Previous Article | Next Article ![]()
Applied and Environmental Microbiology, October 2003, p. 5839-5848, Vol. 69, No. 10
0099-2240/03/$08.00+0 DOI: 10.1128/AEM.69.10.5839-5848.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Molecular Genetics, Biomolecular Sciences and Biotechnology Institute, University of Groningen, Groningen,1 Corporate Research, Friesland Coberco Dairy Foods, 7400 AB Deventer, The Netherlands2
Received 31 March 2003/ Accepted 1 August 2003
|
|
|---|
|
|
|---|
Circularin A, a circular bacteriocin produced by C. beijerinckii ATCC 25752, is active against a broad range of gram-positive bacteria (25). The circularization of the peptide involves a head-to-tail peptide bond formation between the fourth and last amino acid of the precursor peptide (25). Circularin A shares limited sequence homology with enterocin AS-48 (also known as Bac21), a cyclic bacteriocin from Enterococcus faecalis (34, 56), but its precursor lacks the long leader present in the enterocin AS-48 precursor. The circularin A gene cluster is chromosomally located, while the enterocin AS-48 operon is located on a plasmid.
Both circularin A and enterocin AS-48 belong to the recently defined class V bacteriocins of ribosomally synthesized, nonmodified, head-to-tail-ligated cyclic antibacterial peptides (25). Other class V bacteriocins are microcin J25 and gassericin A (4, 22). Microcin J25, peptide of 21 amino acid residues produced by Escherichia coli, is the only circular peptide known so far that is produced by a gram-negative bacterium (4). The genes involved in the production of microcin J25 are located in an operon immediately downstream of the structural gene (51). Gassericin A is produced by Lactobacillus gasseri as a 91-amino-acid precursor peptide that is circularized after removal of a leader peptide of 33 amino acids (22, 23). The coding regions of enterocin AS-48 of two strains have been sequenced and determined to be almost identical (35, 56). Enterocin AS-48 is a tightly packed peptide containing five
-helices and is structurally related to NK-lysin, a cytotoxic peptide from human natural killer or T cells (16).
Most bacteriocins require processing of a precursor peptide in order to become (fully) active. For many bacteriocins, the genes encoding processing, secretion, and immunity functions flank the structural gene. Processing can involve modification of amino acids, as is the case in lantibiotics, leader peptide removal, or, in the case of circular peptides, circularization. The mechanisms underlying these modifications are poorly understood, although the proteins involved are generally known. The secretion of most bacteriocins occurs via dedicated ABC transporters (26), while some can be secreted via the general secretion pathway (5, 21, 33).
Immunity systems for bacteriocins are poorly characterized, but it has been demonstrated that specialized immunity proteins confer immunity to bacteriocin action on cells by blocking access to a putative receptor, as is the case for the lactococcin A immunity protein LciA (58). In some cases ABC transporters have been shown to be involved, e.g., the NisFEG system in nisin resistance and McbFE in microcin B17 resistance (13, 43). Little homology exists among bacteriocin immunity proteins, even those that are involved in immunity against bacteriocins of the same class.
In this study we identified the genes required for functional heterologous expression of circularin A and showed that two independent mechanisms confer reduced circularin A sensitivity, one of which is based on the expression of cirE and the other on the combined expression of cirB and cirD. As such, this study will further the field of clostridial bacteriocins and that of class V (circular) bacteriocins in particular.
|
|
|---|
was grown for 16 h in tryptone-yeast (TY) broth at 37°C with vigorous agitation (250 rpm). For growth on plates, medium containing 1.5% agar was used. Ampicillin (Sigma, Zwijndrecht, The Netherlands) and chloramphenicol (Sigma) were used at 100 and 10 µg/ml, respectively, for E. coli. Chloramphenicol and erythromycin (Sigma) were used at 5 µg/ml each for L. lactis NZ9000 and at 20 and 2 µg/ml, respectively, for E. faecalis JH2-2. When used together, chloramphenicol and erythromycin were employed at 2.5 µg/ml each for Lactobacillus saké ATCC 15521 or at 10 µg/ml (chloramphenicol) and 2 µg/ml (erythromycin) for E. faecalis JH2-2. |
View this table: [in a new window] |
TABLE 1. Bacterial strains and plasmids used in this study
|
Bacteriocin assays.
