one-peptide bacteriocin. However,no obvious gene for a potential immunity determinant wasfound to immediately follow this gene. Instead, plnN is fol-lowed by an ORF (plnO) which encodes a large (399-aa)polypeptide with a rather acidic pI and low hydrophobicity.Nevertheless, the two remaining ORFs on this operon, plnMand -P, both encode hydrophobic cationic polypeptides resem-bling immunity proteins. PlnM has a length of 66 aa, which isin the range (51 to 113 aa) observed for most immunity pro-teins associated with one-peptide bacteriocins (3, 35, 48, 57–60), whereas the size of PlnP (247 aa) exceeds this range. PlnP,however, shows a significant similarity to PlnI (see Results),which is thought to be the immunity protein associated with theputative two-peptide bacteriocin encoded by plnEF. Thus, ei-ther of these two ORFs might encode the immunity protein forthis putative bacteriocin system.The transcription pattern of these three operons is also

consistent with that observed for bacteriocin production: all ofthese operons were repressed when the bacteriocin productionwas turned off, i.e., when cultures either were depleted forplantaricin A or were in the stationary growth phase (8). How-ever, transcription of all became activated when the bacterio-cin production was on, which in Bac1 C11 cultures took placethroughout the exponential and early stationary growth phases(8).The plantaricin A peptide clearly resembles nonlantibiotics

(9, 40). However, the genetic organization of the plantaricinA-encoding operon clearly differs from what is observed formost known bacteriocin-encoding operons and also from thatof the three putative bacteriocin-encoding operons describedabove. No potential immunity gene was present on the plan-taricin A-encoding operon; instead, determinants (plnBCD)

FIG. 4. Primer extension analysis of plnJ (using oligoJ as the primer; A), plnM (oligoM; B), plnN (oligoN; C), plnA (S14; D), plnE (oligoE; E), and plnG (oligoG;F). The nucleotide sequences of the primers are given in Materials and Methods. The nucleotide sequence for each gene was generated by sequencing appropriateplasmid clones, using the same oligonucleotides as primers; lane order is GATC. See Fig. 3 and 5 for genetic locations of these 59-end sites.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

for a signal transducing system were found (9). These results,combined with the fact that plantaricin A lacks antimicrobialactivity (8), suggest strongly that plantaricin A is, per se, not abacteriocin but rather serves as a regulatory peptide.Indication of a common regulatory mechanism in agr-like

regulons. It is known that most transcription regulators andfactors control their target operons by direct binding to a siteclose to or within the promoter region, thereby stimulating orinhibitory the binding of RNA polymerase to promoter 235and 210 sites. Features such as inverted repeats (e.g., in lswitch [46]) or direct repeats (e.g., in arabinose regulation [5,23, 34]) are frequently observed as binding sites for transcrip-tion regulators. They often bind as dimers, attaching to theirbinding sites in a cooperative manner, one subunit on eachrepeat. Critical intermolecular bonds are often formed be-tween the transcription regulator and the most conserved nu-cleotides within the repeats (28, 53). Some detailed studies onsuch binding sites (5, 46) have indicated that they must facetoward the same side of the DNA helix in order to achieveoptimal contacts with the DNA binding domains of the regu-lators. The direct repeats found within the conserved regula-tory-like boxes (CRLBs) in the pln promoters seem to followthis pattern when the 30-bp regions spanning them are dis-played on a DNA double-helix structure (10.5 bp per turn)(data not shown). On the basis of this information, we stronglybelieve that the CRLBs in the pln promoters serve as bindingsites for a common transcription regulator.Interestingly, when we examined the operons regulated by

agr (26) and another agr-related system in Staphylococcus lug-dunensis (61) and the operons or operon-like structures found