Colony overlayer assays were performed as described previously (25). Bacteriocin activity in C. beijerinckii ATCC 25752 supernatant was quantified in triplicate by a critical dilution assay as described by Geis et al. (14), with the modification that assays were performed in microtiter plates. To 50 µl of serially diluted, bacteriocin-containing samples, 150 µl of medium containing the indicator strain Lactobacillus saké ATCC 15521(pMG36e; pMG36c), diluted 100-fold from a stationary-phase overnight culture, was added unless mentioned otherwise. Resistance to bacteriocin was determined by plating strains on plates containing 4 or 10% (vol/vol) filter-sterilized C. beijerinckii ATCC 25752 supernatant containing circularin A. Alternatively, a critical-dilution assay was performed with the strain of interest as an indicator strain. In these critical-dilution assays, 100 µl of bacteriocin-containing sample was mixed with 100 µl of freshly diluted (1,000-fold) indicator strains.
Cloning methods and materials.
Molecular cloning techniques were performed essentially as described by Sambrook et al. (45). Restriction enzymes, T4 DNA ligase, and Expand DNA polymerase were obtained from Roche (Roche Diagnostics GmbH) and used as specified by the supplier. L. lactis NZ9000 was transformed as described by Shepard and Gilmore (19) with 1% glycine (Merck). E. faecalis JH2-2 was transformed as described previously (47; http://w3.ouhsc.edu/enterococcus) with 8% glycine. After transformation, both strains were plated on G[1/2]M17 medium containing 0.5 M sucrose and the appropriate antibiotics. Plasmids from L. lactis NZ9000, Escherichia coli DH5
, and E. faecalis JH2-2 were isolated according to Birnboim (3), with the following modifications for E. faecalis: mutanolysin (1 U/ml; Sigma) was added to the suspension buffer to facilitate lysis, and plasmids isolated from 50 ml of culture were, after RNase (0.5 mg/ml; Sigma) treatment, further purified with the High Pure PCR product purification kit (Roche Diagnostics GmbH).
Cloning of the circularin A determinant.
The region encompassing cirA to cirE was amplified with primers located just upstream of cirA (B51426, 5'-ACGCGTCGACTCATGAGTTTTTCAAAAGGAGGTGATTAATT ATGTTTTTATTGCAGG-3') and downstream of cirE (B51427, 5'-CGCGGATCCGTCGACCTCTCCCACTTTAACATTAGTTATTGCTC-3'). SalI-RcaI and BamHI-SalI sites, respectively, in the two primers are underlined. All enzymes (Roche) were used according to the manufacturer's instruction. The PCR product was cloned with the Zero-Blunt Topo PCR cloning kit (Invitrogen, Breda, The Netherlands), creating pCRAE. The plasmid pCRAE was digested with SpeI and XhoI. The fragment carrying cirA was ligated into pMG36c digested with XbaI and SalI, and the ligation mixture was used to transform E. coli DH5
. Transformants were identified by growth on TY agar with chloramphenicol. The correct plasmid, pMGAE1, was isolated as described above and introduced into E. faecalis JH2-2. Three consecutive selection steps of clean streaking and testing for a strain with a high and stable bacteriocin expression phenotype with the colony overlay assay yielded E. faecalis JH2-2 carrying a pMGAE1 derivative labeled pCir.