in gene clusters responsible for the production of sakacin A(sap) (3), sakacin P (spp) (20), and carnobacteriocins A, B2,and BM1 (cbn) (47, 48, 63), we also found similar CRLBsupstream of their promoters (Fig. 5). This observation suggeststhat all of these systems might be regulated by a commonmechanism. It is noteworthy that all of these bacteriocin sys-tems (including pln) and agr show high amino acid sequencesimilarity between their HPKs and RRs. Regarding the possi-ble role of these RRs as transcription regulators, we also no-ticed a high occurrence of basic residues at their C termini(Fig. 6B) which might play a role in binding to the negativelycharged phosphate groups on the DNA backbone. Moreover,the resemblance of these systems extends to their genetic or-ganization: the HPK and RR genes are organized in tandem,and the HPK gene is preceded by a small ORF (plnA in pln,orf4 in sap, orfY in spp, and orf6 in cbn) encoding a nonlantibi-otic-like peptide precursor (Fig. 6A). As found with plantaricinA (matured from PlnA), the mature gene product of orfY hasalso been found to serve as an induction factor in bacteriocinproduction (10). The roles of the other two small ORFs (orf4and orf6) have not been reported. However, both might havesimilar regulatory functions as well. Finally, the synthesis of thevirulence factors in S. aureus which is regulated by the agr locusis triggered by an octapeptide (encoded by agrD) via a two-component regulatory, signal transduction system (agrA andagrC) similar to what has now been suggested for bacteriocinsystems (42).It was noted that the rate of transcription of plnGHSTUV

appeared to decrease somewhat earlier than that of otheroperons (see Results), indicating a selective regulation. This

FIG. 5. Alignment of the promoter sequences from operons or operon-like structures thought to be regulated by agr and other agr-like systems. The L. plantarumC11 pln promoters (this study and reference 9), plnA, -J, -M, -E, and -G; the L. sake spp promoters (20), orfY, sppA, sppT, and orfX; the S. aureus agr promoters (26),agrD and hld; the S. lugdunensis agr-related promoter (61), orfD (similar to agrD); the L. sake sap promoters (3), sapA and orf4; the Carnobacterium piscicola cbnpromoters (47, 48, 63), orf5, orf6, cbnB2, cbnBM1, and cbnA. L. sake orfY and orfX were identified with coordinates 443 to 551, and 7489 to 7597, respectively, in theDNA sequence obtained (20). The 235 and 210 sites are underlined; the conserved bases within these sites are indicated by boldface letters. Similarly, CRLBscontaining direct (L and R) repeats are boxed, and the invariant bases within the repeats are indicated by boldface letters. The consensus sequence of the direct repeatsis indicated for each promoter group. The transcriptional starts (TS; indicated by boldface letters) in pln or agr in S. aureus have been experimentally determined (thisstudy and reference 26, respectively). The first codons (ATG/TTG) are indicated at the end of each sequence, whereas the preceding figures refer to the number ofbase pairs in the region between. Note that the 235 sites are, almost in all cases, less conserved than the 210 sites and that the plnG promoter contains an extra directrepeat (R9; indicated by italic letters) located just downstream of its 235 site.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

reduction took place in late exponential growth phase prior tothe gradual drop in bacteriocin production (8). This operonencodes an ABC transporter (PlnG) and its accessory protein(PlnH), which are believed to constitute the dedicated process-ing and secretion machinery of plantaricin A and the other fivenew bacteriocin-like peptides with double-glycine-type leaders.It is possible that the repression of transcription of this regu-latory operon is the initial cause in decline of bacteriocin pro-duction. The promoter preceding this operon differs from theothers by containing three consecutive direct repeats: the R9repeat flanking the 39 end of the promoter 235 site (Fig. 5). Itis feasible that the R9 repeat could be occupied by an RRsubunit during the late exponential growth phase, when thelevel of RR will be high. A bound RR on this site (R9) couldinterfere with RNA polymerase binding and thereby poten-tially limit transcription. If this model is correct, an RR occu-pying the R9 repeat might explain the early repression of thisoperon. Such a regulatory mechanism involving both gene ac-tivation and a subsequent repression has also been character-ized for the genetic switch of l, which is controlled by the twocentral transcription regulators, l Cro and cI repressor (46).This work provides new information on the genetics and

regulatory mechanism underlying the bacteriocin productionin L. plantarum C11. Plantaricin A serves as an induction signalfor bacteriocin production by triggering transcription of therelevant genes. The gene encoding plantaricin A is located onthe same transcription unit as the signal transducing system(PlnBCD). It is therefore likely that this regulatory system isinvolved in regulation of bacteriocin production as well. Thus,we propose the hypothesis that the HPK PlnB acts as a recep-tor for plantaricin A and that PlnB, through the RRs (PlnCD),activates transcription of genes involved in bacteriocin produc-tion.Finally, it should be mentioned that in lantibiotics, two-

component regulatory systems have also been found and theregulation of nisin synthesis has been most studied (31). Whilethe pln system and most probably other nonlantibiotic systems(Fig. 6) produce separate bacteriocin-inducing peptides whichare definitely different from the peptides exerting bacteriocin

activity, nisin serves as an induction factor regulating its ownsynthesis (11, 31).