In-frame deletions of cirA through cirD were made by amplifying pCir by PCR with appropriate outward-facing primers, creating a PCR product of the entire plasmid but lacking the gene of interest. The primers used were 5'-AGTATGGCAAGAGCTATAGC-3' and 5'-CACGCCTAGTGCTCCTGC-3' for
cirA, 5'-TAATTATGCCTGTATCATACC-3' and 5'-CCAAGAGTTATAGTTTGAGT CG-3' for
cirB, 5'-GTGCACATAGGTAGGATTTTAAG-3' and 5'-GAAACATTCC AACAATAATACC-3' for
cirC, and 5'-GAACTTAATCTAGTTAACGGAAG-3' and 5'-AGTTATCTCTAGCATAGGCTTC-3' for
cirD.
Each PCR product was kinase treated with T4 polynucleotide kinase (Amersham Pharmacia Biotech) in T4 ligase buffer (Roche Diagnostics GmbH) and subsequently self-ligated with T4 ligase, creating the plasmids pCir
A, pCir
B, pCir
C, and pCir
D. Derivatives of pCir with a deletion of cirE and a deletion of one of the other genes cirABCD, pCir
AE, pCir
BE, pCir
CE, or pCir
DE, were made by the same method with the primers 5'-CATATATTCTACTACCTTTC-3' and 5'-GTAATTAAAGGCTCTAATAAG-3' for
cirE and the plasmids carrying the respective single deletions as templates in the PCR. A plasmid with a triple deletion, pCir
ACE, was constructed likewise by deleting cirC with pCir
AE as a template and the primers used for the single deletion of cirC. Based on pCir
ACE, pCir
ABCE and pCir
ACDE were made with the primers employed for the single knockouts of cirB and cirD, respectively. All plasmids were isolated with E. faecalis JH2-2 as the cloning host.
The cirE gene was cloned behind the lactococcal chromosomal P32 promoter by digesting the PCR product obtained with primers B51426 and B51427 and HpaI and SalI and ligating the cirE-carrying fragment into SmaI- and SalI-digested pMG36e, leading to pMG-E. Colonies obtained after transformation of L. lactis NZ9000 were replica streaked onto G[1/2]M17 plates with 4% (vol/vol) C. beijerinckii ATCC 252752 supernatant to screen for circularin A immunity. The plasmid was isolated and used to transform E. faecalis JH2-2. In order to make pCir
E, a fragment with the immunity gene cirE behind the P32 promoter was first cloned in pIL253 to avoid possible lethal effects of bacteriocin expression without immunity. This was done by digesting pMG-E with EcoRI and SalI and ligating the cirE-carrying fragment into pIL253 digested with the same enzymes. The resulting plasmid (pIL-E) was introduced into E. faecalis JH2-2. pCir
E was subsequently made with pCir as a template and the appropriate primers. E. faecalis JH2-2(pIL-E) was used as the host for construction of pCir
E.
Computational analyses.
Open reading frames were identified with the Glimmer 2.0 program (6). Predictions by the Glimmer 2.0 program were checked manually for validity. Homology comparisons were performed with the basic logical alignment tool (Blast) as described by Altschul et al. (1). Blast searches were performed against the NCBI nonredundant protein database and the NCBI microbial genomes database (http://www.ncbi.nlm.nih.gov/BLAST/). Homologies with conserved domains from the Pfam database (http://www.sanger.ac.uk/Software/Pfam/) (2) were also identified with Blast searches. Putative signal peptides were identified with signalP (http://www.cbs.dtu.dk/services/SignalP/) (38). Putative transmembrane helices were identified with the TMHMM2.0 program (http://www.cbs.dtu.dk/services/TMHMM-2.0/) (28). Dyad symmetries, isoelectric points, and molecular weights were determined with the program Clonemanager 4 (SEcentral; Scientific & Educational Software). Sequence alignments were performed with the ClustalW program available at http://www2.ebi.ac.uk/clustalw/ (54).
Nucleotide sequence accession number.
The cirA sequence is available under GenBank accession number AJ566621.
|
|
|---|
![]() View larger version (27K): [in a new window] |
FIG. 1. (A) Physical map of the region surrounding the circularin A structural gene cirA of C. beijerinckii ATCC 25752. Solid arrows indicate genes; bent arrows show putative promoters; lollipops represent predicted regions of dyad symmetry ( G0 < -10 kcal/mol); dotted arrows show possible polycistronic messengers. Map units are base pairs. (B) Schematic representation of the cir DNA fragments in the indicated plasmids and locations of the deletions (indicated by the thin lines). Open arrowheads indicate a deletion in cirE. Promoters are shown by bent arrows. CirA+, circularin A production; CirR, circularin A resistance, denoted as ++, full protection against CirA [>24-fold increase relative to E. faecalis JH2-2(pMG36c)]; +, partial protection [2- to 16-fold increase relative to E. faecalis JH2-2(pMG36c)]; -, sensitive.
|
![]() View larger version (44K): [in a new window] |
FIG. 2. (A) Nucleotide sequences of promoter and translation initiation regions in the cir gene cluster. Putative -35, -10, and ribosome-binding site (RBS) sequences are underlined. Deduced amino acid sequences are indicated below the nucleotide sequences, and gene names are given below the amino acid sequences. Putative start codons are indicated in boldface and are numbered when more than one possibility exists. Termination codons are underlined and in italic.
|
|
View this table: [in a new window] |
TABLE 2. Characteristics of predicted proteins specified by the C. beijerinckii ATCC 25752 cir gene cluster
|
![]() View larger version (58K): [in a new window] |
FIG. 3. Alignment of the putative ATP binding proteins involved in circularin A (CirD, CirH) and enterocin AS-48 (AS-48D, BacH) production with LolD, a protein involved in lipoprotein secretion. Identical residues are indicated by an asterisk. Colons and periods indicate conserved and semiconserved amino acid substitutions, respectively, according to the ClustalW grouping of amino acids. Dashes indicate gaps introduced in the sequence to maximize alignment. Walker A, Walker, and ABC transporter B motifs are indicated.
|
![]() View larger version (65K): [in a new window] |
FIG. 4. Alignment of CirI with BacG (56), FtsX (6), and LolC and LolE (37). The consensus motif G-X9-F-X10-G for ortholog group 3-1 type ABC transporters is indicated. Predicted transmembrane domains are indicated in bold. The region constituting the predicted DUF214 domain is indicated by a line above the sequences. Identical residues are indicated by an asterisk, whereas colons and periods indicate conserved and semiconserved amino acid substitutions, respectively, according to the ClustalW grouping of amino acids. Gaps were introduced in the sequence to maximize alignment.
|
![]() View larger version (122K): [in a new window] |
FIG. 5. Heterologous production of circularin A by E. faecalis JH2-2, as visualized in a colony overlayer assay with Lactobacillus saké ATCC 15521 as the indicator strain. (A) E. faecalis JH2-2(pMGAE1); (B) E. faecalis JH2-2(pCir); (C) E. faecalis JH2-2(pMG36c).
|
Removal of cirE from the cirABCDE cluster in pCir could not be achieved in several attempts with different cloning hosts [E. faecalis JH2-2 and E. faecalis JH2-2(pMGE)], while simultaneous deletion of cirA and cirE (pCir
AE) was possible. This observation suggested that cirE is involved in bacteriocin immunity, an assumption that will be discussed further below. E. faecalis JH2-2(pCir
AE) showed reduced sensitivity to circularin A present in filter-sterilized culture supernatant of C. beijerinckii ATCC 25752 but was clearly more resistant to circularin A than E. faecalis(pMG36c).
To determine which gene(s) is involved in this partial resistance towards circularin A, single deletions of cirB, cirC, and cirD were combined with a deletion in cirE. A mutation in either cirB or cirD in combination with deletion of cirE led to the loss of the circularin A-resistant phenotype, whereas cells carrying a pCir derivative with a deletion in cirA or cirC in combination with cirE remained partially resistant to circularin A (Fig. 1). These results indicate that both CirB and CirD are required for partial resistance in the absence of CirE. To confirm this hypothesis, three additional deletion constructs were made. E. faecalis JH2-2(pCir
ACE), specifying only CirB and CirD, still showed partial resistance to CirA. E. faecalis JH2-2 expressing only CirB (pCir
ACDE) or CirD (pCir
ABCE) was bacteriocin sensitive, confirming that both CirB and CirD are needed for the partial resistance phenotype in the absence of CirE.