ACKNOWLEDGMENTS

This work was supported by grants from the Norwegian ResearchCouncil (NRC) and the Commission of European Communities (con-tract BIO2 CT 920137). B. D. Diep was supported by a fellowship fromNRC.We thank K. Andersen for critical reading of the manuscript.

REFERENCES

1. Alam, J., R. A. Whitaker, D. W. Krogman, and S. E. Curtis. 1986. Isolationand sequence of the gene for ferredoxin I from the cyanobacteriumAnabaena sp. strain PCC7120. J. Bacteriol. 168:1265–1271.

2. Allison, G., C. Fremaux, C. Ahn, and T. R. Klaenhammer. 1994. Expansionof bacteriocin activity and host range upon complementation of two peptidesencoded by the lactacin F operon. J. Bacteriol. 176:2235–2241.

3. Axelson, L., and A. Holck. 1995. The genes involved in production of andimmunity to sakacin A, a bacteriocin from Lactobacillus sake Lb706. J.Bacteriol. 177:2125–2137.

4. Bourret, R. B., K. A. Borkovich, and M. I. Simon. 1991. Signal transductionpathways involving protein phosphorylation in prokaryotes. Annu. Rev. Bio-chem. 60:401–441.

5. Cara, J. H., and R. F. Schleif. 1993. Variation of half-site organization andDNA looping by AraC protein. EMBO J. 12:35–44.

6. Church, G. M., and W. Gilbert. 1984. Genomic sequencing. Proc. Natl. Acad.Sci. USA 81:1991–1995.

7. Daeschel, M. A., M. C. McKenney, and L. C. McDonald. 1990. Bacteriocidalactivity of Lactobacillus plantarum C11. Food Microbiol. 7:91–99.

8. Diep, D. B., L. S. Håvarstein, and I. F. Nes. 1995. A bacteriocin-like peptideinduces bacteriocin synthesis in L. plantarum C11. Mol. Microbiol. 18:631–639.

9. Diep, D. B., L. S. Håvarstein, J. Nissen-Meyer, and I. F. Nes. 1994. The geneencoding plantaricin A, a bacteriocin from Lactobacillus plantarum C11, islocated on the same transcription unit as an agr-like regulatory system. Appl.Environ. Microbiol. 60:160–166.

10. Eijsink, V. G. H., M. B. Brurberg, P. H. Middelhoven, and I. F. Nes. 1996.Induction of bacteriocin production in Lactobacillus sake by a secreted pep-tide. J. Bacteriol. 178:2232–2237.

11. Engelke, G., Z. Gutowski-Eckel, P. Kiesau, K. Siegers, M. Hammelmann,and K.-D. Entien. 1994. Regulation of nisin biosynthesis and immunity ofLactococcus lactic 6F3. Appl. Environ. Microbiol. 60:814–825.

12. Fath, M. J., and R. Kolter. 1993. ABC transporters: bacterial exporters.Microbiol. Rev. 57:995–1017.

13. Fath, M. J., L. H. Zhang, J. Rush, and R. Kolter. 1994. Purification andcharacterization of colicin V from Escherichia coli supernatants. Biochem-istry 33:6911–6017.

14. Fremaux, C., C. Ahn, and T. R. Klaenhammer. 1993. Molecular analysis ofthe lactacin F operon. Appl. Environ. Microbiol. 59:3906–3915.

15. Håvarstein, L.-S., G. Coomaraswamy, and D. A. Morrison. 1995. An uni-dentified heptadecapeptide pheromone induces competence for genetictransformation in Streptococcus pneumoniae. Proc. Natl. Acad. Sci. USA92:11140–11144.

16. Håvarstein, L.-S., B. D. Diep, and I. F. Nes. 1995. A family of bacteriocinABC transporters carry out proteolytic processing of their substrates con-comitant with export. Mol. Microbiol. 16:229–240.

17. Håvarstein, L.-S., H. Holo, and I. F. Nes. 1994. The leader peptide of colicinV shares consensus sequences with leader peptides that are common amongpeptide bacteriocins produced by Gram-positive bacteria. Microbiology 140:2383–2389.

18. Hoch, J. A., and T. J. Silhavy (ed.). 1995. Two-component signal transduc-tion. ASM Press, Washington, D.C.

19. Holo, H., Ø. Nilssen, and I. F. Nes. 1991. Lactococcin A, a new bacteriocinfrom Lactococcus lactis subsp. cremoris: isolation and characterization of theprotein and its gene. J. Bacteriol. 173:3879–3887.

20. Huehne, K., A. Holck, L. Axelson, and L. Kroeckel. 1995. Cloning andnucleotide sequence of genes involved in regulation and secretion of sakacinP (sakacin 674), a bacteriocin from Lactobacillus sake Lb674. PublishedDNA sequence with accession no. z48542.

21. Hui, F. M., and D. A. Morrison. 1991. Genetic transformation in Streptococ-cus pneumoniae: nucleotide sequence analysis shows comA, a gene requiredfor competence induction, to be a member of the bacterial ATP-dependenttransport protein family. J. Bacteriol. 173:372–381.

22. Hui, F. M., L. Zhou, and D. A. Morrison. 1995. Competence for genetictransformation in Streptococcus pneumoniae: organization of a regulatorylocus with homology to two lactococcin A secretion genes. Gene 151:25–31.

23. Huo, L., K. J. Martin, and R. Schleif. 1988. Alternative loops regulate thearabinose operon in Escherichia coli. Proc. Natl. Acad. Sci. USA 85:5444–5448.

FIG. 6. (A) Conserved genetic organization of four agr-like signal transduc-ing systems: pln (9), sap (3), spp (20), and cbn (47). Small unfilled arrows (plnA,orf4, orfY, and orf6) indicate ORFs encoding precursor peptides with double-glycine-type leaders; filled arrows indicate HPK genes (plnB, sapK, sppK, andcbnK); hatched arrows indicate RR genes (plnCD, sapR, sppR, and cbnR). (B)Alignment of the C termini of the four above-mentioned agr-like RRs and AgrAitself. Conserved residues are indicated by capital letters; basic residues areindicated by boldface letters. Note the high occurrence of basic residues at theirextreme ends.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

24. Igo, M. M., and R. Losick. 1986. Regulation of a promoter that is utilized byminor forms of RNA polymerase holoenzyme in Bacillus subtilis. J. Mol.Biol. 191:615–624.

25. Janzon, L., and S. Arvidson. 1990. The role of the d-lysin gene (hld) in theregulation of virulence genes by the accessory gene regulator (agr) in Staph-ylococcus aureus. EMBO J. 9:1391–1399.

26. Janzon, L., S. Lofdahl, and S. Arvidson. 1989. Identification and nucleotidesequence of the delta-lysin gene, hld, adjacent to the accessory gene regu-lator (agr) of Staphylococcus aureus. Mol. Gen. Genet. 219:480–485.

27. Ji, G., R. C. Beavis, and R. P. Novick. 1995. Cell desity control of staphylo-coccal virulence mediated by an octapeptide pheromone. Proc. Natl. Acad.Sci. USA 92:12055–12059.

28. Jordan, S. R., and C. O. Pabo. 1988. Structure of the lambda complex at 2.5Å resolution: details of the repressor-operator interactions. Science 242:893–899.

29. Klaenhammer, T. R. 1993. Genetics of bacteriocins produced by lactic acidbacteria. FEMS Microbiol. Rev. 12:39–86.

30. Kornblum, J., B. Kreiswirth, S. J. Projan, H. Ross, and R. P. Novick. 1990.agr: a polycistronic locus regulating exoprotein synthesis in Staphylococcusaureus, p. 373–402. In R. P. Novick, (ed.), Molecular biology of the staphy-lococci. VCH Publishers, New York.