Heterologous expression of circularin A in L. lactis NZ9000 was attempted, but pCir and any other vector containing cirB did not give transformants, while all control plasmids did. Apparently, an intact cirB gene is lethal to this host. Plasmids with a deletion in cirB could be stably maintained in L. lactis. L. lactis(pCir
B) did not produce active CirA and was 2.5-fold more resistant to the bacteriocin than L. lactis(pNG8048e), a strain carrying an empty cloning vector.
cirE gene confers circularin A immunity.
To prove that cirE can confer bacteriocin resistance independent of the combination cirB and cirD, the gene was cloned downstream of the lactococcal promoter P32 in pMG36e (pMG-E). Unlike L. lactis(pMG36e), L. lactis NZ9000(pMG-E) was able to grow in a medium with culture supernatant (up to 50%, vol/vol) of C. beijerinckii ATCC 25752 (CirA+), as determined by serial dilution assay. L. lactis NZ9000(pMG-E) also formed normal colonies on plates containing filter-sterilized C. beijerinckii ATCC 25752 culture supernatant (4%, vol/vol), whereas the control strain did not grow at all. These results indicate that cirE alone gives rise to circularin A resistance. E. faecalis JH2-2(pMG-E) was also immune to the bacteriocin present in C. beijerinckii ATCC 25752 supernatant, as determined in a plate assay.
As mentioned above, initial attempts to remove cirE from cirABCDE with either E. faecalis JH2-2 or E. faecalis JH2-2(pMG-E) as the cloning host failed. This problem was circumvented by cloning cirE downstream of the P32 promoter in pIL253. The resulting strain, E. faecalis JH2-2(pIL-E), was immune to C. beijerinckii ATCC 25752 culture supernatant, as determined in a plate assay, and did not produce bacteriocin (Fig. 1), as determined by serial dilution and overlay assays. In E. faecalis JH2-2(pIL-E), we were able to introduce the cirE deletion plasmid pCir
E (CirA+). E. faecalis JH2-2(pIL-E, pCir
E) was immune to C. beijerinckii ATCC 25752 culture supernatant and produced bacteriocin (Fig. 1). Attempts to introduce pCir
E in E. faecalis JH2-2 alone were unsuccessful, indicating that cirE is required for proper bacteriocin immunity.
The levels of resistance conferred by pMG-E and pIL-E in E. faecalis JH2-2, which is able to grow in medium containing filter-sterilized C. beijerinckii ATCC 25752 culture supernatant (at the most 3 to 6%, vol/vol), were only two- to fourfold increased compared to the control strain carrying only pMG36e, able to grow in medium containing at the most 1.5% (vol/vol) filter-sterilized C. beijerinckii ATCC 25752 culture supernatant. This level is lower than the resistance level of E. faecalis JH2-2(pCir
B) or E. faecalis JH2-2(pCir
D), both also expressing only cirE as a functional immunity system. Complementing pIL-E with pCir
DE in E. faecalis JH2-2 did not restore the resistance level to that of E. faecalis JH2-2(pCir
D), indicating that the lower resistance level is not due to the absence of auxiliary factors not present on pMG-E or pIL-E. Taken together, these results confirm that CirE is the dedicated circularin A immunity protein and that expression of cirE alone is sufficient for immunity.
|
|
|---|
We show here that the region cirABCDEGHI is involved in bacteriocin production and secretion. The genes cirABCDE represent the minimal region required for bacteriocin processing and secretion in the heterologous host E. faecalis JH2-2, as deletion of only a single gene from this cluster causes loss of either bacteriocin production or cell viability. The genetic organization of the region cirABCDEGHI seems rather compressed, as several genes overlap. This set-up suggests that translational coupling, a gene-regulatory mechanism often used in operons in which the stoichiometry of gene expression is important (36), may occur.