31. Kuipers, O. P., M. M. Beerthuyzen, P. G. G. A. deRuyter, E. J. Luesink, andW. M. deVos. 1995. Autoregulation of nisin biosynthesis in Lactococcus lacticby signal transduction. J. Biol. Chem. 270:281–291.

32. Kyte, J., and R. F. Doolittle. 1982. A simple method for displaying thehydropathic character of a protein. J. Mol. Biol. 157:105–132.

33. Lillehaug, D., B. Lindqvist, and N. K. Birkeland. 1991. Characterization ofLC3, a Lactococcus lactis subsp. cremoris temperate bacteriophage withcohesive single-stranded DNA ends. Appl. Environ. Microbiol. 57:3206–3211.

34. Martin, K., L. Huo, and R. F. Schleif. 1986. The DNA loop model for ararepression: AraC protein occupies the proposed loop sites in vivo and re-pression-negative mutations lie in these same sites. Proc. Natl. Acad. Sci.USA 83:3654–3658.

35. Marugg, J. D., C. F. Gonzales, B. S. Kunka, A. M. Ledeboer, M. J. Pucci,M. Y. Toonen, S. A. Walker, L. C. M. Zoetmulder, and P. A. Vandenbergh.1992. Cloning, expression, and nucleotide sequence of genes involved inproduction of pediocin PA-1, a bacteriocin from Pediococcus acidilacticiPAC1.0. Appl. Environ. Microbiol. 58:2360–2367.

36. Morfeldt, E., L. Janzon, S. Arvidson, and S. Lofdalh. 1988. Cloning of achromosomal locus (exp) which regulates the expression of several exopro-tein genes in Staphylococcus aureus. Mol. Gen. Genet. 211:435–440.

37. Nes, I. F., L. S. Håvarstein, and H. Holo. 1995. Genetics of non-lantibioticsbacteriocins, p. 645–651. In J. J. Ferretti, M. S. Gilmore, T. R. Klaenhammer,and F. Brown (ed.), Genetics of streptococci, enterococci and lactococci.Karger, New York.

38. Nissen-Meyer, J., L.-S. Håvarstein, H. Holo, K. Sletten, and I. F. Nes. 1993.Association of the lactococcin A immunity factor with the cell membrane:purification and characterization of the immunity factor. J. Gen. Microbiol.139:1503–1509.

39. Nissen-Meyer, J., H. Holo, L.-S. Håvarstein, K. Sletten, and I. F. Nes. 1992.A novel lactococcal bacteriocin whose activity depends on the complemen-tary action of two peptides. J. Bacteriol. 174:5686–5692.

40. Nissen-Meyer, J., A. G. Larsen, K. Sletten, M. Daeschel, and I. F. Nes. 1993.Purification and characterization of plantaricin A, a Lactobacillus plantarumbacteriocin whose activity depends on the action of two peptides. J. Gen.Microbiol. 139:1973–1978.

41. Novick, R., J. Kornblum, B. Kreiswirth, S. Projan, and H. Ross. 1989. agr: acomplex locus regulating post-exponential phase exoprotein synthesis inStaphylococcus aureus, p. 495–510. In L. O. Butler and B. E. B. Moseley(ed.), Genetic transformation and expression. Intercept, Ltd., Andover,United Kingdom.

42. Novick, R. P., S. J. Projan, J. Kornblum, H. F. Ross, G. Ji, B. Kreoswirth, F.Vandenesch, and S. Moghazeh. 1995. The agr P2 operon: an autocatalytic

sensory transduction system in Staphylococcus aureus. Mol. Gen. Genet.248:446–458.

43. Ojcius, D. M., and J. D. Young. 1991. Cytolytic pore-forming proteins andpeptides: is there a common structural motif? Trends Biochem. Sci. 16:225–229.

44. Parkinson, J. S., and E. C. Kofoid. 1992. Communication modules in bac-terial signaling proteins. Annu. Rev. Genet. 26:71–112.

45. Peng, H., R. P. Novick, B. Kreiswirth, J. Kornblum, and P. Schlievert. 1988.Cloning, characterization, and sequencing of an accessory gene regulator(agr) in Staphylococcus aureus. J. Bacteriol. 170:4365–4372.

46. Ptashne, M. 1992. The genetic switch. Cell Press & Blackwell ScientificPublications, Cambridge, Mass.

47. Quadri, L. E. N., K. L. Roy, J. C. Vederas, and M. E. Stiles. 1995. Charac-terization of four genes involved in the production of antimicrobial peptidesby Carnobacterium piscicola LV17B. Published DNA sequence with acces-sion no. 147121.em_bl.new.

48. Quadri, L. E. N., M. Sailers, K. L. Roy, J. C. Vederas, and M. E. Stiles. 1994.Chemical and genetic characterization of bacteriocins produced by Car-nobacterium piscicola LV17B. J. Biol. Chem. 269:12204–12211.

49. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

50. Schiffer, M., and A. B. Edmundson. 1967. Use of helical wheels to representthe structures of proteins and to identify segments with helical potential.Biophys. J. 7:121–135.

51. Schnell, N., G. Engelke, J. Augustin, R. Rosenstein, V. Ungermann, F. Gotz,and K.-D. Entian. 1992. Analysis of genes involved in the biosynthesis oflantibiotic epidermin. Eur. J. Biochem. 204:57–68.

52. Schnell, N., K.-D. Entian, U. Schneider, F. Gotz, H. Zahner, R. Kellner, andG. Jung. 1988. Prepeptide sequence of epidermin, a ribosomally synthesizedantibiotic with four sulphide-rings. Nature (London) 333:276–278.

53. Shimon, L. J. W., and S. C. Harrison. 1993. The phage 434 OR2/R1-69complex at 2.5 Å resolution. J. Mol. Biol. 232:826–838.

54. Skaugen, M. 1995. Lactocin S. Ph.D. thesis. Agricultural University of Nor-way, Ås, Norway.

55. Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989. Protein phosphorylation andregulation of adaptive responses in bacteria. Microbiol. Rev. 53:450–490.

56. Stoddard, G. W., J. P. Petzel, M. J. Van Belkum, J. Kok, and L. L. McKay.1992. Molecular analyses of the lactococcin A gene cluster from Lactococcuslactis subsp. lactis biovar diacetylactis WM4. Appl. Environ. Microbiol. 58:1952–1961.

57. Tichaczek, P. S., R. F. Vogel, and W. P. Hammes. 1993. Cloning and se-quencing of curA encoding curvacin A, the bacteriocin produced by Lacto-bacillus curvatus LTH1174. Arch. Microbiol. 160:279–283.

58. Tichaczek, P. S., R. F. Vogel, and W. P. Hammes. 1994. Cloning and se-quencing of sakP encoding sakacin P, the bacteriocin produced by Lactoba-cillus sake LTH673. Microbiology 140:361–367.

59. van Belkum, M. J., B. J. Hayema, R. E. Jeeninga, J. Kok, and G. Venema.1991. Organization and nucleotide sequences of two lactococcal bacteriocinoperons. Appl. Environ. Microbiol. 57:492–498.

60. van Belkum, M. J., J. Kok, and G. Venema. 1992. Cloning, sequencing, andexpression in Escherichia coli of lcnB, a third bacteriocin determinant fromthe lactococcal bacteriocin plasmid p9B4-6. Appl. Environ. Microbiol. 58:572–577.

61. Vandenesch, F., S. J. Projan, B. Kreiswirth, J. Etienne, and R. Novick. 1993.Agr-related sequences in Staphylococcus lugdunensis. FEMS Microbiol. Lett.111:115–122.

62. Venema, K., R. E. Haverkort, T. Abee, A. J. Haandrikman, K. J. Leenhouts,L. D. Leij, G. Venema, and J. Kok. 1994. Mode of action of LciA, thelactococcin A immunity protein. Mol. Microbiol. 14:521–532.

63. Worobo, R. W., T. Henkel, M. Sailer, K. L. Roy, J. C. Vederas, and M. E.Stiles. 1994. Characteristics and genetic determinant of a hydrophobic pep-tide bacteriocin, carnobacteriocin A, produced by Carnobacterium piscicolaLV17A. Microbiology 140:517–526.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

characterizationofthelocusresponsibleforthebacteriocin...

Documents