The minimal requirements for extracellular circularin A activity are production, processing, circularization, and secretion of the bacteriocin, while the producer cell should be immune to the bacteriocin. All these features should in principle be encoded by cirABCDE. Here, we show that resistance to circularin A is acquired via at least two independent systems. First, expression of cirE confers a certain level of immunity to the expressing strain, which is essential for the bacteria to be able to produce and withstand CirA. CirE has a very high and contains two possible transmembrane helices, which make membrane localization of the protein very likely. Its small size, high isoelectric point, and two predicted transmembrane helices are characteristics that CirE has in common with AS-48D1, the immunity protein of the circular bacteriocin enterocin AS-48, and with the proteins PepI, EciI, LasJ, and DviA, which have all been shown or postulated to be involved in immunity to the unrelated bacteriocins Pep5, epicidin 280, lactocin S, and divergicin A, respectively (18, 41, 44, 50, 59). The immunity mechanism of these proteins is unknown, but PepI has been suggested to inhibit pore formation by Pep5 (46).
The second system conferring reduced sensitivity to CirA depends on the combined activity of CirB and CirD. Together, these proteins form a putative ABC transporter in which CirB is the transporter and CirD provides the nucleotide-binding domain. The putative transporter CirBD confers a basal level of CirA resistance, which is, however, insufficient to support bacteriocin production by the heterologous host that we used; a viable clone of E. faecalis JH2-2(pCir
E) could not be obtained. CirBD (most likely) also function in CirA secretion, as ABC transporters are often implicated in bacteriocin secretion (10). The fact that proteins required for secretion of a bacteriocin can be involved in resistance has been shown for McbE and McbF, which are involved in microcin B17 production (13). The secretion proteins of the lantibiotic nisin were suggested to fulfill a similar role, but involvement of NisI and/or NisFEG, via a regulatory loop inducing expression of the respective genes, cannot be excluded (29, 43). Resistance is most probably obtained by the pumping out of the bacteriocin (42, 43, 46). In conclusion, CirBD confer low-level resistance by virtue of their ability to secrete CirA, while CirE shows structural homology to other bacteriocin immunity proteins, which identifies it as the dedicated CirA immunity protein.
Based on homology studies, CirGHI could constitute another transporter. CirG probably has an auxiliary function, as it is homologous to the HlyD family of proteins, many of which are accessory proteins in the export of drugs or toxic proteins, such as hemolysin (52), lactococcin A (53), and colicin V (15). CirH and CirI probably form an ABC transporter of the LolCDE type (60): CirH is homologous to LolD, while CirI is homologous to both LolC and LolE. The CirGHI homologues in the enterocin AS-48 system (BacGHI/AS-48FGH) enhance the expression of enterocin AS-48 (56) and the resistance to exogenous enterocin AS-48 (8), roles we have not investigated for CirGHI yet. The homology to the LolCDE system furthermore suggests that active transport, perhaps from the outer leaflet of the membrane, as shown for LolCDE (60), is involved in this enhancing effect by making more bacteriocin available. The NisFEG system fulfills such a function in enhancement of nisin secretion (43), but this system is not very homologous to CirGHI or LolCDE.
No experimental evidence has been obtained to identify the protein(s) involved in the processing and/or circularization of the CirA prepeptide. As CirBD together form a putative ABC transporter and cirE confers bacteriocin immunity, the essential protein CirC is a likely candidate to perform this function(s), either alone or together with CirB and/or CirD. This notion seems to be supported by the fact that the only other CirC homologue is AS-48C, encoded by the enterocin AS-48 gene cluster of E. faecalis.
In conclusion, we have identified five genes essential for circularin A production and have shown that three of these genes (cirBDE) are involved in bacteriocin resistance. Future studies will be performed to determine the mechanism of circularization and the possible role of CirC therein.
We are grateful to G. Dunny (Institute of Food Research, Colney, United Kingdom) for kindly providing the E. faecalis JH2-2 strains used in this study.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»